1 INFO-VAX	Thu, 31 Aug 2006	Volume 2006 : Issue 477       Contents:9 Re: $analyze/audit. What does ILLEGAL mean as a username? 5 $analyze/audit. What does ILLEGAL mean as a username? 5 Re: A true pre-emptive multi-tasking operating system 5 Re: A true pre-emptive multi-tasking operating system   Re: Alernative to Ctrl-Z in TYPE  Re: Alernative to Ctrl-Z in TYPE Re: Alpha remembrance day  Re: Alpha remembrance day  Re: Alpha remembrance day ) Another Itanium high roller departs Intel G Anyone out there have any older Dec Tape Drives  they want to part with K Re: Anyone out there have any older Dec Tape Drives  they want to part with 5 Re: Authentication for outgoing TCP/IP Services mail?  Re: Back at work Re: Back at work Re: Back at work Re: Back at work Re: Back at work$ Changes to OpenVMS Patch Kit Formats( Re: Changes to OpenVMS Patch Kit Formats( Re: Changes to OpenVMS Patch Kit Formats( Re: Changes to OpenVMS Patch Kit Formats Re: Difference? $ Distributed Netbeans 5.0 field test?( Re: Distributed Netbeans 5.0 field test?; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time ; Re: Funniest Ask-the-Wizard question and answer of all time  goodbye from me too  Re: goodbye from me too  Re: goodbye from me too  Re: goodbye from me too  Re: goodbye from me too $ HP OpenVMS BASIC boot-camp training! Re: HSG80 LED Codes  HSG80 LED Codes  Re: HSG80 LED Codes 9 Intermixing expected and unanticipated serial comms reads = Re: Intermixing expected and unanticipated serial comms reads  Is the SAIC Relay down?  Re: Is the SAIC Relay down?  Re: Is the SAIC Relay down? & Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS& Re: Locating early versions of VAX/VMS5 MicroVAX I / VT180 / LSI-11/23 / LA100 in Miford, NH. , Re: Mystery multiple TCP connections dropped, Re: Mystery multiple TCP connections dropped, Re: Mystery multiple TCP connections droppedB NEW Partner application System to Map-Document Search Applications+ Re: No more Oracle Standard Edition for VMS + Re: No more Oracle Standard Edition for VMS + Re: No more Oracle Standard Edition for VMS + Re: No more Oracle Standard Edition for VMS + Re: No more Oracle Standard Edition for VMS + Re: No more Oracle Standard Edition for VMS ) OpenVMS Integrity Graphics card questions - Re: OpenVMS Integrity Graphics card questions - Re: OpenVMS Integrity Graphics card questions - Re: OpenVMS Integrity Graphics card questions - Re: OpenVMS Integrity Graphics card questions - Re: OpenVMS Integrity Graphics card questions - Re: OpenVMS Integrity Graphics card questions - Re: OpenVMS Integrity Graphics card questions  Re: Personal note - goodbye !  Re: Personal note - goodbye !  Re: Personal note - goodbye ! 	 Re: phone 	 Re: phone 	 Re: phone 	 Re: phone 	 Re: phone 	 Re: phone 	 Re: phone 	 Re: phone 	 Re: phone 	 Re: phone 	 Re: phone  Problems with Python tests Re: Problems with Python tests Re: Problems with Python tests6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN6 Re: Reading UAF Disuser Flag Across Nodes from FORTRAN2 Re: SEPPUCLU bugcheck introducing new cluster node2 Re: SEPPUCLU bugcheck introducing new cluster node2 Re: SEPPUCLU bugcheck introducing new cluster node2 Re: SEPPUCLU bugcheck introducing new cluster node2 Re: SEPPUCLU bugcheck introducing new cluster node Re: Shadow copy stops at n%  Re: Shadow set problem Re: Shadow set problem Re: Shadow set problem4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC4 Re: Thoughts on the book: DEC is dead, long live DEC Unix to VMS users  Re: Unix to VMS users  Re: Unix to VMS users  Re: Unix to VMS users  Re: Unix to VMS users  Re: Unix to VMS users  Re: Unix to VMS users / Unix/VMS Admin Job Opening in Jacksonville, FL. 3 Re: Unix/VMS Admin Job Opening in Jacksonville, FL. 3 Re: Unix/VMS Admin Job Opening in Jacksonville, FL. 3 Re: Unix/VMS Admin Job Opening in Jacksonville, FL. . VAX 4000/600 free to good home - Manchester UK2 Re: VAX 4000/600 free to good home - Manchester UK2 Re: VAX 4000/600 free to good home - Manchester UK2 Re: VAX 4000/600 free to good home - Manchester UK2 Re: VAX 4000/600 free to good home - Manchester UK2 Re: VGA video adapter for Vaxstation 4000 model 60 VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS 4.4 and file security  Re: VMS in The DA  RE: VMS in The DA  Re: VMS in The DA  Re: VMS in The DA * We're not dead yet, we're feeling better !. Re: We're not dead yet, we're feeling better !. Re: We're not dead yet, we're feeling better ! wwidmgr  Re: wwidmgr  Re: wwidmgr  Re: wwidmgr  Re: wwidmgr 2 Re: X-terminal connects get hosed, requires reboot [OpenVMS Patches] Difference?   F ----------------------------------------------------------------------    Date: 31 Aug 2006 09:48:35 -0700/ From: "Volker Halle" <volker_halle@hotmail.com> B Subject: Re: $analyze/audit. What does ILLEGAL mean as a username?C Message-ID: <1157042915.903262.292150@i42g2000cwa.googlegroups.com>   C This looks like a network task access for an undefined network task G object. You are probably running DECnet-IV software, so try MC NCP SHOW B KNO OBJECT and you might see the TASK object having a user name of ILLEGAL.   --- , Volker Halle, Invenate GmbH, OpenVMS Support  $ An OpenVMS crashdump analysis a day # makes the Windows headaches go away    ------------------------------    Date: 31 Aug 2006 09:35:58 -0700 From: ddujon@gmail.com> Subject: $analyze/audit. What does ILLEGAL mean as a username?B Message-ID: <1157042158.435199.14080@m79g2000cwm.googlegroups.com>   Hi  3 The following is the result of $analyze/audit. What E does illegal mean within the last line? Its in the 'username' column. ; XXXSL2 is the node name.o/s is OpenVMS 7.3-1 on Alphaserver  ES47.Thanks.    8       Date / Time                 Type           Subtype# Node    Username       ID      Term  ------------------------------------------------------------------------------------------------------------------------------------  >   8-OCT-2002 15:26:25.26 RIGHTSDB   RDB_ADD_ID   FISBLD SYSTEM 00000203=   8-OCT-2002 15:31:00.88 SYSUAF     SYSUAF_ADD  FISBLD SYSTEM  00000213   OPA0:@ 11-AUG-2003 11:03:59.89 BREAKIN    NETWORK       XXXSL2  ILLEGAL 00001060   ------------------------------  # Date: Mon, 28 Aug 2006 16:04:38 GMT % From: Rob Brown <mylastname@gmcl.com> > Subject: Re: A true pre-emptive multi-tasking operating systemE Message-ID: <Pine.LNX.4.61.0608280956570.10653@localhost.localdomain>   - On Thu, 24 Aug 2006, Steven M. Schweda wrote:     > From: briggs@encompasserve.org > G >> So if we put time on the x axis and knowledge on the y axis and say  = >> that "foo has a steep learning curve", this can mean that:  >> ... > E >   A graph of knowledge versus time does not have X and Y axes.  It  G > has knowledge and time axes.  Look up "ordinate" and "abscissa" when   > you get home.   D In the best tradition of beating dead horses, I suggest that "steep G learning curve" does not refer to a knowledge vs time graph, but to an  B effort vs knowledge graph.  IE, Knowledge is the abscissa (X) and  effort is the ordinate (Y).   E If the learning curve is steep, then you must put in a lot of effort  $ to gain a small amount of knowledge.     --    B Rob Brown                        b r o w n a t g m c l d o t c o m6 G. Michaels Consulting Ltd.      (780)438-9343 (voice)4 Edmonton                         (780)437-3367 (FAX)2                                   http://gmcl.com/   ------------------------------  # Date: Mon, 28 Aug 2006 22:01:00 GMT + From: "Villy Madsen" <Villy.Madsen@shaw.ca> > Subject: Re: A true pre-emptive multi-tasking operating system. Message-ID: <wIJIg.479610$iF6.341684@pd7tw2no>   Tis hot today Rob  !!   M I would be more inclined to define it as a situation where you have to put a  J lot of time and effort into learning something before you can do anything  useful..  H e.g.  Object orientated Pascal running on a Mac, in the words of an old L aquaintence who used to live in St Albert & flew airplanes:  Took me a week M to learn how to write "hello world"  on the screen, and a week and a half to   conquer the world!!    Hot in Beaumont AB today!    Villy     3 "Rob Brown" <mylastname@gmcl.com> wrote in message  ? news:Pine.LNX.4.61.0608280956570.10653@localhost.localdomain... J > If the learning curve is steep, then you must put in a lot of effort to # > gain a small amount of knowledge.  >  >  > --   >    ------------------------------  % Date: Sun, 27 Aug 2006 22:00:02 +0100 0 From: Tony Arnold <tony.arnold@manchester.ac.uk>) Subject: Re: Alernative to Ctrl-Z in TYPE = Message-ID: <pan.2006.08.27.21.00.02.429129@manchester.ac.uk>   6 On Sun, 20 Aug 2006 11:40:37 -0400, Dave Froble wrote:  , > I just discovered something new.  (To me.) > H > My applications accept the backslash, "/E-xit", and Ctrl-Z as an exit 8 > signal.  Ok, that's specific to the code I've written. > - > Ctrl-Z is an exit signal for VMS utilities.  > G > I was using TYPE/PAGE and inadvertantly typed a backslash instead of  J > Ctrl-Z, and the utility exited to the DCL prompt.  Brain takes a bit to 5 > assimilate this, then I try it again.  Exits again.  > H > This is on VAX/VMS V7.2.  Can anybody comment on when this capability  > was introduced?   J I think it's been there for as long as the /page qualifier has been there!   Regards, Tony.    ------------------------------    Date: 27 Aug 2006 20:42:38 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> ) Subject: Re: Alernative to Ctrl-Z in TYPE B Message-ID: <1156736558.220177.23480@i42g2000cwa.googlegroups.com>   Tony Arnold wrote:8 > On Sun, 20 Aug 2006 11:40:37 -0400, Dave Froble wrote: > I > > This is on VAX/VMS V7.2.  Can anybody comment on when this capability  > > was introduced?  > L > I think it's been there for as long as the /page qualifier has been there! > 
 > Regards, > Tony.   F Yep. It's handy, yet annoying.  I don't know how many times I've hit aF space (In Un*x that makes  the "more" command just go to the next pageF - return moves to the next line), realized my mistake, then hit returnD to get the next page and been dumped out.  It's had that behavior as long as I can remember.      John H. Reinhardt    ------------------------------    Date: 29 Aug 2006 09:22:57 -0700- From: "Andrew" <andrew_harrison@symantec.com> " Subject: Re: Alpha remembrance dayB Message-ID: <1156868577.280064.291470@74g2000cwt.googlegroups.com>   Bill Todd wrote: > Andrew wrote:  >  > ...  >  > >>> Sorry wrong again.L > >> At the risk of being pedantic, to be wrong 'again' I would already haveD > >> had to have been wrong in the current discussion at least once. > >> > > J > > Well if you can point to one single part of this thread where you have7 > > been right then I will be happy to buy you a sweet.  > D > How about the fact that Win2K, contrary to your assertions, indeedJ > *does* support a full 64 GB of RAM?  Not that you appear to have managedF > to understand even that simple a correction below, so I won't botherH > creating a full (and rather lengthy) catalog of the other instances inG > this thread where not only I have been right but you have been either : > dead wrong or utterly irrelevant to the subject at hand. >   E Does it really. Windows 2003 does we know that but can you  provide a . Microsoft reference for 64GB support in Win2K.   > ...  > 9 > >>   The Pentium may well have supported 64GB of RAM in 1 > >>> 1996 but but Windows NT Server 4 never did. L > >> By George, it appears that you may be right for once:  I thought that IJ > >> remembered some trans-4GB facilities (which Win2K later cleaned up asK > >> its AWE mechanisms) being introduced in some flavor(s) of NT V4, but I . > >> can't find any mention of them right now. > >> > >>> Neither did Windows 2000, M > >> Wrong (again).  While Microsoft *nominally* supports only up to 32 GB of I > >> RAM in Win2K Datacenter, had you even investigated the link that you L > >> yourself provided you would have seen that this was because of the lackG > >> of available hardware to test on 64 GB systems, and indeed "Inside M > >> Windows 2000" confirms that the Win2K code supports up to the full 64 GB  > >> system size.  >  > Yup - there it is.  C Err no wrong yet again. Sequent and laterly IBM built Pentium based E systems which physically supported 64GB of RAM as did Unisys. However D the Win2k release notes say that Win2k only supports 32gb of RAM. So> the reason for official support does not relate to the lack of available systems.  A Though its difficult to see how not having 64GB systems available A supports your line of argument since you were the person who very B misguidely introduced the whole 64GB on Pentium point in the firstE place. Why suggest that NT supported 64GB in 1997 when you apparently D knew that no manufaturer build Pentium based boxes with 64GB of RAM.     >  > > H > > How amusing, it was you who claimed that 64GB was addressable by NT. > J > Yes, and (as I readily admitted earlier) it appears that my recollectionH > that the last version of NT had support for extended RAM may have beenL > in error (at least I wasn't able to find corroboration with a quick look). >  >   IfF > > you will note I pointed out that this wasn't true for either NT or	 > > Win2K  > F > And, as I noted above, while you may have been correct about NT, youJ > were wrong about Win2K, which (according to the reasonably authoritativeH > "Inside Windows 2000") in fact *does* support up to 64 GB of RAM.  ForE > that matter, according to a link from the very Web page which *you* I > provided as support for your contention that Win2K supported only up to I > 32 GB of RAM that nominal limit was set only because larger systems did D > not exist at the time on which to validate the full 64 GB support. >   D If you can find a microsoft document that refers to Win2K supportingE 64GB of RAM then please publish it. Reasonably authoritive given your $ track record hardly cuts it does it.  I > I've finally placed you, Andrew:  you're the socially-challenged kid on I > the playground who's determined to get attention even if he has to goad H > others into beating him up to get it.  The kind who thinks he's scoredI > some kind of moral victory by adamantly parroting the same empty drivel F > ("I know *you* are, but what am *I*?!" is the meaningless schoolyardD > comeback that springs to mind) no matter how many times he's first- > corrected, then jeered, then kicked around.  >   = Really fascinating have you looked in the mirror today Bill ?   * BTW thanks for saving the best until last.   Regards  Andrew Harrison    ------------------------------    Date: 31 Aug 2006 04:12:53 -0700- From: "Andrew" <andrew_harrison@symantec.com> " Subject: Re: Alpha remembrance dayA Message-ID: <1157022773.778593.4430@m73g2000cwd.googlegroups.com>    Hoff Hoffman wrote:  > Andrew wrote:  > I > > Does it really. Windows 2003 does we know that but can you  provide a 2 > > Microsoft reference for 64GB support in Win2K. > P >    Here are two such references, to Microsoft Windows 2000 with 64 GB support,N > currently available at the Microsoft web site, with the requisite quotations% > from these pages directly provided:  > K > http://www.microsoft.com/presspass/press/2001/oct01/10-02mecreevespr.mspx  > R >    "In particular, Windows 2000 Datacenter Server allows enterprises to scale upK > by supporting as many as 32 processors and up to 64 gigabytes of memory."  > T > http://www.microsoft.com/technet/prodtechnol/windows2000serv/evaluate/servroi.mspx > N >    "Microsoft has addressed these specific issues in Windows 2000 DatacenterQ > Server and Windows 2000 Advanced Server with improved Symmetric Multiprocessing Q > and Enterprise Memory Architecture (EMA). By utilizing Intel's Physical Address L > Extension (PAE) on Intel 32 bit (IA-32) platforms, Windows 2000 Datacenter9 > Server supports up to 64 gigabytes of physical memory."  >  >    --  > M >    Your favorite search engine can locate rather more of these.  Many more.   E Sure they can but they are all based on Microsoft Press and Marketing C material. The Windows Server Hardware Developers guide on the other D hand lists Windows 2000 Datacenter Server as supporting 32GB and theE documentation for PAE which is the mechanism used by Intel to support  more than 4GB.  D http://www.microsoft.com/whdc/system/platform/server/PAE/PAEdrv.mspx  7 Last time I looked you were a engineer not a marketeer.   G Of course Bills original point refered to Windows NT in 1997 supporting E 64GB of RAM and it was obviously wrong the PAE support required to do @ this is only Supported in XP, 2000 and 2003. Windows 2000 became  publically available in Feb 200.  E Ironically he sort of described how you might manage segmented memory < but unerringly chose the one major OS platform for which hisF description had no relevance at all. To compound the confision neitherB NT or 2000 can actually address 4GB in one flat address space. TheG maximum is either 2 or 3GB and as you can see from the documentation if E you enable 3GB addressing then you restrict the maximum memory on the G box to 16GB you can only have 32GB on Windows 2000 with 2GB addressing.   F There are good reasons why Windows 2000 didn't support >32GB one beingC that Windows 2000 had no support for NUMA platforms, all the >8 way C 32bit Intel servers were NUMA and they were built using quad boards ) with a maximum of 4-8GB of RAM per board.   B Since you couldn't build a >32GB Intel based box running 2000 that! performed at all no one bothered.   F Of course other OS's did have NUMA support, Dynix for example allowingB working, performing ish systems to be built with >32GB of RAM. TheA limitations with Dynix being the speed and latency of the Sequent # interconnect and not the OS itself.    regards  Andrew Harrison    ------------------------------    Date: 31 Aug 2006 04:39:01 -0700- From: "Andrew" <andrew_harrison@symantec.com> " Subject: Re: Alpha remembrance dayC Message-ID: <1157024341.230895.162930@m73g2000cwd.googlegroups.com>    Hoff Hoffman wrote:  > Andrew wrote:  > > Hoff Hoffman wrote: M > >> Now it's "balance"?  This being the fall-back marketeering position, eh?  > >>N > >>    Having 64-bit virtual addressing was and is worthwhile out of the box. > >  > > Sort of. > > J > > The biggest benefit is being able to address more than 4GB of RAM withI > > a single process. There are other benefits but they and being able to @ > > address more then 4GB may come with an expense tag attached. > I >    Processes have always been able to address more than 1GB, 2GB or 4GB F > (depending on details of the virtual memory and the operating systemN > implementation), though it requires local processing -- even on JF's favored1 > twenty-year-old 8086, you could implement this.   
 Of course. > R >    Having a flat address space is easier, simpler, and it transfers the overhead6 > off into a code path that can and will be optimized.     > D > > 64bit executables are larger, use more memory and are less cacheK > > friendly their 32bit equivalents. Some tests have shown up to 20% worse J > > performance for 64bit versions of common utilities compared with their > > 32bit counterparts.  > P >    Some tests have shown I have a zit.  Some tests have shown I haven't.  SomeR > tests matter.  Some tests don't matter.  Some tests have shown storage increasesM > due to the use of 64-bit addressing (due to larger variables) and some show Q > minimal increases (due to mixtures of pointer sizes), and some show that having N > less engineering and support overhead for a given application is beneficial.M > Alpha traded off increased memory use for increased speed, as most any RISC J > platform and most any newer platform in general has made as a trade-off. > M >    And as for speed, Alpha operates 32 bit addressing as a subset of 64-bit M > addressing, so you're always running 64-bit, whether you realize it or not.   D Of course.  Which goes a long way to explain why your zit example is not a good one.   A HP-PA, MIPS, POWER, x86-64 and SPARC on the other hand are either B running 32bit or 64bit so they have the opportunity to measure the@ performance of a 32bit  app really running 32bit or the same app running 64bit.  F To my knowledge IBM, Sun etc have not pushed their ISV's and customersD into a 64bit porting frenzy far from it. Odd when you think about itF because being able to drop 32bit support would reduce costs both at anF OS level and at a HW level. Not so odd when you realise that a large %2 of apps do not benefit from being ported to 64bit.  F Alpha was introduced as the desktop to super computer CPU in 1992. TheF worlds largest supplier of OS's for the desktop and the 1-4 way serverE market Microsoft only provided 64bit server support in April 2003. So F the dominant OS player in the two largest markets Alpha wanted to leadA didn't see enough demand from customers or loss of share from not D supporting 64bit to bother doing it until 2003. Thats 11 years after Alpha's introduction.       O >    In conclusion, your statement that Sun believed that there was no need for L > 64-bit addressing and that this capability was discounted by Sun is, well,M > seemingly somewhat odd, and at odds with what limits were clearly occurring 0 > across most any of the platforms a decade ago.  D No that isn't what I said. Sun did see the need for 64bit addressingC because for large systems moving to midd range as platforms grew in F capacity it was becoming usefull and in some cases not having it was aF showstopper. UltraSPARC 1 introduced in 1995 was a 64bit processor andF Sun would hardly have bothered doing SPARC v9 and adding 64bit support7 to UltraSPARC if theyr thought that 64bit had no value.   G Sun's conclusion was that Digital had whipped up a storm over 64bit but F had forgotten to develop a server platform that could make it usefull.G Digital entered the market with a 64bit OS and CPU but without the rest C of the ecosystem. It wasn't until the GS160/GS320 that Compaq had a 1 platform that had caught up.  Entirely different.    Regards  Andrew Harrison    ------------------------------  % Date: Tue, 29 Aug 2006 07:50:24 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> 2 Subject: Another Itanium high roller departs Intel< Message-ID: <44f428a2$0$24194$9a6e19ea@news.newshosting.com>  7 Another Itanium high roller (Nimish Modi) departs Intel   5 http://www.theinquirer.net/default.aspx?article=33998   3 I don't know if this news is good, bad, or neutral.   
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada." http://www3.sympatico.ca/n.rieck/    ------------------------------    Date: 29 Aug 2006 20:37:35 -0700 From: scsirocks@comcast.net P Subject: Anyone out there have any older Dec Tape Drives  they want to part withB Message-ID: <1156909055.682263.269300@i3g2000cwc.googlegroups.com>  ? I am looking for some older Dec TLZ09-VA, TL07-VA, TLZ09-AA and  TLZ07-AA  = If you have any of these that are working please let me know.      Thanks   ------------------------------  + Date: Wed, 30 Aug 2006 10:54:44 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)T Subject: Re: Anyone out there have any older Dec Tape Drives  they want to part with$ Message-ID: <ed3qpk$tus$3@online.de>  B In article <1156909055.682263.269300@i3g2000cwc.googlegroups.com>, scsirocks@comcast.net writes:   A > I am looking for some older Dec TLZ09-VA, TL07-VA, TLZ09-AA and 
 > TLZ07-AA > ? > If you have any of these that are working please let me know.    Where are you located?   ------------------------------  + Date: Mon, 28 Aug 2006 08:20:51 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)> Subject: Re: Authentication for outgoing TCP/IP Services mail?$ Message-ID: <ecu913$fbl$1@online.de>  5 In article <44D5784C.E24C2894@teksavvy.com>, JF Mezei ' <jfmezei.spamnot@teksavvy.com> writes:    H > The VMS SMTP server was designed as a "real" SMTP server that deliversJ > to many destinations.  In such a context, authentication is meaningless.I >  But in a context where you route all mail to a gateway server (such as @ > your ISPs), then authentication can become realistic (a singleG > user/password needed to send all mail via such server). But the TCPIP . > Services server is not setup to handle this. > F > And usually, an ISP would allow computers using its own IP blocks to. > access its SMTP server without restrictions. > G > The question of authentication has been asked a number of times here. I > And assuming there are people left at TCPIP Services engineering, it is D > doubtful they would be given priority on this feature since paying > customers don't request it.  > E > Your best bet is to find an ISP that doesn't require authentication > > and/or block port 25 so your own SMTP server can do its job.  H For folks with a dynamic IP address, the situation is a bit different.  H First, the service doing the dynamic-DNS update might be independent of A the ISP assigning the IP address.  Yes, at least one dynamic-DNS  D provider, http://www.dynaccess.com/, knows your current IP and will D allow mail through their SMTP relay server based on that IP (and of H course take appropriate steps if you send spam).  Also, even if port 25 G is not blocked, this might not be much use, since in the fight against  G spam many email servers don't accept email from dynamic IP addresses.   7 Yes, it is a coarse filter, but is sometimes necessary.   I Personally, after I moved from fixed IP addresses with ISDN to a dynamic  G IP address with DSL, I could TECHNICALLY do everything myself, but ran  G into the "we don't accept email from dynamic IP addresses" problem.  I  E then found http://www.dynaccess.com/ and that solved my problem.  In  E fact, it's even better than doing everything myself, since if due to  H some technical problem my cluster is offline with respect to the world, I a backup MX server will kick in, HTTP requests will get redirected, I'll  H get informed by email (obviously, also to an address not on my cluster) > and/or SMS etc.  And the prices are quite reasonable.  Highly E recommended.  (Note: I don't receive any commission for recommending  H them; I'm just a satisfied customer and, since they offer many services F "similar" dynamic-DNS providers don't, I think many folks here with a G VMS machine at home might be interested.  Additional note: one can run  E their Perl script for dynamic-DNS updates (regular updates, not just  I when the IP address changes), but I wrote my own update client using DCL   and LYNX in batch mode.)   ------------------------------    Date: 30 Aug 2006 14:45:24 -0700) From: "Sue" <susan_skonetski@hotmail.com>  Subject: Re: Back at work C Message-ID: <1156974324.893124.228740@m79g2000cwm.googlegroups.com>   F Well I would not bet on a marathon but Technical update days in Europe are a sure bet ;')  $ And early in 2007 in the Pac Rim ;')  D Look like my kicking butt future may be a little slow but I am still working on it.   Thanks so much   Sue    Steve Matzura wrote:C > On 7 Aug 2006 15:55:59 -0700, "Sue" <susan_skonetski@hotmail.com>  > wrote: > A > >I just wanted to let everyone know that I am now back at work.  > D > Ah, peace in the valley!  It's good to hear you're back online andG > in-the-house!  Friends of mine from a previous employer have informed D > me that your email newsletter has cranked up again, so I knew it'd; > only be a matter of time before you were spotted in here.  > G > Hopefully all went well and you'll be in next year's marathon, right?    ------------------------------    Date: 30 Aug 2006 17:26:59 -0700 From: sean@obanion.us  Subject: Re: Back at work C Message-ID: <1156984019.821981.310030@h48g2000cwc.googlegroups.com>   8 It's always good to have goals.  While you're in Europe:  ! "A chocolate crawl through Paris" X http://www.ivillage.co.uk/travel/inspiration/hobbies/articles/0,,563217_572641-2,00.html  @ "In Paris, Boutiques and Cafes Where Chocolatiers Raise the Bar"q http://www.nytimes.com/2004/12/12/travel/12tab.html?ex=1260507600&en=1a2debc6f5e03deb&ei=5090&partner=rssuserland    And if the timing fits- E "Le Salon du Chocolat": Carrousel de Louvre; 28 October to 1 November  2006 www.chocoland.com        Sean  
 Sue wrote:H > Well I would not bet on a marathon but Technical update days in Europe > are a sure bet ;') > & > And early in 2007 in the Pac Rim ;') > F > Look like my kicking butt future may be a little slow but I am still > working on it. >  > Thanks so much >  > Sue  >  > Steve Matzura wrote:E > > On 7 Aug 2006 15:55:59 -0700, "Sue" <susan_skonetski@hotmail.com> 
 > > wrote: > > C > > >I just wanted to let everyone know that I am now back at work.  > > F > > Ah, peace in the valley!  It's good to hear you're back online andI > > in-the-house!  Friends of mine from a previous employer have informed F > > me that your email newsletter has cranked up again, so I knew it'd= > > only be a matter of time before you were spotted in here.  > > I > > Hopefully all went well and you'll be in next year's marathon, right?    ------------------------------  % Date: Wed, 30 Aug 2006 20:57:54 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: Back at work , Message-ID: <44F6340F.20336BD7@teksavvy.com>  
 Sue wrote: > H > Well I would not bet on a marathon but Technical update days in Europe > are a sure bet ;')  ? Sue, are you now at a point where your new-and-improved knee is H performing better than the old one ? Or is there still pain and weakness= in the area with the new product not quite yet up to expected  performance standards ?   B If you still need a cane to walk, you do realise that you can makeG effective use of the cane to "motivate" people to do what you want them  to do :-) :-) :-) :-)    ------------------------------    Date: 30 Aug 2006 20:27:51 -0700) From: "Sue" <susan_skonetski@hotmail.com>  Subject: Re: Back at work C Message-ID: <1156994870.972736.285470@i42g2000cwa.googlegroups.com>   G Well lets put it this way MTBF is still greater than before surgery but F starting to get better.  I am so looking forward to seeing everyone inC Europe and maybe even a weekend in Paris, so I am motivated to keep  walking    sue    JF Mezei wrote:  > Sue wrote: > > J > > Well I would not bet on a marathon but Technical update days in Europe > > are a sure bet ;') > A > Sue, are you now at a point where your new-and-improved knee is J > performing better than the old one ? Or is there still pain and weakness? > in the area with the new product not quite yet up to expected  > performance standards ?  > D > If you still need a cane to walk, you do realise that you can makeI > effective use of the cane to "motivate" people to do what you want them  > to do :-) :-) :-) :-)    ------------------------------    Date: 30 Aug 2006 20:28:59 -0700) From: "Sue" <susan_skonetski@hotmail.com>  Subject: Re: Back at work A Message-ID: <1156994939.026858.72020@i3g2000cwc.googlegroups.com>   $ And I have given up my cane as well!    
 Sue wrote:I > Well lets put it this way MTBF is still greater than before surgery but H > starting to get better.  I am so looking forward to seeing everyone inE > Europe and maybe even a weekend in Paris, so I am motivated to keep 	 > walking  >  > sue  >  > JF Mezei wrote:  > > Sue wrote: > > > L > > > Well I would not bet on a marathon but Technical update days in Europe > > > are a sure bet ;') > > C > > Sue, are you now at a point where your new-and-improved knee is L > > performing better than the old one ? Or is there still pain and weaknessA > > in the area with the new product not quite yet up to expected  > > performance standards ?  > > F > > If you still need a cane to walk, you do realise that you can makeK > > effective use of the cane to "motivate" people to do what you want them  > > to do :-) :-) :-) :-)    ------------------------------    Date: 30 Aug 2006 05:08:52 -0700; From: "george.pagliarulo@hp.com" <george.pagliarulo@hp.com> - Subject: Changes to OpenVMS Patch Kit Formats A Message-ID: <1156939732.626228.76300@74g2000cwt.googlegroups.com>   G There are several changes that are being implemented around the formats D for OpenVMS Patch kits.  Following is the Customer Advisory that was released about these changes.       OpenVMS Patch Kit Format Changes CUSTOMER NOTICE    Description D Changes are being made to the packaging and compression formats used for OpenVMS patch kits.      Details E The following changes are being made to formats used for OpenVMS PCSI E patch kits and for the compressed files used for shipping kits to the . ITRC patch distribution site and to customers:   =B7	PCSI$COMPRESSED   A Patch kits will now be produced as KIT_NAME.PCSI$COMPRESSED files F rather than the usual PATCH_KIT_NAME.PCSI files.   The PCSI$COMPRESSEDA format uses the PCSI utility's integrated compression capability, ? producing smaller files that will conserve disk space on user's E systems.  From a user standpoint, there is no difference from the way @ PCSI$COMPRESSED files and PCSI files are installed.  All PRODUCTF commands can be used and function the same on PCSI$COMPRESSED files as they do on PCSI files.  
 =B7	ZIPEXE  E OpenVMS patch kit packages have been produced in a variety of formats ? - DCX as well as ZIPEXE.    Patch kit formats for all supported @ versions of OpenVMS are  being standardized and will now ship asD self-extracting ZIPEXE packages.   When run, the ZIPEXE package will0 expand into the PCSI$COMPRESSED installable kit.  $ =B7	OpenVMS V8.3 and Secure Delivery  F OpenVMS V8.3 will include Secure Delivery capability allowing shipmentG of digitally signed patch kits.   OpenVMS Secure Delivery automatically F ensures that software installed on OpenVMS was not tampered with priorD to installation.  When run, V8.3 ZIPEXE package will expand into twoD files - the digitally signed PCSI$COMPRESSED installable kit and theD PCSI$COMPRESSED_ESW patch manifest or digital signature file.   BothD files will expand into the same directory and must be present in theB same directory at installation time for patch validation to occur.G When the patch kit is installed, PCSI will check for the existence of a B manifest for any kits that are being installed.  If no manifest isF found, PCSI will issue a warning and ask if you want to proceed.  If aF manifest is found but does not match the kit, the installation will beF aborted.  The PCSI database will contain an indication as to whether a) kit used Secure Delivery on installation.   @ For additional information on Secure Delivery and PCSI, refer toE section "5.2 CDSA for OpenVMS and Secure Delivery" in the "HP OpenVMS 4 Version 8.3 New Features and Documentation Overview"  D These format changes are being implemented immediately.  Since patchB kits that are already in production will not be re-manufactured toD update the formats, for a short time, users may see a combination ofF PCSI and PCSI$COMPRESSED, as well as DCX and ZIPEXE patch kit formats.   ------------------------------    Date: 30 Aug 2006 23:14:09 -02006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)1 Subject: Re: Changes to OpenVMS Patch Kit Formats , Message-ID: <44f61bc1$1@news.langstoeger.at>  \ In article <44F5E15A.CBD6023D@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:" >"george.pagliarulo@hp.com" wrote:C >> versions of OpenVMS are  being standardized and will now ship as G >> self-extracting ZIPEXE packages.   When run, the ZIPEXE package will 3 >> expand into the PCSI$COMPRESSED installable kit.  > F >While I assume this is a done deal which cannot be changed, was thereK >any consideration that double compression often results in a larger file ?   H Have you ever tried it? It never resulted in larger files for me so far.M In fact, a ZIPped .PCSI$COMPRESSED file was always smaller than the original.   F >If you're going to be zipping the package for transport, why generate >PCSI$COMPRESSED packages ?   D Because after the transport, you keep your kits unZIPped on disk ;-)M And a .PCSI$COMPRESSED files _is_ way smaller, while not losing functionality H (in regard to a .PCSI file). You could also translate them from .PCSI toE .PCSI$COMPRESSED (and vice versa) with standard DCL PRODUCT commands.   G And the reason why you download a .ZIP* file and not a .PCSI$COMPRESSED M file is: a) ZIP has checksums and you see easily if the archive is corrupted  < b) the VMS attributes and dates are not lost during transfer  J btw On VAX there is still no .PCSI$COMPRESSED, and also VMS ECOs are stillL VMSINSTALlable. I personally don't think, this will ever change, but I hope.  D >Also, will there be any way for cross platform extraction from that >ZIPEXE file ?  K This has never been a problem. It was used in (not only) VMS freeware world H for over a decade now, was then used in other software for VMS world and> has now finally reached VMS engineering. And I thank for that.  I >For instance, if I download from an Alpha and wish to manage the patches I >there and distribute the .PSCI files to various other nodes of different E >architectures, will there be some tool on the ALPHA to unzip a patch # >meant for VAX or that IA64 thing ?  > I >(can the standard ZIP skip over the executable part of a ZIPEXE file and ) >get to the actual compressed contents ?)   E ZIPEXE is ZIPSFX. You can unzip it with your own UNZIP program on any I platform you like (eg. you unpack the IA64 or VAX ECO on your MARVEL) and I you can unpack it on the intended platform without any UNZIP.EXE (because   it is part of the archive file).   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Wed, 30 Aug 2006 21:11:07 -0500 6 From: "David J. Dachtera" <djesys.no@spam.comcast.net>1 Subject: Re: Changes to OpenVMS Patch Kit Formats 0 Message-ID: <44F6453B.EA9E8B67@spam.comcast.net>   JF Mezei wrote:  > # > "george.pagliarulo@hp.com" wrote: D > > versions of OpenVMS are  being standardized and will now ship asH > > self-extracting ZIPEXE packages.   When run, the ZIPEXE package will4 > > expand into the PCSI$COMPRESSED installable kit. > G > While I assume this is a done deal which cannot be changed, was there L > any consideration that double compression often results in a larger file ?  B DCX is not real efficient. I believe that's what PCIS uses to makeB PCSI$COMPRESSED output. DCXed archives will typically squeeze down! another 33% of the DCX-only size.   G > If you're going to be zipping the package for transport, why generate  > PCSI$COMPRESSED packages ?  H Because UNZIP can detect a damaged archive, and may ecven be able to fixF it, depending on the dgree/type of damage. DCX has no such provisions.  E > Also, will there be any way for cross platform extraction from that  > ZIPEXE file ?   D Any version of UNZIP should be smart enough to detect/ignore the SFXB stub prepended to the archive and extract the content, even if theH offsets were not adjusted - UNZIP simply notifies about unexpected stuff' before the archive content was located.    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------    Date: 31 Aug 2006 09:22:35 -02006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)1 Subject: Re: Changes to OpenVMS Patch Kit Formats , Message-ID: <44f6aa5b$1@news.langstoeger.at>  i In article <44F6453B.EA9E8B67@spam.comcast.net>, "David J. Dachtera" <djesys.no@spam.comcast.net> writes: C >DCX is not real efficient. I believe that's what PCIS uses to make C >PCSI$COMPRESSED output. DCXed archives will typically squeeze down " >another 33% of the DCX-only size.  E DCX has also had big problems with not detecting corrupted downloads.   K I often had .PCSI files produced from incomplete downloads without any hint H and wondered why I couldn't install them successfully. It was neccessaryJ to add an additional step (looking for checksums, extracting release notesJ or unpacking the whole .PCSI kit) just to detect the corrupted .PCSI file.L This is no longer neccessary (but could still be done if you're paranoid ;-)  E >Any version of UNZIP should be smart enough to detect/ignore the SFX C >stub prepended to the archive and extract the content, even if the I >offsets were not adjusted - UNZIP simply notifies about unexpected stuff ( >before the archive content was located.  H btw: I once had problems with WINZIP not successfully unpacking archivesK where VMS attributes were saved ("-V"). But this is a done deal nowadays...    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 30 Aug 2006 05:06:12 -0700; From: "george.pagliarulo@hp.com" <george.pagliarulo@hp.com>  Subject: Re: Difference?C Message-ID: <1156939572.464716.276730@p79g2000cwp.googlegroups.com>    Peter,  F The only difference is that a typo in the documentation was corrected.E The only time a kit is re-issued without a bump in the version number E is if there was a non-functional change.  What *should* have happened G is that a new master list released with a comment explaining the change F to the SHADOWING kit.  The master list was created but never completedD the release process.  I'm fixing that now.  Sorry for the confusion.H >From this point forward, when something like this happens again the kit> documentation will also reflect the reason for the re-release.  D I had planned on implementing the ZIPEXE and PCSI$COMPRESSED files aB while ago but then decided to wait until we could implement SecureE Delivery at the same time.  Hopefully customers have already received G the Customer Advisory about this change.  If case not, I'm adding it as  a note here.   George Pagliarulo  ECO Release Process  OpenVMS Sustaining Engineering Hewlett-Packard Company         Peter 'EPLAN' LANGSTOEGER wrote:, > Can anybody tell me the difference between > > > VAXSHAD03_073.A		26-JUN-2006 09:15:00.40		I have since weeks > and > > VAXSHAD03_073.A		25-AUG-2006 13:50:38.69		I downloaded today >  > ?? > N > Why can different kits (date, content, MD5 is different, don't know funtion) > not get different names? > E > btw: Something similar happened to VMS732_MP-V0100 a few weeks ago. D > (But there was also a hint regarding UPDATE kit requirement added) > / > Where is QA when someone needs it? Pleeeaaase  >  > -EPLAN > L > PS: Many thanks to the folks which are responsible for the change of DCXedM > .PCSI files to ZIPSFXed .PCSI[$COMPRESSED] files. I waited (and lobbied ;-) E > for this many years, but finally it's done (totally unexpected ;-).  > -- > Peter "EPLAN" LANGSTOEGER ' > Network and OpenVMS system specialist  > E-mail  peter@langstoeger.atH > A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 30 Aug 2006 16:24:23 -0700 From: sean@obanion.us - Subject: Distributed Netbeans 5.0 field test? C Message-ID: <1156980263.678124.309350@p79g2000cwp.googlegroups.com>   D Back in July, Meg Watson of Distributed NetBeans Engineering stated:6 "We are about to release Distributed NetBeans T5.0..."< I had asked to be in the field test, but have heard nothing.D I also have a client that is looking forward to the Cobol support in 5.0.  E Has anybody gotten the field test?  How's it going (that you can talk  about)?        Sean   ------------------------------    Date: 30 Aug 2006 20:52:12 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> 1 Subject: Re: Distributed Netbeans 5.0 field test? B Message-ID: <1156996332.140571.71510@h48g2000cwc.googlegroups.com>   sean@obanion.us wrote:F > Back in July, Meg Watson of Distributed NetBeans Engineering stated:8 > "We are about to release Distributed NetBeans T5.0..."> > I had asked to be in the field test, but have heard nothing.F > I also have a client that is looking forward to the Cobol support in > 5.0. > G > Has anybody gotten the field test?  How's it going (that you can talk 	 > about)?  >  >  >  > Sean  A You need to check out the "What's new on our site" section of the E OpenVMS pages.  The field test was just announced and there are links  to the details.   C http://h71000.www7.hp.com/openvms/products/ips/netbeans/distnb.html    ------------------------------  + Date: Tue, 29 Aug 2006 08:44:38 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)D Subject: Re: Funniest Ask-the-Wizard question and answer of all time$ Message-ID: <ed0upl$6br$4@online.de>  @ In article <slrnef23gc.sng.usenet@zappy.catbert.org>, Dan Foster <usenet@evilphb.org> writes:    0 > http://h71000.www7.hp.com/wizard/wiz_5690.html >  > :-)  > J > I normally summarize the contents of URLs that I post, but in this case,( > I don't want to give away the joke. :) > E > Offense not meant at all for the questioner or the 'answerer'; just > > honestly found it amusing, and wish I'd thought to ask that!  G Reminds me of Steve Lionel's story about someone asking about the !$DEC G directives in Fortran and suggesting a similar change.  He assured him  E that it stood for "Directive Enhanced Compilation" and should not be  
 changed.  :-)    ------------------------------  % Date: Tue, 29 Aug 2006 07:58:47 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> D Subject: Re: Funniest Ask-the-Wizard question and answer of all time< Message-ID: <44f42a96$0$24208$9a6e19ea@news.newshosting.com>  2 "Dan Foster" <usenet@evilphb.org> wrote in message/ news:slrnef23gc.sng.usenet@zappy.catbert.org... # > In my honest opinion, this is it:  > 0 > http://h71000.www7.hp.com/wizard/wiz_5690.html >  > :-)  > J > I normally summarize the contents of URLs that I post, but in this case,( > I don't want to give away the joke. :) > E > Offense not meant at all for the questioner or the 'answerer'; just > > honestly found it amusing, and wish I'd thought to ask that! >  > -Dan  5 I stand guilty of that instance of post-merger humor.    :-)   K p.s. now that we're seeing product name changes like "HP BASIC for OpenVMS" H and "HP C for OpenVMS" shouldn't they start calling the product "OpenVMS BASIC" and "OpenVMS C"?   
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.! http://www3.sympatico.ca/n.rieck/    ------------------------------    Date: 29 Aug 2006 08:42:57 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) D Subject: Re: Funniest Ask-the-Wizard question and answer of all time3 Message-ID: <dMMWAtLKLHAz@eisner.encompasserve.org>   h In article <44f42a96$0$24208$9a6e19ea@news.newshosting.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes: > M > p.s. now that we're seeing product name changes like "HP BASIC for OpenVMS" J > and "HP C for OpenVMS" shouldn't they start calling the product "OpenVMS > BASIC" and "OpenVMS C"?   @    Good idea.  I can understand DEC wanting to call them DEC (orF    digital) C, BASIC, ...; since DEC had an ephasis on having VMS and     Tru64 compilers compatable.  E    But the HP-UX C compiler and the VMS C compiler, ... are different G    beasts.  C++ very much so.  The emphasis is on compatability between     Alpha and IA64 running VMS.  H    But whatever you call it, I still want an F95 compiler on my hobbyistF    VAX.  Now if I can just find a big commercial VAX shop to make that    same request.   ------------------------------  % Date: Tue, 29 Aug 2006 08:56:15 -0700 ( From: Jeff Cameron <roktsci@comcast.net>D Subject: Re: Funniest Ask-the-Wizard question and answer of all time0 Message-ID: <C119B1AF.24C50%roktsci@comcast.net>  9 When Compaq first bought DEC, we started spelling it DEQ.      On 8/29/06 4:58 AM, in article< 44f42a96$0$24208$9a6e19ea@news.newshosting.com, "Neil Rieck" <n.rieck@sympatico.ca> wrote:    > 4 > "Dan Foster" <usenet@evilphb.org> wrote in message1 > news:slrnef23gc.sng.usenet@zappy.catbert.org... $ >> In my honest opinion, this is it: >>  1 >> http://h71000.www7.hp.com/wizard/wiz_5690.html  >>   >> :-) >>  K >> I normally summarize the contents of URLs that I post, but in this case, ) >> I don't want to give away the joke. :)  >>  F >> Offense not meant at all for the questioner or the 'answerer'; just? >> honestly found it amusing, and wish I'd thought to ask that!  >>   >> -Dan  > 7 > I stand guilty of that instance of post-merger humor.  >  > :-)  > M > p.s. now that we're seeing product name changes like "HP BASIC for OpenVMS" J > and "HP C for OpenVMS" shouldn't they start calling the product "OpenVMS > BASIC" and "OpenVMS C"?  >  > Neil Rieck > Kitchener/Waterloo/Cambridge,  > Ontario, Canada.# > http://www3.sympatico.ca/n.rieck/  >  >  >    ------------------------------  % Date: Tue, 29 Aug 2006 14:41:40 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> D Subject: Re: Funniest Ask-the-Wizard question and answer of all time, Message-ID: <44F48A64.71F822DF@teksavvy.com>   Neil Rieck wrote: M > p.s. now that we're seeing product name changes like "HP BASIC for OpenVMS" J > and "HP C for OpenVMS" shouldn't they start calling the product "OpenVMS > BASIC" and "OpenVMS C"?   C I agree, especially since HP has its own set of compilers which are D distinct from those that run on VMS, so there are multiple different products bearing the same name.    ------------------------------  + Date: Wed, 30 Aug 2006 09:40:19 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)D Subject: Re: Funniest Ask-the-Wizard question and answer of all time$ Message-ID: <ed3me3$lu0$2@online.de>  3 In article <dMMWAtLKLHAz@eisner.encompasserve.org>, > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:   J >    But whatever you call it, I still want an F95 compiler on my hobbyistH >    VAX.  Now if I can just find a big commercial VAX shop to make that >    same request.  I Of course, there was no chance of this happening on VAX even back in the  G DEC days.  There was a company which had a "compiler" which translated  F Fortran90 code to VAX FORTRAN.  (Not necessarily a bad idea, when you 7 consider that some Fortran "compilers" translate to C.)   G As a hobbyist with 15 VAXes at home and also a Fortran95 afficianado, I A can appreciate the request, but ask "why not add an ALPHA to your H cluster?".  In the past few weeks, I've manage to collect 7 EV4 and EV45H ALPHAs for free (or perhaps a beer and a lunch or whatever), so they are there if you look.     ------------------------------    Date: 30 Aug 2006 04:47:43 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) D Subject: Re: Funniest Ask-the-Wizard question and answer of all time3 Message-ID: <xMR745uAa9nw@eisner.encompasserve.org>   w In article <ed3me3$lu0$2@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:   K > Of course, there was no chance of this happening on VAX even back in the  I > DEC days.  There was a company which had a "compiler" which translated  H > Fortran90 code to VAX FORTRAN.  (Not necessarily a bad idea, when you 9 > consider that some Fortran "compilers" translate to C.)   D There is even rumor of a Fortran compiler that translates to machine language :-)   ------------------------------    Date: 30 Aug 2006 07:06:35 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) D Subject: Re: Funniest Ask-the-Wizard question and answer of all time3 Message-ID: <GFNvQzetv7Hn@eisner.encompasserve.org>   w In article <ed3me3$lu0$2@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:  > I > As a hobbyist with 15 VAXes at home and also a Fortran95 afficianado, I C > can appreciate the request, but ask "why not add an ALPHA to your J > cluster?".  In the past few weeks, I've manage to collect 7 EV4 and EV45J > ALPHAs for free (or perhaps a beer and a lunch or whatever), so they are > there if you look.    C    I have an Alpha in my cluster.  But I was working on code that I D    wanted to run on all my systems, had worked NAM$ vs. NAML$ issuesJ    by using Fortran-9x features (VAXen don't have NAML$), only to realize %    my VAX couldn't compile it at all.      G    I tend to develope first on my Alpha, then port to my VAX.  This one     wasn't a recompile.   ------------------------------  % Date: Wed, 30 Aug 2006 15:11:38 +0200 / From: Paul Sture <paul.sture.nospam@hispeed.ch> D Subject: Re: Funniest Ask-the-Wizard question and answer of all timeJ Message-ID: <paul.sture.nospam-8FA36D.15113830082006@mac.sture.homeip.net>  3 In article <GFNvQzetv7Hn@eisner.encompasserve.org>, =  koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:   P > In article <ed3me3$lu0$2@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip + > Helbig---remove CLOTHES to reply) writes:  > > K > > As a hobbyist with 15 VAXes at home and also a Fortran95 afficianado, I E > > can appreciate the request, but ask "why not add an ALPHA to your L > > cluster?".  In the past few weeks, I've manage to collect 7 EV4 and EV45L > > ALPHAs for free (or perhaps a beer and a lunch or whatever), so they are > > there if you look.   > E >    I have an Alpha in my cluster.  But I was working on code that I F >    wanted to run on all my systems, had worked NAM$ vs. NAML$ issuesL >    by using Fortran-9x features (VAXen don't have NAML$), only to realize ' >    my VAX couldn't compile it at all.  >     I >    I tend to develope first on my Alpha, then port to my VAX.  This one  >    wasn't a recompile.  I I am curious. Do many other people develop on Alpha, then port to VAX? I  C came across one company that did this about a decade ago; they had  I plenty of spare resources for development work on Alpha, where the VAXes  & were pretty full with production work.   --  
 Paul Sture   ------------------------------    Date: 30 Aug 2006 11:09:49 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) D Subject: Re: Funniest Ask-the-Wizard question and answer of all time3 Message-ID: <B7S0ZX030cyF@eisner.encompasserve.org>   | In article <paul.sture.nospam-8FA36D.15113830082006@mac.sture.homeip.net>, Paul Sture <paul.sture.nospam@hispeed.ch> writes:5 > In article <GFNvQzetv7Hn@eisner.encompasserve.org>, ? >  koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:   J >>    I tend to develope first on my Alpha, then port to my VAX.  This one >>    wasn't a recompile.  > K > I am curious. Do many other people develop on Alpha, then port to VAX? I  E > came across one company that did this about a decade ago; they had  K > plenty of spare resources for development work on Alpha, where the VAXes  ( > were pretty full with production work.  B It depends on where the development tools work better.   Sometimes( that is Alpha, but more often it is VAX.   ------------------------------    Date: 30 Aug 2006 12:32:27 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) D Subject: Re: Funniest Ask-the-Wizard question and answer of all time3 Message-ID: <$0hLHPfZnS95@eisner.encompasserve.org>   | In article <paul.sture.nospam-8FA36D.15113830082006@mac.sture.homeip.net>, Paul Sture <paul.sture.nospam@hispeed.ch> writes: > K > I am curious. Do many other people develop on Alpha, then port to VAX? I  E > came across one company that did this about a decade ago; they had  K > plenty of spare resources for development work on Alpha, where the VAXes  ( > were pretty full with production work.  G    I do it because compilers tend to be compute intensive.  And I do it     both at home and at work.   ------------------------------  % Date: Wed, 30 Aug 2006 13:54:57 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> D Subject: Re: Funniest Ask-the-Wizard question and answer of all time* Message-ID: <44F5D0F1.4030908@comcast.net>   Bob Koehler wrote:  ~ > In article <paul.sture.nospam-8FA36D.15113830082006@mac.sture.homeip.net>, Paul Sture <paul.sture.nospam@hispeed.ch> writes: > K >>I am curious. Do many other people develop on Alpha, then port to VAX? I  E >>came across one company that did this about a decade ago; they had  K >>plenty of spare resources for development work on Alpha, where the VAXes  ( >>were pretty full with production work. >  > I >    I do it because compilers tend to be compute intensive.  And I do it  >    both at home and at work.  A If, like many of us, you have spent half your life waiting for a  I computer, you use the fastest iron you can afford!  For many years, that   meant Alpha.  E Porting, in either direction, is not that big a deal unless you have  C hardware dependencies built into your code.  And, if you must grub  H around in places where things like page size make a difference, you can  query VMS for the page size.   ------------------------------  % Date: Wed, 30 Aug 2006 13:55:41 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> D Subject: Re: Funniest Ask-the-Wizard question and answer of all time* Message-ID: <44F5D11D.8090207@comcast.net>   Bob Koehler wrote:  ~ > In article <paul.sture.nospam-8FA36D.15113830082006@mac.sture.homeip.net>, Paul Sture <paul.sture.nospam@hispeed.ch> writes: > K >>I am curious. Do many other people develop on Alpha, then port to VAX? I  E >>came across one company that did this about a decade ago; they had  K >>plenty of spare resources for development work on Alpha, where the VAXes  ( >>were pretty full with production work. >  > I >    I do it because compilers tend to be compute intensive.  And I do it  >    both at home and at work.  A If, like many of us, you have spent half your life waiting for a  I computer, you use the fastest iron you can afford!  For many years, that   meant Alpha.  E Porting, in either direction, is not that big a deal unless you have  C hardware dependencies built into your code.  And, if you must grub  H around in places where things like page size make a difference, you can  query VMS for the page size.   ------------------------------  % Date: Wed, 30 Aug 2006 13:56:32 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> D Subject: Re: Funniest Ask-the-Wizard question and answer of all time: Message-ID: <rridnZtD7J9QTGjZnZ2dnUVZ_qmdnZ2d@comcast.com>   Bob Koehler wrote:  ~ > In article <paul.sture.nospam-8FA36D.15113830082006@mac.sture.homeip.net>, Paul Sture <paul.sture.nospam@hispeed.ch> writes: > K >>I am curious. Do many other people develop on Alpha, then port to VAX? I  E >>came across one company that did this about a decade ago; they had  K >>plenty of spare resources for development work on Alpha, where the VAXes  ( >>were pretty full with production work. >  > I >    I do it because compilers tend to be compute intensive.  And I do it  >    both at home and at work.  A If, like many of us, you have spent half your life waiting for a  I computer, you use the fastest iron you can afford!  For many years, that   meant Alpha.  E Porting, in either direction, is not that big a deal unless you have  C hardware dependencies built into your code.  And, if you must grub  H around in places where things like page size make a difference, you can  query VMS for the page size.   ------------------------------  % Date: Tue, 29 Aug 2006 11:33:58 +0100 0 From: Chris Sharman <chris.sharman@sorry.nospam> Subject: goodbye from me too4 Message-ID: <ed156m$314$1$8300dec7@news.demon.co.uk>  I After 21 years on VMS, I'm finishing on Friday to retrain as a secondary   maths teacher.  E I've learned a lot here, and contributed occasionally - I'll miss it.    Good luck to everyone.
 Chris Sharman    ------------------------------  # Date: Tue, 29 Aug 2006 21:34:36 GMT L From: winston@SSRL.SLAC.STANFORD.EDU (Alan Winston - SSRL Central Computing)  Subject: Re: goodbye from me too6 Message-ID: <00A5AF00.B76EA1D7@SSRL.SLAC.STANFORD.EDU>  g In article <ed156m$314$1$8300dec7@news.demon.co.uk>, Chris Sharman <chris.sharman@sorry.nospam> writes:   J >After 21 years on VMS, I'm finishing on Friday to retrain as a secondary  >maths teacher.  > F >I've learned a lot here, and contributed occasionally - I'll miss it. >  >Good luck to everyone.   # And to you.  Enjoy your new career!    -- Alan    ------------------------------  % Date: Tue, 29 Aug 2006 20:26:32 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>   Subject: Re: goodbye from me too, Message-ID: <44F4DB20.58CB7CA5@teksavvy.com>   Chris Sharman wrote:J > After 21 years on VMS, I'm finishing on Friday to retrain as a secondary > maths teacher.  F You do realise that you are under an obligation to teach your studentsF about VMS.... The long term future of VMS rests on your shoulders, youF need to shape those young adolescents on the advantages of VMS so that# they keep their minds opened to it.    :-)    Good luck on your endeavours !   ------------------------------  % Date: Tue, 29 Aug 2006 21:12:49 -0500 6 From: "David J. Dachtera" <djesys.no@spam.comcast.net>  Subject: Re: goodbye from me too0 Message-ID: <44F4F421.FC5E38FA@spam.comcast.net>   Chris Sharman wrote: > J > After 21 years on VMS, I'm finishing on Friday to retrain as a secondary > maths teacher. > G > I've learned a lot here, and contributed occasionally - I'll miss it.  >  > Good luck to everyone. > Chris Sharman    Atta-boy, Chris!  & Best of everything in your new career!   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------    Date: 29 Aug 2006 19:39:05 -0700$ From: "AEF" <spamsink2001@yahoo.com>  Subject: Re: goodbye from me tooC Message-ID: <1156905544.976758.115890@p79g2000cwp.googlegroups.com>    Chris Sharman wrote:J > After 21 years on VMS, I'm finishing on Friday to retrain as a secondary > maths teacher. > G > I've learned a lot here, and contributed occasionally - I'll miss it.  >  > Good luck to everyone. > Chris Sharman   " I hope you get to teach calculus!    Bye.   AEF    ------------------------------    Date: 29 Aug 2006 18:49:44 -0700) From: "Sue" <susan_skonetski@hotmail.com> - Subject: HP OpenVMS BASIC boot-camp training! C Message-ID: <1156902584.285359.146060@p79g2000cwp.googlegroups.com>   ? Dear Internal and External distribution lists (and Newsgroups),   E This BASIC boot camp is different than the OpenVMS Advanced technical F boot camp.  This boot camp is highly recommended for the newer OpenVMS@ user and is being done in response to customer requests for more6 introductory training.  Please forward as you see fit.   OK to post.   
 Warm Regards,    Sue     4 ____________________________________________________8 Subject: FW: Attend HP OpenVMS Basic Boot-camp training!  C Course Title:  HP OpenVMS   Basic  boot-camp ( HP Education  Course  #U3585S)E This   10-day HP OpenVMS basic boot camp course is primarily designed > to provide HP Customers, Selling  Partners ,  Presales SystemsA Engineers and  Services Employees with an intense and accelerated ? training to help them gain basic-to-intermediate OpenVMS system E management skills. Detailed  course outline and pricing available at: / http://www.hp.com/education/courses/u3585s.html B Please  Note: This 10-day OpenVMS Basic boot camp training is at aA different level from and should not be confused with the 5-day HP > OpenVMS Advanced Technical boot camp offered each year in May.? Date:  October 2nd- 13th, 2006       Enroll    Location: Denver . U3585 -HP OpenVMS basic boot camp  Objectives:A To learn enough to perform specific HP OpenVMS system and network  management functionsE To refresh and update HP OpenVMS knowledge for experienced HP OpenVMS C system managers or HP Tru64 system managers returning to HP OpenVMS @ To  prepare for HP OpenVMS certification through coverage of theC certification competency model  enabling  most students to pass the E three HP OpenVMS exams. Labs will take place both in and after class.     Pre-requisites:9  HP  OpenVMS Fundamentals (U3716) or equivalent knowledge                   OR >  Minimum 1 year of basic HP OpenVMS system management hands-on
 experience                 ORG The ability to write hp OpenVMS Command procedures to the level covered B in the latest hp OpenVMS fundamentals course. Students without theF pre-requisite HP OpenVMS knowledge  will find it difficult to maintain the instructional pace. C To Register:   www.hp.com/education   or   Call 1-800 HP CLASS   or  click on the    Enroll  linkA Please bookmark the URL below for  HP OpenVMS Curriculum path and  additional courses: 1 http://www.hp.com/education/sections/openvms.html           7 From: Hewlett-Packard [mailto:us-bulletins@your.hp.com] 6 Subject: Attend HP OpenVMS  Basic  Boot-camp training!    ? Make sure your HP emails always go straight to your inbox.  Add : us-bulletins@your.hp.com to your address book!  Need Help?    & HP OpenVMS  Basic  Boot-camp training!  A Welcome to Education & Training, US & Canada's training bulletin!    August 2006   9 Our training bulletin provides you information about HP's F industry-leading IT training classes.  You can always check the latest schedules on HP's web site.   ' Visit us online at www.hp.com/education  or call  800-472-5277 (US)  800-563-5089 (Canada)   
 Did you know?   0 Print HP Education's OpenVMS  training schedule:  C Printable HP Education OpenVMS training schedule available for your   training & development planning!   Save on your Training!  ? You can save with new HP Care Pack Services for Education!   HP @ CarePack Total Education provides you with a credit of education@ services you can use for all our scheduled courses in any of ourD education centers and online.    More than 1000 courses, from  HP-UXF to Storage!   It is a flexible, economical, and simple way to purchase	 training.   C The HP CarePack Total Education is available directly through HP by E calling 1-800 HPCLASS, your HP Education Sales Representative or your  HP Reseller.      7 Reserve your spot for the HP OpenVMS hands-on training!   E Attend the 10 day HP OpenVMS boot camp (U3585) training course during * October 2nd- October 13th, 2006 in Denver.   Course Objectives:E This  hands-on HP OpenVMS Boot camp training course has the following  objectives: &  =B7 Describe the OpenVMS environment.      Packaging.       Hardware overview.       Software overview.       Associated products. 1  =B7 OpenVMS operating system technical concepts. -  =B7 Installing OpenVMS and layered products. B  =B7 Choosing and configuring and managing cluster configurations.2  =B7 OpenVMS and OpenVMS cluster trouble-shooting..  =B7 Choosing and configuring OpenVMS storage.-  =B7 Configuring and managing OpenVMS Galaxy. C  =B7 Configuring OpenVMS and networking software (LAN, LAT, TCP/IP, 
 DECnet Plus). ?  =B7 Performing day to day OpenVMS and network management tasks 	 (TCP/IP). &  =B7 Trouble shooting TCP/IP Services.    * For a detailed course outline, click here.  
 Registration: > To register, click on the enroll link below or visit our site.  , Date: 10/02-10/13   Enroll  Location: Denver  D Please bookmark the URL below for the HP OpenVMS Curriculum path and additional courses: 1 http://www.hp.com/education/sections/openvms.html    ------------------------------    Date: 27 Aug 2006 11:03:45 -0700/ From: "Volker Halle" <volker_halle@hotmail.com>  Subject: Re: HSG80 LED CodesB Message-ID: <1156701825.853820.162520@75g2000cwc.googlegroups.com>   Tom,   how about this ?   http://h20000.www2.hp.com/bizsupport/TechSupport/DocumentIndex.jsp?contentType=SupportManual&locale=en_US&docIndexId=179911&taskId=101&prodTypeId=12169&prodSeriesId=433888    Volker.    ------------------------------  % Date: Sun, 27 Aug 2006 10:30:52 -0700 * From: "Tom Linden" <tom@kednos-remove.com> Subject: HSG80 LED Codes) Message-ID: <op.teydlquotte90l@hyrrokkin>   = Does anybody have these codes?  Better yet the Service Guide?    Tom    ------------------------------  % Date: Sun, 27 Aug 2006 11:20:14 -0700 * From: "Tom Linden" <tom@kednos-remove.com> Subject: Re: HSG80 LED Codes) Message-ID: <op.teyfv0a5tte90l@hyrrokkin>   3 On Sun, 27 Aug 2006 11:03:45 -0700, Volker Halle  =   ! <volker_halle@hotmail.com> wrote:    > Tom, >  > how about this ? > I > http://h20000.www2.hp.com/bizsupport/TechSupport/DocumentIndex.jsp?con= I tentType=3DSupportManual&locale=3Den_US&docIndexId=3D179911&taskId=3D101= ) &prodTypeId=3D12169&prodSeriesId=3D433888  > 	 > Volker.  >   I Thanks, Volker. guess I was too impatient in my googling.   I guess it i=  s  =   only available( in pdf, but that is better than nothing.   Tom  -- =  E Using Opera's revolutionary e-mail client: http://www.opera.com/mail/    ------------------------------  % Date: Tue, 29 Aug 2006 17:40:08 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> B Subject: Intermixing expected and unanticipated serial comms reads, Message-ID: <44F4B42A.B1FDFE4C@teksavvy.com>   Serial communications with QIO  D I send a packet and then must read a packet which contains an ACk orC NAK. (and resent if it is a NAK or if there is no response within a H timeout period, the serial driver is great because of timeout support in a read operation).  B Or, I await a list of packets (after having sent the remote unit aD command to send me a list of stuff), and with each packet, I have toG acknowledge with an ACK or NAK packet, and if the data stops before the F end, I have to detect the timeout and send out a NAK to try to get the< remote unit to resend a packet and restart the flow of data.  1 In both cases, the timeout on reads is important.   6 So far, all this can be neatly accomplished with QIOW.  E HOWEVER, there may be some unanticipated data being sent by the unit. B (for instance, the GPS can be put in a mode where it transmits itsH position every second, and this can be intermixed with more orderly dataF requests. Or it could be sending a NAK at regular intervals when idle.    D I can handle this with the following when doing "anticipated stuff":  3 $CANCEL any pending AST-driven unanticipated reads.  	$QIOW to write a packet) 	$QIOW to read the ACK/NAK (with timeout) A $QIO to set a read operation with AST for any unanticipated data.  Without timeout.    H If I remember correctly the $CANCEL will unfortunatly invoke the AST for@ the outstanding read (with an SS$CANCEL or something similar) as condition code.   < Does the above $CANCEL/$QIO for every series of "structured"* communications involve a lot of overhead ?    H The other option would be to transform all reads into a single read thatG triggers an AST, and then implement the timeouts seperately. This woudl H change the structure of the program quite a bit. Is it worth that effort
 ? Any ideas ?   F Odeally, if I could MODIFY a pending QIO (for instance, add/remove theB timeout value) instead of $CANCELing it and re-issuing it with the- changed parameters, it would be much cleaner.    ------------------------------  % Date: Wed, 30 Aug 2006 02:33:26 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> F Subject: Re: Intermixing expected and unanticipated serial comms reads, Message-ID: <44F53105.43EE3474@teksavvy.com>   Ulrich Bellgardt wrote: E > device can transmit unsolicited data at any time, it also can do so D > while your $QIOW is waiting for the expected ACK/NAK, and send theC > anticipated data set afterwards. In this case, your $QIOW will be @ > surprised by unanticipated bits, and the $QIO followed by thisE > unexpectedly receives the anticipated data set if the remote device  > really sends them.    D Yeah. You are right. I've begun to (re) structure that bit to be ASTD driven and have the AST do some of the packet assembly work and thenD dispatch a routine based on what type of packet it is. This makes itD more complex because of the timeout (requires even flags, converting@ time into binary time in a quadword calling/cancelling the timerF routines etc.) and of course setting of a dispatcher based on the type2 of packet received, but as Olsen would have said :   "Do the right thing" :-)   ------------------------------  % Date: Wed, 30 Aug 2006 16:14:49 -0400 , From: <Barry.Treahy@EmersonNetworkPower.com>  Subject: Is the SAIC Relay down?M Message-ID: <63A4454BFCE1C048B2683DBB63A3633363C69F@ETP-CIN-US-EX01.etp1.com>   E It's been way too quiet, did I get bounced or is the SAIC relay down?    Barry Treahy, = A Jr=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=20  Vice President/CIO Midwest Microwave, Inc. . Emerson=A0Network Power Connectivity Solutions, E-mail: Barry.Treahy@EmersonNetworkPower.com Phone: 480/314-1320  Cell:=A0=A0=A0=A0 480/216-9568 Fax:=A0 =A0 =A0480/661-7028  =A0 H =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 ... = but it's a DRY HEAT!=A0    ------------------------------  % Date: Wed, 30 Aug 2006 21:38:23 -0400 ' From: Dave Froble <davef@tsoft-inc.com> $ Subject: Re: Is the SAIC Relay down?9 Message-ID: <hY2dnWjYoPF2pmvZnZ2dnUVZ_r-dnZ2d@libcom.com>   + Barry.Treahy@EmersonNetworkPower.com wrote: G > It's been way too quiet, did I get bounced or is the SAIC relay down?   H I'm thinking that it possibly is.  I had a friend ask the same question.   --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Thu, 31 Aug 2006 11:05:45 -0400  From: norm.raphael@metso.com$ Subject: Re: Is the SAIC Relay down?Q Message-ID: <OF5AFD52F3.DCE60A96-ON852571DB.0052D735-852571DB.0052ECBA@metso.com>   H I just got 184 messages in a burst, so I guess whatever was is now okay.  B Dave Froble <davef@tsoft-inc.com> wrote on 08/30/2006 09:38:23 PM:  - > Barry.Treahy@EmersonNetworkPower.com wrote: I > > It's been way too quiet, did I get bounced or is the SAIC relay down?  > J > I'm thinking that it possibly is.  I had a friend ask the same question. >  > --6 > David Froble                       Tel: 724-529-0450@ > Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com > DFE Ultralights, Inc.  > 170 Grimplin Road  > Vanderbilt, PA  15486    ------------------------------  % Date: Sun, 27 Aug 2006 11:29:47 +0200 + From: "Michel HERRSCHER" <mhc@herrscher.fr> / Subject: Re: Locating early versions of VAX/VMS 0 Message-ID: <ecromn$2m8n$1@cabale.usenet-fr.net>  # Dans un message Dan Foster disait :    ....   > J > I would be especially interested in locating a copy of VMS 1.0, as well.  F I am not sure it was published by Digital, I think it was an internal 	 release .   ' The first one I got was ..... V1.3 ....  ouups so long time ago ....    --   Michel HERRSCHER     ------------------------------  % Date: Sun, 27 Aug 2006 11:13:15 -0500 % From: Dan Foster <usenet@evilphb.org> / Subject: Re: Locating early versions of VAX/VMS 5 Message-ID: <slrnef3h4r.hok.usenet@zappy.catbert.org>    Bonjour, Monsieur Herrscher,  [ In article <ecromn$2m8n$1@cabale.usenet-fr.net>, Michel HERRSCHER <mhc@herrscher.fr> wrote: % > Dans un message Dan Foster disait :  > K >> I would be especially interested in locating a copy of VMS 1.0, as well.  > H > I am not sure it was published by Digital, I think it was an internal  > release .   C Ahh, yes, you are right. I had forgotten about that. Thanks for the  clarification (and reminder).   ) > The first one I got was ..... V1.3 ....  > ouups so long time ago ....    :-)    Regards,   -Dan   ------------------------------  % Date: Mon, 28 Aug 2006 01:47:48 -0500 % From: Dan Foster <usenet@evilphb.org> / Subject: Re: Locating early versions of VAX/VMS 5 Message-ID: <slrnef54ck.hok.usenet@zappy.catbert.org>   \ In article <slrnef23ak.sng.usenet@zappy.catbert.org>, Dan Foster <usenet@evilphb.org> wrote:H > As a member of the hobbyist program, I am interested in locating earlyG > versions of VMS since I would like to see if I can boot them with the 6 > SIMH emulator, solely out of intellectual curiosity.  F I have made arrangements for versions 1.5 through 4.x or so, includingF locating a wide array of tape drives -- everything prior to a DLT8000,. thanks to very generous offers for assistance.  E If anyone has any versions prior to 1.5 on tapes, please let me know. * This seems to be the really hard part. :-)  E (You would not need to read the tapes or have working tape drives and ? old systems; that -- and reading tapes -- can be arranged for.)   C A concern is that unless the oldest versions are read off tape, and F copied to newer media, it may be lost forever once tape drives, tapes,F and the systems attached to them experiences further hardware failuresC over time. 25+ years is pretty good, but why further tempt fate? :)   F It will certainly be strange to see a version without the LMF, as wellH as all the RSX stuff, heavy reliance on PDP compatibility mode for earlyC utilities, a different DIRECTORY utility, amongst other things. Was ) ODS-1 supported in early versions of VMS?   D Having originally started with VMS at 5.5-2, was startled to see how9 different it was at 4.4... 1.5 should be mind blowing. :)   E But very impressively, the (SIMH) emulated 11/780 has actually booted B and run OpenVMS/VAX 7.2. Very few manufacturers can claim that itsE earliest machine would ever be capable of running a recent OS version  nearly 3 decades later.    -Dan   ------------------------------  % Date: Mon, 28 Aug 2006 09:15:48 +0200 + From: "Michel HERRSCHER" <mhc@herrscher.fr> / Subject: Re: Locating early versions of VAX/VMS / Message-ID: <ecu57d$coh$1@cabale.usenet-fr.net>   # Dans un message Dan Foster disait :   H > It will certainly be strange to see a version without the LMF, as wellJ > as all the RSX stuff, heavy reliance on PDP compatibility mode for earlyE > utilities, a different DIRECTORY utility, amongst other things. Was + > ODS-1 supported in early versions of VMS?   M oh yaehhh and at that time we also have to backup RP06 drive to RP06 disk as   tape were only 800/1600 BPI ... E adjust the heads of RM03 every one or 2 month.... RP06 every month...    > F > Having originally started with VMS at 5.5-2, was startled to see how; > different it was at 4.4... 1.5 should be mind blowing. :)   G amazing ... I just prefer not to remeber these old days when the cobol  M compiler was run in RSX mode while the generated code was in native oucchhhh  E debugging was not easy at all... and like that for lot of products...    > G > But very impressively, the (SIMH) emulated 11/780 has actually booted D > and run OpenVMS/VAX 7.2. Very few manufacturers can claim that itsG > earliest machine would ever be capable of running a recent OS version  > nearly 3 decades later.   , and also may coexist with recent hardware...     Funny old days.... --   Michel HERRSCHER CONSULTANT  Tel : +33450870912 http://www.mhc.herrscher.fr ; Prsident WINDASSO - Association des utilisateurs WxxDEV(c)  http://www.windasso.org    ------------------------------  # Date: Mon, 28 Aug 2006 14:56:59 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>/ Subject: Re: Locating early versions of VAX/VMS - Message-ID: <%uDIg.17$wx6.6@news.cpqcorp.net>    Dan Foster wrote: H > As a member of the hobbyist program, I am interested in locating earlyG > versions of VMS since I would like to see if I can boot them with the 6 > SIMH emulator, solely out of intellectual curiosity. > J > I would be especially interested in locating a copy of VMS 1.0, as well.E > (SIMH now supports the 11/780 as well as the various 780 busses and A > peripherals, so I believe booting V1 and V2 on it should work.)   P    We looked at booting a (real) VAX-11/780 under V1.0 as part of a trade show, P and ended up scrapping the idea because of the hardware requirements.  It's not M just a VAX-11/780, it's a very specific set of VAX-11/780 storage and system  Q hardware configurations that are required.  There was no way we were going to be  I able to locate the appropriate (and working) hardware for the trade show.   Q    That written, an emulator could be coded to have the correct MA780 memory and  Q the particular disks, et al.   But it would not operate on a "modern" VAX-11/780  P configuration, which is probably what a typical emulator is presently emulating.  Q    There were copies of V1.0 in the field, but I have no idea if any still exist  O outside the local archival media cabinets -- nor if any of the kits around are  O readable.  (And more to the point, I'm not currently in a position to dig into  N the archives and to look for and try to read and to replicate thirty year old K software kits for a request -- and no offense -- such as this one.  Sorry.)    ------------------------------  % Date: Mon, 28 Aug 2006 10:53:38 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> / Subject: Re: Locating early versions of VAX/VMS : Message-ID: <wpWdnekUY-Vtnm7ZnZ2dnUVZ_oidnZ2d@comcast.com>   Dan Foster wrote:   ^ > In article <slrnef23ak.sng.usenet@zappy.catbert.org>, Dan Foster <usenet@evilphb.org> wrote: > H >>As a member of the hobbyist program, I am interested in locating earlyG >>versions of VMS since I would like to see if I can boot them with the 6 >>SIMH emulator, solely out of intellectual curiosity. >  > H > I have made arrangements for versions 1.5 through 4.x or so, includingH > locating a wide array of tape drives -- everything prior to a DLT8000,0 > thanks to very generous offers for assistance. > G > If anyone has any versions prior to 1.5 on tapes, please let me know. , > This seems to be the really hard part. :-) > G > (You would not need to read the tapes or have working tape drives and A > old systems; that -- and reading tapes -- can be arranged for.)  > E > A concern is that unless the oldest versions are read off tape, and H > copied to newer media, it may be lost forever once tape drives, tapes,H > and the systems attached to them experiences further hardware failuresE > over time. 25+ years is pretty good, but why further tempt fate? :)  > H > It will certainly be strange to see a version without the LMF, as wellJ > as all the RSX stuff, heavy reliance on PDP compatibility mode for earlyE > utilities, a different DIRECTORY utility, amongst other things. Was + > ODS-1 supported in early versions of VMS?  > F > Having originally started with VMS at 5.5-2, was startled to see how; > different it was at 4.4... 1.5 should be mind blowing. :)  > G > But very impressively, the (SIMH) emulated 11/780 has actually booted D > and run OpenVMS/VAX 7.2. Very few manufacturers can claim that itsG > earliest machine would ever be capable of running a recent OS version  > nearly 3 decades later.  >  > -Dan  H The biggest problem is likely to be the age of the tapes.  Twenty-eight G year old tapes are probably mostly unreadable if they even exist.  The  I binder holding the magnetic oxide tends to break down after four or five  5 years and the oxide just wipes off the backing. . . .   G It's possible that VMS Engineering could make such a tape, even today,  @ except that they probably have much more important things to do.   ------------------------------  % Date: Mon, 28 Aug 2006 17:55:56 +0200 / From: Paul Sture <paul.sture.nospam@hispeed.ch> / Subject: Re: Locating early versions of VAX/VMS J Message-ID: <paul.sture.nospam-E7385D.17555628082006@mac.sture.homeip.net>  5 In article <slrnef54ck.hok.usenet@zappy.catbert.org>, '  Dan Foster <usenet@evilphb.org> wrote:   H > It will certainly be strange to see a version without the LMF, as wellJ > as all the RSX stuff, heavy reliance on PDP compatibility mode for earlyE > utilities, a different DIRECTORY utility, amongst other things. Was + > ODS-1 supported in early versions of VMS?   D IIRC ODS-1 was indeed supported. Do check out PIP and the other RSX 
 utilities.   F > Having originally started with VMS at 5.5-2, was startled to see how; > different it was at 4.4... 1.5 should be mind blowing. :)  >   G V3.0 brought quite a few useful goodies: MONITOR, BACKUP, CONVERT (and  = IIRC, FDL), CTRL-T, SPAWN/ATTACH. I didn't have much use for  B SPAWN/ATTACH myself, but the others were major advances. Prior to I CONVERT you were stuck with RSX utilities (e.g. IFL for Index File Load,  F which had the nasty tendency to bork before it was finished with some D not very informative message such as "Trap to 10"). It also brought  Prolog 3 indexed files  H V4.0 brought ACLs, RIGHTSLIST.DAT, command line recall and editing (how B we lived without that I don't know, but we did), new logical name H structures (including logical name tables), 39.39 length filenames, and 8 not to be forgotten, the ability to change your prompt.   F V4.0 also broke various bit of code that we had at the time, although I that was actually a Good Thing in terms of future maintainability of our  G own code (I include DCL there, as that had quite a bit of work done to  E it). In charge of our software distribution packaging at the time, I  E rewrote our procedures to use VMSINSTAL* (and said a firm Goodbye to  A using COPY to distribute files from tape - that had been causing  G customers grief with tape errors), and took the opportunity to rewrite  C most of our Installation and Configuration manual, with a chunk of  E recommendations on how to upgrade from V3.n to V4.0 as painlessly as  	 possible.   C * I used a DECUS article to unravel the mysteries of VMSINSTAL. My  C belated thanks to whoever wrote it (sadly I've forgotten his name).    --  
 Paul Sture   ------------------------------    Date: 28 Aug 2006 14:57:59 -04003 From: Rich Alderson <news@alderson.users.panix.com> / Subject: Re: Locating early versions of VAX/VMS . Message-ID: <mdd4pvwhemg.fsf@panix5.panix.com>  5 "Richard B. Gilbert" <rgilbert88@comcast.net> writes:   J > The biggest problem is likely to be the age of the tapes.  Twenty-eight I > year old tapes are probably mostly unreadable if they even exist.  The  K > binder holding the magnetic oxide tends to break down after four or five  7 > years and the oxide just wipes off the backing. . . .   L I frequently need to read 1/2in tapes that are more than 30 years old.  TheyK rarely give me any problems.  DECtapes more than 40 years old are even more M easily read and written.  There are times when care must be exercised, if the O tape has been badly treated and mishandled, but in general basic safeguards (no @ direct sunlight, lightly controlled temps and humidity) suffice.  N Your experience with Costco-special DAT drives does not apply to media created to a higher standard.    --  L Rich Alderson                                       | /"\ ASCII ribbon     |L news@alderson.users.panix.com                       | \ / campaign against |L "You get what anybody gets. You get a lifetime."    |  x  HTML mail and    |L                          --Death, of the Endless    | / \ postings         |   ------------------------------  % Date: Mon, 28 Aug 2006 14:46:48 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> / Subject: Re: Locating early versions of VAX/VMS , Message-ID: <44F33A17.7402B9C1@teksavvy.com>   Hoff Hoffman wrote: R >    That written, an emulator could be coded to have the correct MA780 memory andR > the particular disks, et al.   But it would not operate on a "modern" VAX-11/780R > configuration, which is probably what a typical emulator is presently emulating.    F Out of curiosity, at what level does an OS need to know about hardwareF properties of memory ? Doesn't it just see a set of continugous memoryB locations that contain data accessed transparently via the various- assembly opcodes that provide memory access ?   H Or was MA780 memory more akin to the early PCs with segmented blocks etcF etc and the OS needs to issue special assembler instructions to choose< which bank of memory the next operation will be working on ?   ------------------------------  % Date: Mon, 28 Aug 2006 18:10:00 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> / Subject: Re: Locating early versions of VAX/VMS : Message-ID: <iZWdna241pSq927ZnZ2dnUVZ_t2dnZ2d@comcast.com>   Rich Alderson wrote:  7 > "Richard B. Gilbert" <rgilbert88@comcast.net> writes:  >  > J >>The biggest problem is likely to be the age of the tapes.  Twenty-eight I >>year old tapes are probably mostly unreadable if they even exist.  The  K >>binder holding the magnetic oxide tends to break down after four or five  7 >>years and the oxide just wipes off the backing. . . .  >  > N > I frequently need to read 1/2in tapes that are more than 30 years old.  TheyM > rarely give me any problems.  DECtapes more than 40 years old are even more O > easily read and written.  There are times when care must be exercised, if the Q > tape has been badly treated and mishandled, but in general basic safeguards (no B > direct sunlight, lightly controlled temps and humidity) suffice. > P > Your experience with Costco-special DAT drives does not apply to media created > to a higher standard.  >   G Err. . . .  I was speaking of the seven and/or nine track reel-to-reel  E tapes that were in use in those days.  I would be surprised if there  F were many tapes from 1978 that were still in usable condition.  And I D think that more recent tapes just might be manufactured to a higher H standard; I hope the industry has learned a thing or two about magnetic / tape technology in the last twenty-eight years!   I It's *possible* to preserve a tape for twenty-eight years but damned few  C tapes get the kind of royal treatment required.  Not only must the  B temperature and humidity be controlled but also the tapes must be  rewound every couple of years.   ------------------------------  % Date: Mon, 28 Aug 2006 18:01:02 -0400 2 From: Bob Willard <BobwBSGS@TrashThis.comcast.net>/ Subject: Re: Locating early versions of VAX/VMS : Message-ID: <D9KdnfEuP5ic9W7ZnZ2dnUVZ_rydnZ2d@comcast.com>   JF Mezei wrote:  > Hoff Hoffman wrote:  > R >>   That written, an emulator could be coded to have the correct MA780 memory andR >>the particular disks, et al.   But it would not operate on a "modern" VAX-11/780R >>configuration, which is probably what a typical emulator is presently emulating. >  >  > H > Out of curiosity, at what level does an OS need to know about hardwareH > properties of memory ? Doesn't it just see a set of continugous memoryD > locations that contain data accessed transparently via the various/ > assembly opcodes that provide memory access ?  > J > Or was MA780 memory more akin to the early PCs with segmented blocks etcH > etc and the OS needs to issue special assembler instructions to choose> > which bank of memory the next operation will be working on ?  G Memory is memory, but the CSRs for any implementation vary quite a bit. F And the OS must know the specific layout and content of CSRs to detectJ and correct errors, amongst other details; CSR detail is also used to size  and init. memory during booting. --   Cheers, Bob    ------------------------------  % Date: Mon, 28 Aug 2006 15:32:11 -0400 , From: Hoff Hoffman <hoff-remove-this@hp.com>/ Subject: Re: Locating early versions of VAX/VMS * Message-ID: <44f344bd@usenet01.boi.hp.com>   JF Mezei wrote:   H > Out of curiosity, at what level does an OS need to know about hardware > properties of memory ?  H    An operating system always needs to know about the properties of its B hardware, and older hardware tends to be far more controller- and O platform-specific.  And there tend to be far larger variations from one system  P and one controller to the next than is now more commonly seen a quarter-century  later.  O    The memory controller itself is the issue here, and OpenVMS needs to detect  L and to establish the physical memory configuration, and the physical memory Q mapping.   And (as it was then known) VAX/VMS V1.0 was rather limited in what it  3 would be expected to recognize for system hardware.   Q    The emulator would have to present the A-series memory interface, IIRC.  (And  H this is all rather limited.  C-series was limited to addressing 4 MB of J installed memory IIRC, and involved filling the cabinet with 256KB memory J boards, as well as the controller modules and associated frames.  The 8MB Q VAX-11/780 with C-series I was working with was probably twenty feet long before  O you got to the disks and associated cabinets, and it was a comparatively small  Q configuration.  A 4MB upgrade filled an expansion cabinet, and was comparatively  Q cheap at US$8000 -- a replacement 8MB configuration with a new memory controller   was US$36,000, IIRC.)   P    Operating systems of the vintage of older OpenVMS VAX required a new release P for this sort of thing -- new I/O or new controllers often required non-trivial O effort, and new VAX systems were almost always going to require a new release.  ; This requirement was common up until the V5.* range, or so.   1 > Doesn't it just see a set of continugous memory D > locations that contain data accessed transparently via the various/ > assembly opcodes that provide memory access ?   R    Once configured, yes, application and system code uses transparent data access.   ------------------------------  % Date: Mon, 28 Aug 2006 15:37:04 -0400 , From: Hoff Hoffman <hoff-remove-this@hp.com>/ Subject: Re: Locating early versions of VAX/VMS * Message-ID: <44f345de@usenet01.boi.hp.com>   Rich Alderson wrote:7 > "Richard B. Gilbert" <rgilbert88@comcast.net> writes:  > K >> The biggest problem is likely to be the age of the tapes.  Twenty-eight  J >> year old tapes are probably mostly unreadable if they even exist.  The L >> binder holding the magnetic oxide tends to break down after four or five 8 >> years and the oxide just wipes off the backing. . . . > N > I frequently need to read 1/2in tapes that are more than 30 years old.  TheyM > rarely give me any problems.  DECtapes more than 40 years old are even more O > easily read and written.  There are times when care must be exercised, if the Q > tape has been badly treated and mishandled, but in general basic safeguards (no B > direct sunlight, lightly controlled temps and humidity) suffice. > P > Your experience with Costco-special DAT drives does not apply to media created > to a higher standard.     K    30 year old good-quality 9-track tape, semi-properly stored, is usually  Q entirely readable.  At least, what I've tried of it has been.  Equivalently-aged  P and bargain-quality 9-track tape usually isn't -- and I've seen the oxide flake P off that stuff -- regardless of how it was stored.  And I remember seeing oxide A falling off some of the bargain stuff when it was then-new stuff.   J    If it clogged your tape heads when you wrote the tape 30 years ago, it  probably didn't last.    ------------------------------    Date: 28 Aug 2006 17:48:11 -0700 From: davidc@montagar.com / Subject: Re: Locating early versions of VAX/VMS C Message-ID: <1156812491.802169.130810@b28g2000cwb.googlegroups.com>    Hoff Hoffman wrote: Q >    We looked at booting a (real) VAX-11/780 under V1.0 as part of a trade show, Q > and ended up scrapping the idea because of the hardware requirements.  It's not N > just a VAX-11/780, it's a very specific set of VAX-11/780 storage and systemR > hardware configurations that are required.  There was no way we were going to beK > able to locate the appropriate (and working) hardware for the trade show.   A But didn't you do that at a long past DECUS event, in the OpenVMS B campground?  Pat Jankowiak even provided a couple of pieces of theA system.  Maybe it was V1.5, but I though it might have been V1.0.    ------------------------------    Date: 29 Aug 2006 08:30:33 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) / Subject: Re: Locating early versions of VAX/VMS 3 Message-ID: <RKUN7ezKFwTr@eisner.encompasserve.org>   \ In article <44F33A17.7402B9C1@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: > H > Out of curiosity, at what level does an OS need to know about hardwareH > properties of memory ? Doesn't it just see a set of continugous memoryD > locations that contain data accessed transparently via the various/ > assembly opcodes that provide memory access ?   F    No.  Memory is mapped from virtual to physical addresses and the OSD    contains the code to maintain that map, including all the paging,H    swapping, remapping, and other support code.  That code needs to know=    almost everything about the hardware properties of memory.   B    The parts of the OS that don't maintain that map see contiguous@    locations in system space, and whatever locations the currentD    (possibly interrupted) process happens to have mapped in process F    space (OS routines are often entered via interrupt, which means the5    OS can't know which process happens to be mapped).   J > Or was MA780 memory more akin to the early PCs with segmented blocks etcH > etc and the OS needs to issue special assembler instructions to choose> > which bank of memory the next operation will be working on ?  H    That's segmentation.  VAX and Alpha never did segmentation, they only    did flat addressing.   F    VMS can work with physically non-contiguous memory spaces.  It justG    never maps virtual addresses to physical addresses that don't exist, ,    or are known not to be working correctly.  -    I/O space is almost always non-contiguous.    ------------------------------    Date: 29 Aug 2006 09:18:52 -0500 From: briggs@encompasserve.org/ Subject: Re: Locating early versions of VAX/VMS 3 Message-ID: <WdsHpUUgWb2p@eisner.encompasserve.org>   q In article <RKUN7ezKFwTr@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: ^ > In article <44F33A17.7402B9C1@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: >>  I >> Out of curiosity, at what level does an OS need to know about hardware I >> properties of memory ? Doesn't it just see a set of continugous memory E >> locations that contain data accessed transparently via the various 0 >> assembly opcodes that provide memory access ? > H >    No.  Memory is mapped from virtual to physical addresses and the OSF >    contains the code to maintain that map, including all the paging,J >    swapping, remapping, and other support code.  That code needs to know? >    almost everything about the hardware properties of memory.   C The VAX page table layout is architecturally specified.  One can do D paging, swapping and remapping without any knowledge of the physical$ properties of the underlying memory.  A Once you've figured out how much memory is present, what physical A addresses ranges are available and turned interleaving on or off, F there's not a whole lot for the O/S to do in terms of paging, swappingD and memory management that's not covered by the architecture manual.  A I/O using DMA...  you may need some details there.  I don't think H the Unibus mapping registers are covered in the VAX architecture manual.  A And cache coherency...  You may need to interact with that as you F switch process context or update page table entries.  I don't rememberG to what extent that was architecturally specified.  And you may want to @ tune your page replacement strategy using things like TBSKIPWSL.  @ Still, the devil is in the details.  If Hoff says that the MA780@ behavior is critical to the correct operation of VMS 1.0 then he% surely knows what he's talking about.    ------------------------------  % Date: Tue, 29 Aug 2006 11:38:15 -0400 , From: Hoff Hoffman <hoff-remove-this@hp.com>/ Subject: Re: Locating early versions of VAX/VMS * Message-ID: <44f45f68@usenet01.boi.hp.com>   davidc@montagar.com wrote: > Hoff Hoffman wrote: R >>    We looked at booting a (real) VAX-11/780 under V1.0 as part of a trade show,R >> and ended up scrapping the idea because of the hardware requirements.  It's notO >> just a VAX-11/780, it's a very specific set of VAX-11/780 storage and system S >> hardware configurations that are required.  There was no way we were going to be L >> able to locate the appropriate (and working) hardware for the trade show. > C > But didn't you do that at a long past DECUS event, in the OpenVMS D > campground?  Pat Jankowiak even provided a couple of pieces of theC > system.  Maybe it was V1.5, but I though it might have been V1.0.     Q    We did scrounge pieces (and it would not surprise me that Pat J. was involved  Q in that, as he very likely was), but we couldn't scrounge enough parts that were  M old enough.  (And some individual scrapped a reference system we'd carefully  O tagged and set aside, and without telling anyone.  "Hey, it's big and it's old  N -- we'll ignore the signs on the boxes and just crush it!"  Grumble.)  We did J end up showing a (fully working) VAX-11/780, but it had a couple of minor O problems -- the DECtape controller was incompatible with (IIRC) the DEUNA, and  P there were concerns with the disk (and we obviously didn't want to incur a head Q crash), and the memory we had wasn't old enough for VAX[-11]/VMS V1.0.  (Neither   was the DEUNA, obviously.)  Q    And yes, it was several years back now -- but it's not like there's some tree  N growing old VAX-11/780 memory controllers somewhere, and we just have to wait  for another one to sprout...  P    Head crashes on newer disks just aren't as interesting as those of old, back ) when the disks involved had serious mass.    ------------------------------  # Date: Wed, 30 Aug 2006 02:33:29 GMT + From: "Villy Madsen" <Villy.Madsen@shaw.ca> / Subject: Re: Locating early versions of VAX/VMS . Message-ID: <ZN6Jg.484148$iF6.469899@pd7tw2no>  : "Hoff Hoffman" <hoff-remove-this@hp.com> wrote in message $ news:44f45f68@usenet01.boi.hp.com... > davidc@montagar.com wrote: >> Hoff Hoffman wrote:H >>>    We looked at booting a (real) VAX-11/780 under V1.0 as part of a  >>> trade show, J >>> and ended up scrapping the idea because of the hardware requirements.  >>> It's notJ >>> just a VAX-11/780, it's a very specific set of VAX-11/780 storage and 
 >>> systemI >>> hardware configurations that are required.  There was no way we were   >>> going to be H >>> able to locate the appropriate (and working) hardware for the trade 	 >>> show.  >>D >> But didn't you do that at a long past DECUS event, in the OpenVMSE >> campground?  Pat Jankowiak even provided a couple of pieces of the D >> system.  Maybe it was V1.5, but I though it might have been V1.0. >  > I >   We did scrounge pieces (and it would not surprise me that Pat J. was  L > involved in that, as he very likely was), but we couldn't scrounge enough I > parts that were old enough.  (And some individual scrapped a reference  J > system we'd carefully tagged and set aside, and without telling anyone. I > "Hey, it's big and it's old -- we'll ignore the signs on the boxes and  F > just crush it!"  Grumble.)  We did end up showing a (fully working) C > VAX-11/780, but it had a couple of minor problems -- the DECtape  M > controller was incompatible with (IIRC) the DEUNA, and there were concerns  J > with the disk (and we obviously didn't want to incur a head crash), and K > the memory we had wasn't old enough for VAX[-11]/VMS V1.0.  (Neither was   > the DEUNA, obviously.) > M >   And yes, it was several years back now -- but it's not like there's some  M > tree growing old VAX-11/780 memory controllers somewhere, and we just have  & > to wait for another one to sprout... > L >   Head crashes on newer disks just aren't as interesting as those of old, 0 > back when the disks involved had serious mass. >  >   H I remember once when we had a head crash on an 11/70.  I think that the  drives were rp06's.   J anyway,   the operator had to swap packs - remember those bad old days ?? F He mounted the new pack - and he couldn't read it.  So he moved it to M another drive - couldn't read it there either, so he mounted another pack on  K the drive - and when he couldn't read that, moved that other pack to.,,,,,,   L once all was said & done, I think there were 5 packs gone & 6 sets of heads.  M We looked at one of the packs & it was kind of neat.  The top head crashed &  I as the crud migrated down the platters - gradually taking out all of the  	 heads....   J Not oxide left at all on the top surface (well the bottom of the top) and * the aluminum itself was nicely chewed.....  I I can't remember what happened to the operator, I do remember that there  E weren't enough head sets available in Alberta to repair the damage...    Villy    ------------------------------    Date: 30 Aug 2006 07:30:22 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) / Subject: Re: Locating early versions of VAX/VMS 3 Message-ID: <VaCv7Ls8bt+t@eisner.encompasserve.org>   Y In article <44f45f68@usenet01.boi.hp.com>, Hoff Hoffman <hoff-remove-this@hp.com> writes:  > S >    And yes, it was several years back now -- but it's not like there's some tree  P > growing old VAX-11/780 memory controllers somewhere, and we just have to wait  > for another one to sprout...  C    IIRC, I saw pictures of the "first 11/780" running at a national G    event that I didn't get to several years ago.  It was all spread out F    on a bench, not in the nice blue cabinets production models shipped    in.  H    Unfortunately I understand Compaq tossed that, but I suspect it could@    have been running 1.0, or pre-1.0, if it really was the firstC    hardware.  That may have been confused with a later display of a *    production 11/780 at a different event.   ------------------------------  % Date: Sun, 27 Aug 2006 13:08:57 -0400 - From: "John E. Malmberg" <wb8tyw@qsl.network> > Subject: MicroVAX I / VT180 / LSI-11/23 / LA100 in Miford, NH.; Message-ID: <xN-dndoC_c4xTGzZnZ2dnUVZ_tKdnZ2d@adelphia.com>    Basement cleaning time.   < For the first person that can pick these up from Milford NH:  = LA100 with stand, and some spare ribbons in sealed container.   B VT-180 with some software, 4 floppy drives, and some spare parts. 2 Software appears to be CP/M and some applications.  H LSI-11/23 in narrow pedestal mount.  I never powered it up, so I do not 4 know if it works.  It appears to have a deqna in it.  F Microvax I, was formerly a VAXstation I, however the smoke got out of ? the graphics card.  Then the RD53 died and I never installed a  F replacement RD54.  Dead graphics card set and monitor also available. H Delqua and replacement RD54s (unknown condition) are included.  This is I also in the narrow pedistal.  It has a RX50 that is appears bad.  I have  H two other RX50s drives that are of unknown condition, and pre-formatted < floppies for it.  I think I have an RQDX3 controller for it.  ; De-mung my e-mail address to contact me to arrange pick up.    -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------    Date: 30 Aug 2006 12:57:33 -0700$ From: "AEF" <spamsink2001@yahoo.com>5 Subject: Re: Mystery multiple TCP connections dropped C Message-ID: <1156967853.575565.161870@b28g2000cwb.googlegroups.com>    JF Mezei wrote:  > AEF wrote:C > > Being that this happened on two MicroVAX systems, three Windows E > > multifeed systems, with all six connections between them dropping D > > within a tenth of a second of each other, I can't see any of theJ > > systems being at fault. It must be some external entity. But why these$ > > two pages and not the other two? >  > I > If pages 1 and 2 are sent at 08:00 and pages 3 and 4 are sent at 08:01, I > and a problem happens between 07:59 and 08:00:58, then failure would be J > noticed for pages 1 and 2, but by the time the systems try to send pagesA > 3 and 4, the link si re-established and those attempts succeed.   ? But this all happened withing the same 0.1-second period in two C separate incidents! (+/- any NTP-error time differences between the 3 system times of the two involved MicroVAX systems).   E > From the application point of view, you need to look at how the app H > reacts to essentially the ethernet plug being pulled from the machine.  ? The same as when the problem happened, though there is a 5 min. D time-out between when the plug is pulled and when the error happens.B The 0.1-second time windows occuring on two separate days makes me# doubt that a time-out was involved.   J > (equivalent of a switch going down etc etc). Does the app rely on TNxxx:? > devices or are they simple GB devices ? How is the TCPIP link : > established by the application ? What options are used ?  F This appears to be done in the Pascal code so I'll report back after I7 talk to our developer who is far more familiar with it.   J > You may wish to look at your TCPIP stack's documemtation to see if thereJ > are logicals you xcan set to lengthen any timeout values for the type ofH > connexction you are using, which would allow your apps to survive some" > sporadic switch/router problems.  C Unless both pages were sending time-stamps withing 0.1 sec. of each E other, it looks like a bona fide close connection command occurred at E the multifeed end, except how can three separate boxes conspire to do F that to the same two ports withing 0.1 sec. of each other? The timeout* is already 5 min. for a broken connection.   Thanks for your efforts.   AEF    ------------------------------    Date: 30 Aug 2006 13:39:13 -0700$ From: "AEF" <spamsink2001@yahoo.com>5 Subject: Re: Mystery multiple TCP connections dropped B Message-ID: <1156970353.528136.80440@h48g2000cwc.googlegroups.com>  
 AEF wrote: > JF Mezei wrote:  > > AEF wrote:E > > > Being that this happened on two MicroVAX systems, three Windows G > > > multifeed systems, with all six connections between them dropping F > > > within a tenth of a second of each other, I can't see any of theL > > > systems being at fault. It must be some external entity. But why these& > > > two pages and not the other two? > >  > > K > > If pages 1 and 2 are sent at 08:00 and pages 3 and 4 are sent at 08:01, K > > and a problem happens between 07:59 and 08:00:58, then failure would be L > > noticed for pages 1 and 2, but by the time the systems try to send pagesC > > 3 and 4, the link si re-established and those attempts succeed.  > A > But this all happened withing the same 0.1-second period in two E > separate incidents! (+/- any NTP-error time differences between the 5 > system times of the two involved MicroVAX systems).  > G > > From the application point of view, you need to look at how the app J > > reacts to essentially the ethernet plug being pulled from the machine. > A > The same as when the problem happened, though there is a 5 min. F > time-out between when the plug is pulled and when the error happens.D > The 0.1-second time windows occuring on two separate days makes me% > doubt that a time-out was involved.  > L > > (equivalent of a switch going down etc etc). Does the app rely on TNxxx:A > > devices or are they simple GB devices ? How is the TCPIP link < > > established by the application ? What options are used ? > H > This appears to be done in the Pascal code so I'll report back after I9 > talk to our developer who is far more familiar with it.   F He says he just gives it the name of the driver and uses some standard call.    > L > > You may wish to look at your TCPIP stack's documemtation to see if thereL > > are logicals you xcan set to lengthen any timeout values for the type ofJ > > connexction you are using, which would allow your apps to survive some$ > > sporadic switch/router problems. > E > Unless both pages were sending time-stamps withing 0.1 sec. of each G > other, it looks like a bona fide close connection command occurred at G > the multifeed end, except how can three separate boxes conspire to do H > that to the same two ports withing 0.1 sec. of each other? The timeout, > is already 5 min. for a broken connection. >  > Thanks for your efforts. >  > AEF    ------------------------------    Date: 31 Aug 2006 09:16:00 -0700- From: "Doug Phillips" <dphill46@netscape.net> 5 Subject: Re: Mystery multiple TCP connections dropped C Message-ID: <1157040960.120123.222490@m79g2000cwm.googlegroups.com>   
 AEF wrote: > JF Mezei wrote:  > > AEF wrote:E > > > Being that this happened on two MicroVAX systems, three Windows G > > > multifeed systems, with all six connections between them dropping F > > > within a tenth of a second of each other, I can't see any of theL > > > systems being at fault. It must be some external entity. But why these& > > > two pages and not the other two? > >  > > K > > If pages 1 and 2 are sent at 08:00 and pages 3 and 4 are sent at 08:01, K > > and a problem happens between 07:59 and 08:00:58, then failure would be L > > noticed for pages 1 and 2, but by the time the systems try to send pagesC > > 3 and 4, the link si re-established and those attempts succeed.  > A > But this all happened withing the same 0.1-second period in two E > separate incidents! (+/- any NTP-error time differences between the 5 > system times of the two involved MicroVAX systems).  > G > > From the application point of view, you need to look at how the app J > > reacts to essentially the ethernet plug being pulled from the machine. > A > The same as when the problem happened, though there is a 5 min. F > time-out between when the plug is pulled and when the error happens.D > The 0.1-second time windows occuring on two separate days makes me% > doubt that a time-out was involved.  > L > > (equivalent of a switch going down etc etc). Does the app rely on TNxxx:A > > devices or are they simple GB devices ? How is the TCPIP link < > > established by the application ? What options are used ? > H > This appears to be done in the Pascal code so I'll report back after I9 > talk to our developer who is far more familiar with it.  > L > > You may wish to look at your TCPIP stack's documemtation to see if thereL > > are logicals you xcan set to lengthen any timeout values for the type ofJ > > connexction you are using, which would allow your apps to survive some$ > > sporadic switch/router problems. > E > Unless both pages were sending time-stamps withing 0.1 sec. of each G > other, it looks like a bona fide close connection command occurred at G > the multifeed end, except how can three separate boxes conspire to do H > that to the same two ports withing 0.1 sec. of each other? The timeout, > is already 5 min. for a broken connection. >   C Maybe a restatement of the problem would help. Is this about right?   2  - There are four MicroVax systems (MV) in London.E  - There are four MultiFeed systems (MF); one in London, three in NY. 8  - Each MV has a TELNET connection open to all four MFs.@  - On Tuesday morning two of the MVs lost communication with NY.)  - All other connections remained intact. 1  - Restarting the application restored the links. F  - On Thursday morning this happened again, within an hour of the same time as on Tuesday.    If so:  : First: Why did only those two MVs lose connection with NY? Second: What caused the loss?   E Maybe you'll never know the second, but if you identify the first you 2 can take steps to prevent it from happening again.  G As I suggested before; these two systems probably share something along > the link between their London location and NY that you haven'tF identified. Time to pull out (or create) your infrastructure diagrams,G validate all the boxes, trace the lines and lightning bolts and explore  any clouds.   A Increasing some timeouts might just mask the problem for a while.    ------------------------------    Date: 29 Aug 2006 17:50:04 -0700) From: "Sue" <susan_skonetski@hotmail.com> K Subject: NEW Partner application System to Map-Document Search Applications C Message-ID: <1156899004.511742.293510@i42g2000cwa.googlegroups.com>   ) Dear Distribution lists (and Newsgroups),   D It is very rare that I send out partner application information as a> Pearl but this application is very interesting.  This is a NEWC application for OpenVMS and is available on NonStop, UNIX, Java and @ others.  I thought that it would be extremely useful for several reasons;  G 1=2E considering the size of some of our customers it will increase the ? search speed. 2. the fact that most of you are currently moving 7 platforms 3. clusters 4. I love NEW applications on VMS   % Please take a moment to look at this.    OK to post.    Have a great day,    Sue       K =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3DPANORAMA INFORMATION=3D=3D=3D=3D=3D=3D= ' =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D    NEW FOR OPENVMS   @ ITP-PANORAMA: The only OpenVMS Application Information System To  Map-Document-Search Applications9 PANORAMA info: http://www.emperex.com/panorama/index.html 6 Contact info: www.emperex.com or freedland@emperex.com  D PANORAMA helps organizations gain visibility into their applicationsG across the enterprise, enables sub-second searches, and documents every @ element within applications on many platforms including OpenVMS, NonStop, UNIX, Java and others.   E PANORAMA will not impact the development schedule in any way, because  it is ready-to-use in one day.  
 Relevant Uses ' =B7Day-to-day search work by developers 0 =B7Transition projects (such as move to Itanium)! =B7Staff learning new application  =B7Verifying outsourced work =B7Documenting old applications    Productivity Benefits:  =B7Search time reduced by 90-95%( =B7Accelerate application time-to-market8 -Staff learn unfamiliar applications in hours not months. =B7Outsourced work can be verified immediately< =B7Ready-to-use in one day (incl. installation and training)( =B7Dramatic reduction in Mainframe usage0 =B7Projects done in 2/3 of the time (33% faster)$ =B7Relationship bugs reduced to zero5 =B7Instant documentation for old and new applications   L ---------------------------------------------------------------------------=, -------------------------------------------- Useful Background Links  Management Summary info at: : http://www.itp-panorama.com/pdf/management_summary_eng.pdf Technical Summary at: 3 http://www.itp-panorama.com/pdf/at_a_glance_eng.pdf - Technical Tutorials with real examples is at: . http://www.itp-panorama.com/download_e_vp.php4   Example Features: E - go down to the very detail of every line of code of any application  scanned 6 - shows all errors or missing pieces of an application3 - produce customized reports and detailed documents C - handle heterogeneous IT environments (OpenVMS,  HP NonStop, UNIX,  Java, and more)   - support hundreds of developers - read tens of mill. of LoC B - scan, compile, and cross-check 20 mill. LoC in less then 90 min. - sub-second response time  K Example Savings From Fiducia Bank in Europe =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=  =3D=3D=3D=3D=3D   = Search Type     Typical Effort  PANORAMA Effort Occurrence/yr 5 Simple  2hours           2 minutes                255 6 Medium  16hrs            5 min                     144> Complex          32hrs            2 hrs                     50  + Search time reduced from 4400 to 120 hours. 7 SAVINGS: 4,280 hours saved at $90/hr cost  =3D $385,200 L =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=L =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D      web site: www.emperex.com   ------------------------------    Date: 28 Aug 2006 07:55:56 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett) 4 Subject: Re: No more Oracle Standard Edition for VMS, Message-ID: <PmoAQcz7FU1L@malvm9.mala.bc.ca>  % In article <ecumc3$cs6$1@online.de>,  V     helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:   > = > How much money and work would it be for you to move to Rdb?  >   E     Rdb costs the same as Oracle Enterprise Edition, so there'd be no 2 direct financial benefit to switching back to Rdb.  D     When I stopped running Rdb the SQLNet for Rdb product was prettyF new and had some glitches in it, but assuming they're ironed out now ID could probably convert to Rdb without too much technical difficulty.? Unfortunately it would be cheaper and easier to move the Oracle ; Standard Edition database to a Unix/Linux platform. While I 9 believe that VMS offers an excellent platform to host the < database I doubt I could justify the cost of staying with it: if that means having to move to Enterprise Edition or Rdb.  8     Just another "un-leveling" of the VMS playing field.   ------------------------------    Date: 28 Aug 2006 11:42:34 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> 4 Subject: Re: No more Oracle Standard Edition for VMSB Message-ID: <1156790554.917522.119090@75g2000cwc.googlegroups.com>   Malcolm Dunnett wrote:. > In article <44EC91D1.41927978@teksavvy.com>,4 >    JF Mezei <jfmezei.spamnot@teksavvy.com> writes: >  > > K > > If VMS management do not do something to keep Oracle on VMS and in fact K > > grow its presence on VMS, I think it may in fact signal the true end of & > > VMS. It will send a strong signal. > >  > I >    Well Oracle will be keeping Enterprise Edition on VMS ( at least for D > the time being ), although as usual the port lags far behind other > platforms. > K > > Remember that there is that suppposed 10 billion buck fund to help port J > > software. If HP isn't going to use any of it to help VMS, then it alsoJ > > sends quite a strong signal about HP's intentions with reagrds to VMS. > D >    The fascinating thing is that there is an Oracle 10gR2 StandardE > Edition for Tru64 Unix. So an O/S that HP has declared as dead gets 9 > a port but an O/S HP says has a future doesn't. Hmmm...   E Not only that, but gets released first.  As of 18-Aug-2006 there is a G free download of Oracle 10gR2 for Tru-64 while the VMS version is still   not available as of 28-Aug-2006.   ------------------------------  % Date: Mon, 28 Aug 2006 16:41:22 -0400 ' From: Dave Froble <davef@tsoft-inc.com> 4 Subject: Re: No more Oracle Standard Edition for VMS9 Message-ID: <o92dnYifoavTzm7ZnZ2dnUVZ_o-dnZ2d@libcom.com>   / Phillip Helbig---remove CLOTHES to reply wrote: H > In article <u7v6hp3RVqqH@malvm9.mala.bc.ca>, nothome@spammers.are.scum > (Malcolm Dunnett) writes:  > ; >> Perhaps I'm missing something, but my experience is that 9 >> unless you're running huge databases where things like 8 >> partitioning become critical there's not much need to6 >> use Enterprise Edition and that not having Standard: >> edition available for VMS eliminates the feasibility of; >> being able to use this rock-solid OS as an Oracle server . >> platform unless you have very deep pockets. > = > How much money and work would it be for you to move to Rdb?  >   H I believe that earlier Malcolm state that they moved from RDB to Oracle D Standard to save money.  That Oracle charges the Enterprise Edition  prices for RDB.   H I believe that selling RDB was one of DEC's biggest mistakes.  With the I prices DEC had for RDB, they could be very price competitive with Oracle  6 on other platforms, or, force Oracle into a price war.  F DEC didn't need Oracle classic on VMS nearly as much as Oracle needed B the price of RDB raised, by a factor of 5 if I remember correctly.   --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------    Date: 28 Aug 2006 14:32:50 -0700( From: "joel garry" <joel-garry@home.com>4 Subject: Re: No more Oracle Standard Edition for VMSB Message-ID: <1156800770.469469.28020@i42g2000cwa.googlegroups.com>   Dave Froble wrote:1 > Phillip Helbig---remove CLOTHES to reply wrote: J > > In article <u7v6hp3RVqqH@malvm9.mala.bc.ca>, nothome@spammers.are.scum > > (Malcolm Dunnett) writes:  > > = > >> Perhaps I'm missing something, but my experience is that ; > >> unless you're running huge databases where things like : > >> partitioning become critical there's not much need to8 > >> use Enterprise Edition and that not having Standard< > >> edition available for VMS eliminates the feasibility of= > >> being able to use this rock-solid OS as an Oracle server 0 > >> platform unless you have very deep pockets. > > ? > > How much money and work would it be for you to move to Rdb?  > >  > I > I believe that earlier Malcolm state that they moved from RDB to Oracle E > Standard to save money.  That Oracle charges the Enterprise Edition  > prices for RDB.  > I > I believe that selling RDB was one of DEC's biggest mistakes.  With the J > prices DEC had for RDB, they could be very price competitive with Oracle8 > on other platforms, or, force Oracle into a price war.  @ At the time, DEC was already going down.  It was selling off theE software pieces ("non-strategic operations") in a last-gasp effort to G remain in the hardware business (or perhaps have a marketable entity to ? sell to another clueless hardware maker, to put a more positive F speculative spin on their strategic thinking).  I think it was less of? a DEC mistake than a Larry coup.  DEC had already made too many F mistakes.  'Rdb business has been in a "modest decline,"...The deal isE an example of DEC's previously announced strategy of focusing on core  competencies':O http://calbears.findarticles.com/p/articles/mi_m0NEW/is_1994_Sept_2/ai_16229440    > G > DEC didn't need Oracle classic on VMS nearly as much as Oracle needed D > the price of RDB raised, by a factor of 5 if I remember correctly.  F Phillipe Kahn was already saying "$100/seat" for any software, an idea4 that many found upsetting or ridiculous at the time. http://groups.google.com/group/comp.os.linux.development.system/browse_thread/thread/98907520d6051f6d/ea052004b52e371f?lnk=st&q=%22phillipe+Kahn%22+author%3Agarry&rnum=2&hl=en#ea052004b52e371f   jg -- @home.com is bogus.  Mumps was next: N http://calbears.findarticles.com/p/articles/mi_m0NEW/is_1995_Jan_5/ai_16326089   ------------------------------  + Date: Tue, 29 Aug 2006 09:11:49 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)4 Subject: Re: No more Oracle Standard Edition for VMS$ Message-ID: <ed10cl$96v$4@online.de>  E In article <o92dnYifoavTzm7ZnZ2dnUVZ_o-dnZ2d@libcom.com>, Dave Froble  <davef@tsoft-inc.com> writes:   J > I believe that selling RDB was one of DEC's biggest mistakes.  With the K > prices DEC had for RDB, they could be very price competitive with Oracle  8 > on other platforms, or, force Oracle into a price war. > H > DEC didn't need Oracle classic on VMS nearly as much as Oracle needed D > the price of RDB raised, by a factor of 5 if I remember correctly.  G I think the argument was that if Rdb remained with DEC, Oracle wouldn't G offer Oracle for VMS, giving people (who wanted to run Oracle) a reason G not to use VMS.  Whether would have done more harm than good (to DEC) I  don't know.   H For what it's worth, Rdb continues to be developed, enhanced, have bugs  fixed, new features added etc.   ------------------------------    Date: 31 Aug 2006 09:02:00 -0700" From: dave.baxter@bannerhealth.com4 Subject: Re: No more Oracle Standard Edition for VMSB Message-ID: <1157040120.307732.52930@m73g2000cwd.googlegroups.com>   Sybrand Bakker wrote: H > On Wed, 23 Aug 2006 19:46:27 GMT, winston@SSRL.SLAC.STANFORD.EDU (Alan* > Winston - SSRL Central Computing) wrote: > Q > >Bob, I think your inflammatory accusation of "a lie" is based on a misreading. G > >I belive Sybrand is saying that he won't be surprised if _Oracle_ de  > & > Which was exactly what I was saying.- > Who is using Oracle on OpenVMS? Pretty few.   F This may be the perception of many who operate in the education field,D however I can assure the poster that the number of companies who are/ running Oracle on OpenVMS is NOT insignificant.   E You need to do some research before making these statements.      The E advent of the Electronic Medical Record (EMR) and HIPAA, require that G Healthcare Providers keep their medical records available ALL THE TIME. G     That means NO DOWNTIME for maintenance, NO DOWNTIME for Backups, NO F DOWNTIME for Hardware Failure, etc., etc.     Do you notice the trend, "NO DOWNTIME".  G Now Oracle RAC is very nice for failover, and works well, however there D is more to "NO DOWNTIME" than RAC failover.       The Big HealthcareC providers and Financial institutions realize that, an that probably B explains why such a large proportion of them run on OpenVMS, (even, though the have the option to run on **NIX).  B > And the list of Oracle 9i EE options NOT available on OpenVMS is > pretty long.  " If you don't need them, who cares.  = > What is more: Oracle recommends replacing OpenVMS by Linux.   0 > The future of Oracle on OpenVMS is pretty dim.  G You need to remove the Shades.       Cerner Corp is probably one of the D largest and most successful Medical/Clinical Application provider inG the US, (if not the world), with thousands of clients world-wide, and a F large percentage of their clients (world-wide, not just in the US) areB fully committed to Cerner's App on OpenVMS.      And these clients? constitute a huge financial investment in both the Apps and the 	 Hardware.   E I think the the influence of HP and such large corporations as Cerner C should not be ignored.      For example, with Cerner, the alternate E path for clients would not be Oracle on some other platform, it would B be DB2 on AIX, or something like.      So do you think that OracleG would be happy with the prospect of losing such a large customer base??   6 And I am only talking about one company, in one arena.   Anyway, just my opinion.         Dave   ------------------------------    Date: 29 Aug 2006 17:32:19 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> 2 Subject: OpenVMS Integrity Graphics card questionsC Message-ID: <1156897939.666811.157390@b28g2000cwb.googlegroups.com>   E I can't seem to find the discussions at the moment but it seems to me C that it's been said that OpenVMS Integrity will not support the DVI C port on an ATI Radeon 7000 graphics card.  I also think it was said G that it won't support a second head on the same card ( Mine has one VGA & and one DVI).  Can anyone verify this?  B If you get an Integrity system without the Management Processor, IB assume you can put a Radeon in one of the PCI slots.  Will OpenVMSC Integrity support anything besides the 7000?  Doesn't the 7500 also  work?   F I've read that the AGP slot (on the otherwise unsupported ZX6000) willC not work and I'm pretty sure it has been said that it won't be.  Is A there any work planned for any kind of graphics solutions for the ; Integrity systems in VMS (Other than the Radeon 7000/7500)?    Thanks for your time.      John H. Reinhardt    ------------------------------  # Date: Wed, 30 Aug 2006 12:19:07 GMT 5 From: rdeininger@mindspringdot.com (Robert Deininger) 6 Subject: Re: OpenVMS Integrity Graphics card questions[ Message-ID: <rdeininger-3008060818570001@dialup-4.233.173.154.dial1.manchester1.level3.net>   C In article <1156897939.666811.157390@b28g2000cwb.googlegroups.com>, < "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> wrote:  : Fred may chime in with more details if he sees the thread.  F >I can't seem to find the discussions at the moment but it seems to meD >that it's been said that OpenVMS Integrity will not support the DVID >port on an ATI Radeon 7000 graphics card.  I also think it was saidH >that it won't support a second head on the same card ( Mine has one VGA' >and one DVI).  Can anyone verify this?   J VMS supports the Radeon 7500 PCI plug-in card on the entry-level IntegrityI servers.  I don't recall the product number for the card, but it's in the D configuration guides and QuickSpecs for the systems.  It is the sameI hardware supported on various recent Alpha systems, but the order number, 7 packaging and accompanying documentation are different.   C VMS only supports 1 head per Radeon 7500 card.  You can use the VGA I output.  I think, but I'm not positive, that you can use the DVI output.   You CAN'T use both.   K HP-UX supports the same card, on roughly the same set of Integrity servers.   H If the available configuration documents don't answer these questions toG your satisfaction, the documents are due for a round of repairs.  There J are feedback mechanisms provided on the web, and they are even supposed to be working.   G Radeon 7000 seems to be one of the names used for the built-in graphics G chip used in the Management Processor.  The chip implements a subset of J the Radeon 7500.  VMS supports graphics on the MP of the Integity servers,J but only a "2D" subset of the Radeon 7500 functionality.  I don't know the+ details well enough to describe the subset.   E You can use the Radon 7500 with or without the MP graphics on systems  where MP is optional.   C >If you get an Integrity system without the Management Processor, I C >assume you can put a Radeon in one of the PCI slots.  Will OpenVMS D >Integrity support anything besides the 7000?  Doesn't the 7500 also >work?  J VMS doesn't support a "Radeon 7000" plug-in card, if such a thing exists. J If you find one, VMS might recongnize it and it might "work", but it isn't
 supported.  G >I've read that the AGP slot (on the otherwise unsupported ZX6000) will D >not work and I'm pretty sure it has been said that it won't be.  IsB >there any work planned for any kind of graphics solutions for the< >Integrity systems in VMS (Other than the Radeon 7000/7500)?  C I don't think the AGP graphics card was even tried with VMS on this I system.  VMS doesn't support the system (which has been out of production I for a while), and the industry is moving away from AGP as far as I know.  ? I don't expect any additional VMS work on AGP graphics support.   C Radeon 7500 is likely to be the only add-on graphics solution until E PCI-express ships and matures somewhat on Integrity servers.  There's I currently no timetable for possible support of a future PCI-express card, ( but I believe folks are looking into it.   ------------------------------    Date: 30 Aug 2006 08:13:33 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> 6 Subject: Re: OpenVMS Integrity Graphics card questionsB Message-ID: <1156950813.681613.231440@i3g2000cwc.googlegroups.com>   Robert Deininger wrote: E > In article <1156897939.666811.157390@b28g2000cwb.googlegroups.com>, > > "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> wrote: > < > Fred may chime in with more details if he sees the thread. > H > >I can't seem to find the discussions at the moment but it seems to meF > >that it's been said that OpenVMS Integrity will not support the DVIF > >port on an ATI Radeon 7000 graphics card.  I also think it was saidJ > >that it won't support a second head on the same card ( Mine has one VGA) > >and one DVI).  Can anyone verify this?  > L > VMS supports the Radeon 7500 PCI plug-in card on the entry-level IntegrityK > servers.  I don't recall the product number for the card, but it's in the F > configuration guides and QuickSpecs for the systems.  It is the sameK > hardware supported on various recent Alpha systems, but the order number, 9 > packaging and accompanying documentation are different.  >   D Since this is a hobbyist endeavor I'm planning on using a retail ATIG Radeon 7500 I picked up on Ebay (I picked up 3 new, sealed, in a retail E box for $35 each plus shipping - they also work great in the XP1000). G The ZX6000 workstation I aquired has an NVidia Quadro2 Pro AGP card and 7 no MP so I was looking into OpenVMS compatible devices.   E > VMS only supports 1 head per Radeon 7500 card.  You can use the VGA J > output.  I think, but I'm not positive, that you can use the DVI output. > You CAN'T use both.   E Unfortunately the 7500's I have only have the VGA (plus S-Video and a F TV-tuner).   All of the 7500's I've found with DVI are AGP rather than PCI.   > M > HP-UX supports the same card, on roughly the same set of Integrity servers.  > J > If the available configuration documents don't answer these questions toI > your satisfaction, the documents are due for a round of repairs.  There L > are feedback mechanisms provided on the web, and they are even supposed to
 > be working.  >   G I've read the Install guide for the AGP Radeon 7500 (which is available C on the HP website) and gotten the PCI version to work in my XP1000. B Since this wandering to the Integrity side is all on non-supportedA hardware and hobbyist related, the range of knowledge required is : somewhat outside that available from the HP documentation.  I > Radeon 7000 seems to be one of the names used for the built-in graphics I > chip used in the Management Processor.  The chip implements a subset of L > the Radeon 7500.  VMS supports graphics on the MP of the Integity servers,L > but only a "2D" subset of the Radeon 7500 functionality.  I don't know the- > details well enough to describe the subset.  >   C I have a (two, actually) OEM PC Radeon 7000 with the dual output (1 D VGA, 1 DVI).  Since I want to use this system as a master controllerG for my other VMS sytems and anticipate using DECwindows extensively I'd E like to be able to use it with my 20" LCD and the DVI input.  Since I G can't find any PCI bus 7500's with DVI I may try the 7000 to see if the G DVI output  will work.  I'd rather use a 7500, but since I won't really E be doing much 3D work it may not perform too badly (assuming it works  at all).  G > You can use the Radon 7500 with or without the MP graphics on systems  > where MP is optional.  > E > >If you get an Integrity system without the Management Processor, I E > >assume you can put a Radeon in one of the PCI slots.  Will OpenVMS F > >Integrity support anything besides the 7000?  Doesn't the 7500 also > >work? > K > VMS doesn't support a "Radeon 7000" plug-in card, if such a thing exists. L > If you find one, VMS might recongnize it and it might "work", but it isn't > supported. >   F Okay. I was making the assumption (perhaps a false one) that since theF MP card has a Radeon 7000 equivalent built-in that a separate PCI card+ would work also.  Time for experimentation!   I > >I've read that the AGP slot (on the otherwise unsupported ZX6000) will F > >not work and I'm pretty sure it has been said that it won't be.  IsD > >there any work planned for any kind of graphics solutions for the> > >Integrity systems in VMS (Other than the Radeon 7000/7500)? > E > I don't think the AGP graphics card was even tried with VMS on this K > system.  VMS doesn't support the system (which has been out of production J > for a while), and the industry is moving away from AGP as far as I know.A > I don't expect any additional VMS work on AGP graphics support.  > E > Radeon 7500 is likely to be the only add-on graphics solution until G > PCI-express ships and matures somewhat on Integrity servers.  There's K > currently no timetable for possible support of a future PCI-express card, * > but I believe folks are looking into it.  0 Thanks for all your help and information Robert.     John H. Reinhardt    ------------------------------  % Date: Wed, 30 Aug 2006 14:19:04 -0400 , From: Hoff Hoffman <hoff-remove-this@hp.com>6 Subject: Re: OpenVMS Integrity Graphics card questions* Message-ID: <44f5d697@usenet01.boi.hp.com>   johnhreinhardt@yahoo.com wrote: G > I can't seem to find the discussions at the moment but it seems to me E > that it's been said that OpenVMS Integrity will not support the DVI E > port on an ATI Radeon 7000 graphics card.  I also think it was said I > that it won't support a second head on the same card ( Mine has one VGA ( > and one DVI).  Can anyone verify this?  M    The Radeon 700 PCI cards are not supported, though I've seen reports they  L work on some Alpha boxes -- the particular one I've seen has had some minor P display artifact.  The supported cards for Integrity are the HP Radeon 7500 PCI  cards.  Q    The VGA port is the usual connection, though IIRC, the DVI can potentially be  Q gotten to work in some cases.  (The one I have seen uses the VGA path into an HP  O TFT-class LCD display, though this was on an AlphaStation series and not on an   Integrity server.)  G    Off the top, I don't know that the DVI port is officially supported.   N    NB: there are a bunch of different Radeon 7x00 cards around, and obviously D for AGP and PCI, and obviously from various hardware vendors.  YMMV.    D > If you get an Integrity system without the Management Processor, ID > assume you can put a Radeon in one of the PCI slots.  Will OpenVMSE > Integrity support anything besides the 7000?  Doesn't the 7500 also  > work?   9    The HP Radeon 7500 PCI is the supported graphics card.   K    The Management Processor (MP) is very useful for reasons other than the  P embedded Radeon 7000 graphics, and it's certainly the way I'd go here -- and it ) also obviously doesn't occupy a PCI slot.   H > I've read that the AGP slot (on the otherwise unsupported ZX6000) willC > not work and I'm pretty sure it has been said that it won't be.     N    OpenVMS has no support for the AGP on Integrity, as there are no supported M Integrity boxes with AGP.  (The zx2000 and zx6000 systems are not officially  O supported as they entirely pre-date the OpenVMS port to Integrity, but some of  Q these configurations have been seen to work.  But not the AGP.)  And the bulk of  P the graphics universe is rapidly moving off AGP, and typically over to PCIe 16x  slots.  F > Is there any work planned for any kind of graphics solutions for the= > Integrity systems in VMS (Other than the Radeon 7000/7500)?   P    The current OpenVMS systems are unfortunately PCI only for graphics, and I'm M not in a position to discuss PCI-X or PCIe or graphics futures in general --  P beyond anything that might be included in the OpenVMS roadmap, obviously.  (And P I haven't looked at/for this detail in the current roadmap -- I'm downloading a ; copy of that document from the website right now, however.)   O    If you want support, you will want the HP Radeon 7500 PCI card, or you will  Q want to use the Radeon 7000 embedded in the Integrity Management Processor (MP).  Q   And you will likely want (should this be for use in an office environment) one  M of the so-called office-friendly (eg: quiet) rx2600 or rx2620 series systems.    ------------------------------  % Date: Wed, 30 Aug 2006 14:29:52 -0400 , From: Hoff Hoffman <hoff-remove-this@hp.com>6 Subject: Re: OpenVMS Integrity Graphics card questions* Message-ID: <44f5d91f@usenet01.boi.hp.com>   johnhreinhardt@yahoo.com wrote:   G > Unfortunately the 7500's I have only have the VGA (plus S-Video and a ? > TV-tuner).   All of the 7500's I've found with DVI are AGP...   M    Those would appear to be ATI All-In-Wonder Radeon 7500 cards, and not ATI   Radeon 7500 cards.  L    HP has a selection of HP Radeon 7500 cards available; the retail ATI PCI > versions of the module are now comparatively rarely available.   ------------------------------  # Date: Wed, 30 Aug 2006 21:46:57 GMT % From: Rick Jones <rick.jones2@hp.com> 6 Subject: Re: OpenVMS Integrity Graphics card questions. Message-ID: <lHnJg.19$wZ7.17@news.cpqcorp.net>  7 Peter 'EPLAN' LANGSTOEGER <peter@langstoeger.at> wrote: [ > In article <44f5d697@usenet01.boi.hp.com>, Hoff Hoffman <hoff-remove-this@hp.com> writes: ? >>And you will likely want (should this be for use in an office F >>environment) one of the so-called office-friendly (eg: quiet) rx2600 >>or rx2620 series systems.   F > Any reason why you choose the rx26x0 over the rx16x0 here? Or is theF > 16x0 in regard to the 26x0 too slow/rare/expensive to be considered?  D There simply isn't an OFS version of an rx16x0.  The rx16x0 hardwareD is otherwise performance equivalent (broad handwaving) to its rx26x0@ sibling, modulo it having a smaller memory slot count (8 vs 12),E support for two rather than three internal discs, one PCI-X slot and, / in the case of the 1.6 GHz rx1620 a faster FSB.   
 rick jones --  B firebug n, the idiot who tosses a lit cigarette out his car windowF these opinions are mine, all mine; HP might not want them anyway... :)D feel free to post, OR email to rick.jones2 in hp.com but NOT BOTH...   ------------------------------  % Date: Wed, 30 Aug 2006 22:20:05 -0400 * From: "FredK" <fred.nospam@nospam.dec.com>6 Subject: Re: OpenVMS Integrity Graphics card questions, Message-ID: <44f64755$1@usenet01.boi.hp.com>  L The Radeon 7500 is the only add-in card that will work (it "might" work on aI ZX2000 or a ZX6000 that had different firmware and without the management I option).  The first card (VGA console) on an Alpha will work with the DVI < port provided it is connected when the system is powered up.  H The Radeon 7000 card will, in general, not work on Itanium - despite theI fact that the management option has a R7000 built-in that will work.  The J problem is initialization - which I have not quite figured out yet.  It isH not compatable 100% with the R7500 initialization - which we do by hand.J The card/chip that the firmware initializes works fine.  Which is also whyI the DVI port doesn't work on the R7500 on Itanium - because we don't know - the magic handshake to initialize it by hand.   L We are working on a BIOS emulator to initialize secondary cards for a futureJ release that will "fix" that problem.  In the meantime, the best bet is toI order the AB551A (I think that is the order number for the Radeon 7500) - L since it is nearly impossible to find the PCI card on the secondary market -+ although perhaps Island might have some ;-)    Fred    + <johnhreinhardt@yahoo.com> wrote in message < news:1156950813.681613.231440@i3g2000cwc.googlegroups.com... >  > Robert Deininger wrote: G > > In article <1156897939.666811.157390@b28g2000cwb.googlegroups.com>, @ > > "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> wrote: > > > > > Fred may chime in with more details if he sees the thread. > > J > > >I can't seem to find the discussions at the moment but it seems to meH > > >that it's been said that OpenVMS Integrity will not support the DVIH > > >port on an ATI Radeon 7000 graphics card.  I also think it was saidL > > >that it won't support a second head on the same card ( Mine has one VGA+ > > >and one DVI).  Can anyone verify this?  > > D > > VMS supports the Radeon 7500 PCI plug-in card on the entry-level	 Integrity I > > servers.  I don't recall the product number for the card, but it's in  the H > > configuration guides and QuickSpecs for the systems.  It is the sameE > > hardware supported on various recent Alpha systems, but the order  number, ; > > packaging and accompanying documentation are different.  > >  > F > Since this is a hobbyist endeavor I'm planning on using a retail ATII > Radeon 7500 I picked up on Ebay (I picked up 3 new, sealed, in a retail G > box for $35 each plus shipping - they also work great in the XP1000). I > The ZX6000 workstation I aquired has an NVidia Quadro2 Pro AGP card and 9 > no MP so I was looking into OpenVMS compatible devices.  > G > > VMS only supports 1 head per Radeon 7500 card.  You can use the VGA L > > output.  I think, but I'm not positive, that you can use the DVI output. > > You CAN'T use both.  > G > Unfortunately the 7500's I have only have the VGA (plus S-Video and a H > TV-tuner).   All of the 7500's I've found with DVI are AGP rather than > PCI. >  > > F > > HP-UX supports the same card, on roughly the same set of Integrity servers. > > L > > If the available configuration documents don't answer these questions toK > > your satisfaction, the documents are due for a round of repairs.  There K > > are feedback mechanisms provided on the web, and they are even supposed  to > > be working.  > >  > I > I've read the Install guide for the AGP Radeon 7500 (which is available E > on the HP website) and gotten the PCI version to work in my XP1000. D > Since this wandering to the Integrity side is all on non-supportedC > hardware and hobbyist related, the range of knowledge required is < > somewhat outside that available from the HP documentation. > K > > Radeon 7000 seems to be one of the names used for the built-in graphics K > > chip used in the Management Processor.  The chip implements a subset of E > > the Radeon 7500.  VMS supports graphics on the MP of the Integity  servers,J > > but only a "2D" subset of the Radeon 7500 functionality.  I don't know the / > > details well enough to describe the subset.  > >  > E > I have a (two, actually) OEM PC Radeon 7000 with the dual output (1 F > VGA, 1 DVI).  Since I want to use this system as a master controllerI > for my other VMS sytems and anticipate using DECwindows extensively I'd G > like to be able to use it with my 20" LCD and the DVI input.  Since I I > can't find any PCI bus 7500's with DVI I may try the 7000 to see if the I > DVI output  will work.  I'd rather use a 7500, but since I won't really G > be doing much 3D work it may not perform too badly (assuming it works 
 > at all). > I > > You can use the Radon 7500 with or without the MP graphics on systems  > > where MP is optional.  > > G > > >If you get an Integrity system without the Management Processor, I G > > >assume you can put a Radeon in one of the PCI slots.  Will OpenVMS H > > >Integrity support anything besides the 7000?  Doesn't the 7500 also
 > > >work? > > E > > VMS doesn't support a "Radeon 7000" plug-in card, if such a thing  exists. H > > If you find one, VMS might recongnize it and it might "work", but it isn't  > > supported. > >  > H > Okay. I was making the assumption (perhaps a false one) that since theH > MP card has a Radeon 7000 equivalent built-in that a separate PCI card- > would work also.  Time for experimentation!  > K > > >I've read that the AGP slot (on the otherwise unsupported ZX6000) will H > > >not work and I'm pretty sure it has been said that it won't be.  IsF > > >there any work planned for any kind of graphics solutions for the@ > > >Integrity systems in VMS (Other than the Radeon 7000/7500)? > > G > > I don't think the AGP graphics card was even tried with VMS on this B > > system.  VMS doesn't support the system (which has been out of
 productionL > > for a while), and the industry is moving away from AGP as far as I know.C > > I don't expect any additional VMS work on AGP graphics support.  > > G > > Radeon 7500 is likely to be the only add-on graphics solution until I > > PCI-express ships and matures somewhat on Integrity servers.  There's G > > currently no timetable for possible support of a future PCI-express  card, , > > but I believe folks are looking into it. > 2 > Thanks for all your help and information Robert. >  >   John H. Reinhardt  >    ------------------------------  # Date: Thu, 31 Aug 2006 12:59:30 GMT 5 From: rdeininger@mindspringdot.com (Robert Deininger) 6 Subject: Re: OpenVMS Integrity Graphics card questionsY Message-ID: <rdeininger-3108060859280001@dialup-4.233.173.7.dial1.manchester1.level3.net>   9 In article <lHnJg.19$wZ7.17@news.cpqcorp.net>, Rick Jones  <rick.jones2@hp.com> wrote:   8 >Peter 'EPLAN' LANGSTOEGER <peter@langstoeger.at> wrote:: >> In article <44f5d697@usenet01.boi.hp.com>, Hoff Hoffman! <hoff-remove-this@hp.com> writes: @ >>>And you will likely want (should this be for use in an officeG >>>environment) one of the so-called office-friendly (eg: quiet) rx2600  >>>or rx2620 series systems. > G >> Any reason why you choose the rx26x0 over the rx16x0 here? Or is the G >> 16x0 in regard to the 26x0 too slow/rare/expensive to be considered?  > E >There simply isn't an OFS version of an rx16x0.  The rx16x0 hardware E >is otherwise performance equivalent (broad handwaving) to its rx26x0 A >sibling, modulo it having a smaller memory slot count (8 vs 12), F >support for two rather than three internal discs, one PCI-X slot and,0 >in the case of the 1.6 GHz rx1620 a faster FSB.  I rx1600 and rx1620 have 2 PCI-X slots, not 1.  But 1 of them is limited to E "short" cards because of the space occupied by the power supply.  The F Radeon 7500 graphics card is "long", so only 1 will fit in the system.  F rx1600 and rx1620 are MUCH louder than a normal rx2600 or rx2620.  TheI Office-Friendly rx2620 is much quieter than the regular rx2620.  During a J demo, folks typically ask us when we'll turn it on.  It's already on.  :-)  H If you like skinny systems like the DS10L, the rx16x0 systems are better5 pretty much across the board -- except for the noise.    ------------------------------  + Date: Mon, 28 Aug 2006 10:43:39 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)& Subject: Re: Personal note - goodbye !$ Message-ID: <ecuhcr$409$2@online.de>  6 In article <44e9e4d8@usenet01.boi.hp.com>, "Guy Peleg". <guy.peleg@remove_this_header@hp.com> writes:   L > I wanted to let you know that I have decided to leave HP (I chose to leave > onK > my own). I have been part of DEC/CPQ/HP for almost 10 years, 6 of them in  > VMS engineering. > I > I want to thank this forum for all the feedback provided to me over the  > years.J > Look at recent VMS releases (V8.2 and upcoming V8.3) and I'm sure you'llH > be able to see the tremendous impact this forum had (especially in the > utilities  > area).  C A big thanks to YOU for all your effort.  It's a shame that you're  > leaving, though, since DCL benefitted a lot from your efforts.  0 > I'm going to work with Bruce Ellis at BRUDEN.   A Either Bruce offered you a lot of money (or something else) or HP @ employment isn't very attractive or there were other reasons for
 leaving.  :-)      > The company is growingE > and I'll be responsible for the business in Europe. BRUDEN provides H > training, services and consulting for VMS and other O/S's (guess whichI > area I'll be working on ;-) so there is a good chance you'll be hearing  > more from me.   > The blurb on Bruce's THE HITCHHIKER'S GUIDE TO VMS, IIRC (the I corresponding bookshelf is blocked by some new ALPHAs I just collected),  G states that Bruce used to provide VMS training to DEC employees.  Only  9 very good preachers have a reason to preach to the choir!   ! Is Bruce still a strong Cubs fan?    ------------------------------    Date: 28 Aug 2006 14:40:15 -04003 From: Rich Alderson <news@alderson.users.panix.com> & Subject: Re: Personal note - goodbye !. Message-ID: <mddac5ohfg0.fsf@panix5.panix.com>  J helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:   # > Is Bruce still a strong Cubs fan?    Is there any other kind?   --  L Rich Alderson                                       | /"\ ASCII ribbon     |L news@alderson.users.panix.com                       | \ / campaign against |L "You get what anybody gets. You get a lifetime."    |  x  HTML mail and    |L                          --Death, of the Endless    | / \ postings         |   ------------------------------    Date: 29 Aug 2006 19:05:50 -0700$ From: "AEF" <spamsink2001@yahoo.com>& Subject: Re: Personal note - goodbye !B Message-ID: <1156903550.302732.62670@h48g2000cwc.googlegroups.com>   Guy Peleg wrote: > Dear community,  > L > I wanted to let you know that I have decided to leave HP (I chose to leave > onK > my own). I have been part of DEC/CPQ/HP for almost 10 years, 6 of them in  > VMS engineering. > I > I want to thank this forum for all the feedback provided to me over the  > years.J > Look at recent VMS releases (V8.2 and upcoming V8.3) and I'm sure you'llH > be able to see the tremendous impact this forum had (especially in the > utilities  > area). > F > I'm going to work with Bruce Ellis at BRUDEN. The company is growingE > and I'll be responsible for the business in Europe. BRUDEN provides H > training, services and consulting for VMS and other O/S's (guess whichI > area I'll be working on ;-) so there is a good chance you'll be hearing  > more from me.  > C > My new email starting September 1st will be guy.pelegATbruden.com  > 
 > Thank you !  >  > Guy   $ Bye Guy! Thanks for working with us.   AEF    ------------------------------  % Date: Mon, 28 Aug 2006 16:48:54 -0400 ' From: Dave Froble <davef@tsoft-inc.com>  Subject: Re: phone9 Message-ID: <o92dnYufoauMyG7ZnZ2dnUVZ_o-dnZ2d@libcom.com>   / Phillip Helbig---remove CLOTHES to reply wrote: 5 > In article <kU6IxVj254Ux@eisner.encompasserve.org>, @ > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:  > z >> In article <eb0ce1$tvg$1@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:1 >>> Any chance that PHONE will get IP capability?  >>> K >>    The corresponding IP tool is usually known as talk.   Most people use + >>    IM for this now.  Why invent another?  > E > So that PHONE works on a cluster without DECnet.  AFAIK talk isn't  . > installed by default on VMS, while PHONE is. >   A DECnet will work over any transport the cluster traffic will use.   5 You've been fighting this worthless battle for years.    Why?   Why not just install DECnet.   --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  + Date: Tue, 29 Aug 2006 08:56:46 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: phone$ Message-ID: <ed0vge$96v$2@online.de>  E In article <o92dnYufoauMyG7ZnZ2dnUVZ_o-dnZ2d@libcom.com>, Dave Froble  <davef@tsoft-inc.com> writes:   C > DECnet will work over any transport the cluster traffic will use.  > 7 > You've been fighting this worthless battle for years.  >  > Why? >  > Why not just install DECnet.  I Personally, I probably will.  However, it is no longer an essential part  H of VMS.  It would be nice if PHONE worked out-of-the-box even if DECnet  isn't installed.   ------------------------------    Date: 29 Aug 2006 08:37:56 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: phone3 Message-ID: <cnpaB2Z7CZ$8@eisner.encompasserve.org>   w In article <ed0vge$96v$2@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:  > K > Personally, I probably will.  However, it is no longer an essential part  J > of VMS.  It would be nice if PHONE worked out-of-the-box even if DECnet  > isn't installed.  F    I'm a little confused.  I though you first asked for PHONE over IP,F    from which we all assumed you already had IP installed.  If so, why    not use talk?  H    OBTW, although not officially supported, I know of one site which ranF    VAXclusters (pre-Alpha) from the very earliest days without DECnet.   ------------------------------    Date: 29 Aug 2006 08:35:37 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: phone3 Message-ID: <5f$QP3mMUUHN@eisner.encompasserve.org>   c In article <o92dnYufoauMyG7ZnZ2dnUVZ_o-dnZ2d@libcom.com>, Dave Froble <davef@tsoft-inc.com> writes:  > C > DECnet will work over any transport the cluster traffic will use.   =    You cannot do host-to-host DECnet communication over DSSI.    ------------------------------  % Date: Tue, 29 Aug 2006 11:30:34 -0400 ' From: Dave Froble <davef@tsoft-inc.com>  Subject: Re: phone9 Message-ID: <0dqdnTOD7fxqxmnZnZ2dnUVZ_v6dnZ2d@libcom.com>    Bob Koehler wrote:e > In article <o92dnYufoauMyG7ZnZ2dnUVZ_o-dnZ2d@libcom.com>, Dave Froble <davef@tsoft-inc.com> writes: D >> DECnet will work over any transport the cluster traffic will use. > ? >    You cannot do host-to-host DECnet communication over DSSI.  >   D Ok, that's correct.  It isn't relavant to the issue.  If he's using H DSSI, then he can have ethernet connectivity without going through some 8 router that won't support DECnet.  Keyword may be 'can'.  I My point, perhaps poorly worded, was that he cannot run SCS over routers  @ that wouldn't support SCS and/or DECnet.  Therefore, other than 3 stubbornness, there is no reason not to run DECnet.    --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------    Date: 29 Aug 2006 13:50:37 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: phone3 Message-ID: <EWLna5Jkhwex@eisner.encompasserve.org>   c In article <0dqdnTOD7fxqxmnZnZ2dnUVZ_v6dnZ2d@libcom.com>, Dave Froble <davef@tsoft-inc.com> writes:  > K > My point, perhaps poorly worded, was that he cannot run SCS over routers  B > that wouldn't support SCS and/or DECnet.  Therefore, other than 5 > stubbornness, there is no reason not to run DECnet.   D    I can certainly configure a router to route SCS and block DECnet,C    since those things are typically configurable by protocol number .    and those protocols have different numbers.  C    But I cringe at the thought.  A cluster without DECnet is like a 0    Fortran compiler without tab formated source.   ------------------------------  % Date: Tue, 29 Aug 2006 12:17:45 -0400 , From: Hoff Hoffman <hoff-remove-this@hp.com> Subject: Re: phone* Message-ID: <44f468a9@usenet01.boi.hp.com>   Bob Koehler wrote:e > In article <o92dnYufoauMyG7ZnZ2dnUVZ_o-dnZ2d@libcom.com>, Dave Froble <davef@tsoft-inc.com> writes: D >> DECnet will work over any transport the cluster traffic will use. > ? >    You cannot do host-to-host DECnet communication over DSSI.   P    Nor can you do many other things, like SET HOST, run MONITOR, send SMTP mail M to the Internet, or...  And most systems that had host-to-host DSSI also had   networking.   P    Barring older Q-bus systems with the KFQSA (which doesn't have host-to-host) H sans a DELQA or DESQA NIC obviously, and I'd expect the numbers of such " configurations to be rather small.  P    Most any of the IRC clients available are the current solution.  (I'll admit O to having just recently used PHONE a couple of times, however these cases were  P the first time in years.)  I've been up on IRC far more often than on PHONE, as   are most other folks these days.  L    I don't expect to see an IP transport retrofit for PHONE, barring use of Q DECnet-Plus over IP.  And I'd certainly push for an integrated and supported IRC  N client before the work required to retrofit IP into PHONE, as a supported IRC P client certainly provides more widespread compatibility, and as DECnet-Plus can : be brought to bear for other configurations needing PHONE.   ------------------------------    Date: 29 Aug 2006 22:49:16 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)  Subject: Re: phone3 Message-ID: <udd9ccwUJ7Pm@eisner.encompasserve.org>   w In article <ed0vge$96v$2@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: G > In article <o92dnYufoauMyG7ZnZ2dnUVZ_o-dnZ2d@libcom.com>, Dave Froble   > <davef@tsoft-inc.com> writes:  > D >> DECnet will work over any transport the cluster traffic will use. >>  8 >> You've been fighting this worthless battle for years. >>   >> Why?  >>   >> Why not just install DECnet.  > K > Personally, I probably will.  However, it is no longer an essential part  J > of VMS.  It would be nice if PHONE worked out-of-the-box even if DECnet  > isn't installed.  8 Phone works fine without DECnet, just not between nodes.  : But IP is no more of an essential part of VMS than DECnet.   ------------------------------  + Date: Wed, 30 Aug 2006 10:01:17 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: phone$ Message-ID: <ed3nlc$poa$1@online.de>  3 In article <cnpaB2Z7CZ$8@eisner.encompasserve.org>, > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:   F > In article <ed0vge$96v$2@online.de>, helbig@astro.multiCLOTHESvax.de5 > (Phillip Helbig---remove CLOTHES to reply) writes:   > > M > > Personally, I probably will.  However, it is no longer an essential part  L > > of VMS.  It would be nice if PHONE worked out-of-the-box even if DECnet  > > isn't installed. > H >    I'm a little confused.  I though you first asked for PHONE over IP,H >    from which we all assumed you already had IP installed.  If so, why >    not use talk? > J >    OBTW, although not officially supported, I know of one site which ranH >    VAXclusters (pre-Alpha) from the very earliest days without DECnet.  H Let me explain in more detail.  These days, it is not unusual to have a E VMS cluster with no DECnet.  I think it is then rather annoying that  = PHONE, which is there out-of-the-box, then gives the message:   3    No one with that name is available at this time.   E unless the user requested is logged in to the same node.  Especially  E since some utilities, such as MONITOR, which normally use DECnet for  F communication between nodes, can now use TCPIP (normally for the case  that no DECnet is running).   D Yes, talk would be an option, but a bad one in my view (even if the  application itself is fine):  $    o  it is extra work to install it  E    o  the interface, HELP (man?) etc don't have the VMS look and feel   G    o  it casts a bad light on VMS for a new user (hey! why doesn't this        work?)  D One of the things I like about VMS is that most of what one needs isE there out of the box.  I realise that with things which of their very A nature imply communication with remote systems, file exchange etc C (TCPIP, UUENCODE, ZIP, TAR etc), it makes sense to have a unix port G installed on VMS, and indeed that would be the case for talk as well if F the goal were to talk to a remote non-VMS system via talk (perhaps the@ only option).  However, I am talking about PHONE within the sameH cluster.  "IP" was just shorthand for "it should work in a cluster even 8 if DECnet is not running" (as is the case with MONITOR).  I Note that if users A and B are logged in to node 1, and then user B logs  E in (by LAT, say) to node 2, and user A PHONEs user B, use B will see    5 NODE1::USER1 is phoning you on NODE1::     (11:58:35)   * But user2, after typing PHONE ANSWER, sees      No one is calling you now.   H Of course, I understand what is happening.  It would just be nice if it  would work without DECnet.   For user1 to do       $ PHONE NODE2::USER2   0 is not an option, since that gives, as expected:  B    error in device name or inappropriate device type for operation   ------------------------------    Date: 30 Aug 2006 07:14:43 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: phone3 Message-ID: <DSAJuTqg1dM8@eisner.encompasserve.org>   w In article <ed3nlc$poa$1@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:  > E  Yes, talk would be an option, but a bad one in my view (even if the   > application itself is fine): > & >    o  it is extra work to install it > G >    o  the interface, HELP (man?) etc don't have the VMS look and feel  > I >    o  it casts a bad light on VMS for a new user (hey! why doesn't this  >       work?)  (    Gee, I guess your not using Multinet?  E    talk is, by default, installed "out of the box" with the IP stack.   ,    The HELP file looks like a VMS help file.  H    The interface wasn't developed by DEC and doesn't look like a native C    VMS tool, but then neither does FTP, rcp, ..., and most other IP     applications.  D    If you don't have DECnet, then none of the DECnet tools work.  DoG    you find yourself embarrased to explain why FAL, DECnet MAIL, CTERM,     ... don't work?   ------------------------------    Date: 30 Aug 2006 12:37:04 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: phone3 Message-ID: <b8rcumJX$++9@eisner.encompasserve.org>   w In article <ed41pj$g0l$1@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:   8 > No, but MAIL works fine within a cluster, for example.  C    MAIL just needs file access, and the cluster provides that.  The F    other networking tools generally need other forms of communication.  D    But since MAIL now understands IP transport as shipped, and thereB    has been a lot of noise over the years from DEC, Compaq, and HPB    to move to "standard" protocols; someone should spend the money>    to get all the utilities that did DECnet mailbox -- mailboxH    communication up to snuff on doing IP socket -- socket communication;    consistency is relavent.    ------------------------------  % Date: Wed, 30 Aug 2006 15:20:27 +0930 < From: "Barratt, Chris \(FMC\)" <Chris.Barratt@fmc.sa.gov.au># Subject: Problems with Python tests R Message-ID: <062C11A882E0C749A3CFD62BA461167A06997AF4@hadaltmail.althad.sa.gov.au>  , This is a multi-part message in MIME format.  ' ------_=_NextPart_001_01C6CBF7.BA345F60  Content-Type: text/plain;  	charset="us-ascii" + Content-Transfer-Encoding: quoted-printable    Hi,  =20 F One of my colleagues wishes to play with Python on VMS, so I installedG the 2.5.0 kit on my workstation (500au, VMS 7.3-2, TCPIP Services 5.4 -  ECO 5). H The installation was a bit flakey, but seems to have worked the 2nd time around.  =20 H When I run the tests, however, I get 10 that return errors - whereas theG documentation says that you should only expect one failure (test_time).  =20 ! The ones I had problems with are,  =20      test_mailbox
     test_mmap      test_old_mailbox     test_optparse      test_pwd     test_shutil      test_socket_ssl 
     test_time      test_timeout     test_set =20 G Is this indicative of a problem with my installation, am I missing some ) required libraries/patches or is all ok ?  =20 F I was able to run the pystone program fine, so I imagine it would justD be programs which use those modules above that might have problems ? =20  Thanks,  Chris  =20     =20  ' ------_=_NextPart_001_01C6CBF7.BA345F60  Content-Type: text/html; 	charset="us-ascii" + Content-Transfer-Encoding: quoted-printable   > <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD>7 <META http-equiv=3DContent-Type content=3D"text/html; =  charset=3Dus-ascii">@ <META content=3D"MSHTML 6.00.2900.2873" name=3DGENERATOR></HEAD> <BODY>; <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial=20   size=3D2>Hi,</FONT></SPAN></DIV>; <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial=20 # size=3D2></FONT></SPAN>&nbsp;</DIV> J <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial size=3D2>One of = my colleagues=20H wishes to play with Python on VMS, so I installed the 2.5.0 kit on my=209 workstation (500au, VMS 7.3-2, TCPIP Services 5.4 - ECO =  5).</FONT></SPAN></DIV> G <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial size=3D2>The =  installation was=20 5 a bit flakey, but seems to have worked the 2nd time =  around.</FONT></SPAN></DIV> ; <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial=20 # size=3D2></FONT></SPAN>&nbsp;</DIV> J <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial size=3D2>When I =
 run the=20I tests, however, I get 10 that return errors - whereas the documentation =  says=20 H that you should only expect one failure (test_time).</FONT></SPAN></DIV>; <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial=20 # size=3D2></FONT></SPAN>&nbsp;</DIV> G <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial size=3D2>The = 
 ones I had=20 & problems with are,</FONT></SPAN></DIV>; <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial=20 # size=3D2></FONT></SPAN>&nbsp;</DIV> @ <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp; <FONT = face=3DArial=20 ) size=3D2>test_mailbox</FONT></SPAN></DIV> @ <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp; <FONT = face=3DArial=20 & size=3D2>test_mmap</FONT></SPAN></DIV>@ <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp; <FONT = face=3DArial=20 - size=3D2>test_old_mailbox</FONT></SPAN></DIV> @ <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp; <FONT = face=3DArial=20 * size=3D2>test_optparse</FONT></SPAN></DIV>@ <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp; <FONT = face=3DArial=20 % size=3D2>test_pwd</FONT></SPAN></DIV> @ <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp; <FONT = face=3DArial=20 ( size=3D2>test_shutil</FONT></SPAN></DIV>@ <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp; <FONT = face=3DArial=20 , size=3D2>test_socket_ssl</FONT></SPAN></DIV>@ <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp; <FONT = face=3DArial=20 & size=3D2>test_time</FONT></SPAN></DIV>@ <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp; <FONT = face=3DArial=20 ) size=3D2>test_timeout</FONT></SPAN></DIV> @ <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp; <FONT = face=3DArial=20 % size=3D2>test_set</FONT></SPAN></DIV> ; <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial=20 # size=3D2></FONT></SPAN>&nbsp;</DIV> F <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial size=3D2>Is = this indicative=20E of&nbsp;a problem with my installation, am I missing some required=20 4 libraries/patches or is all ok ?</FONT></SPAN></DIV>; <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial=20 # size=3D2></FONT></SPAN>&nbsp;</DIV> I <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial size=3D2>I was =  able to run=20H the pystone program fine, so I imagine it would just be programs which = use those=20< modules above that might have problems ?</FONT></SPAN></DIV>; <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial=20 # size=3D2></FONT></SPAN>&nbsp;</DIV> ; <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial=20 $ size=3D2>Thanks,</FONT></SPAN></DIV>; <DIV><SPAN class=3D378194205-30082006><FONT face=3DArial=20 " size=3D2>Chris</FONT></SPAN></DIV>9 <DIV><SPAN class=3D378194205-30082006></SPAN>&nbsp;</DIV> ; <DIV><SPAN class=3D378194205-30082006>&nbsp;&nbsp;&nbsp;=20  </SPAN></DIV></BODY></HTML>   ) ------_=_NextPart_001_01C6CBF7.BA345F60--    ------------------------------  % Date: Wed, 30 Aug 2006 14:29:59 +0200 1 From: =?ISO-8859-1?Q?Jean-Fran=E7ois_Pi=E9ronne?= ' Subject: Re: Problems with Python tests 6 Message-ID: <44f584c9$0$27367$ba4acef3@news.orange.fr>   Chris, > Hi,  >   H > One of my colleagues wishes to play with Python on VMS, so I installedI > the 2.5.0 kit on my workstation (500au, VMS 7.3-2, TCPIP Services 5.4 - 	 > ECO 5). J > The installation was a bit flakey, but seems to have worked the 2nd time	 > around.  >     F I have just put online a new Python250 kit which contains a workaroundD for a C compiler bug which can crash the interpreter during a Python script compilation. * It fix some of the problems you mentioned.E It also contain some new libraries like rpdb2 which allow to remotely B debug a Python script running on VMS using a graphical debugger on& Windows. Very useful to debug servlet.  J > When I run the tests, however, I get 10 that return errors - whereas theI > documentation says that you should only expect one failure (test_time).  >   # > The ones I had problems with are,  >    >     test_mailbox >     test_mmap  >     test_old_mailbox >     test_optparse  >     test_pwd >     test_shutil  >     test_socket_ssl  >     test_time  >     test_timeout >     test_set >   I > Is this indicative of a problem with my installation, am I missing some + > required libraries/patches or is all ok ?  >     A Your installation is probably correct, I haven't, yet, update the 2 installation documentation which is for python235.H Some of the tests may successed or failed depend on the hardlink setting of your installation disk.E If I remember correctly test_pwd are know to failed if you have share F UIC between account or don't have the identifier associate to the UIC./ 2 tests succeed on Itanium and failed on Alpha. K You may, also, have some failed tests if your system can't access Internet. C 1 test succeed if your run it from an unprivileged account and fail 
 otherwise.  G You can run any test in verbose mode, for example to run test_shutil in 
 verbose mode:  $ @alltests "test_shutil"   H > I was able to run the pystone program fine, so I imagine it would justF > be programs which use those modules above that might have problems ? >   $ There is a much more complete bench:, python /python_root/tools/pybench/pybench.py         JF   ------------------------------  % Date: Thu, 31 Aug 2006 15:04:59 +0200 1 From: =?ISO-8859-1?Q?Jean-Fran=E7ois_Pi=E9ronne?= ' Subject: Re: Problems with Python tests 6 Message-ID: <44f6de7e$0$25915$ba4acef3@news.orange.fr>    Peter 'EPLAN' LANGSTOEGER wrote: > In article <44f584c9$0$27367$ba4acef3@news.orange.fr>, =?ISO-8859-1?Q?Jean-Fran=E7ois_Pi=E9ronne?= <jf.pieronne@laposte.net> writes:I >> I have just put online a new Python250 kit which contains a workaround G >> for a C compiler bug which can crash the interpreter during a Python  >> script compilation.- >> It fix some of the problems you mentioned.  > 	 > Thanks.  > H > Would you so kind and fix (if still neccessary) the bug which requiresH > one to DEFINE PYTHON_VMS and PYTHON235_ROOT logicals only to deinstallC > the PYTHON (in this case the PYTHON235) kit (later) successfully.  >   / Well, it's not really a bug, it's a feature ;-)   H The problem is that I have a remove procedure to execute which clean allD .pyc and .pyo files in the installation directory. So I need to knowH where is this procedure and this procedure has to know where is the rootH installation directory. The proceduce may, probably, be generated by theH post installation procedure but how to specified where is this procedure in the PCSI kit. Currently I have: I execute install "" remove "@PYTHON_VMS:PYTHON250$REMOVE.COM" interactive; ? execute upgrade "@PYTHON_VMS:PYTHON250$REMOVE.COM" interactive;   C So before upgrade or remove a Python250 kit you have to execute the @ logicals.com procedure which is in the directory contains in the7 PYTHON_VMS logicals or manually define this 2 logicals.   E I haven't, yet, found a better solution, any suggestions are welcome.    JF   ------------------------------    Date: 28 Aug 2006 11:45:40 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN 3 Message-ID: <y6nlgHtrlk4R@eisner.encompasserve.org>     wendzinski@yahoo.com wrote:C > Is it possible to access UAF disuser flag values across nodes via G > FORTRAN?  I believe I can call SYS$GETUAI to to access UAI$V_DISACNT. E > The tricky part is that the UAF may be on a node other than the one 6 > running the program.  Examples would be appreciated.  A    If you have a sufficiently privileged account you can open the D    file via FAL and read it.  Be carefull that you don't open it for    exclusive access.  C    As for the record format, the Datatrieve archives should have an     unsupported definition.  D    But it would be better to use $GETUAI since that is the supported    access method.    ------------------------------  % Date: Mon, 28 Aug 2006 14:51:10 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN , Message-ID: <44F33B1D.C660C011@teksavvy.com>  B Another option which would in fact be better/safer is to have yourC fortran program actually execute on the remote node and deliver the  results to you via decnet.  7 This way, that program will automatically get the local F definition/location of the UAF file (which may or may not be standard)K and you doN't have to worry about where the SYSUAF is located on that node.    ------------------------------    Date: 28 Aug 2006 19:44:18 -0700 From: wendzinski@yahoo.com? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN B Message-ID: <1156819458.342430.77700@m79g2000cwm.googlegroups.com>   JF Mezei wrote:  > wendzinski@yahoo.com wrote: G > > The tricky part is that the UAF may be on a node other than the one 8 > > running the program.  Examples would be appreciated. > J > Not sure if this applies to $GETUAI, but normally the logical SYSUAF canA > be set to point to an alternate file. If that file resides on a G > different node, make sure the file specification is full and that any E > proxies are set properly (or file spec includes username/password).  > J > If a simple DEFINE SYSUAF does not work, you may have to try DEFINE/EXEC > (this needs privileges).      G I'm able to call GETUAI successfully to get the information I need from E the SYSUAF file on the node I am running my program on.  Show logical E SYSUAF does not show that logical as being defined.  Can anybody tell - me how GETUAI knows where to find SYSUAF.DAT?    Thanks for the responses.    ------------------------------  % Date: Tue, 29 Aug 2006 00:17:17 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN , Message-ID: <44F3BFC0.F649C218@teksavvy.com>   wendzinski@yahoo.com wrote: I > I'm able to call GETUAI successfully to get the information I need from G > the SYSUAF file on the node I am running my program on.  Show logical G > SYSUAF does not show that logical as being defined.  Can anybody tell / > me how GETUAI knows where to find SYSUAF.DAT?   G When using RMS services, you can make a complex file specification that @ automatically translates portions of it if it is a logical name.  H For instance, you specify "SYSUAF" as filename, and "SYS$SYSTEM:.DAT" asH default file specification. The system will try to translate "SYSUAF" asB a logical name. If it fails, it keeps SYSUAF and then will use the1 default file specification to fill in the blanks.   F If SYSUAF is a logical name, the later is translated, and if there areF any missing bits, again they come from the default file specification.  G HELP LEX F$PARSE gives you an overview of this file handling technique.    ------------------------------  + Date: Tue, 29 Aug 2006 08:39:51 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN $ Message-ID: <ed0ugn$6br$2@online.de>  @ In article <slrneevgmf.sng.usenet@zappy.catbert.org>, Dan Foster <usenet@evilphb.org> writes:    x > In article <1156562748.043456.197730@b28g2000cwb.googlegroups.com>, wendzinski@yahoo.com <wendzinski@yahoo.com> wrote: > > E > > Is it possible to access UAF disuser flag values across nodes via I > > FORTRAN?  I believe I can call SYS$GETUAI to to access UAI$V_DISACNT. G > > The tricky part is that the UAF may be on a node other than the one 8 > > running the program.  Examples would be appreciated. > G > It depends... if this is for clustered nodes, they would be running a  > cluster-common shared UAF.  = Not necessarily (though in most cases this is probably true).    ------------------------------  + Date: Tue, 29 Aug 2006 08:41:27 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN $ Message-ID: <ed0ujn$6br$3@online.de>  B In article <1156819458.342430.77700@m79g2000cwm.googlegroups.com>, wendzinski@yahoo.com writes:    I > I'm able to call GETUAI successfully to get the information I need from G > the SYSUAF file on the node I am running my program on.  Show logical G > SYSUAF does not show that logical as being defined.  Can anybody tell / > me how GETUAI knows where to find SYSUAF.DAT?   H Presumably the standard method: if the logical is not defined, it looks  in SYS$SYSTEM.   ------------------------------    Date: 29 Aug 2006 08:22:14 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN 3 Message-ID: <53ZU5uKr3tmC@eisner.encompasserve.org>   _ In article <1156819458.342430.77700@m79g2000cwm.googlegroups.com>, wendzinski@yahoo.com writes:  > I > I'm able to call GETUAI successfully to get the information I need from G > the SYSUAF file on the node I am running my program on.  Show logical G > SYSUAF does not show that logical as being defined.  Can anybody tell / > me how GETUAI knows where to find SYSUAF.DAT?   B    It operates as if it simly uses RMS.  RMS allows you to specifyB    the name you want (SYSUAF), and the default name to fill in anyF    missing pieces (SYS$SYSTEM:.DAT).  (RMS also provides for a related.    name, but that doesn't seem to apply here).  D    If SYSUAF is a logical name pointing to the file then the default!    pieces don't need to get used.   G    Note that AUTHORIZE is slightly different, it seems to use a default     name of just .DAT .   ------------------------------  % Date: Tue, 29 Aug 2006 11:49:34 -0400 , From: Hoff Hoffman <hoff-remove-this@hp.com>? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN * Message-ID: <44f46213@usenet01.boi.hp.com>   wendzinski@yahoo.com wrote:   I > I'm able to call GETUAI successfully to get the information I need from ; > the SYSUAF file on the node I am running my program on.     D    Is this a cluster, or is this across a DECnet network of systems?      OpenVMS version(s)?      Architecture(s)?   Q    Some background on the problem and the local application requirements, please?    > Show logicalG > SYSUAF does not show that logical as being defined.  Can anybody tell / > me how GETUAI knows where to find SYSUAF.DAT?   N    It uses entirely standard and typical filename specification parsing, with A the filename being SYSUAF, and with the default specification of  K SYS$SYSTEM:.DAT.  If SYSUAF is defined (in a trusted table), then the open  O translates it.  The parsing then applies the default file specification, using  O pieces for any that are missing.  (If there is no SYSUAF translation, then the  Q results of the filename parsing are SYS$SYSTEM:SYSUAF.DAT.)  This is the typical  M "silent" logical name translation, obviously, and this technique is directly  O available from most OpenVMS languages, and obviously also via direct RMS calls.   O    Since you appear to be considering the use of sys$getuai to access a remote  P node using DECnet and FAL, do realize you could open up a massive security hole Q on the remote node if you're not careful here, as privileges do not traverse the  N network connection.  You'll end up using DECnet proxies, and that has its own J risks.  And the proxy itself has to allow the current user in, which is a  security exposure.  M    As I posted earlier in this thread, the sys$getuai call is the easy part,  N comparatively.  Avoiding the unintended creation of security problems (should T this be involving non-clustered and distributed security) is somewhat more involved.   ------------------------------    Date: 30 Aug 2006 12:12:27 -0700$ From: "Ed Wilts" <ewilts@ewilts.org>? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN C Message-ID: <1156965147.554434.245510@i42g2000cwa.googlegroups.com>   / Phillip Helbig---remove CLOTHES to reply wrote: B > In article <slrneevgmf.sng.usenet@zappy.catbert.org>, Dan Foster > <usenet@evilphb.org> writes: > z > > In article <1156562748.043456.197730@b28g2000cwb.googlegroups.com>, wendzinski@yahoo.com <wendzinski@yahoo.com> wrote: > > > G > > > Is it possible to access UAF disuser flag values across nodes via K > > > FORTRAN?  I believe I can call SYS$GETUAI to to access UAI$V_DISACNT. I > > > The tricky part is that the UAF may be on a node other than the one : > > > running the program.  Examples would be appreciated. > > I > > It depends... if this is for clustered nodes, they would be running a  > > cluster-common shared UAF. > ? > Not necessarily (though in most cases this is probably true).   E Although it's not necessarily true, a supported cluster configuration D requires a single SYSUAF.  Ditto for things like LMF - node-specific- versions may work, but they're not supported.   C Assuming that you have a supported cluster, you can assume a single # cluster-wide version of the sysuaf.   A You may also find really old packages like GETAUI or UAF helpful. B They've done all the hard work for you although both are in C, notA Fortran.  You know when the author quotes a Bitnet address in the % source, it's been around a while :-).   	    .../Ed    ------------------------------  % Date: Wed, 30 Aug 2006 15:22:28 -0400 , From: Hoff Hoffman <hoff-remove-this@hp.com>? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN * Message-ID: <44f5e586@usenet01.boi.hp.com>   Ed Wilts wrote:   G > Although it's not necessarily true, a supported cluster configuration F > requires a single SYSUAF.  Ditto for things like LMF - node-specific/ > versions may work, but they're not supported.   P    Node-specific SYSUAF files are AFAIK supported, so long as the usernames and M UIC values are synchronized.  You can have only one security domain within a  H cluster, but you may need to have different quotas and different nodes.  Maintaining this is, um, ugly.  >    http://h71000.www7.hp.com/doc/82FINAL/6318/6318pro_020.html  ' "11.4.1 Using Multiple SYSUAF.DAT Files   Q Most OpenVMS Clusters are managed with one user authorization (SYSUAF.DAT) file,  Q but you can use multiple user authorization files to limit access for some users  P to certain systems. In this scenario, users who need access to all systems also  need multiple passwords.  N Be careful about security with multiple SYSUAF.DAT files. The OpenVMS VAX and I OpenVMS Alpha operating systems do not support multiple security domains.   Q Reference: See HP OpenVMS Cluster Systems for the list of fields that need to be  D the same for a single security domain, including SYSUAF.DAT entries.  K Because Alpha systems require higher process quotas, system managers often  P respond by creating multiple SYSUAF.DAT files. This is not an optimal solution. N Multiple SYSUAF.DAT files are intended only to vary environments from node to P node, not to increase process quotas. To increase process quotas, HP recommends Q that you have one SYSUAF.DAT file and that you use system parameters to override  H process quotas in the SYSUAF.DAT file with system parameters to control " resources for your Alpha systems."  E > Assuming that you have a supported cluster, you can assume a single % > cluster-wide version of the sysuaf.   N    I certainly tend to assume this, as do all the tools.  Because of this, if H you are using multiple parallel SYSUAF files, they must be synchronized.   ------------------------------  % Date: Wed, 30 Aug 2006 20:53:37 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN , Message-ID: <44F6330F.ED9EAB13@teksavvy.com>   re: multiple UAFs in a cluster.   G It can be done. I have such a setup. There are caveats. And in the end, / you want them to be "mrirrored" to some extent.   H For instance, if the queue manager runs on nodeA, and you are on a NODEBH username and submit a batch job to run on a NODE_B queue, it may fail ifE your username doesn't also exist on NODE-A  because the queue manager H will balk at getting a job with an non existant username (from its pointE of view). The actual submit command will work because it is doen on a   node where that username exists.   ------------------------------  + Date: Thu, 31 Aug 2006 11:30:46 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN , Message-ID: <ed6h96$i2f$1@south.jnrs.ja.net>  j In article <1156965147.554434.245510@i42g2000cwa.googlegroups.com>, "Ed Wilts" <ewilts@ewilts.org> writes:0 >Phillip Helbig---remove CLOTHES to reply wrote:C >> In article <slrneevgmf.sng.usenet@zappy.catbert.org>, Dan Foster  >> <usenet@evilphb.org> writes:  >>{ >> > In article <1156562748.043456.197730@b28g2000cwb.googlegroups.com>, wendzinski@yahoo.com <wendzinski@yahoo.com> wrote:  >> > >H >> > > Is it possible to access UAF disuser flag values across nodes viaL >> > > FORTRAN?  I believe I can call SYS$GETUAI to to access UAI$V_DISACNT.J >> > > The tricky part is that the UAF may be on a node other than the one; >> > > running the program.  Examples would be appreciated.  >> >J >> > It depends... if this is for clustered nodes, they would be running a >> > cluster-common shared UAF.  >>@ >> Not necessarily (though in most cases this is probably true). > F >Although it's not necessarily true, a supported cluster configurationE >requires a single SYSUAF.  Ditto for things like LMF - node-specific . >versions may work, but they're not supported. > D >Assuming that you have a supported cluster, you can assume a single$ >cluster-wide version of the sysuaf. >   I Heterogeneous clusters with system specific sysuaf files have always been 
 supported.5 See for instance the OpenVMS Cluster systems manual     P http://h71000.www7.hp.com/doc/731FINAL/4477/4477pro_006.html#prep_types_environs   section 5.3 examples 4 states    " J 4. To modify records in a computer-specific system authorization file whenM logged in  to another computer that boots from the same cluster common system G disk, you must set your default directory to the specific computer. For H example, if you have setup a computer-specific system authorization fileH (SYSUAF.DAT) for computer JUPITR, you must set your default directory toK JUPITR's computer-specific [SYSEXE] directory before invoking AUTHORIZE, as  follows   ) $ SET DEFAULT SYS$SYSDEVICE:[SYS0.SYSEXE]  $ RUN SYS$SYSTEM:AUTHORIZE   "     
 David Webb Security team leader CCSS Middlesex University      B >You may also find really old packages like GETAUI or UAF helpful.C >They've done all the hard work for you although both are in C, not B >Fortran.  You know when the author quotes a Bitnet address in the& >source, it's been around a while :-). > 
 >   .../Ed >    ------------------------------  # Date: Thu, 31 Aug 2006 13:49:57 GMT ' From: jls <jeffls-nospam@sbcglobal.net> ? Subject: Re: Reading UAF Disuser Flag Across Nodes from FORTRAN 8 Message-ID: <v4qdf218a6fk2niu087hiha91numgmbres@4ax.com>  0 On Wed, 30 Aug 2006 15:22:28 -0400, Hoff Hoffman  <hoff-remove-this@hp.com> wrote:   >Ed Wilts wrote: > H >> Although it's not necessarily true, a supported cluster configurationG >> requires a single SYSUAF.  Ditto for things like LMF - node-specific 0 >> versions may work, but they're not supported. > Q >   Node-specific SYSUAF files are AFAIK supported, so long as the usernames and  N >UIC values are synchronized.  You can have only one security domain within a I >cluster, but you may need to have different quotas and different nodes.   >Maintaining this is, um, ugly.  > ? >   http://h71000.www7.hp.com/doc/82FINAL/6318/6318pro_020.html  > ( >"11.4.1 Using Multiple SYSUAF.DAT Files > R >Most OpenVMS Clusters are managed with one user authorization (SYSUAF.DAT) file, R >but you can use multiple user authorization files to limit access for some users Q >to certain systems. In this scenario, users who need access to all systems also   >need multiple passwords.   E Hmm... while maybe supported, I did find them problematic to manage - D without the utmost care and handling... something which often didn't3 occur consistently over time and personnel changes.   F I always fight for the consolidated UAF and manage resource quotas via* other means (e.g., PQL_M parameters, etc).   ------------------------------    Date: 29 Aug 2006 04:03:21 -0700/ From: "Volker Halle" <volker_halle@hotmail.com> ; Subject: Re: SEPPUCLU bugcheck introducing new cluster node C Message-ID: <1156849401.465335.147170@b28g2000cwb.googlegroups.com>    Tom,  A here is a little bit of information about the resource hash table D (RESHASHTBL) and how it's used for resource lookup. This information@ could be used to judge the advice: just increase your RESHASHTBL  parameter to prevent this crash.  @ Whenever receiving a LOCK request (via the $ENQ system service),C OpenVMS has to decide, if this lock is for a NEW resource or for an D existing resource. To speed up this lookup operation, some fields inC the RSB are hashed to provide a 32-bit hash value (16 bits prior to D V7.1). The same hash algorithm is used on all nodes in a cluster, soD every node would calculate the same hash value for a given resource.G The hash value is store is the RSB - resource block (at RSB$L_HASHVAL).   E When looking up a resource based on its hash value, the hash value is E shifted based on the local resource hash table size and used to index A into the local hash table starting at @LCK$GQ_HASHTBL (previously E LCK$GL_HASHTBL). Each quadword (previously longword) hash table entry D is a header for a chained list of resources (RSBs) - all with a hash3 value pointing to the same hash chain header entry.   E The hash chain is searched sequentially and each RSB found is checked G for a matching resource name, length, access mode, UIC group and parent D RSB. The hash chain is terminated with a zero forward link. The hashB chain header entry is zero, if there is no RSB in this hash chain.  D The hashing algorithm is used to prevent a sequential search throughG all possible resources. Only RSBs within the same hash chain have to be E searched sequentially. This is where RESHASHTBL size comes into play. ? If the resource hash table is 'too small' for a given number of F resources, the individual number of RSBs in each hash chain may become@ quite large and may delay the sequential scanning for a resourceE lookup. If the RESHASHTBL is 'too large', it's just a waste of memory A (although 8 bytes per entry is not that much memory anymore these  days).  D Here is how to look at the related data structures with SDA (example from OpenVMS I64 V8.2):   # $ ANAL/SYS or $ ANAL/CRASH dumpfile   * SDA> READ SYSDEF ! read symbol definitions   SDA> SHOW RESOURCE Resource Database  ----------------- = RSB:         FFFFFFFF.7E3A6440  GGMODE:     NL  Status: VALID  ... F SDA> FORMAT FFFFFFFF.7E3A6440  ! format the whole RSB (resource block)  D SDA> EXA FFFFFFFF.7E3A6440 + RSB$L_HASHVAL ! just examine hash value for resource RSB+00108:  0A000000.00048C04   B Use above hash value to find hash chain header entry in local hash table:  6 SDA> eva (00000000.00048C04@-(^d32-@lck$gl_htblcnt))*8& Hex = 00000000.00000048   Decimal = 72  # SDA> exa @^qlck$gq_hashtbl+00000048 2 FFFFFFFF.7F7B2048:  FFFFFFFF.7E3A6440   "@d:~...."  2 SDA> vali queue/singly/list/quad FFFFFFFF.7F7B2048  $  Entry    Address              Flink$  -----    -------              -----0  Header   FFFFFFFF.7F7B2048    FFFFFFFF.7E3A64400      1.   FFFFFFFF.7E3A6440    00000000.00000000  9 Queue is zero-terminated, total of 1 element in the queue   G The first entry in this hash chain is the RSB looked up above. There is G only one RSB in this hash chain, so a lookup for this resource would be 
 very fast.  E If you are up to it, you could combine the whole operation in one SDA  command line ;-)    SDA> vali queue/singly/list/quad? @^qlck$gq_hashtbl+(00000000.00048C04@-(^d32-@lck$gl_htblcnt))*8     E According to the OpenVMS programming Concepts Manual (chapter 7.2.9.3 C Resource Hash Table), the size of the resource hash table should be G about 4 times the average number of resources used on the system (use $ F MONITOR LOCK to find out). RESHASHTBL has AUTOGEN FEEDBACK attributes,F so an AUTOGEN with feedback should take care of the correct setting of this parameter.    Volker.    ------------------------------    Date: 30 Aug 2006 01:02:29 -0700/ From: "Volker Halle" <volker_halle@hotmail.com> ; Subject: Re: SEPPUCLU bugcheck introducing new cluster node B Message-ID: <1156924949.102620.57900@i42g2000cwa.googlegroups.com>   Tom,  F because I've received questions from another LOCKMGRERR/SEPPUCLU crash: pair, here is some important information on how to further- differentiate these types of generic crashes:   @ If you get a LOCKMGRERR/SEPPUCLU crash pair, you need to furtherE qualify it with the FATALCODE found in the LKMSG @R12 on the SEPPUCLU 4 node or in CDRP$L_VAL13(R5) in the LOCKMGRERR crash.  6 SDA> exa @r12+LKMSG$L_ORIGLKID     ! in SEPPUCLU crash  C SDA> exa @r5+cdrp$l_val13                     ! in LOCKMGRERR crash   C The FATALCODE is set up by the distributed lock manager code on the D node receiving the 'bad' LKMSG packet based on a couple of checks of the received data.  > Only if two SEPPUCLU/LOCKMGRERR crash pairs also have the same8 FATALCODE, you could about a possible 'similar' problem.   --- , Volker Halle, Invenate GmbH, OpenVMS Support  # An OpenVMS crashdump analysis a day $ makes the Windows headaches go away.   ------------------------------  + Date: Wed, 30 Aug 2006 09:34:18 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply); Subject: Re: SEPPUCLU bugcheck introducing new cluster node $ Message-ID: <ed3m2q$lu0$1@online.de>  C In article <1156849401.465335.147170@b28g2000cwb.googlegroups.com>, 2 "Volker Halle" <volker_halle@hotmail.com> writes:   G > According to the OpenVMS programming Concepts Manual (chapter 7.2.9.3 E > Resource Hash Table), the size of the resource hash table should be I > about 4 times the average number of resources used on the system (use $ H > MONITOR LOCK to find out). RESHASHTBL has AUTOGEN FEEDBACK attributes,H > so an AUTOGEN with feedback should take care of the correct setting of > this parameter.   F On a random VAX in my cluster, with no significant change in load and > AUTOGEN from SAVPARAMS to SETPARAMS not that long ago, I have:  K     Total Resources                2780.00    2780.00    2780.00    2780.00   J RESHASHTBL                   2048         64          1   16777216 Entries  D Should the "factor of 4" apply on VAX as well?  Or do I have so few % locks I don't need to worry about it?    ------------------------------    Date: 30 Aug 2006 04:13:40 -0700/ From: "Volker Halle" <volker_halle@hotmail.com> ; Subject: Re: SEPPUCLU bugcheck introducing new cluster node B Message-ID: <1156936420.068523.117780@i3g2000cwc.googlegroups.com>  E > Should the "factor of 4" apply on VAX as well?  Or do I have so few ' > locks I don't need to worry about it?   @ This guideline covers all possible OpenVMS system architectures.  B Look in SYS$SYSTEM:AGEN$PARAMS.REPORT for the RESHASHTBL parameterF information, including the current no. of resources in use at feedback time.   C Increasing RESHASHTBL will only help performance, if your system is D doing heavy locking and would be spending a lot of time sequentially  looking through the hash chains.  F As there is no VALIDATE/SINGLY_LINKED command qualifier in VAX SDA, itE would be kind of hard to check the sizes of your resource hash chains  with SDA manually:  
 $ ANAL/SYS@ SDA> exam @lck$gl_hashtbl;@lck$gl_htblsiz*4 ! this is for VAX or pre-V7.1 Alpha only   D This will show you your system's resource hash table entries. If youC see lots of 0s, your hash table is sparsely populated and you don't B have to worry. If most of the hash chain headers are non-zero, youD could follow the individual hash chains and count the no. of RSBs on each chain:   C SDA> exa 86E20E80            <- use any non-zero address from above  EXAM output D 86E20E80:  00000000   "...."  <- link to next RSB is ZERO, so only 1 RSB in chain  : If you find a non-zero link (RSB$L_HSHCHN), continue with:   SDA> EXA @.   E until the forward link is zero. This will show you the no. of RSBs on  that hash chain.   --- , Volker Halle, Invenate GmbH, OpenVMS Support  # An OpenVMS crashdump analysis a day $ makes the Windows headaches go away.   ------------------------------    Date: 30 Aug 2006 06:03:22 -0700/ From: "Volker Halle" <volker_halle@hotmail.com> ; Subject: Re: SEPPUCLU bugcheck introducing new cluster node A Message-ID: <1156943002.007244.70430@74g2000cwt.googlegroups.com>   9 > Question: is AUTOGEN doing what it is supposed to here?   ? OpenVMS VAX AUTOGEN does not seem to adhere to the factor 4 for F calculating RESHASHTBL size. AUTOGEN on OpenVMS Alpha V8.2 seems to do a better job in this case.  E Unless your systems are involved in heavy locking and the hash chains E would be really large (100 or more entries), this should not have any  negative effect.   --- , Volker Halle, Invenate GmbH, OpenVMS Support  # An OpenVMS crashdump analysis a day $ makes the Windows headaches go away.   ------------------------------  + Date: Mon, 28 Aug 2006 09:20:42 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)$ Subject: Re: Shadow copy stops at n%$ Message-ID: <ecucha$jue$1@online.de>  C In article <1155722950.801695.137000@h48g2000cwc.googlegroups.com>, 1 "Ger_Marsh" <marsh_family@btconnect.com> writes:    D > Just to add to the cov knowledgebase and also some info for anyone" > undergoing a similar exercise... > G > After a great deal of hassle with attempting to move to VW drives (on H > HSZ50's then to HSZ80's), we discovered that some shadow copies get toD > a percentage then hang and lots of mount verifications - and their > completions - get OPCOM'd.  D I experienced a similar problem with EZ (solid-state) disks which I G THINK kicked in after I had upgraded to 7.3-2.  There is a thread here   on this.  C Since the only way to get out of the situation is to reboot, it is  C difficult to do many tests.  However, having recently increased my  B collection of ALPHAs from 6 to 11 (giving away 2 big machines and A collecting 7 new, smaller ones), if I have the proper cables for  ? connecting to a BA356 or whatever I will see if the problem is   reproducible etc.    ------------------------------  + Date: Mon, 28 Aug 2006 09:29:07 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: Shadow set problem $ Message-ID: <ecud13$jue$2@online.de>  3 In article <SkERd2EAu7o9@eisner.encompasserve.org>, ! briggs@encompasserve.org writes:     > > $     MOUNT   N > > DSA0:/CLUSTER/NOASSIST/INCLUDE/NOCOPY/SHADOW=($42$DKA1200:,$42$DKA1300:)  
 > > COMMON > ( > You do not want the /NOCOPY qualifier. > F > When you mount the new member, you want it to be brought up to date.E > If you use /NOCOPY and an a copy operation is required to bring the F > new member up to date, VMS will simply refuse to add the new member. > N > You may want the /CONFIRM qualifier instead.  If a copy operation is needed,F > /CONFIRM will give you a chance to verify that it involves the disksD > you expect in the direction you expect and answer yes or no before! > the copy is actually initiated.   F Note: if you are not familiar with /CONFIRM, don't be surprised by the> -F- (fatal error!) message!  I really think this should be -I- (informational!)    I Does anyone know why a FATAL error message is issued when a full copy is   needed?    ------------------------------  + Date: Mon, 28 Aug 2006 09:38:15 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: Shadow set problem $ Message-ID: <ecudi7$jue$3@online.de>  9 In article <eckp7s$vqq$1@news01.intel.com>, Ken Fairfield ! <my.full.name@intel.com> writes:    B > I think the point is that /INCLUDE /NOCOPY are quite appropriateG > when doing mounts at system startup.  But one does need to understand + > why (I've posted on this in the past...).    Yes, well worth reading.  G Personally, I specify all members at startup with neither of the above  H commands.  I could be bitten by the following situation (each member of 9 the shadow set has a direct connection to only one node):   ! A goes down, B is up-to-date copy  B goes down , A comes up, doesn't see B, A becomes currentG B comes up, its disk gets overwritten by a full copy from A, losing the 4   data written between A going down and B going down  G Having said that, this has never happened to me.  Obviously, the above  F scenario would occur only in the case of severe problems, and in that G case I am going to have an ugly transition anyway, so the data loss is  I probably not that big of an issue.  With systems set to HALT at startup,  B obviously the above cannot happen unless one makes a mistake when E booting by hand.  With systems set to BOOT or RESTART on startup, it  E could, but it means that B would have to go down during A's reboot.    Very unlikely.  G To be on the safe side, set the systems to HALT at startup and have an  F operator always on site, or add a third member to the shadow set on a  third node.   C To me, the alternative is worse: risk running a shadow set with one C member until the problem can be fixed; if that one member then goes G down, I might lose weeks (if I am on holiday, for example---this is my  H hobbyist cluster, so I can't afford an on-site operator) of data rather 6 than the few seconds or minutes in the above scenario.   ------------------------------  + Date: Tue, 29 Aug 2006 09:07:05 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: Shadow set problem $ Message-ID: <ed103o$96v$3@online.de>  D In article <ecudi7$jue$3@online.de>, helbig@astro.multiCLOTHESvax.de3 (Phillip Helbig---remove CLOTHES to reply) writes:    ; > In article <eckp7s$vqq$1@news01.intel.com>, Ken Fairfield # > <my.full.name@intel.com> writes:   > D > > I think the point is that /INCLUDE /NOCOPY are quite appropriateI > > when doing mounts at system startup.  But one does need to understand - > > why (I've posted on this in the past...).  >  > Yes, well worth reading. > I > Personally, I specify all members at startup with neither of the above  J > commands.  I could be bitten by the following situation (each member of ; > the shadow set has a direct connection to only one node):  > # > A goes down, B is up-to-date copy 
 > B goes down . > A comes up, doesn't see B, A becomes currentI > B comes up, its disk gets overwritten by a full copy from A, losing the 6 >   data written between A going down and B going down > I > Having said that, this has never happened to me.  Obviously, the above  H > scenario would occur only in the case of severe problems, and in that I > case I am going to have an ugly transition anyway, so the data loss is  K > probably not that big of an issue.  With systems set to HALT at startup,  D > obviously the above cannot happen unless one makes a mistake when G > booting by hand.  With systems set to BOOT or RESTART on startup, it  G > could, but it means that B would have to go down during A's reboot.    > Very unlikely.  D I should have mentioned that my hobbyist cluster is immune to such aF scenario under normal circumstances, such as multiple power outages atG short intervals.  In the startup sequence, each machine waits until all H the nodes it expects are there (by periodically checking logicals in theC cluster table) or until after a long-enough timeout before mounting : disks with members directly connected to different nodes.   ? I have a non-system disk with SYSUAF etc on it so I mount it in C SYLOGICALS.COM.  What used to happen is that after a cluster reboot E (power outage, say), one node would come up quick enough to mount it  I before all the members were MSCP served to all nodes.  Later on, another  H node would mount it and a full shadow copy would result.  I implemented D the checks to avoid this, but they avoid the above scenario as well.  @ (I also have some code to delay the mounting until a MINICOPY isC complete.  I can thus dismount a member of the shadow set, reboot a H node, and have just a minicopy occur when it comes back up.  This works F even if all members of the shadow set have direct connections only to F VAXes, but only if the DISMOUNT and MOUNT commands are issued from an F ALPHA.  For this reason, on VAX I set SHADOW_MAX_COPY to 0 on startup E and set it to 1 later, since even if the MOUNT command comes from an  F ALPHA the actual copy might be performed by a VAX, and in that case a  full copy would result.)   ------------------------------    Date: 29 Aug 2006 11:14:25 -0500% From: frey@encompasserve.org (Sharon) = Subject: Re: Thoughts on the book: DEC is dead, long live DEC 3 Message-ID: <26swGqVoVSP3@eisner.encompasserve.org>   \ In article <44EE5EB4.1F371648@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:  E > I finished reading it. The first half could have been condensed as:  > G > Olsen was hands-off type, let the engineers battle and decide between ? > themselves. This worked well initially, but ended up creating E > independant kingdoms that where inter-kingdom communication was not G > good. In mid 80s, Olsen became aware of this, but when he tried to do C > something about it, he had lost power within the organisation and J > couldn't force changes onto those independant kingdoms. When he tried toH > reshape the management "matrix" in 1990-1992 period, the various kingsI > didn't accept it. Seems to me that Oslen set himself up to lose control / > by devolving power so much in the early days.   I 	The way I read it, the kingdoms were not a result of Olsen's management  P style per se, but of the natural growth of the company.  As the middle managers L were held responsible for more employees and more production, they began to M spend more time in CYA activities.  As time goes by, this devolves into turf  . wars and extreme protectionism of your people.I 	I found that particularly enlightening because I've always wondered why  I corporate politics, which are usually counter-productive if not outright  M damaging to a company, are so pervasive.  It was helpful to me to understand  K that management is very often given conflicting priorities and expected to  N comply with them to the letter.  Using an example from my own situation, what M is a manager to do when he's instructed to encourage his team to get as much  J training and education as they need to do their job, but then is handed a N miniscule department budget?  Or when your team is tasked with being the sole I producer of a particular suite of high-quality products, but you're only   allowed a "skeleton crew".  I > Manufacturing wasn't too concerned about efficiency, and it seemed that J > throughout DEC, money was no object and they were perfectly happy makingH > high cost items as long as they were also high quality. There were tooE > many plants around the world and no real central controls (remember  > those independant kingdoms).  H 	Indeed, the priority was quality at any cost.  It shows how idealistic  that Olsen was. B 	Along with the lack of central controls, they had no way for the M geographically-disperse groups to communicate.  I'm not sure why that was, I  M don't remember if the author analyzed it.  (After all, was this before WAN's   and All-In-1 or VaxNotes?)  H > When it was decided to battle IBM, DEC hired 26,800 new employees in aG > short period of time. This didn't pan out, but those employees stayed I > on, thus greatly lowering the sales/employee numbers and making it even 2 > harder for DEC to compete. (this was mid 1980s).  F 	Yeah, Olsen felt a great deal of paternalism for his people and took C pride in never laying anybody off.  So the headcount grew and grew.   F > While the book is quite long for what it contains and short of juicyG > interesting details about DEC (more about generalities), it does make B > one wonder about how one can lose control over a large succesfulJ > corporation. And one has to wonder about companies like Yahoo and GoogleG > who are growing by leaps and bounds and going in every direction they  > can.    H 	He implied it was something related to group dynamics.  When a company L is small, the boss has a high degree of control.  Ken was surprised to find M that after the company had tens of thousands of employees and multi-national  4 that it ran itself and not necessarily in good ways.H 	Then too, the author commented that around that time there were enough L new managers around who didn't respect Ken as much as the original team did P that they pretty much dismissed anything he said.  "Yes boss" to his face, then + did what they wanted when he left the room.   I > Also, after reading this book, I think that writing to CEOs or even the E > board is far more valuable than previously thought since it is very G > likely that they are detached and isolated from all the bad news, and J > skipping over all the management within the corporation to reach the top, > guys is probably more important than ever.  G 	If you can do that without being seen as going over your supervisor's  M head...  sometimes corporate politics are so set in that the company's going   down no matter what you do. H 	(On a tangent, did anyone read the article about that man who tried to J warn his company (Lockheed) and the coast guard that some ships they were H refitting for them were full of problems?  According to the story, both J Lockheed and the Coast Guard dismissed his concerns and laid him off as a O troublemaker (although their official reason was budgetary).  He recently told  H his story in a 10-minute video that he placed on Youtube.com for public  consumption.  Ouch!)  	  - Sharon " "Gravity...  is a harsh mistress!"   ------------------------------  % Date: Tue, 29 Aug 2006 21:30:14 -0500 6 From: "David J. Dachtera" <djesys.no@spam.comcast.net>= Subject: Re: Thoughts on the book: DEC is dead, long live DEC 0 Message-ID: <44F4F836.525052C6@spam.comcast.net>  
 Sharon wrote:  > [snip]Q >         I found that particularly enlightening because I've always wondered why J > corporate politics, which are usually counter-productive if not outright+ > damaging to a company, are so pervasive.    C Likewise, I, too, am more oriented toward the result than who looks A good/bad as a result. I've just been attributing it to Asperger's 	 Syndrome.   $ > It was helpful to me to understandL > that management is very often given conflicting priorities and expected toO > comply with them to the letter.  Using an example from my own situation, what N > is a manager to do when he's instructed to encourage his team to get as muchK > training and education as they need to do their job, but then is handed a  > miniscule department budget?    G Unfortunately, a corporate mentality has evolved which believes that if C folks get training, they'll use it to leverage a better position at C another company, possibly a rival firm. So, don't train them on the 6 company's dime - let them pay for their own education.  1 > Or when your team is tasked with being the sole J > producer of a particular suite of high-quality products, but you're only > allowed a "skeleton crew".   Reminds me of the old poster:   H "We, the willing, led by the unknowing, are doing the impossible for theC ungrateful. We have done so much for long with so little we are now ' qualified to do anything with nothing."    > [snip]P >         (On a tangent, did anyone read the article about that man who tried toK > warn his company (Lockheed) and the coast guard that some ships they were I > refitting for them were full of problems?  According to the story, both K > Lockheed and the Coast Guard dismissed his concerns and laid him off as a P > troublemaker (although their official reason was budgetary).  He recently toldI > his story in a 10-minute video that he placed on Youtube.com for public  > consumption.  Ouch!)  G I have a running joke with my boss. His predecessor gave new meaning to D the word "twit", and he particularly resented my informing him of anD alarm situation involving the datacenter environment in front of theD other members of the management team. From then on, he accused me ofF claiming that "the sky is falling". When the new boss came on, it kindH of stuck until he began to realize how "cut to the chase, no bullshit" IG am. Now its both a running joke and way to let him know that I need his B attention at high priority. I just send a message with "The Sky isH Falling" in the subject, and he knows I've got a concern we need to talk about right away.    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------  % Date: Wed, 30 Aug 2006 03:52:24 -0400 ( From: Bill Todd <billtodd@metrocast.net>= Subject: Re: Thoughts on the book: DEC is dead, long live DEC G Message-ID: <Du2dnQo2y5Qk3mjZnZ2dnUVZ_oadnZ2d@metrocastcablevision.com>   
 Sharon wrote:    ...   J >> Manufacturing wasn't too concerned about efficiency, and it seemed thatK >> throughout DEC, money was no object and they were perfectly happy making I >> high cost items as long as they were also high quality. There were too F >> many plants around the world and no real central controls (remember >> those independant kingdoms).  > 0 > 	Indeed, the priority was quality at any cost.  G I'm afraid that while the statement you quoted is flat-out wrong (cost  B was *never* irrelevant), your own response is simply meaningless. D 'Quality' is a continuum, and as any engineer can tell you there is 9 *always* a trade-off with cost made along that continuum.   A Perhaps what you were trying to say was that there was a quality  I threshold below which a product was considered unacceptable (i.e., below  < which quality would not be further traded off for low cost).      It shows how idealistic > that Olsen was.   ? No, it shows what he had learned over more than two decades of  G outstanding success in his chosen field:  that innovative high-quality  H products at fair prices sell well.  The only problem was the appearance G of what people now like to call disruptive technology that changed the  G rules by enabling the creation of lower-quality products at lower cost  H (whereas before you just couldn't save all that much no matter how much  quality you sacrificed).  E Had Olsen had the foresight to embrace the PC for what it was rather  F than forge ahead with what had suddenly become an out-dated vision of H low-end computing, the subsequent history of DEC would likely have been C quite different.  But it's not clear that even then DEC could have  F *built* PCs at competitive prices:  it just didn't know how and would I have found it difficult to learn - OEMing boxes from someone who already  I had that knowledge would have been the most feasible approach, and would  F have allowed DEC to concentrate on the unique value it could *add* to G the PC by integrating it into a business framework backed by DEC minis.   F Just as inexpensive 16-bit PCs were the (disruptive) killer micros of E the first half of the '80s, inexpensive 32-bit workstations were the  I (disruptive) killer micros of the second half.  But while the underlying  H disruptive technology was similar, the guise was sufficiently different G that DEC got blind-sided again:  whereas it was easy to ruefully write  F off its PC debacle as due to the market clout of the IBM brand, these F new workstations were built by no-name upstarts out in Silicon Valley 5 and ran software that could not hold a candle to VMS.   B But they were cheaper and/or faster, both powerful inducements to H workstations users who would cheerfully put up with some idiosyncrasies C if the alternative were to have no workstation at all.  I've never  H looked at the numbers, but I suspect that DEC didn't sell all that many = fewer workstations than ever - just that the total number of  F workstations in the industry exploded, and that DEC's relative market " (and mind) share therefore eroded.  F All this is much easier to see in hindsight, of course, where we have G the Awful Example to make it clear that the so-often-prudent course of  F sticking to one's highly-successful knitting was the wrong choice (at G least in those two significant areas) during that decade.  Even so, if  D DEC had been able to manage its growth and field a competitive RISC E VAX-replacement significantly earlier it *still* could have remained  F healthy despite the major losses its abortive PC efforts cost and the # erosion of its workstation markets.   D But given the degree to which internal turmoil reigned, it could do E neither.  As I said earlier, I have no idea whether Olsen could have  F pulled DEC out of its tailspin had Doriot been around to help - and I G think people despise Palmer not because they think that he destroyed a  E healthy company but because he made so many obviously-bonehead moves  G (also demonstrating a complete lack of appreciation for what he had to  F work with) that accelerated DEC's decline.  Could Olsen, even without I Doriot's help, have done any *worse*?  I tend to doubt it, unless he was  H completely fed up with having to deal not only with the problems of the H company itself but with a BoD that didn't have a clue and wouldn't give ! him room to find a good solution.   F Perhaps I give him too much credit, but I think he earned it over his I first 25 years at the helm, whereas I don't see where the Monday-morning  , quarterbacks ever earned much of any at all.  D > 	Along with the lack of central controls, they had no way for the O > geographically-disperse groups to communicate.  I'm not sure why that was, I  O > don't remember if the author analyzed it.  (After all, was this before WAN's   > and All-In-1 or VaxNotes?)  I No, it was not.  IIRC DEC facilities (and most of the employees in them)  C were connected world-wide by 1980 at the latest.  VAXnotes use was  F rampant in the early (I think *very* early) '80s (an outgrowth of its B predecessor Notes-11, though I don't remember that being all that ? wide-spread).  I tried to Google up exact dates, but (somewhat  @ surprisingly, at least to me) the closest I got was a tentative G reference to 1980 as the birth date of VAXnotes.  All-In-1 was created  E in the same time-frame (didn't find an exact date for that, either -  H again, I was surprised, but I didn't look that hard:  too many memories G of what a pig that product was before hardware advances made that more  H irrelevant - a foretaste of Microsoft bloatware well ahead of its time).   - bill   ------------------------------  % Date: Wed, 30 Aug 2006 05:00:11 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> = Subject: Re: Thoughts on the book: DEC is dead, long live DEC + Message-ID: <44F55360.7B68939@teksavvy.com>    Bill Todd wrote:B > Perhaps what you were trying to say was that there was a qualityJ > threshold below which a product was considered unacceptable (i.e., below> > which quality would not be further traded off for low cost).    G In the book, the image is given that DEC engineers had carte blanche to E develop superior products at any cost. I think that DSSI is a perfect B example of this. Instead of focusing on the existing SCSI standardH (remember that SCSI was in widespread use by Apple back in 1984 when theF MAC was launched), they decided that SCSI was not perfect and designed) their own "improved" version called DSSI.   B The very proprietary buses of the 80s was another issue of certainG people trying to protect their revenues by preventing OEMs from selling 
 to customers.   G The book also mentions that in its early days, DEC relied a lot of EOMs G and thus had more open architectures so that DEC hardware could be used F as building blocks by OEMs. But at one point, DEC decided to dump EOMsF and sell itself, and at that time, developping proprietary stuff was a way to shut out OEMs.   E Note: Apple was succesful with the geeks with its APPLE II because it G was "open" and Apple even incited owners to open the box and add boards F to it. But when the MAC came out, it was a very "DEC" thing, extremelyF proprietary and no way to open the box, while the IBM PC was much moreC open making it possible for 3rd parties to sell all sorts of add on L hardware to make that glorified toy with the IBM logo on it somewhat useful.  H I think that DEC made a very big vision error when it decided to abandonG OEMs, in much the same way Apple erred with its closed Macintosh boxes.   @ > No, it shows what he had learned over more than two decades ofH > outstanding success in his chosen field:  that innovative high-quality$ > products at fair prices sell well.  G When DEC started, the market was wide open, and not developped. DEC was H able to shape it the way it wanted and the market adapted to DEC insteadH of the other way around. Since there was no competition, DEC could set aE price and scientists would then ask for suffic8ient budget to buy DEC B computers because they were THE thing they needed for their tasks.  A DEC's success may not have been so much due to Olsen's vision and G pricing stragegy, it may have been succesful DESPITE it because of lack  of competition.   G When there is no competition, you pay whatever DEC wants. When there is ? some competition such as DG, then the buyers must decide if the D additional quality of DEC products warranted the additional purchase# price. And for a long time, it did.   G There comes a time however where the effective quality difference (from B customer's point of view) doesn't justify the added markup for DECC products.  DSSI may have brought some advantages over SCSI of those G days, but not all customers needed that advantage and thus wouldn't pay  extra for DSSI.   F But back in early/mid 1980s, SCSI only gave you 10/20 meg hard drives,F while DEC was able to sell you 600 meg monsters where you could pack 3E such disks in a rack. If you're a corporation, then you had no choice E but to pay the big bucks for those RA drives built like tanks. But as D SCSI disks progressed, they became contenders for business, at whichG point, there was no longer any justification to pay extra big bucks for H DEC proprietary gear that really didn't give that much more performance.  F If you look at most DEC products during the 1980s, you probably end upF seeing the same story: they startup with some technological advantage,E but that advantage quickly fades as others catch up, and DEC fails to . lower prices to keep its products competitive.    F Consider as that recently as the introduction of DS15/25 machines, theE argument was made by DECies that they cannot compete because they are C costlier to build that equivalent machines. You'd think that Palmer ? would have killed that mentality that everything had to be more G expensive and would have motivated designers to make machines that were E inexpensive to build (while maintaining good quality).  Consider that D the DS15/25 were done during the Compaq/HP era where those engineersD would have had access to Compaq/HP expertise on how to build quality commodity hardware.   E If the "expensive" mentality survived the Palmer era and continued to @ live within Compaq/HP, one has to wonder whether the VMS productH management are still implementing that strategy and not concerned at allD about competing with the rest of the world. You'd think that by now,K they would have learned that to survive and grow, they must be competitive.     D > quite different.  But it's not clear that even then DEC could haveG > *built* PCs at competitive prices:  it just didn't know how and would # > have found it difficult to learn    : That is the role of the leader of a corporation. He has toF task/challenge the troups to find a way to build a competitive machineH that is the same price but of better quality. Or which has significantly) better quality with  a reasonable markup.   G Apple chose the "better quality with reasonable markup", but the markup F ended up being just a bit too big and Apple lost a lot of market shareF because of it.  By the time Palmer took over, he would have known thatF evebn Apple's relatively small markeup (compared to DEC's) just didn't work.   F > neither.  As I said earlier, I have no idea whether Olsen could haveG > pulled DEC out of its tailspin had Doriot been around to help - and I H > think people despise Palmer not because they think that he destroyed aF > healthy company but because he made so many obviously-bonehead movesH > (also demonstrating a complete lack of appreciation for what he had to, > work with) that accelerated DEC's decline.  F I agree with that statement. It gives a good protrait of what actuallyF happened. I am not sure if Olsen really could have brought DEC back toG success though. Based on what I read in the book, his leadership seemed E to have evaporated by the end. It is often easier to sow the seeds to G build a company and set it in a certain direction (which Olsen did very G well), than it is to change the dirtection/mentality of a large company G with lots of innertia and that has become set in its own ways and those  ways are now the wrong way.   F For all his faults, Palmer at least *tried* to change things early on.E But he was nowhere near competant enough to fix DEC. And it is a real G shame that Pfeiffer wasn't given more time to fix DEC, perhaps he could F have breathed new life into DEC.  Curly, like Palmer, was in way aboveC his head and not even close to be able to deal with DEC's problems.    >  Could Olsen, even without( > Doriot's help, have done any *worse*?   F I think Olsen would have managed to keep morale much higher at DEC andD the changes he would have made would have been less drastic. I don'tF think he would have had a slash/burn policy for the software/products.G But financially speaking, I am not sure he would have been able to turn  the company around.   A What is interesting is that Olsen was kicked out after 2 years of G financial losses. Palmer was allowed to stay far longer even though his B total performance was far worse (even though there were short termD financial profits on some quarters here and there, but still lots ofD quarters in the red quickly followed by another shuffling of execs).      H > reference to 1980 as the birth date of VAXnotes.  All-In-1 was createdF > in the same time-frame (didn't find an exact date for that, either -I > again, I was surprised, but I didn't look that hard:  too many memories H > of what a pig that product was before hardware advances made that moreJ > irrelevant - a foretaste of Microsoft bloatware well ahead of its time).  F ALL-IN-1 was the FIRST office product. First to offer corporate email,H word processing etc. DG's CEO came in later (with far less functionalityC (when it actially didn't crash your system).  And yes, it was a big C money maker for DEC because it required lots of computer resources.   G Interesting that people complained about it requiring much memory while G they never complained about moving to PCs that required far more memory A in total since you had to give each employee full hardware memory H whereas with A1, much of the code was shared between processes in memoryH so it ended up requiring far less total physical memory. Problem is thatG DEC's memory was still very expensive/proprietary whereas the PC memory 
 was cheap.   ------------------------------  + Date: Wed, 30 Aug 2006 10:39:09 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)= Subject: Re: Thoughts on the book: DEC is dead, long live DEC $ Message-ID: <ed3psd$tus$1@online.de>  G In article <Du2dnQo2y5Qk3mjZnZ2dnUVZ_oadnZ2d@metrocastcablevision.com>, + Bill Todd <billtodd@metrocast.net> writes:    H > Just as inexpensive 16-bit PCs were the (disruptive) killer micros of G > the first half of the '80s, inexpensive 32-bit workstations were the  K > (disruptive) killer micros of the second half.  But while the underlying  J > disruptive technology was similar, the guise was sufficiently different I > that DEC got blind-sided again:  whereas it was easy to ruefully write  H > off its PC debacle as due to the market clout of the IBM brand, these H > new workstations were built by no-name upstarts out in Silicon Valley 7 > and ran software that could not hold a candle to VMS.  > D > But they were cheaper and/or faster, both powerful inducements to J > workstations users who would cheerfully put up with some idiosyncrasies E > if the alternative were to have no workstation at all.  I've never  J > looked at the numbers, but I suspect that DEC didn't sell all that many ? > fewer workstations than ever - just that the total number of  H > workstations in the industry exploded, and that DEC's relative market $ > (and mind) share therefore eroded.  H Right.  DEC, SUN, SGI, HP, IBM.  In the early 90's, that was the choice.B DEC weren't particularly expensive, and in the early days of ALPHAH (especially after EV5 came out), DEC often held the "fastest workstationE in the world" badge (in fact, the only DEC computer I have bought new E was, once upon a time, the fastest workstation in the world) and thus F often the most flops per currency unit.  At a former job with a couple? of hundred thousand to spend, we tested workstations from all 5 E companies and ended up buying 3 from DEC, even though we had been IBM H (RS/6000 running AIX) only up until then.  The basic reason was bang forF the buck, though quality of compilers was an added plus (most code was number-crunching Fortran).    H What killed DEC with regard to workstations was the price of software.  I On the one hand, of course, you get what you pay for.  After buying said  I workstation (this was at another job), after being strangely regarded at  I first, more and more people wanted an account on my machine, even people  E with no idea what VMS was, because my compilers were better.  On the  H other hand, with hardware becoming so cheap, sites which had previously H not had their own computing facilities were buying them, and only those B places with a DEC Campus license in place could afford to buy DEC < workstations, even though the hardware might have been more  competitively priced.   F I still think DEC's big mistake was letting go of the academic market.E Of course, they might have "lost" money on software, but on the other C hand it was software which was also developed AND PAID FOR by other E markets.  Why essentially give it away?  Because students will become B familiar with it and thus a) want to work with it when they are noB longer students and b) be around to hire when someone wants peopleC familiar with DEC stuff.  The latter-day academic programme was too G little too late.  While it might boast an impressive number of sites (I F have seen a viewgraph with little flags at all the sites), I know fromC personal experience that at least some of those sites used to have  G accounts on DEC machines for all students and now that flag represents  G an older, more grey DEC nostalgist running a machine he saved from the  D skip, under his desk and on his own initiative.  Sic transit gloria  mundi.   ------------------------------  % Date: Wed, 30 Aug 2006 08:03:22 -0400 2 From: Bob Willard <BobwBSGS@TrashThis.comcast.net>= Subject: Re: Thoughts on the book: DEC is dead, long live DEC : Message-ID: <eOKdnRvLI4eX42jZnZ2dnUVZ_radnZ2d@comcast.com>   Bill Todd wrote:   > Sharon wrote:  > H >>     Along with the lack of central controls, they had no way for the I >> geographically-disperse groups to communicate.  I'm not sure why that  J >> was, I don't remember if the author analyzed it.  (After all, was this * >> before WAN's and All-In-1 or VaxNotes?) >  > K > No, it was not.  IIRC DEC facilities (and most of the employees in them)  E > were connected world-wide by 1980 at the latest.  VAXnotes use was  H > rampant in the early (I think *very* early) '80s (an outgrowth of its D > predecessor Notes-11, though I don't remember that being all that A > wide-spread).  I tried to Google up exact dates, but (somewhat  B > surprisingly, at least to me) the closest I got was a tentative I > reference to 1980 as the birth date of VAXnotes.  All-In-1 was created  G > in the same time-frame (didn't find an exact date for that, either -  J > again, I was surprised, but I didn't look that hard:  too many memories I > of what a pig that product was before hardware advances made that more  J > irrelevant - a foretaste of Microsoft bloatware well ahead of its time). >  > - bill  G And, even without VAXnotes, all DEC engineering and manufacturing sites E were connected via DECnet; that was the standard mechanism to deliver   and get early feedback on specs.  F Intergroup communications were so important that, at one time, DEC wasE reported to have the largest private air force (fixed-wing and helos) E in the world.  Taking the chopper from site to site was cheap & easy, " and encouraged within engineering.  A In some PL groups, line managers got monthly phone bills for each A employee, and were expected to review them for excessive use.  In > central engineering, in the '80s and '90s, the boss encouraged? communications by not even giving those bills to line managers.   G Overall, DEC always recognized that intergroup communications were very ? important; probably as a result of the "buy-in" culture of DEC.  --   Cheers, Bob    ------------------------------  % Date: Wed, 30 Aug 2006 08:32:51 -0400 ( From: Bill Todd <billtodd@metrocast.net>= Subject: Re: Thoughts on the book: DEC is dead, long live DEC G Message-ID: <tsGdnV26FoHuGGjZnZ2dnUVZ_oudnZ2d@metrocastcablevision.com>    JF Mezei wrote:  > Bill Todd wrote:C >> Perhaps what you were trying to say was that there was a quality K >> threshold below which a product was considered unacceptable (i.e., below ? >> which quality would not be further traded off for low cost).  >  > I > In the book, the image is given that DEC engineers had carte blanche to ( > develop superior products at any cost.  I Then either the book or your perception of what it portrays is incorrect.       I think that DSSI is a perfect > example of this.  G It cannot possibly be a perfect example, since the situation for which  9 you claim it constitutes a perfect example did not exist.   3   Instead of focusing on the existing SCSI standard J > (remember that SCSI was in widespread use by Apple back in 1984 when the > MAC was launched),  F And whatever sufficed for the MAC would certainly have been more than ; sufficient to support a VAX cluster environment, of course.    Sheesh.   5   they decided that SCSI was not perfect and designed + > their own "improved" version called DSSI.   I Horseshit, as is so typical from you.  The problem was not that SCSI was  G 'not perfect', the problem was that SCSI did not support features that  K DEC systems needed (though I can't remember the specifics after this long).   H SCSI, if you'll remember, stood for "*small* computer system interface" D - and didn't even become an ANSI standard until 1986.  The standard I remained 8 bits X 5 MHz and 8 devices (including initiator) until SCSI-2  H came along in 1994 (doubling all three, which still seems a bit limited # for large VMS systems of that era).    > D > The very proprietary buses of the 80s was another issue of certainI > people trying to protect their revenues by preventing OEMs from selling  > to customers.   C Exactly what part of the difference between seeking perfection (an  D impossible and usually disastrous goal for an engineering firm) and H seeking proprietary advantage do you find difficult to understand?  The K latter may well have been inadvisable, but it is not the same thing at all.    ...   A >> No, it shows what he had learned over more than two decades of I >> outstanding success in his chosen field:  that innovative high-quality % >> products at fair prices sell well.  > I > When DEC started, the market was wide open, and not developped. DEC was J > able to shape it the way it wanted and the market adapted to DEC insteadJ > of the other way around. Since there was no competition, DEC could set aG > price and scientists would then ask for suffic8ient budget to buy DEC D > computers because they were THE thing they needed for their tasks. > C > DEC's success may not have been so much due to Olsen's vision and I > pricing stragegy, it may have been succesful DESPITE it because of lack  > of competition.   I And you may yet manage to achieve 0.1% of the success that he did, but I  I wouldn't bet on it - nor would I bet any more heavily on your assessment   above.  H That's the difference between people who actually accomplish things and 0 people who just babble about them incompetently.   > I > When there is no competition, you pay whatever DEC wants. When there is A > some competition such as DG, then the buyers must decide if the F > additional quality of DEC products warranted the additional purchase% > price. And for a long time, it did.   I In other words, you yourself can see that your previous suggestion (that  H DEC succeeded only due to lack of competition) was absolute garbage.  A D *lot* of competition developed starting around 1970 (DG, Pr1me, HP, E lower-end IBM product lines spring immediately to mind), and yet DEC  ? continued to forge ahead to become the second largest computer  , manufacturer in the world by the early '80s.   > I > There comes a time however where the effective quality difference (from D > customer's point of view) doesn't justify the added markup for DEC > products.   H Horseshit again.  If you had read the post to which you are responding, I you'd have noted that DEC didn't suffer due to added markup, it suffered  > because *the nature of the playing field changed* due to *the , introduction of new, disruptive technology*.  F DEC's problem was not markup, it was its inability to embrace the new I technology with its different price/performance/quality characteristics.  A   It continued to handle its *traditional* competition just fine.    ...   H > Consider as that recently as the introduction of DS15/25 machines, theG > argument was made by DECies that they cannot compete because they are - > costlier to build that equivalent machines.   = Cannot compete with *what*?  They certainly couldn't compete  H head-to-head with machines using x86 processors that *sold* for a small H fraction of what Alpha *cost* to build, and they were never designed to E (Alpha having long since given up that fight).  But they were priced  A reasonably for the people who needed something that x86 couldn't  C provide, and engineering likely sweated bullets to accomplish that.      You'd think that Palmer A > would have killed that mentality that everything had to be more I > expensive and would have motivated designers to make machines that were G > inexpensive to build (while maintaining good quality).  Consider that F > the DS15/25 were done during the Compaq/HP era where those engineersF > would have had access to Compaq/HP expertise on how to build quality > commodity hardware.   F *Alpha* is nothing approaching 'commodity hardware', nitwit.  A great H deal of the rest of the boxes *was* commodity hardware, though, and the  price reflected that.   B (Though that may have been more true for the DS10/20 than for the + DS15/25, if the latter used Rambus memory.)    ...   E >> quite different.  But it's not clear that even then DEC could have H >> *built* PCs at competitive prices:  it just didn't know how and would$ >> have found it difficult to learn  > < > That is the role of the leader of a corporation. He has toH > task/challenge the troups to find a way to build a competitive machine/ > that is the same price but of better quality.   E No, idiot:  that's only obvious in *hindsight*, after the market has  G become well-defined.  Before that, he had to guess whether expensive &  9 solid would beat out cheap & sloppy to define the market.   H Olsen, based on his existing (and extensive) experience in the industry H guessed that it would (just as American automakers thought at the start A of the Japanese onslaught, and they proved largely correct until  I Japanese quality improved a lot - and then, later, when the 'gas shocks'  D of the '70s changed the playing field in a more fundamental sense). E Unfortunately for DEC, PCs turned out to be a new market that didn't  @ function the same way that the existing minicomputer market did.   ...   (   It is often easier to sow the seeds toI > build a company and set it in a certain direction (which Olsen did very I > well), than it is to change the dirtection/mentality of a large company I > with lots of innertia and that has become set in its own ways and those  > ways are now the wrong way.   I You still don't get it.  DEC *didn't need to change from its roots*:  it  F just had to understand where it should compete and where it shouldn't G try to, rather than assume that it could (and should) be all things to  F all customers (something it actually never *had* tried to be until it E grew so large that it seemed to think this had become a requirement).   E It didn't matter that DEC couldn't build PCs competitively, any more  H than it mattered that it couldn't build the trucks that transported its I minicomputers competitively.  As things turned out it didn't even matter  D that it couldn't build low-cost workstations competitively, either: I that market got taken over two decades ago by the killer micros, but the  H high-end proprietary servers are still raking in a huge amount of dough  today.  D As is IBM, without PCs or workstations.  DEC was headed in the same C direction:  immense systems expertise that could put installations  H together using a mix of the equipment that DEC knew how to build better F than anyone else plus the (low-end) equipment that others knew how to E build better than DEC did.  And since DEC had a product set with far  F less sprawl than IBM's (but arguably similar reach in most areas), it H was positioned to compete very effectively with IBM (while virtually no  one else could).   > H > For all his faults, Palmer at least *tried* to change things early on.  D By encouraging customers to migrate from VMS to NT?  By selling off I unique strengths like Rdb?  What, exactly, did Palmer ever do other than  = slash & burn?  Where did he ever try to *create* opportunity?   I DEC was indeed a mess, and some parts needed cutting if they couldn't be  E repositioned into more productive activity.  Olsen would likely have  H tried the latter, and had he been able to straighten out the mess there 3 would have been useful work for those people to do.   C I don't know that he could have, but he sure had more (and better)  " vision than Palmer ever seemed to.  I A lot of us knew what was wrong with DEC back then - we just didn't know  G how to fix it.  The problem was that DEC had stopped trying to "do the  E right thing" (i.e., provide what customers wanted in some reasonable  F priority order based on available resources) and started trying to do C the *strategic* thing (i.e., figure out how to maximize profit and  I market share - which required continued, explosive growth without regard  I to whether it could ultimately be managed).  This change in attitude may  9 have gone hand-in-hand with the "What's in it for *me"?"  H middle-management attitude that someone else reported seeing (that was, I after all, the 'me' decade, though not for all of us) - at least the two  % seem somewhat related in my own view.   @ Olsen had the credibility to try to get the company back on its H traditional course, if that could be done.  Pfeiffer, as you suggested, G might have had the talent and experience to set a similar one.  Palmer   had nothing by comparison.   - bill   ------------------------------  % Date: Wed, 30 Aug 2006 15:36:13 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> = Subject: Re: Thoughts on the book: DEC is dead, long live DEC , Message-ID: <44F5E8A7.1F21E307@teksavvy.com>   Bill Todd wrote:I > SCSI, if you'll remember, stood for "*small* computer system interface" E > - and didn't even become an ANSI standard until 1986.  The standard J > remained 8 bits X 5 MHz and 8 devices (including initiator) until SCSI-2I > came along in 1994 (doubling all three, which still seems a bit limited % > for large VMS systems of that era).   ' My 1991 workstation (3100) has SCSI-2.    G The reason I chose DSSI as an example was that it was an example in the F 1990s of continued development of proprietary stuff.  Had DEC embracedG SCSI in the mid 1980s, it might have worked on elaborating the standard B in such a way that it could be used just as DSSI within a cluster. (which later happened anyways).   A Look at etherhet: DEC got involved and because it was an industry ? effort, it succeeded. It managed to kill off IBM's "token ring" H pathethic attempt at LANs with all those funky proprietary and expensive0 plugs.  DEC should have done the same with SCSI.  E > impossible and usually disastrous goal for an engineering firm) and E > seeking proprietary advantage do you find difficult to understand?    F And thsi is where DEC really failed the paradigm shift. The differenceD in quality between what DEC could do "proprietary" and what could beG obtained as commodity narrowed very fast and DEC decided to continue to D forge ahead with proprietary stuff which costed a lot more but whose@ advantages were less and less dramatic compared to the commodity industry standard stuff.  H Consider that these days, SCSI is considered expensive and IDE more than@ adequate for workstations, to a pont where even VMS supports it.    E > *lot* of competition developed starting around 1970 (DG, Pr1me, HP, F > lower-end IBM product lines spring immediately to mind), and yet DEC@ > continued to forge ahead to become the second largest computer. > manufacturer in the world by the early '80s.  B Because at the time, folks like DG couldn't approach the quality ,G features and performance of Digital. And because back then, the PDP was C the industry standard, just like the "PC" became industry standard.   E DG , Prime and the others did get customers and did survive, but they H got those customers who didn't need those features or whose budgets wereC so small that they could not afford the standard things (DEC gear).   E DEC was king and it didn't think that anyone could copete against its G own technically superior products and that the market would continue to D pay a premium for quality. DEC failed to realise that its premium noG longer matched the quality difference and that the market was no longer H willing to pay such a large premium for quality differences that were no longer so big.    J > you'd have noted that DEC didn't suffer due to added markup, it suffered? > because *the nature of the playing field changed* due to *the . > introduction of new, disruptive technology*.  E That disruptive technology is called COMPETITION.  PC, Risc, whatever G brought in machines with great performance at much lower price and with F capabilities that were enough for the vast majority. DEC was left withH only a subset of the new IT market, that subset that really needed those  few features unique to DEC gear.  D The same is happening with clustering now. The clustering offered byD Unix and even Windows is more than enough for many people who cannotD justify paying a premium for VMS clustering even though it is by farH still superior. You cannot really justify paying more for a product that0 bring your added features which you do not need.    ? > *Alpha* is nothing approaching 'commodity hardware', nitwit.    G Insults aside, Alpha could have been commodity. There is no reason that C they couldn't have built commodity PCs with "Alpha inside" had they D wanted to.  Again, this is a question of leadership. or lack thereof (this one under palmer).  J > You still don't get it.  DEC *didn't need to change from its roots*:  itG > just had to understand where it should compete and where it shouldn't H > try to, rather than assume that it could (and should) be all things toG > all customers (something it actually never *had* tried to be until it G > grew so large that it seemed to think this had become a requirement).     D Sorry, this is the very mentality that brought DEC down and the veryC mentality that is currently hindering VMS. If you feel that you can H retreat to your own small niche market of customers still willing to buyG your products, you have to expect your competitors to eventually invade F your niche and kill you. You cannot afford NOT to compete head to headH against the competition and try to keep as much market share as you can.  F If there is something that prevents your company from competing in theE environment of that day, then it is the responsability of the CEO and B top manmagement to set directives to make that compeny compete andB continue to grow instead of shrink itself to a small size that can' survive on a small number of customers.   J > unique strengths like Rdb?  What, exactly, did Palmer ever do other than? > slash & burn?  Where did he ever try to *create* opportunity?   H In the first 2 years of his helm, Palmer did make big changes to DigitalG that were not yet the "slash and burn". Rememeber that initially, there C were high hopes that he would make Digital lean and mean and get it H competitive again, especially with Alpha. My guess is that Palmer didn'tC even come close to succeeded sufficiently, gave up and went for the H slash/burn policy because at every quarter with disapointing results, he= thought that announcing more cutbacks would please the board.   G The problem with Palmer is that he was insecure with the board. Whereas E Olsen had had 100% freedom to operate his company for decades, Palmer G had a very short honeymoon period where he could show leadership, after H which, the board would demand results/actions. So Palmer didn't gain theG board's confidence and became a puppet of the board instead of a leader 
 of the board.   F But one has to be fair to Palmer, as much as I despise the guy, during= his first couple of years, he fixed up many things internally B (accounting etc). And he brought in new blood to try to change theA cultute to be more profit oriented. It wasn't enough, it wasn't a E success. But he did try. When it didn't work out, then Palmer started C his slash and burn. It is alwasy easier to destroy than to rebuild.   A > Olsen had the credibility to try to get the company back on its I > traditional course, if that could be done.  Pfeiffer, as you suggested, H > might have had the talent and experience to set a similar one.  Palmer > had nothing by comparison.    G I agree that Palmer wasn't even close to being CEO material. He was the D wrong choice for the job. When Compaq outsted Pfeiffer, we heard theG stories of Compaq seeking a new CEO and nobody wanting the job, so , in H the end, they had to give it to the accountant Curly.  I never heard anyE stories on how the DEC board came to hire Palmer as CEO. Was it board F incompetance ?  Was the board unable to find anyone of calibre to take
 on that job ?    ------------------------------  % Date: Wed, 30 Aug 2006 18:36:36 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> = Subject: Re: Thoughts on the book: DEC is dead, long live DEC < Message-ID: <44f61189$0$24196$9a6e19ea@news.newshosting.com>  ; "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message  % news:44F55360.7B68939@teksavvy.com...  >  [...snip...] > H > For all his faults, Palmer at least *tried* to change things early on.G > But he was nowhere near competant enough to fix DEC. And it is a real I > shame that Pfeiffer wasn't given more time to fix DEC, perhaps he could H > have breathed new life into DEC.  Curly, like Palmer, was in way aboveE > his head and not even close to be able to deal with DEC's problems.  > K I'm not saying Olsen was perfect but he did run DEC successfully from 1956  K to 1992 and only posted two losses. But passing control over to Palmer was  M like throwing gasoline on smouldering embers. A more technological guy could  J have turned the company around and my proof is the good job Hurd is doing  cleaning up after Carly.  F We've seen it before and we'll see it again. A technical guy starts a H technical company; gets the push and is replaced by a financial guy who M rides the company into the dirt. Lotus springs to mind but there are others.  K That is why healthy technology companies still have technology guys at the  @ top (Oracle = Ellison, RIM = Lazaridis, Microsoft = Gates, etc.)  K p.s. I wonder what will happen to SUN with McNealy shifting to chairman of  
 the board?  
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.9 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html     ------------------------------  % Date: Wed, 30 Aug 2006 21:55:00 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> = Subject: Re: Thoughts on the book: DEC is dead, long live DEC , Message-ID: <44F6416D.87B9FC9A@teksavvy.com>   Neil Rieck wrote: L > I'm not saying Olsen was perfect but he did run DEC successfully from 1956% > to 1992 and only posted two losses.   G I think there could be a big debate on the exact year when Olsen ceased  to run DEC succesfully.   H What is interesting in the book is that Olsen was, back in the 1989/1990E timeframe, very aware that things were wrong and needed change and he  did try to make changes.    H I'd have no problems saying Olsen rane DEC succesfully until 1980. But IE am not sure where between 1980 and 1990 that I would no longer say he  ran DEC succesfully.  G Just because DEC is still running on steam and still generating profits F doesn't mean that it is being runned succesfully. Decisions at the topP of (and lack thereof) sometimes can take years before there is a visible impact.    V > But passing control over to Palmer was like throwing gasoline on smouldering embers.  G It is a real shame that the book didn't expand on the selection process ( which culminated in Palmer being chosen.    > A more technological guy couldK > have turned the company around and my proof is the good job Hurd is doing  > cleaning up after Carly.  = DEC didn't need a "more technological guy". DEC needed a good F leader/manager/businessman. In the Gerstner book, there is quite a bitD about that. Gerstner at first refused the IBM job because he said heD wasn't qualified, but key board members worked very hard to convinceH Gerstner that he had what it took to lead IBM excactly because he wasn't< a techy and could be a neutral "business" judge of different technolological arguments.  G And remmeber that the DEC book mentioned many times that DEC lacked the D business gene. Hiring another techy may not have fixed that problem.    D In terms of Hurd, I am not sure if he is going to be a Capellas or aE Gerstner. I really haven't heard enouth of him to know if the guy has H any vision, leadership or convictions.  He seems to have done a good jobH or firing people and cutting costs.  But in a hypotetical scenario whereD you'd learn that 50% of VMS engineering had been laid off, would you, still think that Hurd was doing a good job ?  F Hurd is certaintly more serious than La Fiorina and not focused on hisB own fame. And he inherited an HP that had solid products that justE needed tweaking (except the enterprise servers where HP is now losing D market share). But because he is quiet, we don't really know how hisF decisions are impacting VMS, and those who are leaving VMS engineering< aren't exactly making much noise about why they are leaving.  G > We've seen it before and we'll see it again. A technical guy starts a I > technical company; gets the push and is replaced by a financial guy who " > rides the company into the dirt.  H There is a BIG difference between a financial guy (such as Capellas) and( a true business leader such as Gerstner.  E Look at palmer, a techy who was in totally over his head, tried a few H things, they didn't work and then reverted to slash and burn to sell theG company. Capellas also did the same, working hard to find a buying soon H after he officially got the job. (the bankers he hired sollicited IBM to buy Compaq).  E In the end, you need someone of great enough stature to have the full G confidence of the board to enact a serious plan that may take more than ? a quarter to yield benefits and which may involve taking risks.     L > p.s. I wonder what will happen to SUN with McNealy shifting to chairman of > the board?  E Linux is a true disrupive technology, more so than the PC was.  It is = making the OS moot and shifting the business to middlware and H applications. Consider that Oracle is now "more" available on Linux thanF on VMS. Despite VMS being better than Linux, Linux has become a betterD business solution because its software portofolio is growing whereas VMS' is shrinking.  G Sun has seen this and is *trying* to face Linux face to face. HP is not F yet reacting to Linux with regards to its own proprietary products. IfH the strategy is to be a hardware + service company, then HP doesn't careB about losing VMS, HP-UX , NSK if it continues to sell servers with2 Linux/Windows on it and the sell support services.    ? Coming back to Hurd, it is at least good that he didn't jump to C conclusions and started to permanently change HP. He made conscious E decision to not break it up.  But still not clear what his plans with C regares to enterprise servers and OS/software are.  If he waits too H long, it will be too late and he'll have missed the boat. If he acts too5 quickly, he can make harsh decisions that may be bad.    ------------------------------  % Date: Wed, 30 Aug 2006 22:55:14 -0400 ( From: Bill Todd <billtodd@metrocast.net>= Subject: Re: Thoughts on the book: DEC is dead, long live DEC G Message-ID: <EZednZrkspcP0mvZnZ2dnUVZ_rydnZ2d@metrocastcablevision.com>    JF Mezei wrote:  > Bill Todd wrote:J >> SCSI, if you'll remember, stood for "*small* computer system interface"F >> - and didn't even become an ANSI standard until 1986.  The standardK >> remained 8 bits X 5 MHz and 8 devices (including initiator) until SCSI-2 J >> came along in 1994 (doubling all three, which still seems a bit limited& >> for large VMS systems of that era). > ) > My 1991 workstation (3100) has SCSI-2.    E So after SCSI(-1) had actually become an ANSI standard, DEC not only  I embraced that but was willing to take a flyer on SCSI-2 even *before* it  I became an ANSI standard - hardly supporting your suggestion that DEC was  2 trying to tie everyone into proprietary solutions.   > I > The reason I chose DSSI as an example was that it was an example in the H > 1990s of continued development of proprietary stuff.  Had DEC embraced > SCSI in the mid 1980s,  G You mean as opposed to 1988, when it did start supporting SCSI on VAX?  E (Actually, http://www.hpl.hp.com/hpjournal/dtj/vol3num3/foreword.htm  D says 1989, but my impression is that DSSI and SCSI support appeared , together, and DSSI support is dated 1988 in B http://research.microsoft.com/~gbell/Digital/timeline/1988-5.htm )  2 > it might have worked on elaborating the standard  H IIRC it actually tried, but failed - again, taking the wind out of your / argument (and since it was all wind anyway...).   D > in such a way that it could be used just as DSSI within a cluster.! > (which later happened anyways).    No, it did not:   C "For the early product life of MicroVAX and VAX Q-bus systems, the  C DIGITAL Storage Systems Interconnection (DSSI) bus was the storage  G choice typically available from DIGITAL. While the then-current SCSI-1  A bus was about the same performance as the DSSI storage bus, DSSI  I provided dual- and triple-host support, SCS and MSCP capabilities, and a  A number of other capabilities well beyond SCSI. Accordingly, SCSI  G connectivity was a comparatively late feature in Q-bus configurations.  I (Some of the capabilities found in DSSI still do not exist in SCSI. Ever  B SET HOST into a disk controller to directly adjust its settings?)"  5 http://hoffmanlabs.org/openvms/hwvax/hwqbus.shtml (!)   G In fact, even now SCSI is supported only for storage (rather than host  * <-> host) VMS cluster use - see page 3 of 2 http://h18000.www1.hp.com/info/SP2978/SP2978PF.PDF  E While at some point SCSI did grow to support dual (and possibly even  @ more) initiators, ISTR that Tru64 discovered some of these SCSI C limitations when it was starting to investigate fail-over support,  8 though it managed to kludge its way around some of them.   ...   F >> impossible and usually disastrous goal for an engineering firm) andF >> seeking proprietary advantage do you find difficult to understand?  > 9 > And thsi is where DEC really failed the paradigm shift.   H Stop babbling and pay attention for once, JF:  the point above was that I you completely misrepresented one kind of behavior (trying to tie people  F into proprietary technology, although even that assertion is becoming I increasingly dubious since DEC started supporting SCSI soon after it had  A been accepted as an ANSI standard) for another (trying to attain   perfection).   ...   K >> you'd have noted that DEC didn't suffer due to added markup, it suffered @ >> because *the nature of the playing field changed* due to *the/ >> introduction of new, disruptive technology*.  > 3 > That disruptive technology is called COMPETITION.   I But it's a *qualitatively different form* of competition than the simple  E mark-up practices that you stated were responsible, idiot:  that new  I competition could have had mark-ups exceeding DEC's and still shouldered  2 its way in because its *costs* were so much lower.   ...   F > The same is happening with clustering now. The clustering offered byF > Unix and even Windows is more than enough for many people who cannotF > justify paying a premium for VMS clustering even though it is by far > still superior.   
 Horseshit:  F 1.  VMS clustering is by no means 'far superior' in many contexts (in E some ways it's still at least somewhat superior, but what you pay in  G terms of committing to a niche platform that may be on its way out may   far outweigh that).   H 2.  VMS isn't priced to compete because it's a cash cow, not because it + would be impossible to price it to compete.    ...   @ >> *Alpha* is nothing approaching 'commodity hardware', nitwit.  > 1 > Insults aside, Alpha could have been commodity.   D *BUT IT WASN'T*, idiot.  Your argument was not hypothetical in some B never-never land, it was about current reality - and that reality F includes the fact that Alpha is nothing like a commodity product with  commodity costs and pricing.   ...   K >> You still don't get it.  DEC *didn't need to change from its roots*:  it H >> just had to understand where it should compete and where it shouldn'tI >> try to, rather than assume that it could (and should) be all things to H >> all customers (something it actually never *had* tried to be until itH >> grew so large that it seemed to think this had become a requirement). >  > F > Sorry, this is the very mentality that brought DEC down and the very, > mentality that is currently hindering VMS.  H You really are a fucking idiot, JF.  Fortunately, most people recognize B this and ignore you, but every once in a while it's worth letting  newcomers in on the secret.      If you feel that you canJ > retreat to your own small niche market of customers still willing to buy > your products,  G It's not about retreating to a niche, moron:  it's about understanding  G what your strengths and weaknesses are and trying to capitalize on the  : former while minimizing the negative impact of the latter.  F DEC didn't have to *produce* PCs to make a lot of money off them:  it 0 just had to figure out a way to *leverage* them.  :   you have to expect your competitors to eventually invade > your niche and kill you.  ? It's been another 20+ years now and that 'niche' is still very  I lucrative, by the way (not that I was suggesting that DEC confine itself    there, as explained just above).   ...   K >> unique strengths like Rdb?  What, exactly, did Palmer ever do other than @ >> slash & burn?  Where did he ever try to *create* opportunity? > J > In the first 2 years of his helm, Palmer did make big changes to Digital) > that were not yet the "slash and burn".    Such as?   ...      during? > his first couple of years, he fixed up many things internally  > (accounting etc).   G Assuming for the moment that accounting was a real problem, that's one.   2   And he brought in new blood to try to change the% > cultute to be more profit oriented.   H Horseshit.  The culture was *already* too 'profit oriented', and adding F another dose of 'MBA disease' wasn't going to help.  It needed to get < back to being *customer* oriented and *competency* oriented.   - bill   ------------------------------  % Date: Wed, 30 Aug 2006 23:09:44 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> = Subject: Re: Thoughts on the book: DEC is dead, long live DEC + Message-ID: <44F652EC.F6CAE77@teksavvy.com>    Bill Todd wrote:D > "For the early product life of MicroVAX and VAX Q-bus systems, theD > DIGITAL Storage Systems Interconnection (DSSI) bus was the storage* > choice typically available from DIGITAL.  @ Early product life ?.  DSSI came to vax rather late in the game.G Microvaxen existed well before DSSI came into being. Remember RA drives D ? Remember RD53 and RD54 drives ? SCSI controllers for Q-BUS systemsF came from 3rd parties before DEC had DSSI. AND DEC's SCSI controllers,7 when they did come, were rather brain dead and limited.   1 DSSI came at about the time of 4000 series vaxes.   1 > Stop babbling and pay attention for once, JF:   & > You really are a fucking idiot, JF. / > It's not about retreating to a niche, moron:    # Your professionalism is impressive.    ------------------------------  % Date: Thu, 31 Aug 2006 02:11:32 -0400 ( From: Bill Todd <billtodd@metrocast.net>= Subject: Re: Thoughts on the book: DEC is dead, long live DEC G Message-ID: <1smdnVhDpcAJ4GvZnZ2dnUVZ_tidnZ2d@metrocastcablevision.com>    JF Mezei wrote:  > Bill Todd wrote:E >> "For the early product life of MicroVAX and VAX Q-bus systems, the E >> DIGITAL Storage Systems Interconnection (DSSI) bus was the storage + >> choice typically available from DIGITAL.  > B > Early product life ?.  DSSI came to vax rather late in the game.6 > Microvaxen existed well before DSSI came into being.  F You'll really have to take Hoff's wording up with him:  I just quoted H it.  Of course, since VAXen were built right up through Y2K, *anything* I in the '80s could in some ways have been considered their 'early life'...    ...   2 >> Stop babbling and pay attention for once, JF:  ' >> You really are a fucking idiot, JF.  0 >> It's not about retreating to a niche, moron:  > % > Your professionalism is impressive.   E I only treat people like professionals who *act* like professionals.  I Whereas I tend to treat persistent, loud-mouthed idiots like persistent,   loud-mouthed idiots.   - bill   ------------------------------  # Date: Thu, 31 Aug 2006 12:38:12 GMT 5 From: rdeininger@mindspringdot.com (Robert Deininger) = Subject: Re: Thoughts on the book: DEC is dead, long live DEC Y Message-ID: <rdeininger-3108060838100001@dialup-4.233.173.7.dial1.manchester1.level3.net>   G In article <tsGdnV26FoHuGGjZnZ2dnUVZ_oudnZ2d@metrocastcablevision.com>, ) Bill Todd <billtodd@metrocast.net> wrote:    >JF Mezei wrote:   ...    >>  You'd think that Palmer B >> would have killed that mentality that everything had to be moreJ >> expensive and would have motivated designers to make machines that wereH >> inexpensive to build (while maintaining good quality).  Consider thatG >> the DS15/25 were done during the Compaq/HP era where those engineers G >> would have had access to Compaq/HP expertise on how to build quality  >> commodity hardware. > G >*Alpha* is nothing approaching 'commodity hardware', nitwit.  A great  I >deal of the rest of the boxes *was* commodity hardware, though, and the   >price reflected that.  G Commodity hardware to some extent.  But still very much AlphaServers in D design and philosophy.  AlphaServer design was evolving toward lower; costs, but neither merger brought sudden change or a lot of  cross-pollination.  C >(Though that may have been more true for the DS10/20 than for the  , >DS15/25, if the latter used Rambus memory.)  H ES45/DS25/DS15 use pretty standard off-the-shelf memory.  Not Rambus.  IE don't know the specs off the top of my head.  Folks like David Turner 
 likely do.   ------------------------------    Date: 31 Aug 2006 07:59:07 -0700- From: "Andrew" <andrew_harrison@symantec.com> = Subject: Re: Thoughts on the book: DEC is dead, long live DEC A Message-ID: <1157036347.395618.58000@e3g2000cwe.googlegroups.com>    Bill Todd wrote: > JF Mezei wrote:  > > Bill Todd wrote:E > >> Perhaps what you were trying to say was that there was a quality M > >> threshold below which a product was considered unacceptable (i.e., below A > >> which quality would not be further traded off for low cost).  > >  > > K > > In the book, the image is given that DEC engineers had carte blanche to * > > develop superior products at any cost. > K > Then either the book or your perception of what it portrays is incorrect.  > " >   I think that DSSI is a perfect > > example of this. > H > It cannot possibly be a perfect example, since the situation for which; > you claim it constitutes a perfect example did not exist.  > 5 >   Instead of focusing on the existing SCSI standard L > > (remember that SCSI was in widespread use by Apple back in 1984 when the > > MAC was launched), > G > And whatever sufficed for the MAC would certainly have been more than = > sufficient to support a VAX cluster environment, of course.  > 	 > Sheesh.  > 7 >   they decided that SCSI was not perfect and designed - > > their own "improved" version called DSSI.  > J > Horseshit, as is so typical from you.  The problem was not that SCSI wasH > 'not perfect', the problem was that SCSI did not support features thatM > DEC systems needed (though I can't remember the specifics after this long).  > I > SCSI, if you'll remember, stood for "*small* computer system interface" E > - and didn't even become an ANSI standard until 1986.  The standard J > remained 8 bits X 5 MHz and 8 devices (including initiator) until SCSI-2I > came along in 1994 (doubling all three, which still seems a bit limited % > for large VMS systems of that era).  >  > > F > > The very proprietary buses of the 80s was another issue of certainK > > people trying to protect their revenues by preventing OEMs from selling  > > to customers.  > D > Exactly what part of the difference between seeking perfection (anE > impossible and usually disastrous goal for an engineering firm) and I > seeking proprietary advantage do you find difficult to understand?  The M > latter may well have been inadvisable, but it is not the same thing at all.  >  > ...  > C > >> No, it shows what he had learned over more than two decades of K > >> outstanding success in his chosen field:  that innovative high-quality ' > >> products at fair prices sell well.  > > K > > When DEC started, the market was wide open, and not developped. DEC was L > > able to shape it the way it wanted and the market adapted to DEC insteadL > > of the other way around. Since there was no competition, DEC could set aI > > price and scientists would then ask for suffic8ient budget to buy DEC F > > computers because they were THE thing they needed for their tasks. > > E > > DEC's success may not have been so much due to Olsen's vision and K > > pricing stragegy, it may have been succesful DESPITE it because of lack  > > of competition.  > J > And you may yet manage to achieve 0.1% of the success that he did, but IJ > wouldn't bet on it - nor would I bet any more heavily on your assessment > above. > I > That's the difference between people who actually accomplish things and 2 > people who just babble about them incompetently. >  > > K > > When there is no competition, you pay whatever DEC wants. When there is C > > some competition such as DG, then the buyers must decide if the H > > additional quality of DEC products warranted the additional purchase' > > price. And for a long time, it did.  > J > In other words, you yourself can see that your previous suggestion (thatI > DEC succeeded only due to lack of competition) was absolute garbage.  A E > *lot* of competition developed starting around 1970 (DG, Pr1me, HP, F > lower-end IBM product lines spring immediately to mind), and yet DEC@ > continued to forge ahead to become the second largest computer. > manufacturer in the world by the early '80s. >  > > K > > There comes a time however where the effective quality difference (from F > > customer's point of view) doesn't justify the added markup for DEC
 > > products.  > I > Horseshit again.  If you had read the post to which you are responding, J > you'd have noted that DEC didn't suffer due to added markup, it suffered? > because *the nature of the playing field changed* due to *the . > introduction of new, disruptive technology*. > G > DEC's problem was not markup, it was its inability to embrace the new J > technology with its different price/performance/quality characteristics.C >   It continued to handle its *traditional* competition just fine.  >  > ...  > J > > Consider as that recently as the introduction of DS15/25 machines, theI > > argument was made by DECies that they cannot compete because they are / > > costlier to build that equivalent machines.  > > > Cannot compete with *what*?  They certainly couldn't competeI > head-to-head with machines using x86 processors that *sold* for a small I > fraction of what Alpha *cost* to build, and they were never designed to F > (Alpha having long since given up that fight).  But they were pricedB > reasonably for the people who needed something that x86 couldn'tE > provide, and engineering likely sweated bullets to accomplish that.   
 Not quite.  F The DS15 with 1x 1GHz Alpha  with 1GB of RAM and 1 internal drive with Tru64 costs $10900  E The IBM P5 505 1 x 1.65 GHz POWER 5, 1GB of RAM and 2 internal drives  and AIX costs $3500.  D The Sun V210 1 x 1.34 GHz SPARC 1 GB or RAM and 1 internal drive and Solaris costs $3400.  F The HP rx1620 with 1.3 GHz Itanium II and 1 GB of RAM, 1 internal disk and HP-UX costs $6300   G The Sun, HP rx1620 and IBM p5 505 are all faster than the DS on SPECint  and SPECfp.   D The DS is the only single CPU system all the others support 2 CPU's.E The DS supports the smallest amount or RAM 4GB and while it has 4 PCI A slots more than any of the other units it has a poorer basic spec 1 (ethernet etc). It is 3RU and the others are 1RU.   G The 3RU vs 1RU is a pretty serious issue which would rule the DS out of G many HPC bids simply on the space it would require compared with POWER.  Itanium  and x86 alternatives.  B For the price of 1 DS you could buy 3x IBM or SPARC boxes and they9 would deliver >3x the throughput in the same form factor.    Regards  Andrew Harrison      >  >   You'd think that Palmer C > > would have killed that mentality that everything had to be more K > > expensive and would have motivated designers to make machines that were I > > inexpensive to build (while maintaining good quality).  Consider that H > > the DS15/25 were done during the Compaq/HP era where those engineersH > > would have had access to Compaq/HP expertise on how to build quality > > commodity hardware.  > G > *Alpha* is nothing approaching 'commodity hardware', nitwit.  A great I > deal of the rest of the boxes *was* commodity hardware, though, and the  > price reflected that.  > C > (Though that may have been more true for the DS10/20 than for the - > DS15/25, if the latter used Rambus memory.)  >  > ...  > G > >> quite different.  But it's not clear that even then DEC could have J > >> *built* PCs at competitive prices:  it just didn't know how and would% > >> have found it difficult to learn  > > > > > That is the role of the leader of a corporation. He has toJ > > task/challenge the troups to find a way to build a competitive machine1 > > that is the same price but of better quality.  > F > No, idiot:  that's only obvious in *hindsight*, after the market hasH > become well-defined.  Before that, he had to guess whether expensive &; > solid would beat out cheap & sloppy to define the market.  > I > Olsen, based on his existing (and extensive) experience in the industry I > guessed that it would (just as American automakers thought at the start B > of the Japanese onslaught, and they proved largely correct untilJ > Japanese quality improved a lot - and then, later, when the 'gas shocks'E > of the '70s changed the playing field in a more fundamental sense). F > Unfortunately for DEC, PCs turned out to be a new market that didn'tB > function the same way that the existing minicomputer market did. >  > ...  > * >   It is often easier to sow the seeds toK > > build a company and set it in a certain direction (which Olsen did very K > > well), than it is to change the dirtection/mentality of a large company K > > with lots of innertia and that has become set in its own ways and those  > > ways are now the wrong way.  > J > You still don't get it.  DEC *didn't need to change from its roots*:  itG > just had to understand where it should compete and where it shouldn't H > try to, rather than assume that it could (and should) be all things toG > all customers (something it actually never *had* tried to be until it G > grew so large that it seemed to think this had become a requirement).  > F > It didn't matter that DEC couldn't build PCs competitively, any moreI > than it mattered that it couldn't build the trucks that transported its J > minicomputers competitively.  As things turned out it didn't even matterE > that it couldn't build low-cost workstations competitively, either: J > that market got taken over two decades ago by the killer micros, but theI > high-end proprietary servers are still raking in a huge amount of dough  > today. > E > As is IBM, without PCs or workstations.  DEC was headed in the same D > direction:  immense systems expertise that could put installationsI > together using a mix of the equipment that DEC knew how to build better G > than anyone else plus the (low-end) equipment that others knew how to F > build better than DEC did.  And since DEC had a product set with farG > less sprawl than IBM's (but arguably similar reach in most areas), it I > was positioned to compete very effectively with IBM (while virtually no  > one else could). >  > > J > > For all his faults, Palmer at least *tried* to change things early on. > E > By encouraging customers to migrate from VMS to NT?  By selling off J > unique strengths like Rdb?  What, exactly, did Palmer ever do other than? > slash & burn?  Where did he ever try to *create* opportunity?  > J > DEC was indeed a mess, and some parts needed cutting if they couldn't beF > repositioned into more productive activity.  Olsen would likely haveI > tried the latter, and had he been able to straighten out the mess there 5 > would have been useful work for those people to do.  > D > I don't know that he could have, but he sure had more (and better)$ > vision than Palmer ever seemed to. > J > A lot of us knew what was wrong with DEC back then - we just didn't knowH > how to fix it.  The problem was that DEC had stopped trying to "do theF > right thing" (i.e., provide what customers wanted in some reasonableG > priority order based on available resources) and started trying to do D > the *strategic* thing (i.e., figure out how to maximize profit andJ > market share - which required continued, explosive growth without regardJ > to whether it could ultimately be managed).  This change in attitude may: > have gone hand-in-hand with the "What's in it for *me"?"I > middle-management attitude that someone else reported seeing (that was, J > after all, the 'me' decade, though not for all of us) - at least the two' > seem somewhat related in my own view.  > A > Olsen had the credibility to try to get the company back on its I > traditional course, if that could be done.  Pfeiffer, as you suggested, H > might have had the talent and experience to set a similar one.  Palmer > had nothing by comparison. >  > - bill   ------------------------------    Date: 31 Aug 2006 09:12:34 -0700- From: "Andrew" <andrew_harrison@symantec.com> = Subject: Re: Thoughts on the book: DEC is dead, long live DEC B Message-ID: <1157040754.683276.73180@b28g2000cwb.googlegroups.com>   Neil Rieck wrote: < > "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message' > news:44F55360.7B68939@teksavvy.com...  > >  > [...snip...] > > J > > For all his faults, Palmer at least *tried* to change things early on.I > > But he was nowhere near competant enough to fix DEC. And it is a real K > > shame that Pfeiffer wasn't given more time to fix DEC, perhaps he could J > > have breathed new life into DEC.  Curly, like Palmer, was in way aboveG > > his head and not even close to be able to deal with DEC's problems.  > > L > I'm not saying Olsen was perfect but he did run DEC successfully from 1956L > to 1992 and only posted two losses. But passing control over to Palmer wasN > like throwing gasoline on smouldering embers. A more technological guy couldK > have turned the company around and my proof is the good job Hurd is doing  > cleaning up after Carly. > G > We've seen it before and we'll see it again. A technical guy starts a I > technical company; gets the push and is replaced by a financial guy who N > rides the company into the dirt. Lotus springs to mind but there are others.L > That is why healthy technology companies still have technology guys at theB > top (Oracle = Ellison, RIM = Lazaridis, Microsoft = Gates, etc.) > L > p.s. I wonder what will happen to SUN with McNealy shifting to chairman of > the board?  E They seem to be doing Ok. They posted a 29% revenue hike for the last G quarter and were profitable before charges. They also grew market share # in the UNIX and x86 Server markets.   D Its a bit difficult to attribute this improvement to his replacementG who has only been onboard for 90 days. However in reality he was making ? many of the strategy decisions before taking over from McNealy.    Regards  Andrew Harrison  >  > Neil Rieck > Kitchener/Waterloo/Cambridge,  > Ontario, Canada.: > http://www3.sympatico.ca/n.rieck/links/cool_openvms.html   ------------------------------    Date: 31 Aug 2006 10:17:56 -0700- From: "Doug Phillips" <dphill46@netscape.net> = Subject: Re: Thoughts on the book: DEC is dead, long live DEC C Message-ID: <1157044676.448111.137240@b28g2000cwb.googlegroups.com>   
 Andrew wrote:  > Bill Todd wrote: > > JF Mezei wrote: L > > > Consider as that recently as the introduction of DS15/25 machines, theK > > > argument was made by DECies that they cannot compete because they are 1 > > > costlier to build that equivalent machines.  > > @ > > Cannot compete with *what*?  They certainly couldn't competeK > > head-to-head with machines using x86 processors that *sold* for a small K > > fraction of what Alpha *cost* to build, and they were never designed to H > > (Alpha having long since given up that fight).  But they were pricedD > > reasonably for the people who needed something that x86 couldn'tG > > provide, and engineering likely sweated bullets to accomplish that.  >  > Not quite. > H > The DS15 with 1x 1GHz Alpha  with 1GB of RAM and 1 internal drive with > Tru64 costs $10900 >   C For VMS base, make that closer to $12,000. Now, to bring it in line E with what comes with Tru64 we need to add unlimited users for $12,000 E more and NAS-300 for $16,643. Or maybe we only need 32 users for only B $6,739 and NAS-250 for only $12,231 additional. No doubt we forgotF something, like the fact that we need to exist in a Microsoft network,G so better allocate a few thousand more. Anyway, consider a pretty basic # DS15 will likely cost over $30,000.   E Can you get it cheaper? Sure. But not enough to make anyone choose to / move away from another smaller system platform.   G IA64 licensing is finally in line with reality; hope it's not too late.     G > The IBM P5 505 1 x 1.65 GHz POWER 5, 1GB of RAM and 2 internal drives  > and AIX costs $3500. > F > The Sun V210 1 x 1.34 GHz SPARC 1 GB or RAM and 1 internal drive and > Solaris costs $3400. > H > The HP rx1620 with 1.3 GHz Itanium II and 1 GB of RAM, 1 internal disk > and HP-UX costs $6300  > I > The Sun, HP rx1620 and IBM p5 505 are all faster than the DS on SPECint 
 > and SPECfp.  > F > The DS is the only single CPU system all the others support 2 CPU's.G > The DS supports the smallest amount or RAM 4GB and while it has 4 PCI C > slots more than any of the other units it has a poorer basic spec 3 > (ethernet etc). It is 3RU and the others are 1RU.  > I > The 3RU vs 1RU is a pretty serious issue which would rule the DS out of I > many HPC bids simply on the space it would require compared with POWER.   > Itanium  and x86 alternatives. > D > For the price of 1 DS you could buy 3x IBM or SPARC boxes and they; > would deliver >3x the throughput in the same form factor.  >   G Or, if you were running OpenVMS on, say an AS1000, and were looking for E an upgrade to the DS15 level, the 6 to 8 times price difference could ? justify the cost of moving to one of the non-OpenVMS platforms.   G Not everyone can move to IA64 now because some software isn't yet there G (or at least, not ready for prime-time) or might never be. Much of that B software is now and has been available for the aforementioned *nix; platforms. Pretty obvious stuff, except to DECHPQ, I guess.    ------------------------------    Date: 27 Aug 2006 17:49:50 -0700 From: tomarsin2015@comcast.net Subject: Unix to VMS usersC Message-ID: <1156726190.567870.241600@m73g2000cwd.googlegroups.com>    Hello D Does anybody have/know of a good Unix to VMS login script. What I am looking for isE something that when a former/new Unix user logs in, they will feel at 
 home in a VMS  system.  Thanks   ------------------------------    Date: 27 Aug 2006 20:01:58 -0700/ From: "David B Sneddon" <dbsneddon@bigpond.com>  Subject: Re: Unix to VMS usersC Message-ID: <1156734118.631020.141570@i42g2000cwa.googlegroups.com>    tomarsin2015@comcast.net wrote:  > Hello F > Does anybody have/know of a good Unix to VMS login script. What I am > looking for isG > something that when a former/new Unix user logs in, they will feel at  > home in a VMS 	 > system.  > Thanks   Which *nix? Which shell? Why?    Dave   ------------------------------  # Date: Mon, 28 Aug 2006 15:24:53 GMT " From:   VAXman-  @SendSpamHere.ORG Subject: Re: Unix to VMS users0 Message-ID: <00A5AE1D.0BDAADAD@SendSpamHere.ORG>  p In article <ca-dnSpD4Z94nG7ZnZ2dnUVZ_vKdnZ2d@comcast.com>, "Richard B. Gilbert" <rgilbert88@comcast.net> writes: >  >   >tomarsin2015@comcast.net wrote: >> HelloG >> Does anybody have/know of a good Unix to VMS login script. What I am  >> looking for is H >> something that when a former/new Unix user logs in, they will feel at >> home in a VMS
 >> system.	 >> Thanks  >>   > 
 >You mean: > 
 >$ LS :== DIR  >$ LP :== PRINT  >$ CD :== SET DEFAULT  >$ MKDIR :== CREATE /DIRECTORY >$ PS :== SHOW SYSTEM 4 >$ GREP :== SEARCH   ! My VMS port of grep is better
 >$ . . . . > J >Roll your own.  When the former Unix users ask "How do I . . ." then add  >to it.  > D >(The VMS port of grep was mostly adding include files and function D >declarations to get the compiler to shut up its complaining. . . .) > J >Or you could just tell them to use HELP.  It's quite practical for a new  >VMS user to learn via HELP.  I I cringe at the thought of people doing this just as I don't try to alias 1 unix shell commands with VMS/DCL command names.     K I think it's best to learn the nuances and differences of each environment. K When you create symbols like CD, the user will expect the command syntax to J be the same and will try something like CD /etc/local/bin and then get an  error like:   G %DCL-W-NOQUAL, qualifiers not allowed - supply only verb and parameters  \USR\   " and say, "WTF! This VMS unix sux!"     --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Mon, 28 Aug 2006 10:40:16 -0500 % From: Dan Foster <usenet@evilphb.org>  Subject: Re: Unix to VMS users5 Message-ID: <slrnef63j0.hok.usenet@zappy.catbert.org>   m In article <00A5AE1D.0BDAADAD@SendSpamHere.ORG>, VAXman-  @SendSpamHere.ORG <VAXman-@SendSpamHere.ORG> wrote: [ > In article <ca-dnSpD4Z94nG7ZnZ2dnUVZ_vKdnZ2d@comcast.com> tomarsin2015@comcast.net wrote: G >> Does anybody have/know of a good Unix to VMS login script. What I am E >> looking for is something that when a former/new Unix user logs in, * >> they will feel at home in a VMS system.  K > I cringe at the thought of people doing this just as I don't try to alias 3 > unix shell commands with VMS/DCL command names.    > M > I think it's best to learn the nuances and differences of each environment.   2 Sound advice, and worth the slight learning curve.  F At best, UNIX-like symbols in VMS is a crutch. Really doesn't take too9 long to learn VMS basics because the OS is well designed.   F My only minor concession is to use an enhanced 'CD' (change directory)A utility on VMS because the freeware one I have is quite powerful.   D Otherwise, there is no trace of UNIX-isms in my VMS setups. But I'veD also used VMS long enough to be comfortable using 'SET DEFAULT' (VMS* equivalent of UNIX's cd) without thinking.  G I did, however, also put in an alias in my UNIX shell to copy one of my  VMS symbols. :)    -Dan   ------------------------------    Date: 28 Aug 2006 11:53:20 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: Unix to VMS users3 Message-ID: <KJnvu1RxcH0x@eisner.encompasserve.org>   d In article <1156726190.567870.241600@m73g2000cwd.googlegroups.com>, tomarsin2015@comcast.net writes: > Hello F > Does anybody have/know of a good Unix to VMS login script. What I am > looking for isG > something that when a former/new Unix user logs in, they will feel at  > home in a VMS 	 > system.       Just give htem this one:   *    $define/user_mode sys$input sys$command    $help   ------------------------------  # Date: Mon, 28 Aug 2006 19:28:25 GMT A From: "Charlie McCutcheon" <charlie.mccutcheon.removetext@hp.com>  Subject: Re: Unix to VMS users- Message-ID: <ttHIg.26$NK6.5@news.cpqcorp.net>   B More information: http://h71000.www7.hp.com/portability/GNV.html .   ------------------------------  # Date: Mon, 28 Aug 2006 18:59:23 GMT A From: "Charlie McCutcheon" <charlie.mccutcheon.removetext@hp.com>  Subject: Re: Unix to VMS users. Message-ID: <f2HIg.23$NK6.10@news.cpqcorp.net>  J >What I am looking for is something that when a former/new Unix user logs  >in,( >they will feel at home in a VMS system.  E See GNV, an open source series of tools for VMS which implement Unix  E commands at http://h71000.www7.hp.com/opensource/opensource.html#gnv.   I This was started by a 3rd party vendor for their Unix based build system  J when they ported to VMS.  HP employees have since been contributing to it.   Charlie    ------------------------------  % Date: Sun, 27 Aug 2006 10:47:28 -0400 7 From: "Tom Simpson" <thomas.simpson1@fubar.comcast.net> 8 Subject: Unix/VMS Admin Job Opening in Jacksonville, FL.: Message-ID: <p86dnYn_JvUeLWzZnZ2dnUVZ_sOdnZ2d@comcast.com>  M We have a job opening for a UNIX/VMS system administrator at the Mayo Clinic   in Jacksonville, FL.  3 The job number is 3843.  Apply on Mayo's web site:  : http://www.mayoclinic.org/jobs-jax/ if you are interested.  - The duties are typical system manager duties:  - monitoring - tuning - problem solving  - system updates( - disaster recovery planning and testing  L You would be required to be on-call after hours for one week every 5 weeks. # The after-hours call load is light.   C We have capable VMS managers, but marginally qualified Unix system  L administrators, so our focus is on Unix admin skills for this position.  We B have a wide variety of equipment from GS1280 to MicroVAX systems. E Experience in SAN environments is also a plus.  The Unix systems are  F typically smaller dedicated application systems running Red Hat Linux.  J I'm not the hiring manager, I work in this group. I'm just trying to help I out by finding qualified candidates.  Once you read the job description,  = you'll understand why we are not getting many applications...    Regards, Tom Simpson    ------------------------------    Date: 27 Aug 2006 13:37:56 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) < Subject: Re: Unix/VMS Admin Job Opening in Jacksonville, FL.3 Message-ID: <i3aZyHtz3X1V@eisner.encompasserve.org>   t In article <p86dnYn_JvUeLWzZnZ2dnUVZ_sOdnZ2d@comcast.com>, "Tom Simpson" <thomas.simpson1@fubar.comcast.net> writes:  5 > The job number is 3843.  Apply on Mayo's web site:  < > http://www.mayoclinic.org/jobs-jax/ if you are interested.  K > out by finding qualified candidates.  Once you read the job description,  ? > you'll understand why we are not getting many applications...   F Presumably you mean the part about requiring people to "register" just to read the job description.   ------------------------------  % Date: Sun, 27 Aug 2006 15:06:59 -0400  From: "Ray" <no@spam.me>< Subject: Re: Unix/VMS Admin Job Opening in Jacksonville, FL.- Message-ID: <u3mIg.1289$md5.342@newsfe02.lga>   H > Presumably you mean the part about requiring people to "register" just > to read the job description.  E If you click on [Not Registered Yet] you can search and look at their  openings without registering.   , I didn't register. Here is the whole posting  # Category  *SEARCH ALL CATEGORIES/IT  Location  Jacksonville Work Status  40 hrs/week   Systems Administrator - 3843   Description   G Technical resource performing advanced systems support and maintenance, G integration testing and configuration, and problems escalation. Project : resource providing system and cost analysis documentation.  * Department: Technical Engineering Services  K Qualifications: Associates Degree required. Bachelors Degree and MCSE, CCNA @ or other IT related certifications preferred. 5 years experienceL administering a multiple operating system environment and implementing IS/ITL projects. An additional 4 years (9 total years) relavent experience required  in lieu of an Associates Degree.   Benefit Eligibility: Yes   Posting Begin Date: 06/01/2006   Posting End Date: Until Filled  " Employment Type: Full-time/ Exempt   Hours/2-Weeks: 80   K Schedule: Monday-Friday, some weekends and evenings required. Participation " in 24x7 on-call rotation required.   Shift: Days    Building: Stabile    Salary:    ------------------------------  % Date: Sun, 27 Aug 2006 19:26:29 -0400 7 From: "Tom Simpson" <thomas.simpson1@fubar.comcast.net> < Subject: Re: Unix/VMS Admin Job Opening in Jacksonville, FL.: Message-ID: <l6ydnUpv6IKzt2_ZnZ2dnUVZ_omdnZ2d@comcast.com>  M As I alluded to in my original post, the job description is worthless.  I'll  K post the job the job description from the web site here if you really want  G to see it.  If you are not interested in the job, it's a waste of time.   J The MCSE and CCNA is total BS and not at all useful or necessary for this L job...  Someone simply cut and paste the description from another unrelated  job posting.   Regards, Tom   ; "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message  - news:i3aZyHtz3X1V@eisner.encompasserve.org... K > In article <p86dnYn_JvUeLWzZnZ2dnUVZ_sOdnZ2d@comcast.com>, "Tom Simpson"  - > <thomas.simpson1@fubar.comcast.net> writes:  > 5 >> The job number is 3843.  Apply on Mayo's web site: = >> http://www.mayoclinic.org/jobs-jax/ if you are interested.  > K >> out by finding qualified candidates.  Once you read the job description, @ >> you'll understand why we are not getting many applications... > H > Presumably you mean the part about requiring people to "register" just > to read the job description.     ------------------------------  % Date: Sun, 27 Aug 2006 20:43:07 +0100 0 From: Tony Arnold <tony.arnold@manchester.ac.uk>7 Subject: VAX 4000/600 free to good home - Manchester UK = Message-ID: <pan.2006.08.27.19.43.06.348906@manchester.ac.uk>   I I have a VAX 4000 model 600 surplus to requirements which I would like to I see go to a good home rather than be scrapped. I'm in Manchester, UK. The H machine is rather bulky, so any takers would have to come and collect it" or arrange shipping at their cost.  F It has something like 512MB of memory and three DSSI disks and a TK-70I drive all mounted in the same chassis. I don't have access to the machine C right now, so I'm going by memory here. Actual specification may be 
 different.  J The only known fault that I am aware of is that I get error messages aboutH the fuse having blown on one of the DSSI channels. I've not been able to locate said fuse so far!   There are no manuals with it.   F If anyone is interested then please contact me personally by e-mail atG tony.arnold@manchester.ac.uk. I no longer subscribe to these groups, so % I'm unlikely to see a response there.    Regards, Tony.    ------------------------------  + Date: Tue, 29 Aug 2006 08:55:05 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply); Subject: Re: VAX 4000/600 free to good home - Manchester UK $ Message-ID: <ed0vd9$96v$1@online.de>  B In article <pan.2006.08.27.19.43.06.348906@manchester.ac.uk>, Tony. Arnold <tony.arnold@manchester.ac.uk> writes:   K > I have a VAX 4000 model 600 surplus to requirements which I would like to K > see go to a good home rather than be scrapped. I'm in Manchester, UK. The J > machine is rather bulky, so any takers would have to come and collect it$ > or arrange shipping at their cost.  H > If anyone is interested then please contact me personally by e-mail atI > tony.arnold@manchester.ac.uk. I no longer subscribe to these groups, so ' > I'm unlikely to see a response there.   G I used to be in Manchester and I used to use this machine.  I have seen A it.  It is not desktop, but neither is it the size of a full-size B refrigerator.  It will fit under a desk.  It should fit in any car4 (except perhaps something like a Smart or Fiat 500).  E 512 MB is a huge amount of memory for a VAX.  This would make a good  E hobbyist machine since it is quite powerful (for a VAX), can let one   learn about DSSI etc.   I I would take it myself but I have to watch my space (I have 15 VAXes and  I 11 ALPHAs at home now; it's ALMOST time to stop!) and power bills (since  E I have 3 or 4 (mostly small, but nevertheless) machines running full  A time).  However, for a hobbyist running just this machine, power  D consumption shouldn't be too bad (you don't need three-phase power, F special fuses or whatever.)  The machine is certainly in good working ; order; fellow hobbyists: don't let it go to the scrap heap!    ------------------------------  % Date: Tue, 29 Aug 2006 14:39:44 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ; Subject: Re: VAX 4000/600 free to good home - Manchester UK , Message-ID: <44F489F0.815166AC@teksavvy.com>  / Phillip Helbig---remove CLOTHES to reply wrote: G > special fuses or whatever.)  The machine is certainly in good working = > order; fellow hobbyists: don't let it go to the scrap heap!   @ If it weren't for trans-pond transportantion costs, I would mostB certaintly take it.  But if nobody else wants it, i would hope theG person would at least offer parts of it. I would take the memory boards C for instance and perhaps some DSSI drives and perhaps teh CPU board  (backup) etc etc.    ------------------------------  + Date: Wed, 30 Aug 2006 10:53:05 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply); Subject: Re: VAX 4000/600 free to good home - Manchester UK $ Message-ID: <ed3qmh$tus$2@online.de>  F In article <howard-66DF8E.15045229082006@news.supernews.com>, Howard S! Shubs <howard@shubs.net> writes:    G > If it's like my 4000 m 500A, it's 29" high, 19" wide, and around 17"  E > deep.  And will weigh in at something like 100lbs or so.  I've not  D > weighed my machine, so I can't say for sure.  It was heavy, but I G > managed to get it up the stairs by myself, though it took me a while.    Sounds about right.   I > > 512 MB is a huge amount of memory for a VAX.  This would make a good  I > > hobbyist machine since it is quite powerful (for a VAX), can let one   > > learn about DSSI etc.  > I > It's an /insane/ amount of memory for a VAX.  This is a problem?? :-)   J > What the heck was this machine doing??  Sound like someone simply maxed 	 > it out!   " I think it used to do quite a lot.  M > > I would take it myself but I have to watch my space (I have 15 VAXes and  M > > 11 ALPHAs at home now; it's ALMOST time to stop!) and power bills (since  I > > I have 3 or 4 (mostly small, but nevertheless) machines running full  
 > > time). > G > I was gonna ask, but okay, you're excused.  You don't have a low end  D > machine this could replace, though?  C'mon.  You know you want it.  E You're talking to someone (me!) who recently gave away an ALPHAserver E 2000 and an ALPHAserver 2100.  Actually, small, relatively fast VAXes G (e.g. VAXstation 4000/90A or VAX 4000/105A) are good enough for most of C what I do at home, can have SCSI disks and don't need much power or ? space.  I have an ALPHAserver 1200 (a.k.a. Digital Server 5305) G configured as a satellite which I boot up when I need to use CSWB which C is my most powerful machine (though now that I have two new (to me) C smaller EV5 machines, I'll probably use those for CSWB).  I have an H ALPHA in the cluster so that I can let it have the cluster alias most ofG the time to make use of TCPIP 5.4 features (mainly the anti-spam stuff) H and more advanced Fortran.  At home, at most I would need more CPU powerH if I do any number crunching, but EV5 and EV56 ALPHAs are what I need toE switch on there (and they will need less power and space than the big  VAX).    ------------------------------  % Date: Wed, 30 Aug 2006 08:59:07 -0700 * From: "Tom Linden" <tom@kednos-remove.com>; Subject: Re: VAX 4000/600 free to good home - Manchester UK ) Message-ID: <op.te3tctuotte90l@hyrrokkin>   G On Wed, 30 Aug 2006 08:50:39 -0700, Howard S Shubs <howard@shubs.net>    wrote:  & > In article <ed3qmh$tus$2@online.de>,F >  helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to >  reply) wrote: > H >> You're talking to someone (me!) who recently gave away an ALPHAserver  >> 2000 and an ALPHAserver 2100. > J > Okay, so this would /be/ the low end machine you'd upgrade.  Must really > suck to be you! :-D  > H > Wish you were on this side of the pond, or visa versa.  Oh well.  I'veF > honestly got more than I need in VAXes.  I could use a more powerfulF > Alpha, though.  My Alphastation 255 /works/ but it's, y'know, small. > G > I've found that my main CPU-bound task doesn't work well on Alpha.  I H > run the distributed.net client, you see.  Compared to a PPC-based Mac,I > or a Pentium-based Windows box, Alpha doesn't measure up.  I understand G > this is because Alpha doesn't have bit-wise shift in its architecture D > while the others do.  That operation is fairly central to the dnetI > client.  I haven't tried it on VAX due to lack of a client, though I've + > considered trying to build such a client.  > I Even the first Alphas had a shift instruction, what they were lacking was . sign-extended moves for 8 and 16 bit integers.     --  E Using Opera's revolutionary e-mail client: http://www.opera.com/mail/    ------------------------------    Date: 28 Aug 2006 12:38:54 -0700 From: "Verne" <verne@wvnet.edu> ; Subject: Re: VGA video adapter for Vaxstation 4000 model 60 C Message-ID: <1156793934.769414.235580@b28g2000cwb.googlegroups.com>    Michael Kraemer wrote: > George Cook schrieb:^ > > In article <ecdf46$308$01$1@news.t-online.com>, Michael Kraemer <M.Kraemer@gsi.de> writes: > > K > > The 3 pin connector is called a 3W3.  Some SUN and RS6000 (I don't know L > > about Apollo) machines used a 13W3 connector which has ten DB style pins/ > > between pin 1 and 2 of the three coax pins.  > ) > thanks for clarifying the nomenclature. - > 3W3 certainly sounds more professional than ( > "that strange DEC-style connector" :-) >   > > The one IBM 3W3 to BNC cableK > > I have (possibly came with an RT system) is non-standard in that it has  > > Red and Blue reversed. > H > Might give funny colours, but as long as the Green is not affected ... > O > > FWIW, the DEC 3000 I am writing this from uses a 3W3 to VGA cable connected K > > to a VGA to 13W3 cable to connect the 3W3 turbochannel video card to my  > > 13W3 monitor.  > D > I sometimes wonder that gfx signals can travel so many transitions- > and still give some picture on the monitor.       E Yes,  29-32549-01 is the proper DEC part (my old Alpha workstation is  using one right now).   F Just last week I had a need to hook up a VGA device to an older Alpha,@ and just this morning received and tested a pigtail adapter fromE digitalconnection.com ... it gives a male VGA connector on one end so E the 12 inch cable screws directly to my VGA monitor, and has 5 female : BNC connectors to hook to my existing DEC cable (5 becauseD DigitalConnection gives two sync wires; I just will ignore those two connectors).  E As always, I have no financial interest in this vendor;  just found a  cheap VGA pigtail. SeeR http://www.digitalconnection.com/store/Product_List.asp?CID=3&CAT=CABLE%20ADAPTERSF (choose Video Adapters under CABLES on their left-hand menu as one wayE to find them) and specifically item HD15M5BF1. Can't remember now but @ last week I did find another vendor for the same kind of pigtailD adapter (was higher priced than this $24 item at DigitalConnection).G They offer other VGA cables too; make sure you get the right genders on % both ends for your exact application.   B By using this pigtail I have one less physical connection than theG 29-32549-01 setup, but I don't think I'll be replacing my other genuine *  d|i|g|i|t|a|l  adapter any time soon  :-)   Verne    ------------------------------  % Date: Tue, 29 Aug 2006 12:52:07 -0500 + From: "Mr. Question" <Question@example.com> " Subject: VMS 4.4 and file security3 Message-ID: <1156873269_11337@sp6iad.superfeed.net>    Hi.   K Way back in 1986, I took a few courses at the local college and got to use  @ my first big system, a VAX 11/785 running VMS 4.4  (I think 4.4)  J Naturally I played around and discovered how easy it was to set up a file B pointer to point to another place.  Even to another person's file.  I All I had to do was take a file that I already had, and use an editor to  H change the pointers in the directory to point somewhere else.  (I don't ! remember details, been too long.)   ; I could even point to another file that belonged to friend.    It was easy.  L I never got around to exploring the idea further.  I had other things to do ! and just didn't get around to it.   " So all this time I've wondered....  , Was that a gaping hole in the file security?  $ Was that system simply set up wrong?  I Would I have been able to point my file header anywhere into the system?  3 Even into the system files and directory structure?   I Since I don't really want to set up a VMS emulator and learn how to mess  I with it and configure it and install the tools etc., just to find out, I   thought I'd ask.   Thanks.           Q ----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------  # Date: Tue, 29 Aug 2006 19:28:25 GMT % From: Rob Brown <mylastname@gmcl.com> & Subject: Re: VMS 4.4 and file securityE Message-ID: <Pine.LNX.4.61.0608291239020.29483@localhost.localdomain>   J   This message is in MIME format.  The first part should be readable text,K   while the remaining parts are likely unreadable without MIME-aware tools.   & --8323328-1757883977-1156879704=:29483; Content-Type: TEXT/PLAIN; charset=ISO-8859-1; format=flowed  Content-Transfer-Encoding: 8BIT   ( On Tue, 29 Aug 2006, Mr. Question wrote:  F > Way back in 1986, I took a few courses at the local college and got D > to use my first big system, a VAX 11/785 running VMS 4.4 (I think  > 4.4) > G > Naturally I played around and discovered how easy it was to set up a  D > file pointer to point to another place.  Even to another person's  > file.  > A > All I had to do was take a file that I already had, and use an  D > editor to change the pointers in the directory to point somewhere 3 > else.  (I don't remember details, been too long.)   F It doesn't sound like the kind of thing you would do in plain-vanilla F VMS.  Perhaps there was some student file system set up that you were 
 manipulating.   G But, yes, in Plain Vanilla VMS, you *can* edit your directory file and  F change the file IDs, but it would be easier with a binary editor than E a text editor.  A few lines of directory in a text editor looks like   this:   % <<LF>^@^@<LF>ACNTRN.SYS^A^@^@I^@^@^@ ( <LF>^@^@        AFTER.TXT^@^A^@So!^@^@^@/ <LF>^@^@^UALARM_I80.TPU$JOURNAL^@^A^@Pi^X^@^@^@   @ In these examples (it can be more complicated), the file ID and E sequence number information are at the end of the line.  In addition  G to being able to interpret the directory format and and convert binary  F to the appropriate ASCII, you would also have to know the File-ID etc  of the target file.    It is much easier to just type  &    $ SET FILE/ENTER=[]NEWPOINTER.DAT -*      [MYFRIENDSDIRECTORY]MYFRIENDSFILE.DAT  > To do this, you need to have read permission to your friend's 
 directory.  F With your suggested method, you do not, but then you have to find out D the File-ID and sequence number of your friend's file by some other  means.  . > Was that a gaping hole in the file security?  3 Well let's check to see what you have accomplished?   B You have an entry in your directory that references your friend's F file.  Can you read the file?  Only if you have been given permission 	 to do so.   ! So is it a security problem?  NO.   	 Consider:       $ dir [secret]       Directory DISK$8:[SECRET]      MANHATTANPROJECT.TXT;1       Total of 1 file. &    $ type [secret]manhattanproject.txtF    %TYPE-W-OPENIN, error opening DISK$8:[SECRET]MANHATTANPROJECT.TXT;1    as input B    -RMS-E-PRV, insufficient privilege or file protection violationC    $ set file/enter=[]stolensecret.txt [secret]manhattanproject.txt     $ type STOLENSECRET.TXT;1D    %TYPE-W-OPENIN, error opening DISK$8:[BROWN]STOLENSECRET.TXT;1 as    inputB    -RMS-E-PRV, insufficient privilege or file protection violation  E So there is a file in a secret directory that I can't read.  I add a  D pointer to that file in my own directory.  I *still* can't read the  file.   G (Should I have been able to find out what was in the secret directory?  D Not if it had been protected as well as the secret file itself was. 7 But then I would have had to find a different example.)   & > Was that system simply set up wrong?   Perhaps.  4 Or perhaps you were expected to be able share files.  F Actually, you did not report that you could read your friend's files, A just that you could create directory entries to them in your own  
 directory.  C > Would I have been able to point my file header anywhere into the  = > system? Even into the system files and directory structure?   C To anything on the same disk as your directory, you could.  But it  E doesn't matter.  You still would not be able read files that you are   not allowed to read.  F > Since I don't really want to set up a VMS emulator and learn how to D > mess with it and configure it and install the tools etc., just to  > find out, I thought I'd ask.  4 For free VMS systems where you could experiment, try  !    telnet://eisner.decuserve.org/ !    telnet://deathrow.vistech.net/      --    B Rob Brown                        b r o w n a t g m c l d o t c o m6 G. Michaels Consulting Ltd.      (780)438-9343 (voice)4 Edmonton                         (780)437-3367 (FAX)2                                   http://gmcl.com/  ( --8323328-1757883977-1156879704=:29483--   ------------------------------    Date: 29 Aug 2006 14:33:44 -0500 From: briggs@encompasserve.org& Subject: Re: VMS 4.4 and file security3 Message-ID: <Sal2KayiltP3@eisner.encompasserve.org>   a In article <1156873269_11337@sp6iad.superfeed.net>, "Mr. Question" <Question@example.com> writes:  > Hi.  > M > Way back in 1986, I took a few courses at the local college and got to use  B > my first big system, a VAX 11/785 running VMS 4.4  (I think 4.4) > L > Naturally I played around and discovered how easy it was to set up a file D > pointer to point to another place.  Even to another person's file. > K > All I had to do was take a file that I already had, and use an editor to  J > change the pointers in the directory to point somewhere else.  (I don't # > remember details, been too long.)  > = > I could even point to another file that belonged to friend.  >  > It was easy. > N > I never got around to exploring the idea further.  I had other things to do # > and just didn't get around to it.  > $ > So all this time I've wondered.... > . > Was that a gaping hole in the file security?  ? The thing that your directory entry points to is a file header. B The file header is protected against unprivileged modification and) is where the file permissions are stored.   A The directory file contains no information that is relied upon by  the trusted computing base.   K > Would I have been able to point my file header anywhere into the system?  5 > Even into the system files and directory structure?   D Yes.  If you want to have SYSUAF.DAT in your directory, that's fine.H You still can't read it and you still can't modify it.  And other people" can still access it in SYS$SYSTEM.  
 However...  G I believe there was at least one VMS vulnerability associated with this H sort of thing in the VMS 4.x time frame.  There was a software component> that made an unsafe assumption about the user's directory file	 contents.   G And then there's always the risk that a system manager will make unsafe 3 assumptions about a user's directory file contents.    ------------------------------  % Date: Tue, 29 Aug 2006 16:00:52 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> & Subject: Re: VMS 4.4 and file security: Message-ID: <O_Gdnd_gjdzvAGnZnZ2dnUVZ_ridnZ2d@comcast.com>   Mr. Question wrote:    > Hi.  > M > Way back in 1986, I took a few courses at the local college and got to use  B > my first big system, a VAX 11/785 running VMS 4.4  (I think 4.4) > L > Naturally I played around and discovered how easy it was to set up a file D > pointer to point to another place.  Even to another person's file. > K > All I had to do was take a file that I already had, and use an editor to  J > change the pointers in the directory to point somewhere else.  (I don't # > remember details, been too long.)  > = > I could even point to another file that belonged to friend.  >  > It was easy. > N > I never got around to exploring the idea further.  I had other things to do # > and just didn't get around to it.  > $ > So all this time I've wondered.... > . > Was that a gaping hole in the file security? > & > Was that system simply set up wrong? > K > Would I have been able to point my file header anywhere into the system?  5 > Even into the system files and directory structure?  >   F No, it was not a security hole.  A directory entry only points to the G file header.  The file header contains the ownership, permissions, etc. D No matter how you get to a file, you only have the access the owner G permitted you, unless you have privilege; "there is no defense against   privilege".    ------------------------------  % Date: Tue, 29 Aug 2006 15:18:40 -0500 + From: "Mr. Question" <Question@example.com> & Subject: Re: VMS 4.4 and file security3 Message-ID: <1156882059_12471@sp6iad.superfeed.net>   , <briggs@encompasserve.org> wrote in message - news:Sal2KayiltP3@eisner.encompasserve.org... / >> Was that a gaping hole in the file security?  > A > The thing that your directory entry points to is a file header. D > The file header is protected against unprivileged modification and+ > is where the file permissions are stored.   E No, I created a fake file header that pointed to where ever I wanted.   E After 20 years, I can't remember all the details.  And since I'm not  B familiar with VMS anymore, I can't even use the right terminology.  L But I could gain access (at least read permission, can't remember if I ever G even tried write permission) to a friends file that I had no rights to  J access.  We were in the same 'group' (Students), but we couldn't normally " gain access to each other's files.  K I think I created a fake file and / or directory entry to point wherever I   wanted.   H The only thing I needed was the 'magic number' (file ID?, sector?  LFN? + what??) to plug into the fake file pointer.   J Then I simply accessed my own 'file' that was pointing to the wrong place.  H I think I could do directories that way too, in which case I could read  their directory.  7 It's been a long time and I can't remember the details.   F I just basically remember that I was able to read files that I had no C permission to access.  The exact details of how are a little fuzzy.   K And I never got around to seeing just how far I could go.  Whether I could  . read system files or the faculty's files, etc.          Q ----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------    Date: 29 Aug 2006 14:31:38 -0700$ From: "AEF" <spamsink2001@yahoo.com>& Subject: Re: VMS 4.4 and file securityB Message-ID: <1156887098.331347.15580@h48g2000cwc.googlegroups.com>   Mr. Question wrote: - > <briggs@encompasserve.org> wrote in message / > news:Sal2KayiltP3@eisner.encompasserve.org... 1 > >> Was that a gaping hole in the file security?  > > C > > The thing that your directory entry points to is a file header. F > > The file header is protected against unprivileged modification and- > > is where the file permissions are stored.  > G > No, I created a fake file header that pointed to where ever I wanted.  > F > After 20 years, I can't remember all the details.  And since I'm notD > familiar with VMS anymore, I can't even use the right terminology.  ( Well, that's going to make it difficult.  M > But I could gain access (at least read permission, can't remember if I ever H > even tried write permission) to a friends file that I had no rights toK > access.  We were in the same 'group' (Students), but we couldn't normally $ > gain access to each other's files. > L > I think I created a fake file and / or directory entry to point wherever I	 > wanted.  > I > The only thing I needed was the 'magic number' (file ID?, sector?  LFN? - > what??) to plug into the fake file pointer.  > L > Then I simply accessed my own 'file' that was pointing to the wrong place. > I > I think I could do directories that way too, in which case I could read  > their directory. > 9 > It's been a long time and I can't remember the details.  > G > I just basically remember that I was able to read files that I had no E > permission to access.  The exact details of how are a little fuzzy.  > L > And I never got around to seeing just how far I could go.  Whether I could0 > read system files or the faculty's files, etc.  F I think that the files you read but think you shouldn't have been ableF to probably had world=read permission but the directories they were inC did not. This means that you could not read the files the usual way G thru the directory. But if you use the file-id I think you could bypass E the directory file and read the file with your "fake file" trick, but @ only if that file had world=read permission. I think this is whyD DIGITAL strongly recommends that to hide files completely you shouldE apply WORLD=no-access for both the files and the directories they are D in, not just the directory.OTOH, if you added file headers, then youG had permission to access INDEXF.SYS, which is not normally w=re. So I'm  still not sure what you did.   AEF    ------------------------------  # Date: Tue, 29 Aug 2006 22:28:44 GMT % From: Rob Brown <mylastname@gmcl.com> & Subject: Re: VMS 4.4 and file securityE Message-ID: <Pine.LNX.4.61.0608291556240.31762@localhost.localdomain>   ( On Tue, 29 Aug 2006, Mr. Question wrote:  4 > "Rob Brown" <mylastname@gmcl.com> wrote in message >>E >> You have an entry in your directory that references your friend's  > >> file.  Can you read the file?  Only if you have been given  >> permission to do so.  >>$ >> So is it a security problem?  NO. > / > I didn't have permission.  But I did read it.   F Unless you actually displayed the permission bits and owner using the C appropriate DIRECTORY command, you don't know that you didn't have  E permission.  And 20 years down the road it is too late to check.  As  D AEF pointed out, your friend's directory could have been protected, * but the files could have been unprotected.  E In my previous example, the [SECRET] directory was readable, but the  ? file in it was not.  I was able to put an entry in my personal  C directory that pointed to the protected file.  But since I did not  . have permission to read the file, I could not.  A I have changed the protection on the file to make it readable by  E anybody.  But I have also changed the protection on the directory so  E that nobody can read the directory except the owner.  So now we have   this:       $ dir [secret] @    %DIRECT-E-OPENIN, error opening DISK$8:[SECRET]*.*;* as inputB    -RMS-E-PRV, insufficient privilege or file protection violation&    $ type [secret]manhattanproject.txt*    %TYPE-W-SEARCHFAIL, error searching for'    DISK$8:[SECRET]MANHATTANPROJECT.TXT; B    -RMS-E-PRV, insufficient privilege or file protection violation    $ type stolensecret.txt    bomb     $  D So it *looks* like I am reading a file that I don't have permission 8 to, but in fact I *do* have permission to read the file:      $ dir/sec stolensecret.txt       Directory DISK$8:[BROWN]   F    STOLENSECRET.TXT;1   [SYSTEM]                         (,RWED,,RWED)      Total of 1 file.   B That second RWED means that *anybody* can Read, Write, Extend, or @ Delete the file.  [SYSTEM] means that the file is owned by user ! [SYSTEM] and not by me ([BROWN]).   F > I suppose it's possible that the directory structure for "Students" F > was set so we couldn't read them, but that the files were set so we F > could if we knew the filename?  Doesn't make too much sense, but it  > might explain things.   ? Sure we could do that too.  I can change the protection on the  F [SECRET] directory so that it can't be searched, but so that it *can* 1 be used to reference filenames known by the user.       $ dir [secret] @    %DIRECT-E-OPENIN, error opening DISK$8:[SECRET]*.*;* as inputB    -RMS-E-PRV, insufficient privilege or file protection violation&    $ type [secret]manhattanproject.txt    bomb   F > But I think I could have done the same kind of thing by pointing it E > to an upper directory that I had access to.  Directories were just   > files. > E > I just point mine to the directory and read it.  I get the file id  C > stuff for the directory I can't access.  I put that into my fake  E > file and I can suddenly read the file header as an actual file.  I  A > get the info from there and plug it into my fake file and I've  B > suddenly bypassed the file permssions and can now read my 'own'  > file.   E I doubt it.  Perhaps the unspecified bug that John Briggs alluded to  G was this very behaviour.  If so it would have been a shortcut added on  A purpose without thinking through the consequences, and obviously  " removed they realized their error.  F And of course, it is not "your own file".  You just have an entry for  it in your directory file.  4 Maybe someone with a V4.4 hobby system can try this.  = > I'm having a lot of trouble remembering ... the proper vms   > terminology.   No comment.  ;-)     --    B Rob Brown                        b r o w n a t g m c l d o t c o m6 G. Michaels Consulting Ltd.      (780)438-9343 (voice)4 Edmonton                         (780)437-3367 (FAX)2                                   http://gmcl.com/   ------------------------------  % Date: Tue, 29 Aug 2006 15:18:38 -0500 + From: "Mr. Question" <Question@example.com> & Subject: Re: VMS 4.4 and file security3 Message-ID: <1156882057_12469@sp6iad.superfeed.net>   2 "Rob Brown" <mylastname@gmcl.com> wrote in message > G > It doesn't sound like the kind of thing you would do in plain-vanilla G > VMS.  Perhaps there was some student file system set up that you were  > manipulating.   > I don't think so, but I'm not quite sure what you are meaning.  M If you did a 'dir DUA0:' (or whatever the vms command is), you could see the  L whole directory structure, including (eventually) our 'Students' section of 
 the files.  K Of course, doing that you couldn't gain access to any files or directories  L that were blocked.  You could just follow the readable, publicly accessable G directory structure.  You still didn't have permission to read anybody  4 else's files or poke into their private directoires.  7 But the way I did it, I did read somebody else's files.   F (Well, actually it was 'my' files.... There were several dozen unused M accounts for students that never showed up, and since they all used the same  L initial password, I just commandered a few extra accounts to play with.  So = I wasn't playing around with anybody else's important files.)   H > But, yes, in Plain Vanilla VMS, you *can* edit your directory file andG > change the file IDs, but it would be easier with a binary editor than F > a text editor.  A few lines of directory in a text editor looks like > this:   K I don't know what editor I used.  Doesn't really matter because the result  
 was the same.    > ' > <<LF>^@^@<LF>ACNTRN.SYS^A^@^@I^@^@^@ * > <LF>^@^@        AFTER.TXT^@^A^@So!^@^@^@1 > <LF>^@^@^UALARM_I80.TPU$JOURNAL^@^A^@Pi^X^@^@^@  > A > In these examples (it can be more complicated), the file ID and F > sequence number information are at the end of the line.  In additionH > to being able to interpret the directory format and and convert binaryG > to the appropriate ASCII, you would also have to know the File-ID etc  > of the target file.   L Or figure out that of the directory, and then read that.  Take the pointers M and put them into my fake file pointer and read the file header thingie that  J I don't normally have access to, get the file id and stuff there and then I poke that back into my fake file and then I have access to a file that I   shouldn't have access to.     > It is much easier to just type > ' >   $ SET FILE/ENTER=[]NEWPOINTER.DAT - + >     [MYFRIENDSDIRECTORY]MYFRIENDSFILE.DAT  > ? > To do this, you need to have read permission to your friend's  > directory.  E I don't know that command, but I didn't have access to his directory.     / >> Was that a gaping hole in the file security?  > 5 > Well let's check to see what you have accomplished?  > C > You have an entry in your directory that references your friend's G > file.  Can you read the file?  Only if you have been given permission  > to do so.  > # > So is it a security problem?  NO.   - I didn't have permission.  But I did read it.   1 It was a simple test file, but I could access it.     ' >> Was that system simply set up wrong?  > 
 > Perhaps. > 6 > Or perhaps you were expected to be able share files.  M Nope.  No sharing was allowed.  Or at least things were set up so we thought  F we couldn't.  We could see other people's main directory name, but we  couldn't go in there.   L I suppose it's possible that the directory structure for "Students" was set M so we couldn't read them, but that the files were set so we could if we knew  H the filename?  Doesn't make too much sense, but it might explain things.    G > Actually, you did not report that you could read your friend's files, B > just that you could create directory entries to them in your own > directory.  L Sorry.  Guess I should have written my original message better.  Yes, I was  able to read the file.  K It was just a simple test file.  Nothing major.  But I shouldn't have been   able to read it.  M I did know the file ID etc. stuff, so I did know where to point my fake file   to.   J But I think I could have done the same kind of thing by pointing it to an C upper directory that I had access to.  Directories were just files.   M I just point mine to the directory and read it.  I get the file id stuff for  F the directory I can't access.  I put that into my fake file and I can L suddenly read the file header as an actual file.  I get the info from there M and plug it into my fake file and I've suddenly bypassed the file permssions   and can now read my 'own' file.   L But some of that is speculation and fuzzy memory.  I don't remember how far ! I went or the exact steps I used.    > C >> Would I have been able to point my file header anywhere into the > >> system? Even into the system files and directory structure? > D > To anything on the same disk as your directory, you could.  But itF > doesn't matter.  You still would not be able read files that you are > not allowed to read.   I did.    F >> Since I don't really want to set up a VMS emulator and learn how toD >> mess with it and configure it and install the tools etc., just to >> find out, I thought I'd ask.  > 6 > For free VMS systems where you could experiment, try > " >   telnet://eisner.decuserve.org/" >   telnet://deathrow.vistech.net/  G It's been 20 years.  I barely even remember how to *spell* VMS...[grin]   L Thanks for the links, but I've forgotten nearly everything about VMS that I M knew.  That's one of the reasons my messages are so fuzzy.  I'm having a lot  C of trouble remembering details and even the proper vms terminology.                 Q ----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------  % Date: Tue, 29 Aug 2006 21:01:47 -0500 + From: "Mr. Question" <Question@example.com> & Subject: Re: VMS 4.4 and file security3 Message-ID: <1156903078_15243@sp6iad.superfeed.net>   0 "AEF" <spamsink2001@yahoo.com> wrote in message < news:1156887098.331347.15580@h48g2000cwc.googlegroups.com... >  > Mr. Question wrote:   G >> After 20 years, I can't remember all the details.  And since I'm not E >> familiar with VMS anymore, I can't even use the right terminology.  > * > Well, that's going to make it difficult.  
 Yeah, I know.   M I hesitated for quite a while before even posting the message, but I figured  ? I wasn't going to remember anymore details, so I might as well.   I Who knows, maybe somebody would know for sure about it and could say for  % certain whether it would work or not.    I had nothing to loose.     H >> And I never got around to seeing just how far I could go.  Whether I  >> could1 >> read system files or the faculty's files, etc.  > H > I think that the files you read but think you shouldn't have been ableH > to probably had world=read permission but the directories they were inE > did not. This means that you could not read the files the usual way    That's possible.  M It's certainly possible the SysOp set things up so only the directories were  
 protected.  E He was a 'professional' SysOp they hired, rather than somebody local   admin'ing it.  But who knows.     I > thru the directory. But if you use the file-id I think you could bypass G > the directory file and read the file with your "fake file" trick, but B > only if that file had world=read permission. I think this is why  I Well, the read/write permissions were stored in the file header, weren't  K they?  (I don't know.)  If so, then my modifying my own 'fake' file header  M to point to the real file data would result in my own file permissions being   used to access the data.  I But I have no idea where the file permissions were stored.  Way too long   ago.  M I long ago threw away all my VMS data that they gave us, so I don't remember   how those things were stored.   F > DIGITAL strongly recommends that to hide files completely you shouldG > apply WORLD=no-access for both the files and the directories they are F > in, not just the directory.OTOH, if you added file headers, then youI > had permission to access INDEXF.SYS, which is not normally w=re. So I'm  > still not sure what you did.  J Unfortunately, I don't really "know" what I did either.  Been a long time.  I I was kind of hoping somebody in here would recognise what I was talking  J about, and say something like "Oh yeah, I remember doing that.  Darn near 0 got fired when they found out what I was doing."          Q ----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------  % Date: Tue, 29 Aug 2006 21:08:27 -0500 + From: "Mr. Question" <Question@example.com> & Subject: Re: VMS 4.4 and file security3 Message-ID: <1156903079_15245@sp6iad.superfeed.net>   3 "Rob Brown" <mylastname@gmcl.com> wrote in message  ? news:Pine.LNX.4.61.0608291556240.31762@localhost.localdomain... * > On Tue, 29 Aug 2006, Mr. Question wrote: > 5 >> "Rob Brown" <mylastname@gmcl.com> wrote in message  >>> L >>> You have an entry in your directory that references your friend's file. L >>> Can you read the file?  Only if you have been given permission to do so. >>> % >>> So is it a security problem?  NO.  >>0 >> I didn't have permission.  But I did read it. > H > Unless you actually displayed the permission bits and owner using the E > appropriate DIRECTORY command, you don't know that you didn't have  
 > permission.   I Yup.   I certainly didn't have access to the actual directory, but it is  H possible that the files were accessible.  That only the directories had % default settings to keep people away.     F > permission.  And 20 years down the road it is too late to check.  As   Very too late.  = And my fuzzy memory from that far back isn't helping, either.   G I had hesitated a couple months before even posting my question, but I  M figured my memory wasn't going to improve any, so I might as well post it as   best as I could.    K >> I suppose it's possible that the directory structure for "Students" was  L >> set so we couldn't read them, but that the files were set so we could if L >> we knew the filename?  Doesn't make too much sense, but it might explain 
 >> things. > J > Sure we could do that too.  I can change the protection on the [SECRET] J > directory so that it can't be searched, but so that it *can* be used to ( > reference filenames known by the user.  I Considering it was a college student enviornment, I would expect them to  M take as much precautions as they could, just to keep the students from doing   too much damage.  L But maybe not.  Maybe he got sloppy or didn't think about it, or no telling  what.   L I do know that the previous semester, they had so many people giving others M their course work by 'mail' that they modified the command to send a copy of  M each message to the faculty, so they could see who was cheating.  That would  I suggest you could get the files by a direct copy.  But maybe a protected  G directory structure was all they did, rather than protecting the files.       K > I doubt it.  Perhaps the unspecified bug that John Briggs alluded to was  D > this very behaviour.  If so it would have been a shortcut added on  G It's possible.  But I certainly can't say whether it was or not.  It's  ; entirely possible my memory is so bad I'm imagining things.    I remember for certain:   G 1) pointing it to a file in my own directory.  Two file header entries  3 pointing to the same file.  That was my first test.   I 2) pointing to a directory.  Certainly my own, since I would have tested  F that first.  This meant the directory became a file I could play with.  K 3) pointing to a file in somebody else's directory.  I don't remember what  4 file it was.  Probably just a simple test text file.  K After that, I'm not sure how much I truely did.  I was a bit preocupied at  L the time, so I may have planned to take the next step but didn't get around 9 to it.  Or maybe I did and can't remember it for certain.   L I know for certain that I did not try to do that for arbitrary system files M or faculty files.  It was near the end of the course and I didn't have time.  ! I was too busy doing other stuff.     K > And of course, it is not "your own file".  You just have an entry for it   > in your directory file.   L I was just meaning that it was a file that I created in my directory that I H had full rights and access to.  As opposed to a file that somebody else - created that I may or may not have rights to.              Q ----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------    Date: 29 Aug 2006 19:34:44 -0700$ From: "AEF" <spamsink2001@yahoo.com>& Subject: Re: VMS 4.4 and file securityB Message-ID: <1156905284.778623.277410@74g2000cwt.googlegroups.com>   Mr. Question wrote: 1 > "AEF" <spamsink2001@yahoo.com> wrote in message > > news:1156887098.331347.15580@h48g2000cwc.googlegroups.com... > >  > > Mr. Question wrote:  > I > >> After 20 years, I can't remember all the details.  And since I'm not G > >> familiar with VMS anymore, I can't even use the right terminology.  > > , > > Well, that's going to make it difficult. >  > Yeah, I know.  > N > I hesitated for quite a while before even posting the message, but I figuredA > I wasn't going to remember anymore details, so I might as well.  > J > Who knows, maybe somebody would know for sure about it and could say for' > certain whether it would work or not.  >  > I had nothing to loose.  >   
 Loose? :-)   > I > >> And I never got around to seeing just how far I could go.  Whether I 
 > >> could3 > >> read system files or the faculty's files, etc.  > > J > > I think that the files you read but think you shouldn't have been ableJ > > to probably had world=read permission but the directories they were inG > > did not. This means that you could not read the files the usual way  >  > That's possible. > N > It's certainly possible the SysOp set things up so only the directories were > protected.  F Perhaps he thought students wouldn't bother looking at files by FID or didn't care.   > F > He was a 'professional' SysOp they hired, rather than somebody local > admin'ing it.  But who knows.  >  > K > > thru the directory. But if you use the file-id I think you could bypass I > > the directory file and read the file with your "fake file" trick, but D > > only if that file had world=read permission. I think this is why > J > Well, the read/write permissions were stored in the file header, weren'tL > they?  (I don't know.)  If so, then my modifying my own 'fake' file headerN > to point to the real file data would result in my own file permissions being > used to access the data.  C The file headers are all in INDEXF.SYS in the master file directory D ([000000]). They are normally protected from modification by regularG users. The .DIR;1 files contain, IINM, only a mapping from filenames to E file id's. All other information about the file is in its file header G in INDEXF.SYS. Why don't you read the description of the Files-11 ODS-2 B disk structure in the Guide to File Applications manual in the VMSD section of the HP Web site to get caught up on the terminology. Note5 that there are no file headers in the directory file.   J > But I have no idea where the file permissions were stored.  Way too long > ago.  A They're stored in the file header. I believe someone posted that.     N > I long ago threw away all my VMS data that they gave us, so I don't remember > how those things were stored.  > H > > DIGITAL strongly recommends that to hide files completely you shouldI > > apply WORLD=no-access for both the files and the directories they are H > > in, not just the directory.OTOH, if you added file headers, then youK > > had permission to access INDEXF.SYS, which is not normally w=re. So I'm   G Oops, make that w=wre. You need write priv to modify a file, of course.     > > still not sure what you did. > L > Unfortunately, I don't really "know" what I did either.  Been a long time. > J > I was kind of hoping somebody in here would recognise what I was talkingK > about, and say something like "Oh yeah, I remember doing that.  Darn near 2 > got fired when they found out what I was doing." >   @ We've made our guesses and explained them. That's all we can do.   AEF    ------------------------------    Date: 29 Aug 2006 22:53:46 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) & Subject: Re: VMS 4.4 and file security3 Message-ID: <UDbQCi71SQ6H@eisner.encompasserve.org>   a In article <1156873269_11337@sp6iad.superfeed.net>, "Mr. Question" <Question@example.com> writes:   K > All I had to do was take a file that I already had, and use an editor to  J > change the pointers in the directory to point somewhere else.  (I don't # > remember details, been too long.)  > = > I could even point to another file that belonged to friend.   . > Was that a gaping hole in the file security? > & > Was that system simply set up wrong? > K > Would I have been able to point my file header anywhere into the system?  5 > Even into the system files and directory structure?   F On a properly set up VMS V4.4 system you could not alter file headers,C but you could alter directory entries, essentially associating them , with some other file header (and thus file).  ? But file protection is in the file header, so manipulating your . directory does not give you undeserved access.   ------------------------------  # Date: Wed, 30 Aug 2006 11:24:24 GMT ( From: Alan Greig <greigaln@netscape.net>& Subject: Re: VMS 4.4 and file security> Message-ID: <IzeJg.182245$9d4.59253@fe2.news.blueyonder.co.uk>   Mr. Question wrote:    > That's possible. > O > It's certainly possible the SysOp set things up so only the directories were   > protected. > G > He was a 'professional' SysOp they hired, rather than somebody local   > admin'ing it.  But who knows.   H Maybe someone has said this (not read entire thread) but it's perfectly H possible that the other user set their own files to be world (or group) * readable no matter what defaults were set.   --  
 Alan Greig   ------------------------------    Date: 30 Aug 2006 06:54:02 -0500 From: briggs@encompasserve.org& Subject: Re: VMS 4.4 and file security3 Message-ID: <NAkAwgUETk8p@eisner.encompasserve.org>   m In article <Pine.LNX.4.61.0608291556240.31762@localhost.localdomain>, Rob Brown <mylastname@gmcl.com> writes: * > On Tue, 29 Aug 2006, Mr. Question wrote: [snip]F >> I just point mine to the directory and read it.  I get the file id D >> stuff for the directory I can't access.  I put that into my fake F >> file and I can suddenly read the file header as an actual file.  I B >> get the info from there and plug it into my fake file and I've C >> suddenly bypassed the file permssions and can now read my 'own'   >> file. > G > I doubt it.  Perhaps the unspecified bug that John Briggs alluded to  I > was this very behaviour.  If so it would have been a shortcut added on  C > purpose without thinking through the consequences, and obviously  $ > removed they realized their error.  B Nope.  Nothing to do with the bug I obliquely referred to.  As farE as I know, that bug only manifested as a DoS attack.  You could crash B the system, but you couldn't gain unauthorized access.  It was not a "gaping hole"   C I think others here have correctly diagnosed the behavior.  We have ? a directory that is protected against access but which contains - files which are not protected against access.   B VMS security does not depend on permissions on the directory file.> You can indeed evade directory file permissions with a file-id> guessing attack.  You can even skip the directory entirely andD open a file by file-id rather than by file name.  But then you stillD have to get past file permissions.  And that's where the trustworthy security guarantees apply.   ------------------------------    Date: 30 Aug 2006 07:22:03 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) & Subject: Re: VMS 4.4 and file security3 Message-ID: <dcDXRDBtcaXP@eisner.encompasserve.org>   a In article <1156882057_12469@sp6iad.superfeed.net>, "Mr. Question" <Question@example.com> writes:  > 9 > But the way I did it, I did read somebody else's files.   G    Then either you had privileges or the file was set to be readable to B    you.  Manipulating directory file entries does not change this.  H    At first glance you may think you somehow worked around the original H    directory file's readability.  On VMS you need at least E access to aE    directory to access files in it who's name you already know, and R E    access to get a listing of the directory such as via the DIRECTORY     command.   F    But without the detail of reading the directory and perhaps some ofJ    the file header data first by legitimate means, you wouldn't know what 1    to put in the directory file you were editing.   E    I suspect these files were all set to at least g:r and in the same D    student group.  Since it was 20 years ago, we will probably never    know.   ------------------------------    Date: 30 Aug 2006 07:24:23 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) & Subject: Re: VMS 4.4 and file security3 Message-ID: <cejWWKWwprOd@eisner.encompasserve.org>   a In article <1156882059_12471@sp6iad.superfeed.net>, "Mr. Question" <Question@example.com> writes: . > <briggs@encompasserve.org> wrote in message / > news:Sal2KayiltP3@eisner.encompasserve.org... 0 >>> Was that a gaping hole in the file security? >>B >> The thing that your directory entry points to is a file header.E >> The file header is protected against unprivileged modification and , >> is where the file permissions are stored. > G > No, I created a fake file header that pointed to where ever I wanted.   E    There is no unprivileged method to edit a file header, it's a data J    structure maintained by the OS executive (probably kernel mode code).  I    If you had the privileges to get around that then the actual path you  8    took to take advantage of that privilege is academic.   ------------------------------  + Date: Mon, 28 Aug 2006 12:17:26 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: VMS in The DA$ Message-ID: <ecumsm$cs6$3@online.de>  A In article <4l5sltFdn9oU1@individual.net>, bill@cs.uofs.edu (Bill  Gunshannon) writes:   < > Surprise, surprise.  I have actually seen a job posting by9 > The Department of the Army looking for someone with VMS < > experience (among many other requirements).  Of course, it= > is in the one last field that hasn't totally abandoned VMS,  > Healthcare.     A One last field?  Mobile-phone backend systems?  Stock exchanges?  H Self-written applications at large shops (ca. 100 top-of-the-line ALPHA * machines)?  Intel processor assembly line?   ------------------------------  % Date: Thu, 31 Aug 2006 11:28:56 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com>  Subject: RE: VMS in The DAT Message-ID: <FA60F2C4B72A584DBFC6091F6A2B8684019BC4D3@tayexc19.americas.cpqcorp.net>   > -----Original Message-----3 > From: Phillip Helbig---remove CLOTHES to reply=20 - > [mailto:helbig@astro.multiCLOTHESvax.de]=20  > Sent: August 28, 2006 8:17 AM  > To: Info-VAX@Mvb.Saic.Com  > Subject: Re: VMS in The DA >=20C > In article <4l5sltFdn9oU1@individual.net>, bill@cs.uofs.edu (Bill  > Gunshannon) writes:=20 >=20> > > Surprise, surprise.  I have actually seen a job posting by; > > The Department of the Army looking for someone with VMS > > > experience (among many other requirements).  Of course, it? > > is in the one last field that hasn't totally abandoned VMS,  > > Healthcare. =20  >=20E > One last field?  Mobile-phone backend systems?  Stock exchanges?=20 6 > Self-written applications at large shops (ca. 100=20 > top-of-the-line ALPHA=20, > machines)?  Intel processor assembly line? >=20  F Including many other big name chip manufacturing companies (and I mean big)...    :-)   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Thu, 31 Aug 2006 08:33:07 -0700 * From: "Tom Linden" <tom@kednos-remove.com> Subject: Re: VMS in The DA) Message-ID: <op.te5mthtttte90l@hyrrokkin>   J On Thu, 31 Aug 2006 08:28:56 -0700, Main, Kerry <Kerry.Main@hp.com> wrote:   >  >  >> -----Original Message----- 1 >> From: Phillip Helbig---remove CLOTHES to reply + >> [mailto:helbig@astro.multiCLOTHESvax.de]   >> Sent: August 28, 2006 8:17 AM >> To: Info-VAX@Mvb.Saic.Com >> Subject: Re: VMS in The DA  >>D >> In article <4l5sltFdn9oU1@individual.net>, bill@cs.uofs.edu (Bill >> Gunshannon) writes: >>? >> > Surprise, surprise.  I have actually seen a job posting by < >> > The Department of the Army looking for someone with VMS? >> > experience (among many other requirements).  Of course, it @ >> > is in the one last field that hasn't totally abandoned VMS, >> > Healthcare. >>C >> One last field?  Mobile-phone backend systems?  Stock exchanges? 4 >> Self-written applications at large shops (ca. 100 >> top-of-the-line ALPHA- >> machines)?  Intel processor assembly line?  >> > H > Including many other big name chip manufacturing companies (and I mean	 > big)...   A And guess which language their major applications are written in?    >  > :-)  >  > Kerry Main > Senior Consultant  > HP Services Canada > Voice: 613-592-4660  > Fax: 613-591-4477  > kerryDOTmainAThpDOTcom > (remove the DOT's and AT)  > 6 > OpenVMS - the secure, multi-site OS that just works.       --  E Using Opera's revolutionary e-mail client: http://www.opera.com/mail/    ------------------------------  % Date: Thu, 31 Aug 2006 19:03:45 +0200 / From: Paul Sture <paul.sture.nospam@hispeed.ch>  Subject: Re: VMS in The DAJ Message-ID: <paul.sture.nospam-E1D1D7.19034531082006@mac.sture.homeip.net>  ) In article <op.te5mthtttte90l@hyrrokkin>, ,  "Tom Linden" <tom@kednos-remove.com> wrote:  L > On Thu, 31 Aug 2006 08:28:56 -0700, Main, Kerry <Kerry.Main@hp.com> wrote: >    > > J > > Including many other big name chip manufacturing companies (and I mean > > big)...  > C > And guess which language their major applications are written in?  >   + I assume you aren't referring to C/C++ ...   ;-)    --  
 Paul Sture   ------------------------------  % Date: Wed, 30 Aug 2006 15:55:13 -0400   From: "DBT" <dbturner@icusc.com>3 Subject: We're not dead yet, we're feeling better ! 0 Message-ID: <12fbr96tlma6nde@news.supernews.com>   Enough said?!?!    --     David B Turner Island Computers US Corp 2700 Gregory St, Suite 180 Savannah GA 31404  Tel: 912 447 6622 X201 Cell: 912 447 6622 X251  Fax: 912 201 0402  Email: dbturner@islandco.com Web: http://www.islandco.com% ===================================== < All orders are subject to the following terms and conditions. of sale. These should be read before ordering.% http://www.islandco.com/warranty.html    ------------------------------    Date: 30 Aug 2006 13:13:28 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> 7 Subject: Re: We're not dead yet, we're feeling better ! C Message-ID: <1156968808.009379.253180@b28g2000cwb.googlegroups.com>   
 DBT wrote: > Enough said?!?!  >  > -- >  > David B Turner > Island Computers US Corp > 2700 Gregory St, Suite 180 > Savannah GA 31404  > Tel: 912 447 6622 X201 > Cell: 912 447 6622 X251  > Fax: 912 201 0402  > Email: dbturner@islandco.com > Web: http://www.islandco.com' > ===================================== > > All orders are subject to the following terms and conditions0 > of sale. These should be read before ordering.' > http://www.islandco.com/warranty.html    Who said you were dead?    ------------------------------  % Date: Wed, 30 Aug 2006 21:11:53 -0500 6 From: "David J. Dachtera" <djesys.no@spam.comcast.net>7 Subject: Re: We're not dead yet, we're feeling better ! 0 Message-ID: <44F64569.C46CD99A@spam.comcast.net>  
 DBT wrote: >  > Enough said?!?!    O.k. What did I miss?    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------  % Date: Tue, 29 Aug 2006 09:01:58 -0700 * From: "Tom Linden" <tom@kednos-remove.com> Subject: wwidmgr) Message-ID: <op.te1ytk1mtte90l@hyrrokkin>   9 Is there a way to display the firmware version of the HBA   from the wwidmgr command? Tom    ------------------------------    Date: 29 Aug 2006 23:44:09 -07001 From: "Bart.Zorn@gmail.com" <Bart.Zorn@gmail.com>  Subject: Re: wwidmgrC Message-ID: <1156920248.964277.261030@b28g2000cwb.googlegroups.com>   ; Not in WWIDMGR (that I know of), but with ANAL/SYS you can:   
 $ anal/sys   OpenVMS (TM) system analyzer   SDA> fc show dev fga0 , FGA0: operational firmware revision HS1.91X1A port_name(adapter_id) = 1000-0000-C93E-7C87, node_name(host_id) =  2000-0000-C93E-7C87  SDA>   HTH,  	 Bart Zorn      Tom Linden wrote: ; > Is there a way to display the firmware version of the HBA  >  from the wwidmgr command? > Tom    ------------------------------    Date: 30 Aug 2006 10:20:11 -0500. From: brooks@cuebid.zko.hp.nospam (Rob Brooks) Subject: Re: wwidmgr, Message-ID: <gbio+EtAJxT5@cuebid.zko.hp.com>  3 "Bart.Zorn@gmail.com" <Bart.Zorn@gmail.com> writes: = > Not in WWIDMGR (that I know of), but with ANAL/SYS you can:  > $ anal/sys >  > OpenVMS (TM) system analyzer >  > SDA> fc show dev fga0 . > FGA0: operational firmware revision HS1.91X1C > port_name(adapter_id) = 1000-0000-C93E-7C87, node_name(host_id) =  > 2000-0000-C93E-7C87    > Tom Linden wrote: F >> Is there a way to display the firmware version of the HBA from the  >> wwidmgr command?   ; From V8.3 (and perhaps to be backported, if I get the time)   < $ write sys$output f$getdvi( "fga0", "fc_hba_firmware_rev" ) HS1.90A4   For the future . . .  6 $ write sys$output f$getdvi( "fga0", "adapter_ident" ) KGPSA-EA (Emulex LP9802)     --    H Rob Brooks    VMS Engineering -- Exec Group     brooks!cuebid.zko.hp.com   ------------------------------  % Date: Wed, 30 Aug 2006 09:29:23 -0700 * From: "Tom Linden" <tom@kednos-remove.com> Subject: Re: wwidmgr) Message-ID: <op.te3uq9vztte90l@hyrrokkin>   1 On Wed, 30 Aug 2006 08:20:11 -0700, Rob Brooks  =   $ <brooks@cuebid.zko.hp.nospam> wrote:  5 > "Bart.Zorn@gmail.com" <Bart.Zorn@gmail.com> writes: > >> Not in WWIDMGR (that I know of), but with ANAL/SYS you can:
 >> $ anal/sys  >> >> OpenVMS (TM) system analyzer  >> >> SDA> fc show dev fga0/ >> FGA0: operational firmware revision HS1.91X1 I >> port_name(adapter_id) =3D 1000-0000-C93E-7C87, node_name(host_id) =3D=    >> 2000-0000-C93E-7C87 >  >> Tom Linden wrote:F >>> Is there a way to display the firmware version of the HBA from the >>> wwidmgr command? > = > From V8.3 (and perhaps to be backported, if I get the time)  > > > $ write sys$output f$getdvi( "fga0", "fc_hba_firmware_rev" )
 > HS1.90A4 >  > For the future . . . > 8 > $ write sys$output f$getdvi( "fga0", "adapter_ident" ) > KGPSA-EA (Emulex LP9802) > G Thanks,  while on the topic (and I have brought this up before) I would I find it useful to have some of the capability of consvar under Tru64, in=   D particular the ability to reset CONSOLE and AUTO_ACTION,  F$SETDVI ?     -- =  E Using Opera's revolutionary e-mail client: http://www.opera.com/mail/    ------------------------------    Date: 30 Aug 2006 10:44:31 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com>  Subject: Re: wwidmgrB Message-ID: <1156959871.207320.27810@p79g2000cwp.googlegroups.com>   Tom Linden wrote: ; > Is there a way to display the firmware version of the HBA  >  from the wwidmgr command? > Tom   A If you're looking for a way to get the information at the console B prompt then depending on your system you might get it from the LFUA utility.  Going into LFU and doing a "List" shows me the firmware F versions on most of the PCI cards in my DS20e, however, in my AS1200 IG just get the SRM and ARC versions and my XP1000 doesn't even have a LFU  command.  A Looks like the other options people have given might be the best.      John H. Reinhardt    ------------------------------  % Date: Sun, 27 Aug 2006 09:00:44 +0200 # From: "H Vlems" <hvlems@freenet.de> ; Subject: Re: X-terminal connects get hosed, requires reboot 4 Message-ID: <ecrfqh$1kj9$1@registered.motzarella.de>  J "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> schreef in bericht- news:wPAEVRFua4Ib@eisner.encompasserve.org... D > In article <1156507515.019462.13290@i42g2000cwa.googlegroups.com>,) "tadamsmar" <tadamsmar@yahoo.com> writes:  > > > > > I get a DECnet communications error "Line open error, LineJ > > communications error" when the X-terminal is loading software (it doesF > > not have to load from the node that host it for Dec$windows by theK > > way.)   Maybe restarting DECnet would do the trick.  I will try that in  > > the future.  > F >    That error is common when a node sees a MOP boot request, but theC >    system making the request is not in that node's MOP data base.  > F >    We used to see it all the time when we had a few LAVC sharing theJ >    same ethernet.  Each boot host would see and reject MOP load requestsH >    from the other boot node's satellites, with the above error message >    posted via OPCOM. > J >    We simply went into the DECnet data base and removed logging for thatD >    message.  It was causing confusion amoungst the uneducated, who; >    thought there "must be" a problem needing to be fixed.  > H >    We kept the message in our development cluster so I could tell whatB >    was going on when I added new satellites.  I was the only one, >    reading OPCOM messages on that cluster. >  > J The "line open" error also occurs on hosts that have the service attribute> disabled on the DECnet circuit that receives the load request.   ------------------------------    Date: 28 Aug 2006 16:39:05 -02006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)& Subject: [OpenVMS Patches] Difference?, Message-ID: <44f31c29$1@news.langstoeger.at>  + Can anybody tell me the difference between    < VAXSHAD03_073.A		26-JUN-2006 09:15:00.40		I have since weeks and < VAXSHAD03_073.A		25-AUG-2006 13:50:38.69		I downloaded today   ??  L Why can different kits (date, content, MD5 is different, don't know funtion) not get different names?  C btw: Something similar happened to VMS732_MP-V0100 a few weeks ago. B (But there was also a hint regarding UPDATE kit requirement added)  - Where is QA when someone needs it? Pleeeaaase    -EPLAN  J PS: Many thanks to the folks which are responsible for the change of DCXedK .PCSI files to ZIPSFXed .PCSI[$COMPRESSED] files. I waited (and lobbied ;-) C for this many years, but finally it's done (totally unexpected ;-).  --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------   End of INFO-VAX 2006.477 ************************                                                                                                             directory.  <<< CWD /info-vaxs1 >>> 250 Connected to /disk$misc/decus/info-vax.s	 <<< PWDo; >>> 257 "/disk$misc/decus/info-vax" is current directory.  <<< TYPE I >>> 200 Type I ok. <<< LIST -la 2005_121.txt  >>> 150 List started./ >>> 226 Transfer completed.< <<< PORT 24,5,236,254,17,85e3 >>> 200 Port 17.85 at Host 24.5.236.254 accepted.6 <<< LIST 2005_121.txtr >>> 150 List started.r >>> 226 Transfer completed.0 <<< PORT 24,5,236,254,17,86o3 >>> 200 Port 17.86 at Host 24.5.236.254 accepted.f <<< RETR 2005_121.txteZ >>> 150 IMAGE retrieve of /disk$misc/decus/info-vax/2005_121.txt (112096 bytes) started.< >>> 226 Transfer completed.  106774 (8) bytes transferred. <<< CWD /info-vax1 >>> 250 Connected to /disk$misc/decus/info-vax.d	 <<< PWDO; >>> 257 "/disk$misc/decus/info-vax" is current directory.. <<< PORT 24,5,236,254,17,87_3 >>> 200 Port 17.87 at Host 24.5.236.254 accepted.r <<< REST 106774 7 >>> 500 I never heard of the REST command.  Try HELP.. <<< CWD /info-vax 1 >>> 250 Connected to /disk$misc/decus/info-vax.o	 <<< PWDc; >>> 257 "/disk$misc/decus/info-vax" is current directory.2 <<< REST 106774d7 >>> 500 I never heard of the REST command.  Try HELP. <<< CWD /info-vax/1 >>> 250 Connected to /disk$misc/decus/info-vax./	 <<< PWDe; >>> 257 "/disk$misc/decus/info-vax" is current directory., <<< REST 106774o7 >>> 500 I never heard of the REST command.  Try HELP.7 <<< CWD /info-vax 1 >>> 250 Connected to /disk$misc/decus/info-vax.i	 <<< PWD; >>> 257 "/disk$misc/decus/info-vax" is current directory.  <<< REST 106774s7 >>> 500 I never heard of the REST command.  Try HELP.7  <<< CWD /info-vax/2005_120.txt9 >>> 550 You are not permitted to access this directory.  <<< CWD /info-vaxs1 >>> 250 Connected to /disk$misc/decus/info-vax.s	 <<< PWDo; >>> 257 "/disk$misc/decus/info-vax" is current directory.  <<< TYPE I >>> 200 Type I ok. <<< LIST -la 2005_120.txt  >>> 150 List started./ 