1 INFO-VAX	Thu, 13 Jun 2002	Volume 2002 : Issue 325       Contents: Re: "Tru64 and OpenVMS Times" 4 Re: A dvdwrite(r)-Program: save 4.3 GB on a DVD-R(W)& Re: Alpha to ia64: where is the issue?. Re: AlphaServer 255 - Firmware upgrade Misfire. Re: AlphaServer 255 - Firmware upgrade Misfire# Re: Carly was here in ZKO yesterday # Re: Carly was here in ZKO yesterday  Re: Copying a file via FID Re: Could linux become VMS?  Re: Could linux become VMS?  Re: Could linux become VMS?  Re: Could linux become VMS?  DECWindows "hang"  Re: DEFINE/TRANS=TERM  Re: DEFINE/TRANS=TERM  Re: DEFINE/TRANS=TERM  Re: DEFINE/TRANS=TERM  Re: Fibre Disk vs. SCSI Disk Re: Fibre Disk vs. SCSI Disk Re: Fibre Disk vs. SCSI Disk  Re: Fibre disks free space drift9 Re: How can I tell which version of DECNet I am running ?  Re: how to renew a vmasmail ?  Re: how to renew a vmasmail ? + Re: howto create self-extracting zip files? + Re: howto create self-extracting zip files? " Re: HP MPE to HP-UX customer plans* Re: Job Opening: Chief Competitive Officer* Re: Job Opening: Chief Competitive Officer+ Re: Looking for Unix commands to do VMS DCL + Re: Looking for Unix commands to do VMS DCL . Mail masquerading using Compaq TCP/IP services, Mail masqurading with Compaq TCP/IP services0 Re: Mail masqurading with Compaq TCP/IP services( My conversation with Linus about VMS ... Need DECNet packet information" Re: Need DECNet packet information" Re: Need DECNet packet information Re: No new Alpha sales Re: Open Letter to HP  Re: Open Letter to HP  Re: Open Letter to HP  Re: Open Letter to HP  Re: Open Letter to HP  Re: Open Letter to HP  Re: Open Letter to HP  RE: Open Letter to HP  Re: Open Letter to HP  Re: Open Letter to HP  Re: OpenVMS Mailing Lists? Re: OpenVMS Mailing Lists?G RE: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading) G Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading) G Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading) " Re: rooted devices and /TRANS=CONC RE: SCSI- Fatal Drive Error 1 Re: Sun is on the way out!  Only HP and IBM left! 1 Re: Sun is on the way out!  Only HP and IBM left! 1 Re: Sun is on the way out!  Only HP and IBM left!  Re: SYSMAN default parameters  Re: SYSMAN default parameters & Re: tpc/ip file server for VMS systems@ VMS Groups and Privileges (was: Carly was here in ZKO yesterday)D Re: VMS Groups and Privileges (was: Carly was here in ZKO yesterday)D Re: VMS Groups and Privileges (was: Carly was here in ZKO yesterday)( Re: What process is using this pagefile?( Re: What process is using this pagefile?( Re: What process is using this pagefile? Re: xtoolkit error  F ----------------------------------------------------------------------  # Date: Wed, 12 Jun 2002 18:44:52 GMT 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> & Subject: Re: "Tru64 and OpenVMS Times"2 Message-ID: <EEMN8.26$ck5.765066@news.cpqcorp.net>   > ? >> "Terry C. Shannon" <terryshannon@attbi.com> wrote in message    >> >L >> > That is a very good reason. I initially planned to call SKC SKHPS as inJ >> > Shannon Knows High Performance Systems. I was advised by a senior HPQF >> > executive to go with another name  because HPS does in fact equal Hewlett  >> > Packard Services. >   J I am also told that in general you can't have a TLA that ends in an "S" asL it is not distingushable from a plural form of a shorter name.  For example,E I wanted to use XUIS as a product name, but was told that there was a / trademarked name XUI - so I ended up with UISX.    ------------------------------  # Date: Thu, 13 Jun 2002 01:15:10 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> = Subject: Re: A dvdwrite(r)-Program: save 4.3 GB on a DVD-R(W) ' Message-ID: <3D07F5D1.495BE221@fsi.net>    Rudolf Wingert wrote:  >  > Hello, >  > Atlant Schmidt wrotes: >  > >>> 4 > Also, yes, DQDRIVER is perfectly happy reading DVD8 > discs. But most real-world DVD discs are written using7 > the UDF file system. Last I knew, VMS didn't know how 8 > to read that file system. But a 4.7GB ODS-2 DVD should > work just fine.  > <<<  > E > Is it possible to mount it with the UCX V4.2 or TCPip 5.1 services?   4 Do you mean when offered as a NFS by another server?  H Otherwise, no - no TCP/IP stack I'm aware of adds any support to VMS for additional filesystems.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Thu, 13 Jun 2002 00:43:26 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> / Subject: Re: Alpha to ia64: where is the issue? ' Message-ID: <3D07EE60.9E937071@fsi.net>    Rob Young wrote: > \ > In article <3D0583B5.2E2E577@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes: > > "Main, Kerry" wrote: > >>
 > >> Arne, > >>I > >> >>>   2) even if it is just a re-compile, then there are still costs ? > >>      (hardware and software purchase, testing etc.etc.)<<<  > >>K > >> While you are correct, something to keep in mind from a developers and H > >> support point of view - assuming a SAN is in place, the same systemI > >> IPF HW will be able to be used in the future to test and support not I > >> only multiple versions of OpenVMS, but also W64 (whenever it becomes I > >> available), HP-UX, or Linux. Simply point the boot device to another  > >> part of the SAN.  > > F > > Of course, the problem there is WhineBloze's nasty little habit ofK > > assuming it can write a "harmless" signature to (i.e., trash the zeroth K > > block of) every disk it sees, unless this is finally corrected by then.  > >  > 5 >         Yes, it is corrected.  It is called zoning.   6 Correction: that's a work-around. The problem remains.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Wed, 12 Jun 2002 16:21:12 -0400 ' From: "Zeni Schleter" <zzb@y12.doe.gov> 7 Subject: Re: AlphaServer 255 - Firmware upgrade Misfire , Message-ID: <ae8afk$rkl$1@sws1.ctd.ornl.gov>  L Thanks for the information.  I have the AlphaServer back as a OpenVMS systemJ .  Was able to put the SRM back using the AlphaBIOS upgrade option for the9 menu screens and could use the Firmware CD as the source.   L I also have saved the link for the EISNER.DECUS.ORG VMSFAQs as well as a few* other sites for AlphaServer 255 specifics.   ------------------------------    Date: 12 Jun 2002 13:57:46 -0700( From: bob@instantwhip.com (Bob Ceculski)7 Subject: Re: AlphaServer 255 - Firmware upgrade Misfire = Message-ID: <d7791aa1.0206121257.3c04fa67@posting.google.com>   [ "Zeni Schleter" <zzb@y12.doe.gov> wrote in message news:<ae7k6k$uei$1@sws1.ctd.ornl.gov>... M > I tried to do a firmware upgrade v5.4 on an AlphaServer 255.  The menu that N > I was given did not appear as I thought it should but I thought that was theM > difference between the 255 and the 2100 that I had done.  I updated the SRM K > .  I was expecting several other items but that didn't happen.  I updated J > the ARC .  There were also some other items that I could have chosen butF > didn't.  I powered off and on.  Now, I am getting a screen saying noK > Operating system selections found.  And when I step through several other F > screens I find that the Operating information it is defaulting to is& > WindowsNT.  This is not what I want. > E > The original system disk is OpenVMS 6.2.  I am trying to replace my M > AlphaStation OpenVMS 7.2-1 with this 255.  When I tried to boot the OpenVMS M > 7.2-1 system , I got the message about upgrading the firmware.  That is why  > I am where I am. > I > I looked at the release notes after the fact.  I still don't see in the M > notes what I saw on the screen.  I have a call out to the original owner to  > get the manual.  > N > I am posting this just to get confirmation that I can get the system back to > the OpenVMS variety.  D I tried a 5.4 upgrade on an Alphastation 500, it failed, and retriedE it again, and it worked ... found out later that 5.4 cd had a problem B updating Alphastation 500's from Q hardware support ... maybe thisA affected more than Alphastation 500's?  I was told not to upgrade  these boxes w/5.4 cd ...   ------------------------------  # Date: Wed, 12 Jun 2002 18:10:09 GMT ( From: Don Sykes <annonymous@pacbell.net>, Subject: Re: Carly was here in ZKO yesterday+ Message-ID: <3D078EE0.8CE3F2F3@pacbell.net>    Larry Kilgallen wrote: > X > In article <3D06978F.4B8B4A60@pacbell.net>, Don Sykes <annonymous@pacbell.net> writes: > T > > Not exactly what I had in mind. I agree ODS5 and COE bring some crucial parts ofK > > Unix into VMS, but even some VMS features could do with a rethink. E.g. R > > privileges are not logical. They seemed to have evolved as needed, so there is > E > "More logical" presumes they are orthogonal, which is not the case. D > Privilege _levels_ (categories) are worth thinking about, and theyH > are the basis on which you should evaluate privileges.  But privilegesH > on a level can be used to bootstrap into a different privilege.  ThinkM > of privilege levels as providing real control and the individual privileges F > within a level as providing protection against typographical errors.  E That's what I meant. More logical meaning architected, and as you say J pivilege_levels are a good way to start looking at such a re-architecting.   > Q > > no uniform architecture to them. Users cannot belong to more than 1 group and  > D > In what way do general identifiers fall short of what you see fromC > Unix groups ?  Certainly giving concepts the same name as in Unix  > is not a real requirement.  L Identifies are fine, but they can be hard to track - e.g. to see how sets of/ identifiers impact sets (groups) of users, etc. M I not suggesting making them the SAME as Unix groups. They don't fit the bill H either. I mean a way for administrator to delegate certain functions andP privileges to group administrators. This is a big topic and would require a goodO deal of thought to properly implement, but its worth it especially where large, # complex enterprises are concerned.     --     Have VMS. Will Travel. Wire Paladin (@alphase.com) 
 San Francisco    ------------------------------  # Date: Wed, 12 Jun 2002 18:26:20 GMT ( From: Don Sykes <annonymous@pacbell.net>, Subject: Re: Carly was here in ZKO yesterday+ Message-ID: <3D0792AB.F1116A6D@pacbell.net>    Rob Young wrote: > X > In article <3D06978F.4B8B4A60@pacbell.net>, Don Sykes <annonymous@pacbell.net> writes: > > Bob Ceculski wrote:  > >>` > >> Don Sykes <annonymous@pacbell.net> wrote in message news:<3D0648CA.936EE87F@pacbell.net>...  > >> > "Terry C. Shannon" wrote: > >> > > > >> >  <snip> > >> > >R > >> > > Hypothetically speaking, if HPQ were able to take HP-UX and endow it withT > >> > > ALL the goodness of VMS (clustering, DLM, security, stability, reliability,U > >> > > enhanced management, dynamic partitioning, Galaxy, etc, etc) would it really 2 > >> > > make any difference if VMS still existed? > >> > > > >> > > Just wondering. > >> >H > >> > Terry. What a question for this group! This could lead to another > >> > never-ending-thread! W > >> > But, I agree in theory. Even more I think it's time for a new OS that would be a P > >> > superset of all Unix & VMS functionality. With such an approach you couldW > >> > support both classic Unix & classic VMS environments under the same umbrella and U > >> > begin to build an even better OS environment (shell, symbolic structures, more K > >> > logical privileges, etc, etc) that folks could gradually migrate to. D > >> > I've mentioned this idea before, but no one seemed to get it. > >>L > >> you already have it!  VMS now w/ODS5 and COE is that meshing of VMS andJ > >> unix, plus, you have all those great features that make VMS the best!/ > >> It already exists, why reinvent the wheel?  > > T > > Not exactly what I had in mind. I agree ODS5 and COE bring some crucial parts ofK > > Unix into VMS, but even some VMS features could do with a rethink. E.g. R > > privileges are not logical. They seemed to have evolved as needed, so there isQ > > no uniform architecture to them. Users cannot belong to more than 1 group and S > > there is no concept of a true "group administrator". Also, super symbols - many M > > different levels of symbols from thread specific, to environment specific P > > (beyond the cluster) could do a lot more than symbols and logicals do now. AP > > true "Shiva" OS would allow Unix & VMS environments to coexist and allow newQ > > ones to develop. Unix users shouldn't have to learn DCL and VMS'ers shouldn't Q > > have to know csh, ksh or the like. It would be nice to be able to choose your E > > development & production environments, maybe depending on a group  > > administrator's parameter.C > > There are lots of possibilities once you begin to open the box.  > >  > F >         Ummm... but unlike POSIX of old (slow performance, etc.) COEB >         will give you ksh.  As Hoff alluded to in another thread > ] > http://groups.google.com/groups?selm=zvdh7.692%24bB1.31613%40news.cpqcorp.net&output=gplain  > @ > : long discussion of the benefits of mount points over volumes > I >   We have added mount point support into OpenVMS as part of the DII COE H >   work, permitting at least some of the same benefits (and, of course,D >   the same problems) that are available with the UNIX environment. >  >  --- > ; >         You will have Unix mount points so I assume that:  >  >                 # cd / >                 # rm -r *  >  >         should work.   ;-) > J >         As I had mentioned several times, it really isn't a big deal forK >         VMS to become a superset of Unix (not negating the work involved, O >         just comparing one paradigm to another and stuffing the Unix paradigm  >         into VMS). > G >         Some of the shells are missing as Tom Kednos points out.  But O >         once all the supporting infrastructure shows up (a true fork() , long G >         discussed) it should be much easier to port Unix apps to VMS.  > @ >         This master OS you are describing is Mica.  Mica died. > I Maybe Mica was ill from the start. I don't know Mica, but a short sighted K project charter or project leader can inadvertently doom a project from the P start. Just because there may be similarities with the OS I'm trying to describe doesn't mean it's the same.  --     Have VMS. Will Travel. Wire Paladin (@alphase.com) 
 San Francisco    ------------------------------    Date: 13 Jun 2002 06:13:19 +0800, From: Paul Repacholi <prep@prep.synonet.com># Subject: Re: Copying a file via FID - Message-ID: <873cvskuww.fsf@prep.synonet.com>   0 Jack.trachtman@vmmc.org (Jack Trachtman) writes:   > More detail:   B > We have an application that uses Spooled LAT devices to send its> > output to printers.  To be able to debug some of the reportsB > generated, we need to capture these reports.  Unfortunately, VMSF > does not assign a file name to these files, only the (required) FID,8 > and the file name shows up in the print que as "[].;".   F > I can get the FID via F$GETQUI and I can look at the directory entryF > using DFU. But I can't figure out (without writing a program) how to. > copy the file within DCL using just its FID.   F > Or maybe there's a way to temporarily direct the output from a print > que to a file???  L ANAL/DISK/REP will drop it into [syslost] for you. Tested on a SYSUAF.DAT :(> ::note to self: next time, PRINT/DEL sysuaf.LIS, not .dat...::   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Thu, 13 Jun 2002 00:07:52 GMT ' From: Jeffrey Chimene <jeff@nospam.net> $ Subject: Re: Could linux become VMS?) Message-ID: <3D07E2D6.F637F3F@nospam.net>    Brass Christof wrote:   C > The commands' parameters and options are a complete mess on UNIX.  > C > The standardised API for UNIX is that low level that tons of code2 > are duplicated with the apps.d >e  O The GNU environment has gone to great lengths to standardize the parameter set.6  : That said, I agree with the fundamentals of your argument.  , > Why isn't there anything like CDL on UNIX? >sF > A mature OS should provide a more complete base and more concise API > than UNIX.  Q Command language parsing is available as an API in the GNU library. One must lookeT farther afield in the Unix environment for solutions than VMS. One doesn't find manyU VMS APIs compared to Unix. It's a difference in philosophy. Vendors in the Unix space T distinguish themselves via their APIs. In the VMS space, there's only one vendor. MyR guess is that the same arguments can be heard on the Non-Stop Kernel and OS/VS, &c listservers.    P That said, it would be nice to have a CDU API in Unix. It's probably do-able via BISON/YACC/Flex.   -- microsoft free by 2003   ------------------------------  # Date: Thu, 13 Jun 2002 00:54:25 GMT:1 From: "David J. Dachtera" <djesys.nospam@fsi.net>C$ Subject: Re: Could linux become VMS?' Message-ID: <3D07F0F1.622152C6@fsi.net>o   "Craig A. Berry" wrote:C > ) > In article <3D058592.1AF7C0BC@fsi.net>,c5 >  "David J. Dachtera" <djesys.nospam@fsi.net> wrote:a >  > > Aristotle SnowNasis wrote:! >  VMS can be so obtuse at times,a > >VD > > Examples, please? Hard to understand the reference without them. > H > Well, how about having to learn how to disable the windowing system soB > you can install the license that allows you to run the windowing
 > system?    I believe that's in the FAQ.  ? > I've also never seen AUTOGEN successfully increase the number  > of global sections;   ? Add "MIN_GBLSECTIONS = value" to MODPARAMS.DAT. See the OpenVMSl$ documentation or view it on-line at:  " http://www.openvms.compaq.com/doc/  4 > it always loops endlessly through multiple reboots: > but never actually changes anything until you whack in a$ > MIN_GBLSECTIONS to modparams.dat.   @ That's the documented method. Every product you install (should)9 indicate any requirements it may have for addition globalu sections/pages.u   > The only thing that makes the I > experience of a first-timer installing VMS reasonable is the experience E > of a first-timer installing Linux.  We already do better than that, ; > which isn't saying much, but we don't do well enough IMO.a > * > > > whereas MOST Unix have configuration' > > > files in centrally located areas.a > J > As does VMS, though the definition of "central" may not be the same, andJ > there are a variety of security, efficiency, and useability reasons thatC > the windoze .ini file approach is not right for every facility or- > application. > - > >> Also, help on VMS is quite brief and not N > > > always very helpful. On the other hand, man pages, while sometimes beingO > > > tautological are full of information, especially that parts that describe.0 > > > the files used in such-and-such a program. > >oH > > That's what documentation is for. However, since none comes with theL > > average UN*X variant, the man pages are about all there is - ANYWHERE! - > G > I think the commercial Unices do a bit better than that, Dave.  Tru64s > certainly does.e  A Really? Does Tru64 ship with the equivalent of "great grey wall"?g  H > >  Even today, there is still a great preference for hard-copy doc.'s. > C > The difference between hard copy and electronic is a pint of ink.S  D ...and several (hundred) tons of offset litho and binding equipment.  O > > > By the way, I'm an administrator on several unix and vms platforms (vax & 
 > > > alpha).M > F > You'd really do well, then, Aristotle, to acquaint yourself with the< > thousands of pages of excellent documentation available atH > <http://www.openvms.compaq.com/doc>.  These are also included in everyF > OpenVMS software distribution.  If you don't know how to install theB > documentation on your OpenVMS system (or any other HTML- or PDF-G > capable system for that matter) please ask for help here or call yourh > support center.  > F > > > Overall, VMS has stood still for years while Unix has marched on > F > Patently false.  Does VMS have room to grow?  Sure.  But how many of > your Unices can do Galaxy?  & Why do I want to say, "Think: VMware"?   -- f David J. Dachterao dba DJE Systemsn http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/A   ------------------------------  # Date: Thu, 13 Jun 2002 00:55:21 GMT-1 From: "David J. Dachtera" <djesys.nospam@fsi.net>:$ Subject: Re: Could linux become VMS?' Message-ID: <3D07F12B.7303EF4E@fsi.net>p  ( Andrew Harrison SUNUK Consultancy wrote: >  > David J. Dachtera wrote: >  > > Aristotle SnowNasis wrote: > H > > That's what documentation is for. However, since none comes with theL > > average UN*X variant, the man pages are about all there is - ANYWHERE! -K > > other than things like the on-line FreeBSD doc.'s and such. Even today,o; > > there is still a great preference for hard-copy doc.'s.a > >r > >  > F > I have to dissagree. Solaris for example comes with something calledC > answerbook which is full doumentation on CD. If you don't want tobI > use that and have a browser connected to the internet then docs.sun.comrD > and or sunsolve also provide full documentation and patch reports.  G ...and if you don't have another computer running, your screwed withoute the hard-copy.   -- u David J. Dachterae dba DJE Systemss http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Thu, 13 Jun 2002 01:11:58 GMTn1 From: "David J. Dachtera" <djesys.nospam@fsi.net>s$ Subject: Re: Could linux become VMS?' Message-ID: <3D07F50E.97CC8F9D@fsi.net>-   Bill Gunshannon wrote: > ) > In article <3D058592.1AF7C0BC@fsi.net>,i6 >  "David J. Dachtera" <djesys.nospam@fsi.net> writes: > |>I > |> That's what documentation is for. However, since none comes with the K > |> average UN*X variant, the man pages are about all there is - ANYWHERE!5 > ( > Haven't been to the bookstore lately??  B Actually, I've been to the bookstores and the computer stores, butC nowhere did I see a Linux or *BSD distro. that comes with three bigc( boxes of hard-copy doc.'s like VMS does.   -- u David J. Dachterau dba DJE SystemsD http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/e   ------------------------------    Date: 12 Jun 2002 21:52:20 -0700" From: cstranslations@msn.com (Joe) Subject: DECWindows "hang"; Message-ID: <d56d1c2d.0206122052.a643fb@posting.google.com>r   OpenVMS 7.1-1H2g DWMotif V1.2-4
 Alpha 4100  A We have a frequent problem with the console that works roughly astE follows: DECWindows is started during boot, next morning no login box 8 - just a "blue-gray screen." Telnet in (or such) and runA sys$manager:decw$startup. Go home at end of the day. Come in nextu3 morning and repeat with sys$manager:decw$startup...e  F This has been mostly an annoyance. If console access is needed we justD login other than via the console restarting DECWindows as above. TheA afternoon during a power outtage when we were on UPC there was nopB other way in so the "blue-gray screen with no login box" was a wee problem.  F I'm guessing we have something that "dumps on the console" so to speakD somewhere along the line during nightly batching (what ever it is itD probably worked well enough when the Alpha was a VAX and the console, was a VT but that was then and this is now).  E Other than search through more source code than I care to looking for C matchs on something like OPA I was thinking put an alarm ACE on the F OPA devices then work backwards from whatever messages might be in the3 operator log to an offending batch job and program.h   $ set sec/class=device -C    /acl=(alarm=security,option=default,access=success+failure) opa0i $ set sec/class=device -C    /acl=(audit=security,option=default,access=success+failure) opa0    and then (for instance)a   $ open/write foo opa0: $ write foo "hello"   F 1) Generates no opcom message and leaves nothing in the audit log as I guess I'm expecting.  F 2) Turns the console into a "dumb terminal" displaying "hello" until IC do a Ctl-F2 to restore the windowing system and doesn't give me the-+ "blue-gray screen" I guess I was expecting.0   Anyone have any ideas?   tiau joeb   ------------------------------    Date: 12 Jun 2002 13:45:02 -0600 From: briggs@encompasserve.org Subject: Re: DEFINE/TRANS=TERM3 Message-ID: <Zj+GqHJvKWiK@eisner.encompasserve.org>   w In article <01KIUS021WWW96WTPR@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes: K >> >> I could be wrong, but I was under the impression that /Trans=Term was J >> >> there to prevent a user from re-directing a logical someplace else.  >> >  K >> > Well, that is a feature (i.e. if DISK$ALPHASYS is DSA0:, a user can't  L >> > define DSA0: to XXX if DISK$ALPHASYS is defined /TRAN=CONC).  However, E >> > this feature would be useful for all types of logicals, whereas lH >> > /TRANS=TERM only works if the target of the definition really is a  >> > physical device.i >> a >> That's patently untrue. > J > Sorry, that's not what I meant.  Of course one can define a logical name> > which is the same as a device name, but in the example aboveF > DISK$ALPHASYS won't translate to it, since the translation stops at A > DSA0:.  So although the logical can be defined, it won't cause s+ > DISK$ALPHASYS to point to somewhere else.   C /TRANS=TERM works whether the equivalence name is a physical device B name, a yes/no parameter value or the number of peanuts in the bagB on my desk expressed in roman numerals.  Its effect is to suppress= further automatic iterative logical name translation.  And itt' is successful in doing that.  It works.i  C If you think that it fails in some other ill defined purpose, you'dr6 better define that purpose before elaborating further.   	John Briggs   ------------------------------  + Date: Wed, 12 Jun 2002 21:15:20 +0100 (MET)l9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>l Subject: Re: DEFINE/TRANS=TERM; Message-ID: <01KIUUXX8ES696WTPR@sysdev.deutsche-boerse.com>e  E > /TRANS=TERM works whether the equivalence name is a physical device-D > name, a yes/no parameter value or the number of peanuts in the bagD > on my desk expressed in roman numerals.  Its effect is to suppress? > further automatic iterative logical name translation.  And ita) > is successful in doing that.  It works.  > E > If you think that it fails in some other ill defined purpose, you'd>8 > better define that purpose before elaborating further.  D I never said it didn't work.  I was just wondering what the initial C reason for introducing it was (save CPU cycles, protect users from fE unexpected behaviour if they define a logical which is the same as a sH device name to which another logical points, stop iterative translation , when this is desired for other reasons etc).  D With regard to TRANS=CONC, my question is why disk names can't have ? iterative translations if (CONC,NOTERM) is involved (TERM I cany understand), e.g.    $ sh log/ful disk$test9 %SHOW-S-NOTRAN, no translation for logical name DISK$TEST  $ sh log/ful disk$fooo8 %SHOW-S-NOTRAN, no translation for logical name DISK$FOO $ sh log/ful disk$alphasys9    "DISK$ALPHASYS" [exec] = "DSA0:" [concealed,terminal] o (LNM$SYSTEM_TABLE)  $ define disk$test disk$alphasys $ dir/grand disk$test:[000000]  % Grand total of 1 directory, 18 files.n $ define disk$foo disk$test, $ dir/grand disk$foo:[000000]e  % Grand total of 1 directory, 18 files.i% $ define disk$foo disk$test/tran=conci@ %DCL-I-SUPERSEDE, previous value of DISK$FOO has been superseded $ dir/grand disk$foo:[000000] : %DIRECT-E-OPENIN, error opening DISK$FOO:[000000] as inputK -RMS-F-DEV, error in device name or inappropriate device type for operatione  & Grand total of 0 directories, 0 files.* $ define disk$test disk$alphasys/tran=concA %DCL-I-SUPERSEDE, previous value of DISK$TEST has been superseded  $ dir/grand disk$test:[000000]; %DIRECT-E-OPENIN, error opening DISK$TEST:[000000] as inputsB -RMS-F-DEV, error in device name or inappropriate device type for 	 operationt  & Grand total of 0 directories, 0 files.  = In other words, why does /TRAN=(CONC,NOTERM) cause RMS-F-DEV.n   One can get round this byy  7 $ define disk$test 'f$trnlnm("disk$alphasys")/tran=concbA %DCL-I-SUPERSEDE, previous value of DISK$TEST has been superseded  $ dir/grand disk$test:[000000]  % Grand total of 1 directory, 18 files.t  E which has the advantage that the only translation is done during the CF definition, but does not allow disk$alphasys to be redefined and have * this propagate automatically to disk$test.  E If disk$test, however, is defined /TRAN=CONC, then one cannot define iE disk$foo to be 'f$trnlnm("disk$test")' and have the dir command work.g   ------------------------------    Date: 12 Jun 2002 16:27:39 -0700. From: SPAMSINK2001@YAHOO.COM (Alan E. Feldman) Subject: Re: DEFINE/TRANS=TERM= Message-ID: <343f30ae.0206121527.32ccd7e2@posting.google.com>m  l "Randy Park" <rjpark@mindspring.nospaam.com> wrote in message news:<ae63ss$9c8$1@slb5.atl.mindspring.net>...= > Here's my understanding of how I used it and why I used it.O > ? > I managed a small VMScluster and network, consisting of about 8 > 10 - 12 disk drives.  Some of the disks would be local? > and some would be cluster and some would be remoted depending ; > upon which system you were on.  The startup files on each = > system would be slightly different but would be similar to: ' >         DEFINE            DISKA DUA0: ' >         DEFINE            DISKB DUA1:l/ >         DEFINE/TRANS=TERM DISKC ROCKY$DKA300: / >         DEFINE/TRANS=TERM DISKD ROCKY$DKA200:  >         :u >         :o/ >         DEFINE/TRANS=TERM DISKI BORIS$DKA400:d > 8 > Now...  RMS, when accessing files, translates logicals= > all the way, so that it can access the file.  RMS will alsot> > return to the program the name of the file after the various@ > logical translations have been performed, but for the purposes9 > of this returned filename, it stops translating when itc5 > encounter a logical with the termination attribute.I > A > For example, if I:  DEFINE MYORDER DISKC:[DATA]ORDER_MASTER.DATe< > and then use the filename MYORDER in a utility program, or? > even a sophisticated application program, RMS will internally : > translate MYORDER to ROCKY$DKA300:[DATA]ORDER_MASTER.DAT< > but RMS will return to the program the translated filename8 > of DISKC:[DATA]ORDER_MASTER.DAT. Given the appropriate; > option or qualifier, the program may display some kind ofi? > confirmation using a filename of DISKC:[DATA]ORDER_MASTER.DAT < > because that's the translated filename given to it by RMS.  t  F That would be true if you had used /TRANSLATION_ATTRIBUTES=CONCEALED!.1 The terminal attribute has nothing to do with it.     = o  CONCEALED - hides the equivalence name from command outputo  F o  TERMINAL - prevents the equivalence name from being translated as aA logical name, i.e., iterative translation is aborted when such an;; equivalence name is encountered in logical name translatione  A The terminal attribute for DISKC prevents ROCKY$DKA300 from beingg/ translated if it happened to be a logical name.i   Example:   $ DEFINE AAA DSA0: /TRAN=TERMy $ DIR AAA:[000000]000000.DIR    Directory DSA0:[000000]f  ?= 000000.DIR;1                4/9       17-OCT-1991 13:30:30.91   a Total of 1 file, 4/9 blocks. $e  D It is DSA0: that appears in the output, not AAA. You need /TRAN=CONC
 to hide DSA0:    $ DEFINE AAA DSA0: /TRAN=CONC0; %DCL-I-SUPERSEDE, previous value of AAA has been superseded  $ DIR AAA:[000000]000000.DIR    Directory AAA:[000000]  C= 000000.DIR;1                4/9       17-OCT-1991 13:30:30.91e    Total of 1 file, 4/9 blocks.  C I think it helps to remember that it is the *equivalence name* that0E has these attrributes, not the logical name. For example, let's check: out SYS$SYSROOT:   $ SHOW LOG/FUL SYS$SYSROOTC    "SYS$SYSROOT" [exec] = "xxx$DKA100:[SYS0.]" [concealed,terminal]a (LNM$SYSTEM_TABLE)         = "SYS$COMMON:"X7 1  "SYS$COMMON" [exec] = "xxx$DKA100:[SYS0.SYSCOMMON.]" ' [concealed,terminal] (LNM$SYSTEM_TABLE)   F Here we see that SYS$SYSROOT's first equivalence name is concealed but= its second equivalence name, SYS$COMMON, isn't. Note that thenF equivalence name *for* SYS$COMMON *does* have the concealed attribute,? but SYS$COMMON as an equivalence name for SYS$SYSROOT does not.n    ? > This allows the designer of the logical to control what level A > of translation the end user will see.  In this case the logicale> > disk DISKC is separated from the physical disk ROCKY$DKA300.      What does "separated" mean here?    > > This is how I have used this feature of logical definitions. >  > F > Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote in message7 > news:01KITFNFA6JK96WTPR@sysdev.deutsche-boerse.com.../K > > What is the real purpose of /TRANS=TERM?  I know what it DOES, I'm justi$ > > wondering why this is necessary. > >eK > > Is it just to tell the OS "hey, I'm telling you this doesn't need to behI > > translated further, so you don't need to waste any resources trying"?n > > K > > Of course, specifying /TRANS=TERM when this isn't the truth leads to anoG > > error.  The opposite, leaving it out when it could be specified: is>: > > there any situation where this could lead to an error? > > C > > Presumably the performance hit caused by trying to translate itVL > > iteratively only to find that one is already at the terminal translationJ > > is negligible on a modern system (but might have been an issue back in > > the 11/780 days).s > >t     Disclaimer: JMHO Alan E. Feldmant" afeldman atski gfigroup dotski com   ------------------------------    Date: 12 Jun 2002 19:26:50 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman) Subject: Re: DEFINE/TRANS=TERM= Message-ID: <b096a4ee.0206121826.59389d64@posting.google.com>o  | Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote in message news:<01KIUUXX8ES696WTPR@sysdev.deutsche-boerse.com>... [snip]F > With regard to TRANS=CONC, my question is why disk names can't have A > iterative translations if (CONC,NOTERM) is involved (TERM I canm > understand), e.g.  >  > $ sh log/ful disk$test; > %SHOW-S-NOTRAN, no translation for logical name DISK$TEST  > $ sh log/ful disk$foo : > %SHOW-S-NOTRAN, no translation for logical name DISK$FOO > $ sh log/ful disk$alphasys; >    "DISK$ALPHASYS" [exec] = "DSA0:" [concealed,terminal] a > (LNM$SYSTEM_TABLE)" > $ define disk$test disk$alphasys  > $ dir/grand disk$test:[000000] > ' > Grand total of 1 directory, 18 files.c > $ define disk$foo disk$testd > $ dir/grand disk$foo:[000000]t > ' > Grand total of 1 directory, 18 files.      Change thisf  ' > $ define disk$foo disk$test/tran=concy                              A to this   (   $ define disk$foo disk$test:/tran=conc   and it will work.   C *Always* include a trailing colon when the equivalence name in yourlE DEFINE command is a device name. Doing so also avoids one of the bugs F in SET DEFAULT. I'm pretty sure one of the manuals tells you to alwaysF do this somewhere, though I don't remember exactly where right now and I don't think it explained why.-    B > %DCL-I-SUPERSEDE, previous value of DISK$FOO has been superseded > $ dir/grand disk$foo:[000000]o< > %DIRECT-E-OPENIN, error opening DISK$FOO:[000000] as inputM > -RMS-F-DEV, error in device name or inappropriate device type for operationy > ( > Grand total of 0 directories, 0 files., > $ define disk$test disk$alphasys/tran=concC > %DCL-I-SUPERSEDE, previous value of DISK$TEST has been superseded   > $ dir/grand disk$test:[000000]= > %DIRECT-E-OPENIN, error opening DISK$TEST:[000000] as input D > -RMS-F-DEV, error in device name or inappropriate device type for  > operation  > ( > Grand total of 0 directories, 0 files. > ? > In other words, why does /TRAN=(CONC,NOTERM) cause RMS-F-DEV.n >  > One can get round this byb > 9 > $ define disk$test 'f$trnlnm("disk$alphasys")/tran=concnC > %DCL-I-SUPERSEDE, previous value of DISK$TEST has been supersededr  > $ dir/grand disk$test:[000000] > ' > Grand total of 1 directory, 18 files.  > G > which has the advantage that the only translation is done during the aH > definition, but does not allow disk$alphasys to be redefined and have , > this propagate automatically to disk$test. > G > If disk$test, however, is defined /TRAN=CONC, then one cannot define dG > disk$foo to be 'f$trnlnm("disk$test")' and have the dir command work..    2 In this example, you are already using /TRAN=CONC.  4 $ DEFINE DISK$TEST 'F$TRNLNM("DISK$DATA1")/TRAN=CONC $ SLF DISK$TESTm5    "DISK$TEST" [super] = "FTMKT2$DKA400:" [concealed]g (LNM$PROCESS_TABLE)F $ DEFINE DISK$FOO DISK$TEST:@ %DCL-I-SUPERSEDE, previous value of DISK$FOO has been superseded $ DIR DISK$FOO:['Z']'Z't    Directory DISK$TEST:[000000]   = 000000.DIR;1                1/9       12-JAN-1999 12:40:33.12r  t Total of 1 file, 1/9 blocks. $r  % (Secret code revealed: Z == "000000")   A Seems to work fine. Can you please post a specific example of thes failure mode? Thanks.    Disclaimer: JMHO Alan E. Feldman " afeldman atski gfigroup dotski com   ------------------------------  # Date: Wed, 12 Jun 2002 21:38:30 GMTt. From: peter@langstoeger.at (Peter LANGSTOEGER)% Subject: Re: Fibre Disk vs. SCSI Diske5 Message-ID: <qbPN8.223503$305.2974836@news.chello.at>l  i In article <3d0705a3_1@news.iprimus.com.au>, "David McKenzie" <david.mckenzie@paradigm-shift.biz> writes:e, >It depends on what you mean by fibre disks.   I thought, I made it clear.e; A _disk_ with a Fibre Channel Interface and not a SCSI one.w  K I don't know which StorageWorks Controller is able to connect Fibre Channel E Disks. Maybe there is still none. I do know that EMC has an offering.   G >                                      Fibre disks on an OpenVMS systemt% >are currently accessed via an HSG80.a  I No. A HSG has the Fibre Channel to the host side and SCSI to the disks !!s  M >                                       In that sense OpenVMS can access themcE >directly, in exactly the same sense that OpenVMS accesses SCSI diskst >directly through an HSJ.     No. The analogy would be the CI.  3 HSC is CI bus to the host and SDI/SCSI to the disksp0 HSJ is CI bus to the host and SCSI to the disks.2 HSD is DSSI bus to the host and SCSI to the disks.P HS[F] is FDDI/network bus to the host and SCSI to the disks (=in fact a VMS box): HSZ is SCSI bus to the host and [Ultra] SCSI to the disks.> HSG is FibreChannel to the host and [Ultra] SCSI to the disks.  M >HSGs allow direct access to all disks in a HBVS set, HSJs do not do the samecJ >thing for SCSI disks. In my experience HSGs are *much* faster than HSJs () >same type of disk behind the controller)   M As far is I understand it then HSZ and HSG (in contrast to HSC,HSJ,HSD,HS[F]) N do not only offer disks over MSCP (=served), they also offer them over SCSI-3.7 But maybe I'm wrong and the HSG can't do MSCP. Anyone ?a   -- w Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialiste E-mail  peter@langstoeger.atP A-1030 VIENNA  AUSTRIA              I'm looking for (a) Network _and_ VMS Job(s)   ------------------------------  % Date: Wed, 12 Jun 2002 16:42:19 -0500n& From: jlsue <jlsuexxxz@screaminet.com>% Subject: Re: Fibre Disk vs. SCSI Diskh8 Message-ID: <nvffgu42getu2fjvv8c3j010pgpn5k9s3o@4ax.com>  2 On Wed, 12 Jun 2002 08:27:06 GMT, "David McKenzie"* <david.mckenzie@paradigm-shift.biz> wrote:  M >It depends on what you mean by fibre disks. Fibre disks on an OpenVMS systemtK >are currently accessed via an HSG80. In that sense OpenVMS can access themwE >directly, in exactly the same sense that OpenVMS accesses SCSI disks  >directly through an HSJ.i    C Minor nit.  The disks attached to an HSG controller are SCSI disks.k= The disks that connect to the HSV controller are Fibre disks.n1 Not speaking for anyone, certainly not DEC/Compaqs- (get rid of the xxxz in my address to e-mail)u   ------------------------------  % Date: Wed, 12 Jun 2002 18:20:47 -0400s6 From: "John.Malmberg" <Malmberg@dskwld.zko.dec.compaq>% Subject: Re: Fibre Disk vs. SCSI Diske4 Message-ID: <3D07C93F.1040900@dskwld.zko.dec.compaq>   Peter LANGSTOEGER wrote: > I thought, I made it clear. = > A _disk_ with a Fibre Channel Interface and not a SCSI one.i > M > I don't know which StorageWorks Controller is able to connect Fibre Channeli > Disks.   HP Enterprise Virtual Arrayw  7 http://www.compaq.com/products/storageworks/enterprise/-   -John-! malmberg@dskwld.zko.dec.compaq.hp2 Personal Opinion Onlym   ------------------------------  # Date: Thu, 13 Jun 2002 01:27:14 GMTZ1 From: "David J. Dachtera" <djesys.nospam@fsi.net>e) Subject: Re: Fibre disks free space drift.' Message-ID: <3D07F8A5.22A8219A@fsi.net>s   Steve McGinley wrote:  > N > Anyone seen free space drift on fibre disks connected to systems running VMSH > V7.2-1.  I'm used to seeing some free space drift on CI disks, but I'mG > talking about several Gb's worth of drift on fibe disks on occasions.c  H Please make sure you've turned off XFC by setting VCC_FLAGS to 1 in your
 MODPARAMS.   --   David J. Dachterae dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/>   ------------------------------    Date: 12 Jun 2002 13:34:17 -0700# From: fs63@volcanomail.com (Tim C.)cB Subject: Re: How can I tell which version of DECNet I am running ?< Message-ID: <399504c6.0206121234.fddf8be@posting.google.com>  \ "Peter Flunger" <p-i-b@gmx.at> wrote in message news:<ae6vd4$fd2$1@newsreader1.netway.at>...1 > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote   F JF and Peter,  Thanks for the help.  Will try next week when we return, to the office.  (attending training offsite)   Timn   ------------------------------  % Date: Wed, 12 Jun 2002 17:35:31 +0200s2 From: martin@radiogaga.harz.de (Martin Vorlaender)& Subject: Re: how to renew a vmasmail ?; Message-ID: <3d076a43.524144494f47414741@radiogaga.harz.de>l  , Bru, Pierre (pierre.bru@spotimage.fr) wrote:D > is modifing the right record of sys$system:vmsmail_profile.dat theH > only way/hach of renewing mails moved/copied into the newmail folder ?   <QUOTE>E= From: vandenheuvel@eps.enet.dec.com (Hein RMS van den Heuvel)E6 Newsgroups: comp.os.vms,comp.sys.dec,vmsnet.mail.misc," 	vmsnet.networks.desktop.pathworks0 Subject: Re: Moving old VAX mail files to eudora Date: 20 NOV 96 00:02:43+ Organization: Digital Equipment Corporationn, Message-ID: <56u3k2$3nc$1@nntpd.lkg.dec.com>  7 karcher@uttan.waisman.wisc.edu (Carl Karcher) writes...a" > nester@ad3100.ada.epa.gov wrote:  :G >> desktops running Eudora(PRO and Light).  Users can get all of their hI >> "new" mail from the VAX using Eudora just fine, but they want to have tC >> all of their old mail messages which are saved on the VAX to be a! >> retrieved using Eudora also.  i  :E > Sure. Use VMS mail's "MOVE/ALL NEWMAIL" command to move them to thehF > newmail folder. Then Eudora will download them. Do this for each VMS > mail folder ands  G That may well do the trick. However... Just being in the NEWMAIL folder I officially does not make it a new mail message. For that it needs a flag!tG Just in case Eudora needs a little more help you may want to look at...e  > $ OPEN/READ/WRITE/SHARE=WRITE file MAIL.MAI ! Verify directory; $ READ/ERRO=No_mail_today/KEY="NEWMAIL"/index=1 file recordR $ new_bit = 48 * 8	 $ new = 0n $ not_new = 0  $ newmail = f$extr(8,8,record) $ loop:Y $ if f$cvsi(new_bit,1,record)  $   then $      new = new + 1 $   else $      not_new = not_new + 1 $      record[new_bit,1]=1& $      write/update/symbol file record $ endifd $ read/erro=done file record2 $ if f$extr(8,8,record).eqs.newmail then goto loop $ done:i/ $ write sys$output new, " new and ", not_new, -S2     " not so new messages found in NEWMAIL folder" $ no_mail_today:
 $ close file 0    : Hope this helps,		+--------------------------------------+F Hein van den Heuvel, Digital.	| All opinions expressed are mine, and |C   "Makers of VMS and other	| may not reflect those of my employer |sD    fine Operating Systems."	+--------------------------------------+ </QUOTE>  ? Probably to be followed by a MAIL> DIR/NEW to update the count.p   Hope it helps,   Martin -- sD                        |  Martin Vorlaender  |  VMS & WNT programmer1   OpenVMS: When you    |  work: mv@pdv-systeme.de.E   KNOW where you want  |     http://www.pdv-systeme.de/users/martinv/n8   to go today.         |  home: martin@radiogaga.harz.de   ------------------------------   Date: 12 Jun 2002 21:38:54 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)& Subject: Re: how to renew a vmasmail ?* Message-ID: <ae8f1e$jcq$1@web1.cup.hp.com>  k In article <1c0e37b1.0206120351.32c62090@posting.google.com>, pierre.bru@spotimage.fr (Bru, Pierre) writes:   C :is modifing the right record of sys$system:vmsmail_profile.dat the G :only way/hach of renewing mails moved/copied into the newmail folder ?i     No.  a  &   Terse questions beget terse answers.  H   What question did you intend to ask?  Please provide us with the usualI   information requests -- platform, version information, some background  B   on the particular problem you are trying to solve, etc.  Thanks!    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Wed, 12 Jun 2002 23:18:35 GMTr0 From: Wayne Morrison <Wayne.Morrison@compaq.com>4 Subject: Re: howto create self-extracting zip files?* Message-ID: <3D07D4C2.51FEDBB6@compaq.com>   "Doc.Cypher" wrote:    <snip>  TD > I'd certainly like to see patches being signed, but we'd need some% > official encryption support in VMS.    <snip>  M Encryption support is part of OpenVMS as of V7.3-1 with the inclusion of CDSAtI (Common Data Security Architecture) in the operating system.  See the newlM manual "Open Source Security for OpenVMS Alpha Volume 1: Common Data Securitya% Architecture" for additional details.   : When/if you'll see signed patches is yet to be determined.   	Wayne Morrisonn 	CDSA Project Leader 	OpenVMS Security Groupe   ------------------------------   Date: 13 Jun 2002 04:22:04 GMT2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com>4 Subject: Re: howto create self-extracting zip files?, Message-ID: <ae96lc025j4@enews2.newsguy.com>  1 Wayne Morrison <Wayne.Morrison@compaq.com> wrote: O > Encryption support is part of OpenVMS as of V7.3-1 with the inclusion of CDSA.K > (Common Data Security Architecture) in the operating system.  See the newmO > manual "Open Source Security for OpenVMS Alpha Volume 1: Common Data Securitye' > Architecture" for additional details.n  J I can't resist asking, where on earth are we supposed to see these manuals8 unless we're part of the Field Test (which I'm not). :^)  = On a possitive note, V7.3-1 is starting to sound interesting!a   		Zane   ------------------------------  % Date: Wed, 12 Jun 2002 18:26:00 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>o+ Subject: Re: HP MPE to HP-UX customer plansn, Message-ID: <3D07CA76.35C35FE3@videotron.ca>   Larry Kilgallen wrote:G > Who would have thought that someone from Gartner would not know abouta& > little-endian vs big-endian issues ?    J I realise that this is an issue that one must consider when doing porting.K However, for most application software, is that a show stopper or just some1" small issue to must be looked at ?  J Writing a program to convert on-disk binary data from little to bin endianI isn't that big a deal, is it ? And if all yor data is stored as character * data, then the endianness is not an issue.  F I realise that if you are doing stuff such as writing a TCPIP or otherK communication stack, this is an issue, and if you have software that parsessE image or other "web" type files (flash, PDF etc), you need to look atuM endianness seriously, but for you average business application is that REALLYW such a big issue ?   ------------------------------  % Date: Wed, 12 Jun 2002 18:20:08 -0400M- From: JF Mezei <jfmezei.spamnot@videotron.ca>Y3 Subject: Re: Job Opening: Chief Competitive Officerc, Message-ID: <3D07C916.B7A7D36A@videotron.ca>   "Terry C. Shannon" wrote:rM > Mr. Palmer's skill set obviously does not include competitive marketing. IfiF > it did, the Alpha debacle would have been handled a bit differently.  L I beg to differ. Mr Palmer has extremely good competitive marketing skills. > He greatly helped market his competitor's products :-) :-) :-)   ------------------------------  # Date: Wed, 12 Jun 2002 22:12:17 GMTo1 From: "Terry C. Shannon" <terryshannon@attbi.com> 3 Subject: Re: Job Opening: Chief Competitive Officerr< Message-ID: <5HPN8.26544$6m5.7581@rwcrnsc51.ops.asp.att.net>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3D07C916.B7A7D36A@videotron.ca... > "Terry C. Shannon" wrote:-L > > Mr. Palmer's skill set obviously does not include competitive marketing. IfH > > it did, the Alpha debacle would have been handled a bit differently. >lE > I beg to differ. Mr Palmer has extremely good competitive marketingf skills.o@ > He greatly helped market his competitor's products :-) :-) :-)  ? Well, now that I look at it that way, you are absolutely right!    ------------------------------  % Date: Wed, 12 Jun 2002 22:43:16 +0200 & From: Bernard Giroud <bgiroud@free.fr>4 Subject: Re: Looking for Unix commands to do VMS DCL' Message-ID: <3D07B264.5A8FDD4C@free.fr>    Arthur Cochrane a crit :a  G >     I am a VMS system manager but to broaden my worth I want to learn M >     Linux/UNIX. I have installed Red Hat Linux on one of my PCs at home and G >     on an Alpha workstation at work, but I use another Alpha with VMS L >     installed for my real work of VMS system management. What I would likeH >     is to define symbols that are Unix commands for VMS commands, i.e.M >     pwd:==show default. Also, maybe some programs to do VMS things but withEK >     UNIX style commands. Grep for search as an example. I have VILE for ayL >     vi editor (just need to start using it, I love TPU/EDT). If I could atI >     least start using UNIX syntax that would help in some transition. IsL >     understand Unix and the file system but getting use to the commands isL >     the next hurdle I think I need now. In DCL you use SHOW so look at theK >     status of a lot of items, (show user, show system, show network, showgL >     default, etc.) but in Unix each of these commands is a different verb.J >     The same with SET, (set process, set default, file, etc.) To save meI >     some work is there a good resource that would have a command file IeL >     could copy to my login.com to set up some good symbols? Are there someI >     command procedures to emulate some of the basic Unix commands? BestrK >     programs to do Unix commands on VMS? This way I can do my work on VMSk6 >     but learn Unix commands also. Thanks in advance. >a >     Arthur CochraneeB >     stop_spamcochrane@stop_spamEisner.Encompasserve.orgstop_spam  >     Remove all stop_spam above  M GNV is a very good start; you have bash and a lot of utilities; you will finde it at:   http://gnv.sourceforge.net   Enjoy,   Bernard    ------------------------------  # Date: Thu, 13 Jun 2002 01:58:59 GMTo1 From: "David J. Dachtera" <djesys.nospam@fsi.net>f4 Subject: Re: Looking for Unix commands to do VMS DCL' Message-ID: <3D080014.27A4A687@fsi.net>    Jim Agnew wrote: > H > There is/was a book UNIX for VMS users, from digital press. I think it3 > may be outta print, but you could check around...  > = > plus, there is a cheat sheet i'll append to this message...    A few minor corrections:  5 There is no direct parallel to logical names in UN*X.0  @ The shells do, however, have the concept of "local" symbols (notD "exported") and global symbols (exported). AFAIK, SHOW SYMBOL/ALL isG most closely akin to "set" with noparameters and SHOW SYMBOL/GLOBAL/ALLn! is most closely akin to printenv.   H INITIALIZE <> FORMAT. INIT is more closely akin to mkfs. Try INIT'ing anF unformatted floppy (make sure it wasn't pre-formatted at the factory).  F tar is a close parallel to BACKUP; however, UN*X does not support ANSI labelled tapes.t  F UN*X-land mount is necessary only for filesystems (floppies, CDs, diskH partitions, ramdrives, etc.). Since UN*X doesn't support labelled tapes,( attempting to mount one seems illogical.  F AFAIK, there is no direct parallel to DIRECTORY/FULL in UN*X land. "ls; -l" is most closely akin to DIRECTORY/SIZE/DATE/PROTECTION.   D SHOW DEVICE differs from df, although SHOW DEVICE/MOUNTED is closelyE akin to the way df displays all mounted filesystems. However, df of a[H specific filesystem *IS* akin to SHOW DEVICE/FULL, but only with respect to MOUNTed disk volumes.  H EXCHANGE has no direct parallel in UN*X, since UN*X lacks a concept suchG as RMS. Files are just streams of bytes/bits. "dd" is most closely akin F to cp-ing from a raw device to a file, or from a file to a raw device.G "dd" from file to file only provides the ability to copy a portion of a D file instead of cp-ing the entire file. On VMS, you can do somethingD similar to dd by MOUNTing a device /FOREIGN (tape, floppy or CD, for' example) and then COPYing it to a file.e  C AFAIK, mv can "move" files across filesystems. RENAME cannot "move"E( files across volume (device) boundaries.  C SET FILE/OWNER is most closely akin to chown. "chgrp" changes groupeB associations of a UN*X file (no parallel in VMS other than setting ACLs).  F "less" goes "more" one step further since it allows scrolling back and; forth and allows searching by regular expression (I think).f  ? Note also that F$TRANSLATE() has been superceded by F$TRNLNM().l   -- b David J. Dachtera  dba DJE SystemsM http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Thu, 13 Jun 2002 01:12:49 GMT0  From: "AWK" <aweisz@bigpond.com>7 Subject: Mail masquerading using Compaq TCP/IP services @ Message-ID: <lkSN8.294081$o66.763673@news-server.bigpond.net.au>  F I'm a Unix guy struggling to find my feet on a VMS host. I'm trying toI configure what I'd call mail masquerading (or at least, sendmail calls itsF that) - I want mail from the host user@y.x.com.au to send out with theI sender address changed to user@x.com.au instead. From what I can see from.K the TCP/IP services manuals, I think what I'm after is a substitute domain, J x.com.au. But when I set up a substitute domain, the mail never leaves theJ host's mail queue (it should be delivered to a smarthost handling the mail for the whole domain)   C What I've tried is to set the subs domain okay using something likeeK tcpip$config set substitute_domain=(NOHIDDEN,name=x.com.au) , then I set upl3 a couple of logicals TCPIP$SMTP_NO_SUBS_INBOUND andnK TCPIP$SMTP_ALTGATE_ALWAYS to tell the system not to treat the mail as local,L and to always use the smarthost for delivery of non-local mail. But the mailJ doesn't get to the smarthost, the VMS host still tries to deliver locally.L No matter what combinations of these logicals, gateway address, altgateways,K HIDDEN or NOHIDDEN on the subs domain I can't get it to go. Can anyone help- me?     K Here's my non-mangled SMTP config without any substitutions. Do I need this 
 zone setting?b   SMTP ConfigurationJ                                                                    OptionsL Initial interval:   0 00:30:00.00       Address_max:    16       NOEIGHT_BITH Retry interval:     0 01:00:00.00       Hop_count_max:  16       NORELAYL Maximum interval:   3 00:00:00.00                                TOP_HEADERS  G Timeout             Initial       Mail    Receipt       Data  Terminate G   Send:                   5          5          5          3         10    Receive:                5s    Alternate gateway:  203.18.241.7  General gateway:    203.18.241.7   Substitute domain:  not definedn Zone:               x.com.au   Postmaster:         TCPIP$SMTPC Log file:           SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_LOGFILE.LOG0  0 Generic queue       Queues   Participating nodes  " TCPIP$SMTP_TITAN_00    1     TITAN   Thanks,    Ari.   ------------------------------    Date: 12 Jun 2002 16:46:18 -0700- From: ari_weisz@hotmail.com (Ari Weisz-Koves)w5 Subject: Mail masqurading with Compaq TCP/IP services = Message-ID: <361d85b2.0206121546.1b6455e7@posting.google.com>   A I'm a Unix guy struggling to find my feet on an OpenVMS host. I'mwA trying to configure what I'd call mail masquerading (or at least,d7 sendmail calls it that) - I want mail from the host "y"-@ (user@y.x.com.au) to send out with the sender address changed toC user@x.com.au instead. From what I can see from the TCP/IP serviceseE manuals, I think what I'm after is a substitute domain but when I settC up a substitute domain, the mail never leaves the host's mail queuekF (it should be delivered to a smarthost handling the mail for the wholeF domain) - I see the mail in the queue with the right sender addresses, but going nowhere.   C What I've tried is to set the subs domain okay using something likeCD tcpip$config -> set substitute_domain=(NOHIDDEN,name=x.com.au) (I've6 also tried HIDDEN), then I set up a couple of logicalsD TCPIP$SMTP_NO_SUBS_INBOUND and TCPIP$SMTP_ALTGATE_ALWAYS to tell theE system not to treat the mail as local and to always use the smarthosta? for delivery of non-local mail. But the mail doesn't get to thefF smarthost, the VMS host still tries to deliver locally. No matter whatD combinations of these logicals, gateway address, altgateways, HIDDEND or NOHIDDEN on the subs domain I can't get it to go. Can anyone help$ me understand where I'm going wrong?  a   F Here's my non-mangled SMTP config without any substitutions. Do I need this zone setting?    SMTP ConfigurationB                                                                    Optionsr@ Initial interval:   0 00:30:00.00       Address_max:    16       NOEIGHT_BITe@ Retry interval:     0 01:00:00.00       Hop_count_max:  16       NORELAYp@ Maximum interval:   3 00:00:00.00                                TOP_HEADERSs  l= Timeout             Initial       Mail    Receipt       Data d	 TerminateeD   Send:                   5          5          5          3         10   Receive:                5   p Alternate gateway:  x.x.x.7a General gateway:    x.x.x.7s  p Substitute domain:  not definedl Zone:               x.com.au    Postmaster:         TCPIP$SMTPC Log file:           SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_LOGFILE.LOG.  .0 Generic queue       Queues   Participating nodes   " TCPIP$SMTP_TITAN_00    1     TITAN    Thanks,:  > Ari.   ------------------------------    Date: 12 Jun 2002 21:07:41 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)b9 Subject: Re: Mail masqurading with Compaq TCP/IP servicesh3 Message-ID: <MeSBs9nPoSUK@eisner.encompasserve.org>E  m In article <361d85b2.0206121546.1b6455e7@posting.google.com>, ari_weisz@hotmail.com (Ari Weisz-Koves) writes:nC > I'm a Unix guy struggling to find my feet on an OpenVMS host. I'm C > trying to configure what I'd call mail masquerading (or at least,l9 > sendmail calls it that) - I want mail from the host "y"hB > (user@y.x.com.au) to send out with the sender address changed to > user@x.com.au instead.  D I am a VMS guy who does not do much with IP.  But I know two vendors4 who provide this as a service using Multinet on VMS.   So it is possible.   ------------------------------    Date: 12 Jun 2002 19:53:57 -0700( From: bob@instantwhip.com (Bob Ceculski)1 Subject: My conversation with Linus about VMS ...n< Message-ID: <d7791aa1.0206121853.b640909@posting.google.com>  A I posted recently in comp.os.linux.advocacy and got this response(> from what appears to be no one other than Linus.  I don't know@ if this was really him, maybe Larry can verify that, but here is what he said so far ...o    = In article <d7791aa1.0206111620.771abf39@posting.google.com>,-) Bob Ceculski <bob@instantwhip.com> wrote:e< >I guess while someone is rewriting linux in Bliss, they can< >fix this kernel problem also ... this is why I will stay on >VMS ...  F Side note for people who don't like VMS (probably the big majority outF there): the real news is not so much that a bug was found (*), but the> fact that AMD got so involved with it.  Which is really good.   H For some of the nastier subtle bugs it really helps to have the kinds ofF resources that sometimes the manufacturer is the only one to have (notB just hardware like ICEs etc, but people who designed the thing and realize some of the pitfalls).    F We usually do reasonably well even without the help, but people should" really take this as positive news.   		Linus?  E (*) I've got news for you: we've got bugs, and they get fixed all the1 time.2    ) Bob Ceculski <bob@instantwhip.com> wrote:e > H there are a lot of people in comp.os.vms who say you are VMS illiterate.E Are you?  Why would one not like the bulletproof security, the worldsoE best clustering, the best realtime platform, the most scalable OS outwI there, the worlds fastest jvm?  Why are you trying to reinvent the wheel?uE If VMS is so lousy, why did Bill Gates hire Cutler and use stolen VMSkG MICA code for NT if it was so lousy?  Why did Intel just grab the AlphakG engineers and secrets so they can build a decent 64 bit platform?  Why -C was it that the only environment left standing from 9/11 was a VMS .I cluster?  Is defcon a bunch of idiots when they claim VMS to be "Cool and-/ unhackable"?  Why is Alpha/VMS so lousy to you?f     Message 18 in thread  5 From: Linus Torvalds (torvalds@penguin.transmeta.com)-2 Subject: Re: Major Linux kernel bug found by AMD! " Newsgroups: comp.os.linux.advocacy View this article only o Date: 2002-06-12 12:08:41 PST    = In article <d7791aa1.0206121019.73a55b96@posting.google.com>,e) Bob Ceculski <bob@instantwhip.com> wrote:e >aI >there are a lot of people in comp.os.vms who say you are VMS illiterate.S	 >Are you?>  H I leart VMS before I learnt UNIX, but I've happily forgotten most of it.  B I've got the big book on VMS internals on the alpha, and I've read through much of it."  @ So call me VMS-illiterate, but I know enough to be dangerous. In0 particular, I know enough to dislike the system.  = > Why would one not like the bulletproof security, the worldsMF >best clustering, the best realtime platform, the most scalable OS out >there, the worlds fastest jvm?n  D You left out "..  the worlds worst interfaces, the most braindamagedF filesystems, and tied to the worlds most expensive dead-end hardware."  G Yes, tastes differ. And maybe it's gotten more useful since I last usedoE it on a VAX (I've never actually used it on an alpha, despite readinglH abotu the internals). But the whole system is designed for databases and7 large embedded usage, and it's positively user-hostile.e  F >If VMS is so lousy, why did Bill Gates hire Cutler and use stolen VMS% >MICA code for NT if it was so lousy?e  C Yeah, that's it. MS knows technology, and would be sure to bee-linet' straight for the best stuff out there. a   NOT.  E > Why did Intel just grab the Alpha engineers and secrets so they can-! > build a decent 64 bit platform?1  B Ehh.. The alpha architecture is wonderfully clean, and those alphaE engineers are studs. Most of the engineers went to AMD, btw, I think.75 Intel got the IP rights and the manufacturing plants.h  B The intel 64-bit people came largely from HP, and nobody sane ever- called the Itanic a "decent 64-bit platform".e  I < Why was it that the only environment left standing from 9/11 was a VMS e	 >cluster?   . Because even terrorists won't touch the thing?   		Linusm   ------------------------------    Date: 12 Jun 2002 14:04:27 -0700# From: fs63@volcanomail.com (Tim C.)i' Subject: Need DECNet packet information = Message-ID: <399504c6.0206121304.474a86d4@posting.google.com>   C I need DECNet packet information to use with our Ethernet sniffer. nF The link below was posted in 1998 in this newsgroup, but is now a dead link.w  + Anyone know a new link to this information?e   Thanks,D Tim)     1 >On Tue, 17 Nov 1998, H. Blakely Williford wrote:  >r" > the dec net documentation is at M > <ftp://gatekeeper.dec.com/pub/DEC/DECnet/PhaseIV/index.html> if that helps.  >2   ------------------------------  % Date: Wed, 12 Jun 2002 14:36:58 -0700R+ From: "Barry Treahy, Jr." <Treahy@mmaz.com>e+ Subject: Re: Need DECNet packet information ' Message-ID: <3D07BEFA.1050200@mmaz.com>s  
 Tim C. wrote:s  D >I need DECNet packet information to use with our Ethernet sniffer. G >The link below was posted in 1998 in this newsgroup, but is now a deadm >link. >e, >Anyone know a new link to this information? >t >Thanks, >Tim >s >  e >  ' >r2 >>On Tue, 17 Nov 1998, H. Blakely Williford wrote: >>" >>the dec net documentation is at M >><ftp://gatekeeper.dec.com/pub/DEC/DECnet/PhaseIV/index.html> if that helps.  >>     >>G I can't believe HP has pull down Gatekeeper!  Hopefully this is just a eB mistake and HP will get this back on-line...  Even Compaq kept it ( running though it had a DEC.com root....  G 220 gatekeeper.research.compaq.com FTP server (Version 5.182 Mon Apr 3 l 17:48:00 PDT 2000) ready.+ Name (gatekeeper.dec.com:treahy): anonymousb+ 331 Guest login ok, send ident as password.o	 Password:  550 Can't set guest privileges.t
 Login failed.    Barry    -- e  @ Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO   A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028-   ------------------------------   Date: 12 Jun 2002 22:26:45 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)+ Subject: Re: Need DECNet packet information-* Message-ID: <ae8hr5$jh1$1@web1.cup.hp.com>  U In article <3D07BEFA.1050200@mmaz.com>, "Barry Treahy, Jr." <Treahy@mmaz.com> writes:e :Tim C. wrote: :NE :>I need DECNet packet information to use with our Ethernet sniffer.  H :>The link below was posted in 1998 in this newsgroup, but is now a dead :>link.  :>- :>Anyone know a new link to this information?o    G   Donno why gatekeeper is offline right now -- the system is apparently J   running, based on the errors I see returned.  I'll pass along a question   to some of the network folks.d  I   I have copies of the documentation around (somewhere), if it takes moreeJ   than a week or so for the folks to figure out why gatekeeper.dec.com is H   returning the ftp 550 error (and resolve it).  (If this drags out too G   long, I'll drop the copies of the documents I have onto the Freeware h)   website or the Ask The Wizard website.)O    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------    Date: 12 Jun 2002 15:10:09 -0600- From: koehler@encompasserve.org (Bob Koehler)  Subject: Re: No new Alpha salesB3 Message-ID: <1DqXGTNTfiQR@eisner.encompasserve.org>    In article <ae7qo7$e6f$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> writes:  $ > are however interested in how well@ > the system will run their DBMS/apps and SPECint by itself does > not tell them that.f  H    DBMS?  Now why would I want an expensive, overloaded, hog like a DBMS    on a production computer?  7    Andrew still lives in the small world of a few apps.m   ------------------------------   Date: 12 Jun 2002 20:17:44 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: Open Letter to HP* Message-ID: <ae8a98$hb3$1@web1.cup.hp.com>  _ In article <3CFDF967.17291AF8@Mvb.Saic.Com>, Mark Berryman <Mark.Berryman@Mvb.Saic.Com> writes:  :Peter LANGSTOEGER wrote:t :> tb :> In article <adh4qr$bqo$6@web1.cup.hp.com>, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes:q :> >In article <9059bf6b.0205310725.73d173d4@posting.google.com>, jodonnell@hrblock.com (Jason O'Donnell) writes:o? :> >:2. Major network television commercials promoting OpenVMS.e :> > :> >  Not gonna happen.Y :> GF :> Bingo. So, OpenVMS will continue to seem a dead product in the head6 :> of the crowd and almost-all/the-uninformed CEO/CIO.    E   Interesting theory.  If you can confirm this using the TV audience sG   surveys, please let me know.  I'm quite sure that there are CEOs and dG   CIOs that watch TV, but also please consider in your calculations if CF   there more economical -- more targeted -- means to reach these same E   folks.  Expending more megabucks than necessary is frowned upon --  <   well, by all but the folks that profit from it, of course.  L :> btw, Hoff, are you in marketing now ? Or else, how do you know for sure ?  E   How do I know this?  (And no, I'm not in marketing.)  I heard Mark yF   Gorham answer this same question -- point-blank, and quite cogently "   -- within the past month or two.  I   Please think of what products you see advertised on (US) TV.  You will *G   see (often long-running) brand-recognition ads, and you will see ads eG   for mass-commodity consumer products.  These ads are all targeted at ,H   the vast numbers of potential consumers for the particular product(s).I   The comumers for these products also happen to be the audience for TV; nJ   TV is a good means of reaching the audience for these consumer products.  *   Know your audience, or waste your money.  H :I'm acquainted with a few CEO/CIO types.  Not one of them gets their ITH :information from TV.  TV would be a huge waste of money to talk about a :particular operating system...g  K   To put this another way, when was the last time you bought an appreciablesL   quantity of commercial two-way radio equipment?  If you are in the market K   for commercial equipment, you might be influenced by brand ads but -- if oJ   you are looking for the gear you will be looking in trade magazines and G   (depending on your experience and familiarity with the gear) with thef6   equipment vendor, or partner or system integrator.    H   When was the last time you saw a TV ad for a commercial two-way radio?E   For a defibrillator?  For crampons?  Most any product area you can CG   imagine -- and more than a few you probably can't imagine -- has one bD   or more trade magazines or periodicals -- these target the product?   audience, and provide for far more economical communications.   F   If I wanted to advertise a "laser hair restoration comb" -- and I'm I   not kidding about this gizmo -- I'd likely want TV ads or infomercials..F   Why?  Because the TV audience IS my target product audience.  But ifF   I were, say, selling two-way radio gear, I'd be advertising in tradeD   magazines targeting radio equipment and radio equipment resellers,7   and in trade magazines targeting my product audience.    9 :Advertising in trade rags would be a big boost, however.g  C   I would think that this advertising route would be far more more oD   cost-effective, and would be quite reasonable to provide -- there C   are and will continue to be various Compaq and HP ads in various  E   trade magazines, and we will continue to spend time and money with aF   the analysts.  TV will likely continue to see the brand recognition G   ads, and product ads that have a high correlation to the TV audience.g   	--a  I   Please consider taking a few business courses -- much of this stuff is  I   actually surprisingly simple and surprisingly clear (once you know the nJ   rules and the general constraints), but there are some very sharp folks J   that have given basic business issues a whole lot of thought.  I've met I   my share of dumb and quite useless MBAs, and I've also met some really eG   sharp ones...  The sharper MBAs are very valuable to a company -- theaF   dumb ones can do substantial damage.  (The same holds for many otherG   jobs -- a dumb engineer can certainly and quite clearly do damage...)n    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  + Date: Wed, 12 Jun 2002 16:24:58 -0500 (CDT)  From: sms@antinode.org Subject: Re: Open Letter to HP) Message-ID: <02061216245852@antinode.org>2  E    I try to avoid these discussions, but, sadly, even I get sucked in  from time to time.  2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)   > [...]5K >   Please think of what products you see advertised on (US) TV.  You will  I >   see (often long-running) brand-recognition ads, and you will see ads nI >   for mass-commodity consumer products.  These ads are all targeted at  J >   the vast numbers of potential consumers for the particular product(s). > [...]a  F    Somewhere on PBS, possibly The Nightly Business Report, I've latelyD been hearing, "Oracle.  Can't break it.  Can't break in".  It may beD that most of the audience will never buy data base software, but nowF they've heard of Oracle.  I'm not convinced that this sort of thing isH worthwhile, but it's not immediately obvious to me that it's worthless. 7 Apparently the folks at Oracle think it has some value.t  : >   Please consider taking a few business courses -- [...]      Not likely.  H ------------------------------------------------------------------------  C    Steven M. Schweda               (+1) 651-699-9818  (voice, home) C    382 South Warwick Street        (+1) 763-781-0308  (voice, work)oG    Saint Paul  MN  55105-2547      (+1) 763-781-0309  (facsimile, work)t9    sms@antinode.org                sms@provis.com  (work)o   ------------------------------   Date: 12 Jun 2002 21:46:11 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: Open Letter to HP* Message-ID: <ae8ff3$jcq$2@web1.cup.hp.com>  B In article <02061216245852@antinode.org>, sms@antinode.org writes:F :   I try to avoid these discussions, but, sadly, even I get sucked in :from time to time.i :r3 :From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman): :  :> [...]L :>   Please think of what products you see advertised on (US) TV.  You will J :>   see (often long-running) brand-recognition ads, and you will see ads J :>   for mass-commodity consumer products.  These ads are all targeted at K :>   the vast numbers of potential consumers for the particular product(s).  :> [...] :0G :   Somewhere on PBS, possibly The Nightly Business Report, I've lately E :been hearing, "Oracle.  Can't break it.  Can't break in".  It may be E :that most of the audience will never buy data base software, but now G :they've heard of Oracle.  I'm not convinced that this sort of thing is I :worthwhile, but it's not immediately obvious to me that it's worthless.  8 :Apparently the folks at Oracle think it has some value.    !   That is a brand-recognition ad.i  A   You'll see Compaq brand ads on various PBS shows, and I expect  >   that there are HP brand ads around.  (A comment which itselfB   shows how much attention I pay to TV.)  Brand ads also regularly:   appear in print campaigns and in other media, of course.    ; :>   Please consider taking a few business courses -- [...]o :  :   Not likely.   A   That decision would be solely to your own detriment, of course.c    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 12 Jun 2002 18:18:02 -0400k- From: JF Mezei <jfmezei.spamnot@videotron.ca>M Subject: Re: Open Letter to HP, Message-ID: <3D07C899.D8E141F5@videotron.ca>   "Terry C. Shannon" wrote:aC > One might surmise some attrition, especially from the Tru64 camp.t  H It would be interesting to find out why each Tru64 customer chose Tru64.  : This would provide some insight on the possible attrition.  K For instance, those who went Tru64 for the raw power of Alpha may stay withoL EV7 systems, but after that, might be forced to go Power or whatever chip is1 the fastest at that time, irrespective of vendor.a  L Those who had chosen Tru64 because it was a technically better platform withN high quality support, well all bets are open on this one because it depends onJ what HP does to HP-UX. the downgrade in performance when forced to move toE IA64 may be a lesser factor if HP managed to do a high quality job onr( improving HP (and supporting customers).   ------------------------------  % Date: Wed, 12 Jun 2002 19:01:32 -0400d- From: JF Mezei <jfmezei.spamnot@videotron.ca>  Subject: Re: Open Letter to HP, Message-ID: <3D07D2C8.679BC542@videotron.ca>   Keith Parris wrote:iB > JF, I don't understand how you interpreted this -- sorry.  TheseC > market share rankings are based on "Worldwide Unix Server Factory.F > Revenue", which is just IDC's way of saying this is the total dollarG > value paid for the Unix servers sold all around the world for a givenr1 > vendor.  So it does include sales to end users.e  K Nop. Factory revenu includes all the stock a company accumulated because itoM isn't being sold. Compaq was quite good at having great production statisticsuJ and would get hammered down by Wall Street the next quarter when inventory levels were way too high.   L Sun, Nortel, Lucent, Cisco, Ericsson also had great numbers for a while, butG they included unsecured sales to .COMs with equipment being returned todF manufacturer after the loan program expired and the .COM couldn't pay.   ------------------------------  % Date: Wed, 12 Jun 2002 19:11:57 -0400i- From: JF Mezei <jfmezei.spamnot@videotron.ca>  Subject: Re: Open Letter to HP, Message-ID: <3D07D538.78BA96C5@videotron.ca>   Bill Todd wrote:N > If Tru64's Q1 market share fell to about 7% (as the numbers quoted appear to= > reflect) there's *already* been a great deal of attrition. i  M If you are told that you are better to remain on the sinking ship because the J rescue ship will arrive "soon" and you won't have to get wet (the water isN mighty cold), then you may stay on the ship thinking that the rescue ship will arrive before your ship sinks.  L And in a way, as long as you continue to have light, music and drinks on theJ slowly sinking ship, isn't it better to stay on it instead of venturing inL vold waters or boarding some rickety old ship without heat, food or drinks ?  K And if the rescue ship ends up sinking before it gets to you, you can still./ board the rickety old ship that is standing by.g   ------------------------------  + Date: Wed, 12 Jun 2002 18:21:43 -0500 (CDT), From: sms@antinode.org Subject: Re: Open Letter to HP) Message-ID: <02061218214301@antinode.org>l  2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)  I > :   Somewhere on PBS, possibly The Nightly Business Report, I've latelyuC > :been hearing, "Oracle.  Can't break it.  Can't break in".  [...]d  # >   That is a brand-recognition ad.n  ; >   You'll see Compaq brand ads on various PBS shows, [...].  A    Of course, Oracle is the name of a company and a _product_.  AeG generic Compaq ad probably does little for VMS, and I have not seen anyn VMS-awareness ads lately.   = > :>   Please consider taking a few business courses -- [...]t > :e > :   Not likely.a > C >   That decision would be solely to your own detriment, of course.D  H    Life is finite.  I'd prefer to catch up on thermodynamics if I had to pick a topic for further study.e  H ------------------------------------------------------------------------  C    Steven M. Schweda               (+1) 651-699-9818  (voice, home)wC    382 South Warwick Street        (+1) 763-781-0308  (voice, work)nG    Saint Paul  MN  55105-2547      (+1) 763-781-0309  (facsimile, work)09    sms@antinode.org                sms@provis.com  (work)N   ------------------------------  % Date: Wed, 12 Jun 2002 22:24:20 -0400o' From: "Main, Kerry" <Kerry.Main@hp.com>t Subject: RE: Open Letter to HPT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF40266077D@kaoexc01.americas.cpqcorp.net>  G >>>    Of course, Oracle is the name of a company and a _product_.  <<<)  ' No more so than Microsoft or Compaq.=20l  F Oracle has a huge number of different software products, so unless youF are in the IT industry, most T.V. viewers would only see the name of aE company .. They would not associate it with a database .. Heck 80% ofr9 T.V. viewers probably don't even know what a database is.e  % As Steve said .. Its brand marketing.    Regards0  
 Kerry Main Senior Consultants Hewlett-Packard Canada! Consulting & Integration Servicesn Voice: 613-592-4660n Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----3 From: sms@antinode.org [mailto:sms@antinode.org]=20k Sent: June 12, 2002 7:22 PM  To: Info-VAX@Mvb.Saic.Coml Subject: Re: Open Letter to HP    2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)  B > :   Somewhere on PBS, possibly The Nightly Business Report, I've latelyC > :been hearing, "Oracle.  Can't break it.  Can't break in".  [...]k  # >   That is a brand-recognition ad.   ; >   You'll see Compaq brand ads on various PBS shows, [...]   A    Of course, Oracle is the name of a company and a _product_.  ArG generic Compaq ad probably does little for VMS, and I have not seen anym VMS-awareness ads lately.e  = > :>   Please consider taking a few business courses -- [...]s > :o > :   Not likely.r >=20C >   That decision would be solely to your own detriment, of course.I  H    Life is finite.  I'd prefer to catch up on thermodynamics if I had to pick a topic for further study.   H ------------------------------------------------------------------------  C    Steven M. Schweda               (+1) 651-699-9818  (voice, home)rC    382 South Warwick Street        (+1) 763-781-0308  (voice, work)BG    Saint Paul  MN  55105-2547      (+1) 763-781-0309  (facsimile, work)d9    sms@antinode.org                sms@provis.com  (work).   ------------------------------    Date: 12 Jun 2002 13:10:02 -0600+ From: young_r@encompasserve.org (Rob Young)k Subject: Re: Open Letter to HP3 Message-ID: <z+Shb8oO1eum@eisner.encompasserve.org>   p In article <6mLN8.216474$Oa1.20252668@bin8.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes: > : > "Rob Young" <young_r@encompasserve.org> wrote in message/ > news:l7io3HPdAddz@eisner.encompasserve.org...h >  > ...f > M >> The shift is being driven by hardware price/performance, Silver said. "The) > SuntI >> boxes cost $10,000 and run at 1 GHz or less, while the PC workstationst > costK >> $2,000 and run at 2-to-3 GHz. You just can't get that performance out of- > the-  >> RISC-based systems," he said. > M > Of course, you won't be able to get IA32-competitive price *or* performanceoN > out of Itanic systems either - including McKinley and its later mini-clones.    = 	And as the article points out , the move is to Linux + IA32.fC 	Itanium is still a distant (2-4 year) thing for many, HP included.J    B > And if you need 64-bit support (let alone a good mixed 32/64-bitF > environment) the price/performance sweet spot will unquestionably beM > Hammer - which means that Intel will have to release Yamhill to counter it,s) > further undermining Itanic's viability.n > K > I.e., your contentions that poor li'l AMD can't mount an effective attacktK > are, as usual, just hot air.  But the evidence will be clear soon enough.f    F 	Which portion?  Be specific.   Offer up some supporting evidence.  I = 	didn't even bother mentioning that Hammer is now most likelysF 	delayed until next year.  Is that the "soon enough" you are referring 	to?  C 	They certainly won't be making any money this year on IA32.  Intelu 	will ensure that.  E 	Also, it appears a 9 cent loss per share for AMD with a -14 percent 9 	growth this quarter:m  # http://biz.yahoo.com/z/a/a/amd.htmlc  C 	But the good news is that the loss will narrow to 1 cent per share$< 	by September.  That isn't the point.  The point is found in= 	the more troubling Decmeber 2002 numbers.  Two months ago its= 	was expected that AMD's earnings were to be 7 cents a share sG 	(concensus) for the quarter ending December 2002.  However , today we >A 	see December 2002 earnings are projected to be -5 cents a share.u  A 	Like I said earlier, Intel is ripping the profits right out fromtC 	underneath them.  Sustained losses usally have a negative business B 	impact.  Let's see if Intel can sustain the losses past December. 	  				Rob    ------------------------------  % Date: Wed, 12 Jun 2002 23:30:39 -0400r- From: JF Mezei <jfmezei.spamnot@videotron.ca>m Subject: Re: Open Letter to HP+ Message-ID: <3D0811C8.85FEC7F@videotron.ca>s   Hoff Hoffman wrote:nB >   You'll see Compaq brand ads on various PBS shows, and I expect( >   that there are HP brand ads around.   C And Compaq had no problem portraying its wintel crap on those brand(L recognition ads. It had no problems featuring wintel crap in its "enterpriseD TV ads". That is more than brand recognition, it is corporate image.  N HP has a corporate image problem with some of its most lucrative customers whoN distrust HP's true intentions with regards to VMS. Those customers are lookingN for a public commitment from HP, something that not only make HP more credibleM to them, but also sends a strong signal to ISVs that HP is serious about VMS.sN And such TV ads are a very great way to achieve this. (athough, by themselves, won't do the job 100%).   N Consider that even if you, the venerable Hoff, don't want to see VMS mentioned4 on TV, how the hell are customers supposed to feel ?  G It isn't a question of the TV ads generating sales. It is a question oftL regaining the trust of customers and ISVs. And the refusal to mention VMS inN public before is the issue, this is why having a VMS in a TV ad would be a bigN step forwards and show customers that HP intends to back its roadmap with realM actions and break the taboos which in the past prevented VMS from succeeding.y   ------------------------------  # Date: Wed, 12 Jun 2002 22:30:45 GMTc. From: peter@langstoeger.at (Peter LANGSTOEGER)# Subject: Re: OpenVMS Mailing Lists?i5 Message-ID: <pYPN8.223930$305.2981215@news.chello.at>d  h In article <88FN8.246$yK6.55754@news1.iquest.net>, skidmore@alternate-venue.org (Barry Skidmore) writes:3 >>>I am looking for a general OpenVMS mailing list.g >>1 >>Why a mailing list and not a USENET newsgroup ?a >-H >I already subscribe to this newsgroup, so I am looking for something in >addition to comp.os.vms.I  ) How about the USENET vmsnet.* hierarchy ?n   -- - Peter "EPLAN" LANGSTOEGERb% Network and OpenVMS system specialists E-mail  peter@langstoeger.atP A-1030 VIENNA  AUSTRIA              I'm looking for (a) Network _and_ VMS Job(s)   ------------------------------  # Date: Wed, 12 Jun 2002 23:04:08 GMT 3 From: skidmore@alternate-venue.org (Barry Skidmore) # Subject: Re: OpenVMS Mailing Lists?c2 Message-ID: <IrQN8.256$yK6.58041@news1.iquest.net>   >i* >How about the USENET vmsnet.* hierarchy ? >  >--  >Peter "EPLAN" LANGSTOEGER& >Network and OpenVMS system specialist >E-mail  peter@langstoeger.at Q >A-1030 VIENNA  AUSTRIA              I'm looking for (a) Network _and_ VMS Job(s)d  K Thanks, I was not aware of these newsgroups.  There are almost TOO many to s choose from.    # -----------------------------------  Barry Skidmore# skidmore at alternate-venue dot org   Hobbyist MicroVAX 3100-40 System Running OpenVMS 7.2g# -----------------------------------i   ------------------------------  % Date: Wed, 12 Jun 2002 14:38:42 -0400o' From: "Main, Kerry" <Kerry.Main@hp.com>sP Subject: RE: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading)T Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF402660777@kaoexc01.americas.cpqcorp.net>   Osmo -  H >>> High availability is important, and OpenVMS has features which wouldD be very usable here, but when OpenVMS is seen as fossil, we must use unix/linux >>>  < Before calling OpenVMS a fossil - keep in mind that the UNIX; architecture is at least 10 years older than OpenVMS ...=20t   :-)-  E >>> There may be some truth in that, but I'd blame mostly MS and it'sS8 good partners like you for that. I stronly disagree withG digital/Compaq/HP, when you try to force us using Windows. (By offeringnH even VMS documents in MS-formats, by declaring VMS as server only OS, by1 advertising and suggesting to use Windows ...)<<<4  E While I will certainly not argue marketing mistakes have been made in G the past, I would also suggest that you will never win any battles withcD Mgmt by suggesting OpenVMS is the best solution for all requirements either.@  F >>> I didn't even think about centralizing in the whole university. We3 (computing center) have already lost the power. <<<   F This was true with many companies as well until senior mgmt decides toF signifgicantly reduce overall IT costs. All of a sudden centralizationC (technology as well as processes and funding) starts to become veryp attractive again.=20  H Bottom line is that politics in any organization will only exist so longF as the $'s (budgets) are there to support it. When the $'s dry up, youD can expect to see more of this power coming back to your department.   Regards,  
 Kerry Main Senior Consultantt Hewlett-Packard Canada! Consulting & Integration Serviceso Voice: 613-592-4660s Fax   : 613-591-4477 Email: Kerry.Main@hp.com     -----Original Message-----4 From: Osmo Kujala [mailto:kujala@tukki.cc.jyu.fi]=20 Sent: June 12, 2002 5:59 AM- To: Info-VAX@Mvb.Saic.Com-G Subject: Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in( Reading)    H Hi Kerry, I'm pleased to see that you have interest in needs of schools.  & Main, Kerry <Kerry.Main@hp.com> wrote: > Osmo -  G >>>> You see, if some of those little servers now crash, it's not a bigs
 > deal.<<<  E > Today perhaps. I also know that schools are beginning (some have=20GH > always) demanded high availability for selected application area's,=20I > because there is always a huge amount of stress when servers go down=206C > right around exam time, when big assignements are due, or when=20d0 > registration for a new session is in progress.  G High availability is important, and OpenVMS has features which would befD very usable here, but when OpenVMS is seen as fossil, we must use=20A unix/linux. (We get the same availability with more work and lesse money.)    > As to the comment - E >>>> But if we integrate to a few big systems, they must have doubledd< > components etc. to get same service availability level.<<<  2 > Perhaps this does not apply to your environment,  D It does, in some degree. Problem is that we don't have feedback fromG decisions in clear numbers (money). Should have measure of productivity 
 or something.e   > but one of the hottestJ > topics for almost all med-large corp's today is IT Consolidation - on=20 > all platforms.   > Big drivers for this are:iD > - lower overall costs. A few big systems properly config'ed can=20E > typically be maintained by less staff (number 1 IT cost for most=20lG > companies). Admittably, these centralized staff are typically more=20a2 > senior experienced and higher trained resources.  F Staff is number 1 cost in our computer center too and we have tendencyF to forget staff costs (especially savings), when planning new systems.  G > - being much more proactive - big driver is to fix problems BEFORE=20t > they impact the business.yF > - increase service levels e.g. availability, performance. Yes, IT=20H > Consolidation does require much more focus on availability features=20@ > like clustering - hence the requirement for more senior and=20 > experienced resources.F > - increased security. Maintaining access to 10 or less servers is=20D > typically much easier to maintain than a hundred. (yes, you can=20I > implement distributed security features like kerberos, but you still=20pH > need to maintain  each system so that an individual system does not=20 > have some back door)G > - business continuity (BC) and disaster recovery (DR). Fire in the=20iH > datacenter is good example where the rest of the facility is fine, but  D > the computer room is gone. Now what? Since 9/11, this is a huge=20H > concern by all med-large businesses. Implementing BC and DR is much=20& > easier with fewer systems than many.  B >>>>> As a school we must have systems, which are common in the=20 >>>>> world.<<<e  @ > Do you mean like teaching Java on a high availability cluster?  G I was talking about OS's. We must have linux/unix-systems, we must have E windows (and centralized systems to support windows) etc. but no needdH for VMS, MVS, OS1100, Tandem... I guess Java and C,C++ are found in most$ of common OS's. I don't know much=20G of computer science teaching, do they need Java in special environment.-  I > While there are still some issues with Java, it might be interesting=20p > to note the following trends::  G > http://news.com.com/2009-1001-868454.html?tag=3Ddd.ne.dtx.nl-sty.0=20a	 > (March,i > 2002) E > "Although recent studies show that older technologies are still thea mainG > languages used by software developers, Java is beginning to close thej > gap."   H > "And market research firm Evans Data has reported that more people are  D > using Java while the number of C and C++ developers is declining."  G > "That trend is reflected in academia as well. At the Massachusetts=20cE > Institute of Technology, the computer science department in 1997=20 H > replaced Microsoft's C++ with Java as the primary software language=20H > that students are required to learn. This spring, the University of=20E > California at Berkeley offered 25 Java courses and only seven on=20i > Microsoft languages.  F > "In my mind, Java is rapidly going to displace all the languages,=20H > particularly C++," MIT associate professor Daniel Jackson said, noting  J > that undergraduate students are increasingly teaching themselves some=20F > Java even before attending his classes. In two years, the College=20I > Board will switch from C++ to Java for the computer science advanced=20oE > placement tests that high school students take to get credit for=20* > college-level courses."t  H >>>> If we had to concentrate to one OS.......the hardest thing would be( > to get rid of Windows. It's so sad.<<<  I > I would suggest that moving to one OS is almost always NOT the right=20a > solution.   F There may be some truth in that, but I'd blame mostly MS and it's goodF partners like you for that. I stronly disagree with digital/Compaq/HP,G when you try to force us using Windows. (By offering even VMS documents E in MS-formats, by declaring VMS as server only OS, by advertising and  suggesting to use Windows ...)  = If that doesn't stop, I won't buy anything from you any more!h  I > However, if indeed the school does decide to centralize, because the=20 G > number of users will be much higher, then the centralized platform=20jI > absolutely will require much higher RASS (reliability, availability,=20rH > scalability and security) than the current environment. This gets even  H > more critical when the school decides to start offering online courses  0 > from remote students from just about anywhere.  E I didn't even think about centralizing in the whole university. We=20v? (computing center) have already lost the power. Only network is 
 completely=20pH in our hands. For online courses we offer the needed net capacity and=20C participate in the projects, but faculties, departments and such dok their=20G own special systems. There I don't see so much problem if Windows or=20o whatever OS is used.  , ---- rest of orig. message deleted----------   Osmo   ------------------------------  # Date: Wed, 12 Jun 2002 18:59:24 GMTh* From: "Bill Todd" <billtodd@metrocast.net>P Subject: Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading)B Message-ID: <gSMN8.145526$4i.13292302@bin2.nnrp.aus1.giganews.com>  2 "Main, Kerry" <Kerry.Main@hp.com> wrote in messageL news:BE56C50EA024184DAF48F0B9A47F5CF402660777@kaoexc01.americas.cpqcorp.net. .. Osmo -  H >>> High availability is important, and OpenVMS has features which wouldD be very usable here, but when OpenVMS is seen as fossil, we must use unix/linux >>>  < Before calling OpenVMS a fossil - keep in mind that the UNIX8 architecture is at least 10 years older than OpenVMS ...   *** D Fossilization is not so much a matter of age as it is of inactivity.  J (And to keep things accurate I should point out that Unix is somewhat less than 10 years older than VMS.)   - bill   ------------------------------  % Date: Wed, 12 Jun 2002 14:29:15 -0500o& From: jlsue <jlsuexxxz@screaminet.com>P Subject: Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading)8 Message-ID: <468fgug5hi21e8sehmscp7uejrgg3kvf42@4ax.com>  B On Wed, 12 Jun 2002 10:48:02 +0000 (UTC), david20@alpha2.mdx.ac.uk wrote:    H >>Sure, but rather than have the CPU on each end process the encryption,E >>I thought IP/SEC was a better solution.  Offload all that work ontoe6 >>something small, simple, and almost management-free. >> >:J >Yes VPNs are a solution. However they either require a box at each end toO >handle the tunnel which is extra cost and also leaves a little gap at each end O >where there is unencrypted traffic. Or you don't offload the CPU and have the iN >VPN software running on the systems themselves (almost required on the clientP >side when using lots of widespread clients). In which case you need to have theK >VPN software you are using available and running on those machines you are-K >going to use - which again is generally another cost. Can you setup a VPN sP >directly from a VMS box using IPSEC or any other VPN using any of the available >TCPIP stacks ?, > N >VPNs may also not work well with other networking components (IPSEC has majorI >problems with NAT whereas SSH has no such problems). SSH is pretty much -N >available for every system and is generally free. It seems to be pretty much $ >the defacto standard at the moment.H >VPNs seem to work best when you need a dedicated link between groups of >machines at two remote sites. >   E VPN is something different from IP/SEC.   IP/SEC is security built-in.F to the IP cards.  And I believe that my $150 Linksys router (NAT) willA allow IP/SEC to flow through (though I have no way to test this).R    1 Not speaking for anyone, certainly not DEC/Compaqe- (get rid of the xxxz in my address to e-mail)O   ------------------------------    Date: 12 Jun 2002 16:47:02 -0700. From: SPAMSINK2001@YAHOO.COM (Alan E. Feldman)+ Subject: Re: rooted devices and /TRANS=CONC3= Message-ID: <343f30ae.0206121547.225f8855@posting.google.com>i  | Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote in message news:<01KIUO0CNSMI96WTPR@sysdev.deutsche-boerse.com>...K > A while back I mentioned that a rooted device cannot be defined in terms aK > of another rooted device.  Actually, it seems that any concealed logical  I > name cannot be defined in terms of another concealed logical name (and aC > work properly---restricting ourselves to disk names here).  With eD > /TRAN=TERM I can see the reason for this, but why with /TRAN=CONC? > J > As one can see reading the other thread (DEFINE/TRAN=TERM), either many G > folks are confusing the effects of TERM and CONC or each of these at tJ > least in some situations also has some effects normally produced by the  > other.    @ It is important to use a trailing colon. Only then does it work:     $ SHOW LOG/FUL AAA9    "AAA" [super] = "BBB:" [concealed] (LNM$PROCESS_TABLE)': 1  "BBB" [super] = "DSA0:" [concealed] (LNM$PROCESS_TABLE) $ DIR AAA:[000000]000000  w Directory AAA:[000000]   = 000000.DIR;1                4/9       17-OCT-1991 13:30:30.91i  c Total of 1 file, 4/9 blocks.   $i $ DEFINE AAA BBB /TRANS=CONC; %DCL-I-SUPERSEDE, previous value of AAA has been supersededu $ SHOW LOG/FUL AAA8    "AAA" [super] = "BBB" [concealed] (LNM$PROCESS_TABLE): 1  "BBB" [super] = "DSA0:" [concealed] (LNM$PROCESS_TABLE) $ DIR AAA:[000000]000000; %DIRECT-E-OPENIN, error opening AAA:[000000]000000 as inputyK -RMS-F-DEV, error in device name or inappropriate device type for operationo $c     Disclaimer: JMHO Alan E. Feldmann" afeldman atski gfigroup dotski com   ------------------------------  # Date: Thu, 13 Jun 2002 03:45:03 GMTp' From: Stuart Fuller <stufuller@usa.net>n$ Subject: RE: SCSI- Fatal Drive Error0 Message-ID: <m639ea.33k.ln@dadsys1.fuller.local>   WILLIAM WEBB wrote:i > 3 > Every BA356 I've ever seen is wide SCSI (16-bit).d1 > That's probably why it doesn't like your ULTRA.o >  > WWWebb >  > -----Original Message-----1 > From: Info-VAX-Request@Mvb.Saic.Com at INTERNETt& > Sent: Tuesday, June 11, 2002 2:40 PMD > To: Webb, William W Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET" > Subject: SCSI- Fatal Drive Error >  > F > We are trying to install an Ultra Wide, 9 Gig, Solid State Disk fromH > BitMicro into a StorageWorks DS-BA356 cabinet.  The drive is seen fromH > the console of a VAX 4106, OpenVMS 7.2, but cannot be Initialized from > VMS. > 1 > The error is %INIT-F-MEDOFL, medium is offline.n > : > The error log indicates "Invalid inquiry data received". > 	 > Thanks,c > Phil    Except, the DS-BA356 *is* Ultra.  I Also, it's likely that the controller on the VAX doesn't do Ultra, so it p3 really doesn't matter if the BA356 is Ultra or not.   J Finally, "Ultra" is not the same concept as "wide SCSI (16-bit)".  "Wide" 2 means 16 bit, while "Ultra" means 20Mhz data rate.           Stum   ------------------------------  % Date: Wed, 12 Jun 2002 19:26:31 -0400r- From: JF Mezei <jfmezei.spamnot@videotron.ca>e: Subject: Re: Sun is on the way out!  Only HP and IBM left!, Message-ID: <3D07D8A1.4DB3B440@videotron.ca>   re: Sun's VARs in disarray.o  K Nice timing for such an article. Don't you think that the combination of HPiJ and Compaq's "channels" combined with the need to return to a direct salesM method to be more competitive isn't causing any disarray at the former HP andb Compaq channels ?   N Has Sun begun or stated intentions to sell more directly ? Or will it continueH to rely on the "channel" even if it means that its products will be less
 competitive ?h  N Gotta give it to Mikey Dell, he is forcing the big guys to rethink their whole sales paradigm.2   ------------------------------    Date: 12 Jun 2002 13:48:30 -0600+ From: young_r@encompasserve.org (Rob Young)P: Subject: Re: Sun is on the way out!  Only HP and IBM left!3 Message-ID: <Bdx+7RvbjwMg@eisner.encompasserve.org>e  o In article <ROLN8.195672$%y.18848494@bin4.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes:d > 7 > "Bob Ceculski" <bob@instantwhip.com> wrote in messagen9 > news:d7791aa1.0206120933.41610cfb@posting.google.com...m: >> I didn't say it, the Inquirer does ... guess we will'nt< >> have poor Andrew around here anymore making bogus claims!- >> No wonder he is anxious to trash Alpha ...0 > M > Learn to read, Bob.  The Inquirer also speculates that HP covertly used the-J > quoted article to spread FUD:  if you believe the one, you really should > believe the other. >   ? 	I think it is a good article in several respects because it is  	fairly factual.  For example:  G http://www.varbusiness.com/sections/columns/columns.asp?ArticleID=35611"  9 The most recent one to fall was Infinite Technology GroupEA (ITG) of Mineola, N.Y., a major Sun VAR whose sales peaked at $80IH million a few years ago. Today, its sales are closer to $20 million. ItsE executives boasted of great relationships with Sun's distributors and-G executives. ITG made a good impression and possessed a strong knowledge-B of the market. As a result of its reliance on Sun and, to be fair,F pressure from more macroeconomic issues, it became burdened with $12.7H million in debt and has only $5.2 million in assets. That, obviously, is not a healthy ratio.    : 	To go from $80 million to $20 million in sales, not good.  = 	If I were a competitor , I would be attempting to turn everya> 	Sun VAR out there.  Call it FUD.  Call it smart business too.   				Robt   ------------------------------  # Date: Thu, 13 Jun 2002 04:27:20 GMTn  From: cjt <cheljuba@prodigy.net>: Subject: Re: Sun is on the way out!  Only HP and IBM left!+ Message-ID: <3D081F27.8539AA87@prodigy.net>t   JF Mezei wrote:n >  > re: Sun's VARs in disarray.r > M > Nice timing for such an article. Don't you think that the combination of HPyL > and Compaq's "channels" combined with the need to return to a direct salesO > method to be more competitive isn't causing any disarray at the former HP andV > Compaq channels ?  > P > Has Sun begun or stated intentions to sell more directly ? Or will it continueJ > to rely on the "channel" even if it means that its products will be less > competitive ?r  N I think I read somewhere that Sun is trimming VARs that don't actually "A" any3 "V," but otherwise continuing to work through VARs.    > P > Gotta give it to Mikey Dell, he is forcing the big guys to rethink their whole > sales paradigm.t   ------------------------------  # Date: Thu, 13 Jun 2002 01:35:02 GMTt1 From: "David J. Dachtera" <djesys.nospam@fsi.net>i& Subject: Re: SYSMAN default parameters' Message-ID: <3D07FA77.9282BE5F@fsi.net>r   David McKenzie wrote:m >  > IIRC it is possible. > 
 > $ mc sysgen 
 > use defaulto > set mumble.... > write defaultt > ) > I haven't tried this for a while thoughe  E In V7.1-2, both SYSGEN and SYSMAN accept (PARA) WRITE DEFAULT without ( error, but the changes are not retained.   --   David J. Dachtera& dba DJE Systemsi http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/-   ------------------------------  # Date: Thu, 13 Jun 2002 03:24:05 GMT,  From: John Santos <JOHN@egh.com>& Subject: Re: SYSMAN default parameters4 Message-ID: <1020612232056.385B-100000@Ives.egh.com>  - On Thu, 13 Jun 2002, David J. Dachtera wrote:t   > David McKenzie wrote:  > >  > > IIRC it is possible. > >  > > $ mc sysgen> > > use defaultm > > set mumble.... > > write default' > > + > > I haven't tried this for a while though  > G > In V7.1-2, both SYSGEN and SYSMAN accept (PARA) WRITE DEFAULT withouta* > error, but the changes are not retained.  C Well, they probably are, but in a file called "default.par" in your2 current default directory ;-)h  @ "use default" sets the current parameters to the default values.= "write default" creates a parameter file called "default.par"pA "use default.par" loads the current parameters from "default.par"V  B "Use" isn't totally symmetric with "write".  According to the fineB online help, "write" takes a file spec (default type .par), or one@ of the keywords "CURRENT" or "ACTIVE".  "Use" takes any of those' parameters or a 3rd keyword, "DEFAULT".e   -- v John Santosv Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  # Date: Thu, 13 Jun 2002 01:28:53 GMTc1 From: "David J. Dachtera" <djesys.nospam@fsi.net>h/ Subject: Re: tpc/ip file server for VMS systems ' Message-ID: <3D07F908.380ED2E6@fsi.net>s  	 rf wrote:g > L > Is it possible/practical to setup LINUX/BSD to provide file/print servicesI > for VMS/AXP (OVMS 6.2) system(s) over IP?  If so some guidance would bea > appreciated.   Possible? Sure.l  D Practical? Questionable - it depends why you think you need/want it.F VMS/Linux synergy could be a good thing if it serves a useful purpose.   -- i David J. Dachterat dba DJE Systemso http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/p   ------------------------------    Date: 12 Jun 2002 13:43:49 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)uI Subject: VMS Groups and Privileges (was: Carly was here in ZKO yesterday)I3 Message-ID: <juQBbxwUTSJO@eisner.encompasserve.org>   V In article <3D078EE0.8CE3F2F3@pacbell.net>, Don Sykes <annonymous@pacbell.net> writes: > Larry Kilgallen wrote: >>  Y >> In article <3D06978F.4B8B4A60@pacbell.net>, Don Sykes <annonymous@pacbell.net> writes:  >> iU >> > Not exactly what I had in mind. I agree ODS5 and COE bring some crucial parts ofkL >> > Unix into VMS, but even some VMS features could do with a rethink. E.g.S >> > privileges are not logical. They seemed to have evolved as needed, so there iss >> eF >> "More logical" presumes they are orthogonal, which is not the case.E >> Privilege _levels_ (categories) are worth thinking about, and they?I >> are the basis on which you should evaluate privileges.  But privilegeseI >> on a level can be used to bootstrap into a different privilege.  Think>N >> of privilege levels as providing real control and the individual privilegesG >> within a level as providing protection against typographical errors.  > G > That's what I meant. More logical meaning architected, and as you saywL > pivilege_levels are a good way to start looking at such a re-architecting.  G There is no "start looking at" -- privilege levels were documented withsI VMS V4.4.  I see no reason why any redesign would be appropriate, and thesI committment by VMS to backward compatibility mandates existing privilegesi stay as they are.2  R >> > no uniform architecture to them. Users cannot belong to more than 1 group and >> dE >> In what way do general identifiers fall short of what you see fromsD >> Unix groups ?  Certainly giving concepts the same name as in Unix >> is not a real requirement.g > N > Identifies are fine, but they can be hard to track - e.g. to see how sets of1 > identifiers impact sets (groups) of users, etc.m  I Certainly, because VMS lets you dismount shared disks on a running systemcI as a normal course of events.  And unlike RACF, security metadata for VMSm is stored on the volume itself.2  O > I not suggesting making them the SAME as Unix groups. They don't fit the billtJ > either. I mean a way for administrator to delegate certain functions andR > privileges to group administrators. This is a big topic and would require a goodQ > deal of thought to properly implement, but its worth it especially where large,t% > complex enterprises are concerned. D  E Control of a disk, control of a queue, etc. can be delegated based ongB general identifiers.  What else would you delegate ?  The listingsB have hints about a previous plan to go beyond the GROUP and GRPPRV> mechanisms for delegation, but I don't believe many things areC appropriate for delegation.  There are system management tools thatoB allow delegation of username creation and deletion to less skilledF individuals, but I don't see any reason for doing that in the base OS.   ------------------------------  # Date: Thu, 13 Jun 2002 00:53:30 GMTi( From: Don Sykes <annonymous@pacbell.net>M Subject: Re: VMS Groups and Privileges (was: Carly was here in ZKO yesterday)a+ Message-ID: <3D07ED6C.1991279A@pacbell.net>    Larry Kilgallen wrote: > X > In article <3D078EE0.8CE3F2F3@pacbell.net>, Don Sykes <annonymous@pacbell.net> writes: > > Larry Kilgallen wrote: > >>[ > >> In article <3D06978F.4B8B4A60@pacbell.net>, Don Sykes <annonymous@pacbell.net> writes:e > >>W > >> > Not exactly what I had in mind. I agree ODS5 and COE bring some crucial parts of,N > >> > Unix into VMS, but even some VMS features could do with a rethink. E.g.U > >> > privileges are not logical. They seemed to have evolved as needed, so there ist > >>H > >> "More logical" presumes they are orthogonal, which is not the case.G > >> Privilege _levels_ (categories) are worth thinking about, and theynK > >> are the basis on which you should evaluate privileges.  But privilegescK > >> on a level can be used to bootstrap into a different privilege.  ThinkhP > >> of privilege levels as providing real control and the individual privilegesI > >> within a level as providing protection against typographical errors.  > >mI > > That's what I meant. More logical meaning architected, and as you say.N > > pivilege_levels are a good way to start looking at such a re-architecting. > I > There is no "start looking at" -- privilege levels were documented with@K > VMS V4.4.  I see no reason why any redesign would be appropriate, and theeK > committment by VMS to backward compatibility mandates existing privilegesn > stay as they are.rO In their current state, privileges are like ACLs. They do not have "levels" and6P they do not relate to one another. I would like a more hierarchial organization.   ...nQ > > I not suggesting making them the SAME as Unix groups. They don't fit the billsL > > either. I mean a way for administrator to delegate certain functions andT > > privileges to group administrators. This is a big topic and would require a goodS > > deal of thought to properly implement, but its worth it especially where large,e& > > complex enterprises are concerned. > G > Control of a disk, control of a queue, etc. can be delegated based on-6 > general identifiers.  What else would you delegate ?N As a enterprise administrator, I would like to allow Joe (a group admin) to beK able to setup new users, give, or deny, them certain privileges without theeO ability to effect Sam (a different group admin). Each has full control over all21 users in his group, but cannot effect each other.t     -- t   Have VMS. Will Travel. Wire Paladin (@alphase.com) 
 San Franciscot   ------------------------------    Date: 12 Jun 2002 21:10:29 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) M Subject: Re: VMS Groups and Privileges (was: Carly was here in ZKO yesterday)u3 Message-ID: <Xv3kfICKJ+h5@eisner.encompasserve.org>   V In article <3D07ED6C.1991279A@pacbell.net>, Don Sykes <annonymous@pacbell.net> writes: > Larry Kilgallen wrote:  J >> There is no "start looking at" -- privilege levels were documented withL >> VMS V4.4.  I see no reason why any redesign would be appropriate, and theL >> committment by VMS to backward compatibility mandates existing privileges >> stay as they are.Q > In their current state, privileges are like ACLs. They do not have "levels" andnR > they do not relate to one another. I would like a more hierarchial organization.  ' So you disagree with the documentation.SB I have had good success relying on the documentation about levels.  H >> Control of a disk, control of a queue, etc. can be delegated based on7 >> general identifiers.  What else would you delegate ?mP > As a enterprise administrator, I would like to allow Joe (a group admin) to beM > able to setup new users, give, or deny, them certain privileges without thePQ > ability to effect Sam (a different group admin). Each has full control over allu3 > users in his group, but cannot effect each other.I  C There are add-on products to do that, but since it crosses securitytB domains it is certainly not a proper part of the operating system.   ------------------------------  % Date: Wed, 12 Jun 2002 15:37:42 -0400I0 From: "Syltrem" <syltremspammenot@videotron.com>1 Subject: Re: What process is using this pagefile?a4 Message-ID: <ZpNN8.3280$H67.17352@tor-nn1.netcom.ca>  & You mentioned SHO PROC/PHD. THAT`S IT!, It tells me what page file the process is on   So the whole thing is: $ SHOW ME/FILE/FULLm! --> Identify the page file numbert  
 $ ANAL/SYS SDA> SET OUTPUT F.LISr SDA> SHOW PROC/PHD ALL SDA> ^Z L $ SEAR F.LIS "extended pid", "Current page file                 4"/OUT=F.TXT $ SEAR F.TXT "Current page filek 4"/OUT=F.LIS/WIN=(1,0)/NOHEADe   Answer is in F.LIS   --   Syltrem=I http://pages.infinit.net/syltrem (OpenVMS related web site - en franais)c> To reply to myself directly, remove .spammenot from my address  E "Richard Brodie" <R.Brodie@rl.ac.uk> a crit dans le message de news:t! ae7v2l$1pf8@newton.cc.rl.ac.uk...a >o= > "Syltrem" <syltremspammenot@videotron.com> wrote in messaget0 > news:AxIN8.3229$H67.17254@tor-nn1.netcom.ca... > L > > I would like to deinstall one pagefile from my system (it's currently inG > > deinstall pending), and would like to know what processes are stillA using0G > > it. If I know what they are and I could kill them (they may be just  users,I > > no system stuff), then I could delete the file without having to do aA > > reboot.O > ? > f$getjpi(pid,"PAGFILLOC") or SHOW PROC/PHD and tie the numberH& > up with SHOW MEM/FILE/FULL displays. >o
 > Quick hack:R >S > $ ctx = ""= > $ temp = F$CONTEXT ("PROCESS", ctx, "USERNAME", "*", "EQL")A > $loop: > $ pid = F$PID(ctx) > $ IF pid .EQS. "": > $ THEN > $     GOTO endloop > $ ELSE& > $     pf = f$getjpi(pid,"PAGFILLOC")- > $     pf = f$integer("%x" + pf) / %X1000000 % > $     write sys$output pid, " ", pfu > $     GOTO loop-	 > $ ENDIF- > $ endloop: exita >  >,   ------------------------------    Date: 12 Jun 2002 15:12:05 -0600- From: koehler@encompasserve.org (Bob Koehler)D1 Subject: Re: What process is using this pagefile?E3 Message-ID: <bmnEGkIe1CGF@eisner.encompasserve.org>S  f In article <ae7mrm$pqc$1@n.ruf.uni-freiburg.de>, gartmann@immunbio.mpg.de (Christoph Gartmann) writes:   >   SHOW DEVICE/FILESX  <    At that level the OS has the file open, not each process.   ------------------------------    Date: 12 Jun 2002 14:02:57 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris) 1 Subject: Re: What process is using this pagefile?,= Message-ID: <cf15391e.0206121302.434b6d44@posting.google.com>   l "Syltrem" <syltremspammenot@videotron.com> wrote in message news:<AxIN8.3229$H67.17254@tor-nn1.netcom.ca>...J > I would like to deinstall one pagefile from my system (it's currently inK > deinstall pending), and would like to know what processes are still usinge > it.   E (I cheated by looking this up in a Google search, because it has beend! discussed in comp.os.vms before.)e  @ $SHOW MEMORY/FILES/FULL will tell you the indices for all of theC pagefiles, and once you've identified the index for the pagefile ofh3 interest then you can sift through the output from:     $ANALYZE/SYSTEM    SDA> SHOW PROCESS/PHD ALLB looking for processes which use that pagefile index, and then killA those processes.  And you can check the "system" process pagefiles assignment with:    SDA> SHOW PROCESS/SYSTEM/PHD'B but if the "system" process is using that pagefile, I think you're looking at a reboot, IIRC.  B As a quick first pass, you could use a bit of DCL to loop over allA processes checking F$GETJPI(pid,"PAGFILLOC") for a match with they? desired index, but that only finds a process if the pagefile ofaF interest is the Current pagefile assignment for that process, and thusD might miss processes which used that pagefile earlier and still have? some pages in it, although they've been subsequently reassigned  another primary pagefile.e: ----------------------------------------------------------: Keith Parris | parris <at> DECUServe <dot> decus <dot> org   ------------------------------  % Date: Thu, 13 Jun 2002 12:44:23 +1000g* From: "Dale King" <dalek@forpresident.com> Subject: Re: xtoolkit error2' Message-ID: <ae90u8$53l$1@lore.csc.com>,  9 "Jesper Naur" <jesper.naur@post.tele.dk> wrote in messagec5 news:3d0227e5$0$238$edfadb0f@dspool01.news.tele.dk...dN > I very much suspect, that the xcursion server is not started on the PC. This > can be verified as follows:   L From your VMS machine, you should also be able to $telnet 192.168.0.123 6000O and get connected.  This will tell you if there is something preventing trafficaO to your X server and ensures it is running - if you get connection refused thenhK your X server is preventing the incoming connection, though you should haver already been told of this.   ------------------------------   End of INFO-VAX 2002.325 ************************