1 INFO-VAX	Sun, 14 Jul 2002	Volume 2002 : Issue 384       Contents:- Re: "Clean" CISC (was Re: McKinley Cometh...) - Re: "Clean" CISC (was Re: McKinley Cometh...)  2 CMS questions  Re: 2 CMS questions  Re: Doing the Math on Alpha  Re: HP Itanium2 benchmarks Re: HP Itanium2 benchmarks$ I wonder if they even know about VMS! Re: Itanium II Another Shoe Drops ! Re: Itanium II Another Shoe Drops  Re: McKinley Cometh...+ Re: McKinley tops SpecFP AND SpecInt charts . Re: OpenSSL and certificates concept questions. Re: OpenSSL and certificates concept questions Re: OpenVMS Ambassadors A Re: OpenVMS on third-party platforms (was: Re: VMS port delayed!) A Re: OpenVMS on third-party platforms (was: Re: VMS port delayed!)  Re: OpenVMS Polls are back!  Re: Oracle RDB on VMS  Re: Oracle RDB on VMS  Re: Oracle RDB on VMS  Re: Oracle RDB on VMS 1 PW600au problem (probably h/w) - long and tedious 1 Re: Quality control problems in VMS Engineering ? 1 Re: Quality control problems in VMS Engineering ? ; Re: TSM (Terminal Server Manager) on DECnet IV and DECnet V ; Re: TSM (Terminal Server Manager) on DECnet IV and DECnet V ; Re: TSM (Terminal Server Manager) on DECnet IV and DECnet V  Re: un-INITing a tape? Re: un-INITing a tape? Re: un-INITing a tape?= Re: VS3100/38: Infinitely looping bugchecks at boot (VMS 552) $ Who said Carly doesn't like OpenVMS?  F ----------------------------------------------------------------------  + Date: Sat, 13 Jul 2002 20:51:18 +0000 (UTC) / From: Sander Vesik <sander@haldjas.folklore.ee> 6 Subject: Re: "Clean" CISC (was Re: McKinley Cometh...)2 Message-ID: <1026593480.83354@haldjas.folklore.ee>  > In comp.arch Stephen Fuld <s.fuld@pleaseremove.att.net> wrote: > 7 > "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message , > news:agk81u$8pk$1@pegasus.csx.cam.ac.uk... >  > snip > ? >> I think that it depends on what you are comparing them with. < >> I agree that they are better than an unclear CISC, but my> >> experience is that they have a much larger number of really) >> nasty problems than the cleaner CISCs.  > I > This brings up an interesting question. What do you consider a "clean"" J > CISC?  Intel (and AMD) has proven that you can take what most consider aJ > really ugly CISC instruction set and make it run very fast, at least forL > integer code.  I suspect that if it wasn't saddled with the byzantine 8087N > compatible floating point model, it could run floating point stuff very fast
 > as well.  O I (well, ok, AMD) have an opteron to sell you - you just need to recompile for  6 Linux or one of the BSDs running it in 64bit mode. 8-)   > K > So the question is, could a well designed CISC architecture out perform a N > well designed RISC architecture?  About the only advantage I can see for theL > CISC design is the perhaps shorter instruction length leading to perhaps aM > higher I cache hit rate and lower required instruction fetch bandwidth.  On H > the other hand, you light lose some optimiziation opportunities as theK > reordering provided by the CISC to RISC translation is probably less than 1 > that what could be provided by a good compiler.   F Its not about architecture - nobody just has built a 6 wide RISC that I does similar runtime optimisations as Athlons. For lots of things having  6 64K+64K 2-way followed by highly associative L2 helps.   > J > What other advantages and disadvantages are there, and what would be the6 > characteristics of a "good clean CISC" architecture?  $ "Don't use 2-operand instructions" ?   >  > -- > - Stephen Fuld, >   e-mail address disguised to prevent spam >  >    --   	Sander    +++ Out of cheese error +++    ------------------------------  + Date: Sat, 13 Jul 2002 21:01:49 +0000 (UTC) / From: Sander Vesik <sander@haldjas.folklore.ee> 6 Subject: Re: "Clean" CISC (was Re: McKinley Cometh...)2 Message-ID: <1026594111.16472@haldjas.folklore.ee>  : In comp.arch Peter Boyle <pboyle@holyrood.ed.ac.uk> wrote: > & > On Thu, 11 Jul 2002, JF Mezei wrote: > # >> One issue about the MVC vs RISC.  >>M >> In cobol and other languages, MVC uis extremely useful because you know in , >> advance how many bytes need to be copied. >>O >> But in Unix/C, a routine such as strcpy must really look at individual bytes O >> to determine where the copy is to stop, so a single assembler instruction to Q >> move large amounts of data doesn't map terribly well to C in this case and you @ >> need a loop with a test for 0 in there and move byte by byte. >  > Absolutely, ;) > + > C null terminated strings are responsible 3 > for a great many optimisation and security evils.   P Well, on the positive side, lots of peopel didn't have to rewrite their softwareO because somebody had assumed string length could safely be 8 bits or 16 bits or  possibly even 32 bits 8-)    >  > Peter  >    --   	Sander    +++ Out of cheese error +++    ------------------------------  % Date: Sat, 13 Jul 2002 19:57:43 -0700 # From: "Tom Linden" <tom@kednos.com>  Subject: 2 CMS questions9 Message-ID: <CIEJLCMNHNNDLLOOGNJIAEMCFFAA.tom@kednos.com>   H Never quite have understood classes and the syntax used to specify them.  H an element may belong to classes (including the null class).  Now if you modifyK an elemnt by the usual RESERVE, REPLACE it won't be replaced in the classes J where it appears.  There you must explicitly REMOVE and INSERT the element using K the correct generation and classes.  This all seems very error prone to me.   F Is there no way to determine all occurences of an element (and current generation) I in the various classes and to do a global replcement?  (Assume you do not  know how many classes there are)   H I may have made an error, but it seems the order of the CMS libraries is
 importast.K In my case I have two libraries, and when I issue CMS SHOW LIB it lists the 
 libraries,J as say A and B in that order.  Now if I try to FETCH or RESERVE an element which is in J B it responds element not found, but if I issue CMS SET LIB B then life is good.   L Finally, it appears that the /GEN qualifier is overloaded, sometimes meaning class, this  is most unfotunate.   A Is there any other interface to the CMS libs that provides a nore  consistent, orthogonal. and simpler syntax, preferrably through emacs?   TIA  Tom  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.372 / Virus Database: 207 - Release Date: 6/20/2002   ------------------------------  # Date: Sun, 14 Jul 2002 04:13:01 GMT ( From: "C.W.Holeman II" <cwhii5@ACM5.org> Subject: Re: 2 CMS questionsF Message-ID: <hT6Y8.19905$A43.2009893@newsread2.prod.itd.earthlink.net>   Tom Linden wrote:   J > Never quite have understood classes and the syntax used to specify them. > J > an element may belong to classes (including the null class).  Now if you > modifyE > an elemnt by the usual RESERVE, REPLACE it won't be replaced in the 	 > classes L > where it appears.  There you must explicitly REMOVE and INSERT the element > using I > the correct generation and classes.  This all seems very error prone to  > me.   B If a class is used to identify which generations of files are in aK release of a product then new generations of a files would correspond to a   different release.   --   C.W.Holeman II cwhii5@ACM5.org  remove the fives http://also.as/cwhii   ------------------------------   Date: 14 Jul 2002 00:50:44 GMT* From: Rick Kelly <rmk@roothog.rmkhome.com>$ Subject: Re: Doing the Math on Alpha5 Message-ID: <3d30cae4$0$53629$75868355@news.frii.net>   ) Nick Maclaren <nmm1@cus.cam.ac.uk> wrote:   E > There is also very good evidence that the fact that the Alpha sales B > figures were poor was not because there was a lack of customers.B > I don't know how many customers were begging to buy, but many of@ > them were wanting to buy in units of dozens, hundreds or more.  ( Indeed. Look at: http://www.microway.com  0 They sell Alpha systems based on Samsung boards.  / And then search on Google: Linux Alpha clusters   C There are a lot of small companies out there that are selling Alpha  workstations...   H Compaq didn't like selling Alpha workstations ahead of their proprietary Intel junk.   A And there are a lot of NetBSD and OpenBSD users running on Alpha.   M NetBSD/alpha and NetBSD StrongARM were partially funded by DEC. At that point > in time it seemed that DEC wanted to sell to the desktop user. --  , rmk@rmkhome.com		http://www.rmkhome.com/~rmk   ------------------------------  + Date: Sat, 13 Jul 2002 21:13:15 +0000 (UTC) / From: Sander Vesik <sander@haldjas.folklore.ee> # Subject: Re: HP Itanium2 benchmarks 3 Message-ID: <1026594796.481262@haldjas.folklore.ee>   ; In comp.arch Larry Kilgallen <Kilgallen@spamcop.net> wrote: O > In article <3d2f2a5f$1@news.meer.net>, lindahl@pbm.com (Greg Lindahl) writes: E >> I was reading the HP Itanium2 performance whitepaper and noticed a   >> couple of interesting points: >>  I >> In the lmbench section, page 7, they state "The zx1 hardware pre-fetch H >> plays a significant role in achieving this score". If this were true,! >> then they broke the benchmark.  > E > But for anybody whose real program benefits, that is not a problem.   L yes - but this does not mean the real memory latency is now lower, just thatJ some more programs can take advantage of prefetching. Which is an entirely different thing...   --   	Sander    +++ Out of cheese error +++    ------------------------------    Date: 13 Jul 2002 18:15:41 -0700$ From: lindahl@pbm.com (Greg Lindahl)# Subject: Re: HP Itanium2 benchmarks & Message-ID: <3d30d0bd$1@news.meer.net>  F In article <loQX8.18113$A43.1822962@newsread2.prod.itd.earthlink.net>,* Norbert Juffa <juffa@earthlink.net> wrote:  ? >http://www.hp.com/products1/itanium/chipset/analyst_report.pdf  > H >The stated bandwidth of the 2-way system is 8.5 GB/s, derived from fourA >PC2100-CL2 (DDR266A) channels, and stated bandwidth of the 4-way  >system is 12.8 GB/s.   E Right. But if you look at the system diagram, the front-side bus is a 5 bottleneck. That's why STREAM numbers would be handy.   B >The description of the zx6000 workstation claims 78ns "open page" >memory latency.  E It says 78 ns "memory system latency", which probably has to have the & 32 ns of Itanium2 latency added to it.   greg   ------------------------------  # Date: Sun, 14 Jul 2002 01:31:28 GMT # From: "John N." <JNixon@cfl.rr.com> - Subject: I wonder if they even know about VMS ; Message-ID: <Qv4Y8.29089$DS.689140@twister.tampabay.rr.com>   F This sounds like an ideal environment for a VMS installation.  Secure,I powerful, flexible, scalable, and think of all the jobs for VMS people it  would open up.  But alas . . .  L http://story.news.yahoo.com/news?tmpl=story&ncid=581&e=3&cid=581&u=/nm/20020' 712/tc_nm/bush_homeland_technology_dc_1   . Computers for New Security Agency Under Review Fri Jul 12, 7:58 PM ET  J WASHINGTON (Reuters) - The White House budget office said on Friday it wasH setting up a review board to assess the technology needs of the proposed  Department of Homeland Security.      K "This new department should have one world-class infrastructure, and on the L books right now are plans totaling between $1 billion and $2 billion. And soI a review board has been constituted to pick the best of those and to move K forward with one plan," White House budget director Mitch Daniels said. . .  .    ------------------------------  # Date: Sat, 13 Jul 2002 18:31:40 GMT # From: "John Smith" <a@nonymous.com> * Subject: Re: Itanium II Another Shoe DropsF Message-ID: <gm_X8.3442$WsS.2384@news01.bloor.is.net.cable.rogers.com>  5 "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message * news:agpoof$o1c$1@pegasus.csx.cam.ac.uk...G > In article <62XX8.3171$WsS.444@news01.bloor.is.net.cable.rogers.com>, $ > John Smith <a@nonymous.com> wrote: > > 8 > >"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message- > >news:agonhf$1i6$1@pegasus.csx.cam.ac.uk...  > >>E > >> But, to be radical, HP could perfectly well be the new Dell.  If D > >> I were Fiorina, I would spin off the HP and Compaq PC divisionsB > >> as a new, wholly owned company.  This would both simplify theB > >> merger, and allow that company to use Dell's model to competeF > >> with Dell head-on.  And, if it flopped, it wouldn't interfere too! > >> badly with the main company.  > > I > >That would be a good solution but it requires a considerable effort to E > >revamp the manufacturing process to one that is almost exclusively  > >'build-to-order'. > > D > >However, I would think that the largest impediment to this is the	 'channel' I > >and how it will react to what would amount to the elimination of their  > >livelihood. ... > > K > >I thinkthe spun-off division really only makes sense to HP if it is sold  via F > >an IPO (burn your bridges?) rather than keeping it as a wholly-ownd > >subsidiary. ... > D > Oh, I agree with your points.  The advantages would be more to getB > rid of a major distraction.  HP/Compaq would then concentrate onC > the 'PC sales' that are done by the hundred and include full 24x7 @ > support, and similar relatively high margin markets.  This is,( > after all, the approach that IBM took. > D > What I REALLY can't see is HP/Compaq as currently organised having3 > a hope of competing with Dell in the bulk market.     	 I concur.   J What this means in a practical sense is that as long as HP 'thinks' it canK compete with Dell, it will maintain the manufacturing infrastructure at the H size it currently is. And that means that they will lose money with eachF machine that they sell. The more they sell, the more they lose, though9 perhaps at a lesser rate per machine as volumes increase.   K The Wintel server space is the place where they can remain compatitive, but J I don't think that I2 will take off there. The only way that it will is ifL Intel finds an opportunity to kill the P3/P4 server-grade chips, most likely$ through a failure of AMD to compete.  L The reality at the desktop is that even running MS 'bloatware', *almost* allL machines currently installed at S&P 1000 companies are more than capable forL the average worker.  Some corporations will replace their machine due to taxL reasons, others due to the desire to have a single supported platform (whichD becomes hard to do after the passage of several years). But for mostH companies, there is no compelling reason to change desktop hardware. How8 much faster does the average worker need to spell-check?  K In the consumer space, the reasons are even less compelling for most users.    ------------------------------  % Date: Sat, 13 Jul 2002 23:18:06 -0400 ( From: David Froble <davef@tsoft-inc.com>* Subject: Re: Itanium II Another Shoe Drops, Message-ID: <3D30ED6E.2040004@tsoft-inc.com>   JF Mezei wrote:    > Bob Ceculski wrote:  > P >>>Consider that IA64 will get most of the HP-UX and NSK customers. It will also' >>>get part of Tru64 and VMS customers.  >>> 3 >>part of VMS customers?  Try all of VMS customers!  >> > N > At this point in time, it is way too early to tell what will happen with the > VMS marketplace.   > M > Lets assume for a moment that HP intends to scale VMS down over the next 10 I > years. If this becomes apparent before VMS on IA64 becomes commercially P > viable, then it is more likely that the remaining VMS customers will just stayP > on their Alpha systems as long as possible and have migration plans to anotherO > vendor over the medium/long term. The Alpha murder last year probably sparked  > quite a few such plans.  > N > Also consider the VAX installed base. With HP having said absolutely nothingP > about VMS on VAX, and whether new versions will appear on VAX,  it is possibleO > that HP intends to stop VMS on VAX ASAP so that it can start the 5 year timer P > for support. Many of those customers would then be forced with a migration andR > they may choose to go to a system whose future is not constantly be in question. >   N Come on JF, that's wild ass speculation, with nothing to suggest that it will  happen.  FUD, pure and simple.   ------------------------------  + Date: Sat, 13 Jul 2002 20:04:29 +0000 (UTC) / From: Sander Vesik <sander@haldjas.folklore.ee>  Subject: Re: McKinley Cometh... 3 Message-ID: <1026590670.590078@haldjas.folklore.ee>   ; In comp.arch JF Mezei <jfmezei.spamnot@videotron.ca> wrote:n > Rupert Pigott wrote:/ >> http://www.apple.com/xserve/performance.htmlC? >> It's also complete shite because they don't compare the same F >> codebases and they don't provide the source so you can verify this. > M > That is a fair criticism. But doesn't the apple comparison represent a moreuM > "real world" approach. You take the application available on platform 1 andSP > compare it with whatever application is available on platform 2 to do the same > work ?  M No - it represents a more real world approach to something. Did they publish sR specweb results? It is very unclear why somthing that involves non-trivial amountsR of disk access or network connectivity is more 'real world' than one that doesn't.O Its just picking a mix - and unless what they publish the suite and guidelines,l@ it is hard to take as benchmarketing of a pretty pointless kind.   > M > What I find significant is that Apple is starting to push its Unix servers.e   -- n 	SanderD   +++ Out of cheese error +++    ------------------------------  # Date: Sat, 13 Jul 2002 22:06:49 GMTeL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")4 Subject: Re: McKinley tops SpecFP AND SpecInt charts8 Message-ID: <00A10E07.608C0F84@SSRL04.SLAC.STANFORD.EDU>  l In article <uEWX8.3147$WsS.1537@news01.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes: > 6 >"Bob Ceculski" <bob@instantwhip.com> wrote in message8 >news:d7791aa1.0207130441.6df5ba6d@posting.google.com... >>H >> I think you are underestimating Carly ... she is a former MIT studentH >> which many ex-DEC techs were, and she understands the DEC superiorityI >> and Ken Olson philosophy, unlike Capellas/Palmer bean counters ... shea >> may just suprise you! >nL >Bob, in the end you may well be correct. However you cannot make a sweepingK >statement that simply because somebody went to MIT has the same philosophyaL >of others who also went to MIT, nor that she understands 'DEC superiority'./ >That kind of statement simply hold no water.  s  E And even further, a lot of HP/Agilent people don't think Ms. Fiorina eK understands _HP's_ considerable tradition of technical superiority or care aH for employees - and that was moving into a company that still existed.    J It's possible that she'll get and take good advice and the results of thisH merger will be surprisingly good, but we can't predict that based on herM having met Ken Olsen once.  I expect Bob Palmer met Ken Olsen more than once.e   -- Alans    O ===============================================================================C0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056eM  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210rO ===============================================================================i   ------------------------------  % Date: Sat, 13 Jul 2002 17:09:50 -0500t( From: Rich Jordan <duodec@speakeasy.net>7 Subject: Re: OpenSSL and certificates concept questionsr, Message-ID: <3D30A52E.1080706@speakeasy.net>   Larry Kilgallen wrote:Y > In article <3D2FB543.9020500@speakeasy.net>, Rich Jordan <duodec@speakeasy.net> writes:- >  > 6 >>Who designed the x.509 certs and the SSL API anyway? >  > F > I don't know about the SSL API, but the x.509 certificates are quiteF > nice to work with.  The trick is that you should not try to code forG > them by hand.  Use an ASN.1 tool that generates code in your favoritev > programming language.   I Are there any such available under OpenVMS?  I'll have more time to look  L come Monday but this weekend is already allocated.  Thanks for the response!   Rich Jordan    ------------------------------  % Date: Sat, 13 Jul 2002 17:57:00 -05005( From: Rich Jordan <duodec@speakeasy.net>7 Subject: Re: OpenSSL and certificates concept questions , Message-ID: <3D30B03C.5040706@speakeasy.net>   david20@alpha1.mdx.ac.uk wrote: h > In article <cc5619f2.0207121235.74601f9c@posting.google.com>, jordan@ccs4vms.com (Rich Jordan) writes: > ^ >>Rich Jordan <duodec@speakeasy.net> wrote in message news:<3D2E5D27.1020106@speakeasy.net>... >>2 >> ..... but still looking for a way to get SSL toA >>read an environment variable (symbol or logical?) to locate theoH >>'random seed file' we provided to get s_client working; changes to the: >>openssl.cnf file also don't seem to make a difference.   >> > K > Don't know if this will help but in order to generate certificates on VMSoP > you need to seed the random number generator - the recommended procedure is :- > 
 > ........  > This is documented on :-d  >F  > http://www.ourservers.net/openvms_ports/openssl/openssl3.html#ss3.1  >   David,E       thanks very much for the info, and especially for that URL.  I  B don't believe google managed to provide that one yet (or I hadn't 3 managed to come up with the right set of keywords).2   Rich Jordan    ------------------------------    Date: 13 Jul 2002 15:27:02 -07001 From: susan_skonetski@hotmail.com (Sue Skonetski)s  Subject: Re: OpenVMS Ambassadors= Message-ID: <857e9e41.0207131427.131cd32e@posting.google.com>   E I just wanted to make a point here, some times when folks are let go,eC they want to leave and would prefer that they leave with a package,e) sometimes it happens to be an Ambassador.d  F From what the Ambassadors have told me there was also a period of timeF when the Ambassadors program was in a state of depression but that was awhile ago now.a   sueo    a JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3D2DE023.17A2CC35@videotron.ca>...r > "Main, Kerry" wrote:L > > In fact, if the individual you are thinking about initials are PBL, then& > > I could not agree with you more !! > > K > > I knew him quite well as he also worked in Ottawa. He was definitely ine > > the role of VMS "guru" ..  > O > His name has escaped me for the past few weeks. I have a picture of his face,pN > his style of voice, but not his name !  When he was fired, he and some otherJ > ex-DECies formed a company that was the opposite of Digital (think of anO > ______ watch) but that didn't seem to go very well and he was hired by one ofc5 > the few remaining large VMS shops in montreal area.s   ------------------------------    Date: 13 Jul 2002 11:25:59 -0700+ From: seanobanion@attbi.com (Sean O'Banion)eJ Subject: Re: OpenVMS on third-party platforms (was: Re: VMS port delayed!)= Message-ID: <f883d5a4.0207131025.514b2a8e@posting.google.com>c  Q david20@alpha1.mdx.ac.uk wrote in message news:<agos7r$ahd$1@aquila.mdx.ac.uk>...ls > In article <20020712191434.10915.qmail@web20204.mail.yahoo.com>, Fabio Cardoso <fabiopenvms@yahoo.com.br> writes:n > >At the end... > >c7 > >I asked here a few days ago, about what do you think 0 > >about a hacker with an Itanium server at home2 > >with OVMS binaries... Nobody hacks OVMS because( > >it runs in Alphas (non easy to buy) ! > >But with Itanium....  > >h
 > >Regards > >i > >FC  > M > I've been getting the impression their are a fair number of people who havei= > obtained hobbyist licenses - they must be finding hardware.e >  > David Webb > VMS and Unix team leader > CCSS > Middlesex University    H AlphaServer 2100 on EBay are around $500.  I've got two, one for spares.     Sean   ------------------------------  # Date: Sat, 13 Jul 2002 18:35:53 GMTy# From: "John Smith" <a@nonymous.com>RJ Subject: Re: OpenVMS on third-party platforms (was: Re: VMS port delayed!)I Message-ID: <dq_X8.44713$WJf1.10735@news01.bloor.is.net.cable.rogers.com>y  5 "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in messageu* news:agp2td$9i1$1@pegasus.csx.cam.ac.uk... >oD > So, what I am saying, is that the hobbyist licence could well takeE > off (especially in dual-boot mode) if boxes that could run it startmB > to be very widespread.  DS10Ls aren't, and few people here wouldB > buy one for their own use - commodity Intel or AMD PCs give more > the the money.  K Which perhaps, in a way, goes back to the failure of Digital to ensure thattJ Microsoft Office products had release parity on Alpha when they negotiated) the BackOffice products parity agreement.o  J Had that occurred, more sales of Alpha running desktop products would haveG resulted, increasing Alpha chip volumes, and most likely would not haveh- allowed the death of 64-bit Windows on Alpha.    ------------------------------  % Date: Sun, 14 Jul 2002 06:10:49 +0200-. From: Dennis Grevenstein <dennis@pcde.inka.de>$ Subject: Re: OpenVMS Polls are back!, Message-ID: <3D30F9C9.67551D49@pcde.inka.de>   Peter da Silva wrote:2 >4J > Is that sentence syntactically sound? I didn't think IA64 and affordable@ > were allowed in the same sentence without a negation operator.  < SPARC's aren't cheap either, but low end Sun's sell for lessA than just the OpenVMS operating system license for a workstation.iC So please tell me how you want to get to 1000$ for a whole VMS box?z   Dennis   -- eC Scientific view: studies show that men tend to be better at dealingeB with visual concepts, while women are better at complex linguistic communication.; computer geek view: man are from Mac's, women are from VMS!o   ------------------------------  % Date: Sat, 13 Jul 2002 12:46:17 -0600c+ From: "Ransom Fitch" <rlfitch@peakpeak.com>  Subject: Re: Oracle RDB on VMS0 Message-ID: <000601c22a9d$930c8ae0$0a00a8c0@w2k>  ? Anyone know where to get (find) and older version of Oracle RDB  compatible with OpenVMS v6.2?d   thanks,c Ransom Fitch   ------------------------------  % Date: Sat, 13 Jul 2002 21:04:42 +0200X9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>b Subject: Re: Oracle RDB on VMS' Message-ID: <3D3079CA.8B90346A@aaa.com>a   What is the problem ?r8 VMS 6.2 is just fine with the supported Rdb 7.0.x.x kit.( It's just Rdb 7.1.x.x that needs VMS 7.2   As can be verifyed on :d  6 http://www.oracle.com/rdb/product_info/html_documents/( index.html?rdb_prod_matrix_020607_a.html  6 So the downloadable kit ("developer kit") would be OK.   Jan-Erik Sderholm     Ransom Fitch wrote:  > A > Anyone know where to get (find) and older version of Oracle RDB  > compatible with OpenVMS v6.2?i > 	 > thanks,  > Ransom Fitch   ------------------------------  % Date: Sat, 13 Jul 2002 13:20:54 -0600-+ From: "Ransom Fitch" <rlfitch@peakpeak.com>r Subject: Re: Oracle RDB on VMS0 Message-ID: <000c01c22aa2$68fcc620$0a00a8c0@w2k>   Excellent! =20 Thanks very much!t   Ransom Fitch       > -----Original Message-----4 > From: Jan-Erik S=F6derholm [mailto:aaa@aaa.com]=20' > Sent: Saturday, July 13, 2002 1:05 PMt > To: Info-VAX@Mvb.Saic.Como  > Subject: Re: Oracle RDB on VMS >=20 >=20 > What is the problem ?t: > VMS 6.2 is just fine with the supported Rdb 7.0.x.x kit.* > It's just Rdb 7.1.x.x that needs VMS 7.2 >=20 > As can be verifyed on :  >=20H http://www.oracle.com/rdb/product_info/html_documents/index.html?rdb_pro d_matrix_020607_a.html  6 So the downloadable kit ("developer kit") would be OK.   Jan-Erik S=F6derholm     Ransom Fitch wrote:i >=20D > Anyone know where to get (find) and older version of Oracle RDB=20 > compatible with OpenVMS v6.2?b >=20	 > thanks,  > Ransom Fitch   ------------------------------  % Date: Sat, 13 Jul 2002 21:28:09 +0200o9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>  Subject: Re: Oracle RDB on VMS' Message-ID: <3D307F49.D776E6D4@aaa.com>a  C FYI, if you'd like to ask and get Rdb questions/answers from a morej> Rdb-specific group, check the Rdb list-server at JCC Inc. at :   www.jcc.com.  ; compleat instructions for the list-server can be found at :.  / http://www.jcc.com/jccs_oracle_list_servers.htm.   Jan-Erik Sderholm.a   Ransom Fitch wrote:4 >  > Excellent! > Thanks very much!> >  > Ransom Fitch >y   ------------------------------  % Date: Sun, 14 Jul 2002 06:44:51 +0930s/ From: Mark Daniel <mark.daniel@wasd.vsm.com.au>.: Subject: PW600au problem (probably h/w) - long and tedious/ Message-ID: <3D30984B.DDB192BF@wasd.vsm.com.au>d  H Unfortunately a hardware maintainence call is not possible in this case.  H This PW 600au restarts ('crashes') without so much as leaving a trace of> a dump or error log entry.  Obviously a hardware issue of someD description.  Stable until there's lots of SCSI activity (backup forE instance) then bang!  I've tried moving everything off the A bus to afC second SCSI controller but to no effect, even though there's now nox activity on that bus.o  D I've included (what I think is) the relevant DIAGNOSE entry and some! systems details from the console.w   Any suggestions?  / Slightly off-topic, I know.  Thanks in advance.v  F ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  , Logging OS                        1. OpenVMS* System Architecture               2. Alpha) OS version                           V7.3g$ Event sequence number            58.9 Timestamp of occurrence              13-JUL-2002 05:07:30p5 Time since reboot                    0 Day(s) 8:20:00 + Host name                            KLAATU   A System Model                         Digital Personal WorkStatione  B Entry Type                       98. Asynchronous Device Attention     ---- Device Profile ----0 Unit                                 KLAATU$PKA0B Product Name                         ISP1020/1040 PCI-SCSI Adapter   ----- SCSI Port -----h# Long Word Count           x00000077i# Error Log Revision              x01r4 Error Type                    x030C  0x0C, Bus Error@                                      0x03, SCDRP due to watchdog# SCSI ID                         x00u  # Command Length                  x0Ae Command & Data#                                 x28t#                                 x00i#                                 x00l#                                 x75 #                                 x61y#                                 x20,#                                 x00D#                                 x00s#                                 x05r#                                 x00i  # Message Length                  x80  Request Entry - ,  Request Entry Type             x01  Command#  Entry Count                    x01n#  Sys_Defined                    x14 #  Flags                          x00o#  Handle                   x27880082e#  Lun                            x00,#  Target                         x00n  Request remaining bytes  >           15--<-12  11--<-08  07--<-04  03--<-00   :Byte OrderE  0000:    20617500  00280001  00180000  4028000A   *..(@......(..ua * E  0010:    00000000  0A00C00E  8C000000  00050000   *................*0E  0020:    00000000  00000000  00000000  00000000   *................*nE  0030:                        00000000  00000000   *        ........*z   Response Entry -+  Response Entry Type            x03  Statusi#  Entry Count                    x01o#  Sys_Defined                    x15u#  Flags                          x00o#  Handle                   x27090082o#  SCSI Status                  x0000m6  Completion Status            x0000  Command Complete.C  State Flags                  x5F00  SCSI Bus Arbitration Occurred.sB                                      Target Selected Succussfully.D                                      CDB Transferred Successfully to Target.t?                                      Data In/Out Phase Transfers	 Occurred.o=                                      Status Phase Successful.e7                                      Transfer Complete.l9  Status Flags                 x0001  Disconnect Occurred.   Response remaining bytesh  >           15--<-12  11--<-08  07--<-04  03--<-00   :Byte OrderE  0000:    04000000  00007C7B  00000000  00000018   *........{|......*oE  0010:    CF040024  00000000  0AF1FFFF  FF2500F0   *..%.........$...*pE  0020:    00000000  00000000  00000000  00000700   *................*     7 SCSI Status                     xFF  No Status Receiveda   --- Chip Registers ---7 Vendor/Device ID          x10201077  Vendor ID = QLogico=                                      Device ID = ISP1020/1040-# Hardware Revision             x0005,6 Mailbox 0 Status              x4000  Command Complete.   ----- Software Info ----- 6 UCB$x_ERTCNT                      4. Retries Remaining6 UCB$x_ERTMAX                      0. Retries Allowable+ IRP$Q_IOSB                x0000000000000000e+ UCB$x_STS                 x10000010  Onliner4 IRP$L_PID                 x00000000  Requestor "PID"5 IRP$x_BOFF                        0. Byte Page Offset = IRP$x_BCNT                        0. Transfer Size In Byte(s) 5 UCB$x_ERRCNT                      4. Errors This Unite4 UCB$L_OPCNT                      38. QIO's This Unit/ ORB$L_OWNER               x00010004  Owners UIC . UCB$L_DEVCHAR1            x0C440000  Available2                                      Error Logging5                                      Capable of Inputb6                                      Capable of Output  F ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~   Firmware4 SRM Console:    V7.2-1          Mar  6 2000 14:47:02 ARC Console:    5.70E PALcode:        OpenVMS PALcode V1.20-16, Tru64 UNIX PALcode V1.22-18e SROM Version:   v5.90e  	 Processor 8 DECchip (tm) 21164A-2   Pass   600 MHz  96 KBytes SCache 2 MB BCaches PYXIS ASIC Pass 257n   MEMORY     Memory Size = 192Mb    Bank      Size/Sets   Base Addr  ------    ----------  ---------     0         64Mb      00000000N    1         64Mb      04000000m    2         64Mb      08000000n     BCache Size = 2Mbc   Tested Memory =  192Mbytes   PCI Bus4D      Bus 00  Slot 03: Digital Semiconductor 21143 Network Controller8                                    ewa0.0.0.3.0          00-00-F8-76-1C-F6o  7      Bus 00  Slot 07: Cypress PCI Peripheral Controller7    )      Bus 00  Slot 07: Function 1: PCI IDE.    )      Bus 00  Slot 07: Function 2: PCI IDE.  H                                    dqb0.0.0.207.0         TOSHIBA CD-ROM XM-62s        Bus 00  Slot 07: PCI USBO6      Bus 00  Slot 11: Digital TGA2 Graphics Controller5      Bus 00  Slot 20: DECchip 21152 PCI to PCI Bridgey/        Bus 01  Slot 04: ISP1040 Scsi Controller F                                    pka0.7.0.1004.0       SCSI Bus ID 7  3        Bus 01  Slot 10: NCR     810 Scsi ControlleraF                                    pkb0.7.0.1010.0       SCSI Bus ID 7?                                    dkb0.0.0.1010.0        RZ29Bn?                                    dkb100.1.0.1010.0      RZ28D ?                                    mkb600.6.0.1010.0      TLZ09m   ISAnG Slot    Device  Name            Type         Enabled  BaseAddr  IRQ    a DMAo 0dB         0       MOUSE           Embedded        Yes     60      12A         1       KBD             Embedded        Yes     60      1xA         2       COM1            Embedded        Yes     3f8     4=A         3       COM2            Embedded        Yes     2f8     3=A         4       LPT1            Embedded        Yes     3bc     7iG         5       FLOPPY          Embedded        Yes     3f0     6      P 2iB         6       EIDE            Embedded        Yes     1f0     14B                                                         3f6     15;                                                         170 ;                                                         376pB         8       USB             Embedded        Yes             10  F ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  D dkb0.0.0.1010.0            DKB0                          RZ29B  0016D dkb100.1.0.1010.0          DKB100                        RZ28D  0010D dqb0.0.0.207.0             DQB0        TOSHIBA CD-ROM XM-6202B  1110 dva0.0.0.0.1               DVA0tD mkb600.6.0.1010.0          MKB600                        TLZ09  0172> ewa0.0.0.3.0               EWA0              00-00-F8-76-1C-F6D pka0.7.0.1004.0            PKA0                  SCSI Bus ID 7  5.57> pkb0.7.0.1010.0            PKB0                  SCSI Bus ID 7> pqa0.0.0.107.0             PQA0                       PCI EIDE> pqb0.0.0.207.0             PQB0                       PCI EIDE  F ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~   pka0_host_id            7  pka0_mode               fast pka0_soft_term          on pkb0_disconnect         1w pkb0_fast               19 pkb0_host_id            7i  F ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~   ------------------------------    Date: 13 Jul 2002 14:44:30 -0600B From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley): Subject: Re: Quality control problems in VMS Engineering ?3 Message-ID: <Y29GN9+MYxJO@eisner.encompasserve.org>l  b In article <agke0t$e62$1@web1.cup.hp.com>, "Sue Skonetski" <susan.skonetski@hp.nospam.com> writes:J > I took Fabio's message and forwarded.  I have removed the names involved+ > since I do not have their ok to use them.-D > __________________________________________________________________ > K > Last October a fix was made to modile ppp_mgmt.c to keep us from matching3H > line TXA0 against TXA1. This fix was triggered by a customer who shallL > reamin namelss. We generated the fix and did the usual in house testing ofL > the fix. We passed it along the the TCP/IP folk and to the unamed customerM > who said it worked. In the normal course of things this fix along with some_J > others found its way into the remedial stream and eventually into a TIMA > kit. > L > The kits were produced and I trusted that they were tested out internally.H > The made their way out to customers. Igot some mail from **** saying aJ > customer had found a problem and asked that the kit be placed on hold. IM > also asked that the folks the original fix was made for confirm that it wasiM > working. They reported that it had in fact broken the disconnect command. IiA > also worked up a fix for the problem that is now in the system.r > L > Did we miss something yes guilty, were we sloppy, no we did all the thingsL > we had done in the past. This one slipped through the cracks it happens itM > should not, but it does. This is not the first nor will it be the last TIMAlB > kit or piece of production code that goes out with a regression. >   K Sue, thank you for forwarding my original message and posting the response.i  K One thing that I want to make very clear is that my interest was _not_ thatoM somebody produced code that was broken. All developers, myself included, makeo0 mistakes and it's unfair to blame them for that.  K My interest was in finding out how it managed to get past the traditionallynK very strong quality control procedures in VMS engineering, as not detecting-L that a once working command has broken suggested that regression testing mayM not have been done or that the testing may have been inadequate, which is not0C the kind of thing that you normally associate with VMS engineering.   L I was also concerned that if by any chance testing had not been done to it'sL normal standards, that this was a oneoff and not the start of a trend due to perhaps cost-cutting..  G Thanks for getting the confirmation that normal testing procedures weret	 followed.i   Simon.   -- vB Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP       + Microsoft: The Lada of the computing world.w   ------------------------------    Date: 13 Jul 2002 15:11:10 -0600B From: clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley): Subject: Re: Quality control problems in VMS Engineering ?3 Message-ID: <OohfdpKBT0ei@eisner.encompasserve.org>   x In article <Y29GN9+MYxJO@eisner.encompasserve.org>, clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) writes: > M > My interest was in finding out how it managed to get past the traditionally@M > very strong quality control procedures in VMS engineering, as not detectingtN > that a once working command has broken suggested that regression testing mayO > not have been done or that the testing may have been inadequate, which is not$  M Just to clarify: When I wondered if regression testing may have not been donenH at all, I meant due to purely accidental causes and _NOT_ as a matter of policy.   E > the kind of thing that you normally associate with VMS engineering.d > N > I was also concerned that if by any chance testing had not been done to it'sN > normal standards, that this was a oneoff and not the start of a trend due to > perhaps cost-cutting.u > I > Thanks for getting the confirmation that normal testing procedures werea > followed.t   Simon.   -- cB Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP       + Microsoft: The Lada of the computing world.    ------------------------------  % Date: Sat, 13 Jul 2002 20:50:33 +0200-9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>0D Subject: Re: TSM (Terminal Server Manager) on DECnet IV and DECnet V' Message-ID: <3D307679.47CDCD20@aaa.com>3  A Note that it's the timeout from *non-responding* terminal serversa@ that is longer under "V" then under "IV". So reducing the number> of terminal servers that don't respond could be a solution, if possible of course.c  F When you do a SHOW SERVER, how many of the 127 servers don't respond ?9 Each of them add 15 sec to the total time for SHO SERVER.o  8 Does also the *responding" servers take longer to list ?  G The errors from your script could be differenct syntax in "IV" and "V".eD Do "SET VER" and rerun to check what the script actualy tries to do.   Regardso Jan-Erik Sderholm.w   Lee Y T Mah wrote: > J > We still have 127 terminal servers on our LAN and have been using DECnetH > IV for many years.  TSM has been used to automate the configuration ofE > the terminal servers via script files and batch jobs.  Yesterday we D > upgraded to DECnet V on an ES45.  The time required to execute one? > particular TSM command under DECnet V has been surprising and-! > disappointing.  The command is:, >  > $ TSM SHOW SERVER ALL  >m   ------------------------------  % Date: Sat, 13 Jul 2002 14:15:55 -0600 $ From: Lee Y T Mah <lytmah@cha.ab.ca>D Subject: Re: TSM (Terminal Server Manager) on DECnet IV and DECnet V) Message-ID: <3D308A7A.C0AC11EF@cha.ab.ca>   # Results on ES45, node D, DECnet IV:    $ SHOW TIME    13-JUL-2002 14:04:06* D _J$DKB0:[RAHSEC] $ $ TSM SHOW SERVER ALL  F   Server           Address            Type     Status        Partition  D   AH206A           08-00-2B-04-6F-4B  DS100    Available     DEFAULTD   C0535A           08-00-87-01-78-D0  DS200    Available     DEFAULTD   C2523C           00-00-C9-00-2F-FB  DS200    Available     DEFAULT ...-D   W6506B           08-00-87-04-7B-79  DS200    Available     DEFAULTD   WB856A           00-00-C9-00-60-73  DS200    Available     DEFAULTD   WB856B           08-00-87-04-84-BB  DS200    Available     DEFAULT  J  TOTAL  127 Servers,  127 Available,    0 Unavailable,    0 Circuit Errors   $ SHOW TIMEo   13-JUL-2002 14:04:06    " Results on ES45, node G, DECnet V:  " G _G$DKA100:[RAHSEC] $ $ SHOW TIME   13-JUL-2002 14:04:32, G _G$DKA100:[RAHSEC] $ $ TSM SHOW SERVER ALL  F   Server           Address            Type     Status        Partition  D   AH206A           08-00-2B-04-6F-4B  DS100    Available     DEFAULTD   C0535A           08-00-87-01-78-D0  DS200    Available     DEFAULTD   C2523C           00-00-C9-00-2F-FB  DS200    Available     DEFAULT ... D   W6506B           08-00-87-04-7B-79  DS200    Available     DEFAULTD   WB856A           00-00-C9-00-60-73  DS200    Available     DEFAULTD   WB856B           08-00-87-04-84-BB  DS200    Available     DEFAULT  J  TOTAL  127 Servers,  125 Available,    2 Unavailable,    0 Circuit Errors   $ SHOW TIME-   13-JUL-2002 14:09:07  # The difference is almost 5 minutes.S. Every display line has a delay of 2-3 seconds.     Jan-Erik Sderholm wrote:   C > Note that it's the timeout from *non-responding* terminal serversiB > that is longer under "V" then under "IV". So reducing the number@ > of terminal servers that don't respond could be a solution, if > possible of course.. > H > When you do a SHOW SERVER, how many of the 127 servers don't respond ?; > Each of them add 15 sec to the total time for SHO SERVER.- >-: > Does also the *responding" servers take longer to list ? > I > The errors from your script could be differenct syntax in "IV" and "V".oF > Do "SET VER" and rerun to check what the script actualy tries to do. >n	 > Regardsd > Jan-Erik Sderholm.i >o > Lee Y T Mah wrote: > > L > > We still have 127 terminal servers on our LAN and have been using DECnetJ > > IV for many years.  TSM has been used to automate the configuration ofG > > the terminal servers via script files and batch jobs.  Yesterday we!F > > upgraded to DECnet V on an ES45.  The time required to execute oneA > > particular TSM command under DECnet V has been surprising andb# > > disappointing.  The command is:s > >  > > $ TSM SHOW SERVER ALL  > >    -- Leea  ; Lee Y T Mah                        Capital Health Authorityc? Email: lytmah@cha.ab.ca            Information Systems, RAH CSC~4 Phone:  (780) 477-4725, 477-4233   10240 Kingsway NW? Fax:      (780) 491-5119, 491-5619    Edmonton, AB, CAN  T5H3V9    ------------------------------  % Date: Sat, 13 Jul 2002 23:47:59 +0200e) From: Bart Zorn <B.Zorn@xs4all.nospam.nl> D Subject: Re: TSM (Terminal Server Manager) on DECnet IV and DECnet V/ Message-ID: <3D30A00F.6050906@xs4all.nospam.nl>    Bob Ceculski wrote:mV > Lee Y T Mah <lytmah@cha.ab.ca> wrote in message news:<3D30286A.EDFF8F3@cha.ab.ca>... > J >>We still have 127 terminal servers on our LAN and have been using DECnetH >>IV for many years.  TSM has been used to automate the configuration ofE >>the terminal servers via script files and batch jobs.  Yesterday werD >>upgraded to DECnet V on an ES45.  The time required to execute one? >>particular TSM command under DECnet V has been surprising and ! >>disappointing.  The command is:d >> >>$ TSM SHOW SERVER ALL  >>' >>Time comparisons running the command:  >>B >>22 seconds    DECnet IV, TSM V2.1-03, on an AS1200 at VMS 7.1-2.
 >>Acceptable. G >>2 seconds      DECnet IV, TSM V2.1-03, on an ES45 at VMS 7.3.  Happy. J >>5 minutes       DECnet V, TSM V2.1-03 and TSM V2.1-06, on an ES45 at VMS >>7.3.  Sad. >>$ >>The DECnet V documentation states: >>5 >>          15.9  SHOW SERVER Slower Under DECnet/OSI  >>I >>              Currently even with the circuit retransmit timer set to 1  >>for I >>              all of the circuits which TSM may be using, TSM will take I >>              longer to timeout on non-responding terminal servers when H >>              doing the SHOW SERVER command in the USE DIRECTORY mode. >>ItC >>              takes approximately 15 * (timer setting) seconds ton
 >>timeout,I >>              and the timer can not be set to less than 1. Note that ifO >>the0F >>              timer is left at the default of 4, each non-responding
 >>terminalF >>              server will take 1 minute to time out. It is recommend
 >>that the/ >>              circuit timers be reduced to 1.3 >>C >>The analyst who performed the migration to DECnet V says that thep) >>circuit timer setting has been set to 1 G >>and the node rebooted.  However, the time required to run the command + >>against the 127 terminal servers is still  >>too long to be acceptable.G >>Is it possible that we did not make the change correctly?  Is there as >>fix for this ugly time delay?t >>I >>I also tried running a batch job under DECnet V to configure a terminal  >>server using TSM commands,% >>and the following errors showed up. 	 >>    ...  >>    Server G6511I is a DS200I >>    %DCL-W-IVOPER, unrecognized operator in expression - check spellingt >>and syntax
 >>     \.27.\ I >>    %DCL-W-IVOPER, unrecognized operator in expression - check spelling  >>and syntax
 >>     \.27.\s	 >>    ...eF >>It looks like the command broke.  (I realize that debug mode will be& >>required to check into the problem.)B >>The DECnet V documentation states that TSM should be compatible.? >>Are there any subtle changes required for TSM under DECnet V?0 >> >>As always, TIA.  >  > @ > Phase V command structure is a nightmare, and I think Phase IV9 > runs much better with fewer commands ... so why switch?   I Here we go again! Phase V runs much better, especially in networks where  G network management has decided that you may only use TCP/IP. And since  : when are you too lazy or whatever, to learn something new?  	 Bart Zorn    ------------------------------  % Date: Sat, 13 Jul 2002 19:55:55 +0100 + From: "antonio.carlini" <arcarlini@iee.org>s Subject: Re: un-INITing a tape? & Message-ID: <3D3077BB.7060105@iee.org>   Dirk Munk wrote:    > You can do that yourself.B  > If you mount the tape as foreign, you can copy the contents to  disk.e    0 You can certainly mount it and you can certainly6 copy data. There won't be any, however. Ever since the4 days of the TK50 (or thereabouts) the mechanism used6 to encode data on the tape has meant that the firmware: on the device cannot go past the EOT mark. It's a feature.  0 Both DAT and DLT devices have this as a feature.  0 Back when DEC had a storage division, this was a8 frequent request ... people did not like the engineering5 team's answer ("Nope. We cannot do that") but I neveru4 saw anyone ever get anything other than that answer.9 The philosophy was that if your data mattered enough that13 you were willing to pay, you could go to one of the 3 3rd party data recovery firms that had the time andP0 the kit to (perhaps) get your data back for you.  C  > Many years ago I used this to copy all kind of data from a tape.a    . Many years ago you had nice simple 9 track (or/ maybe even 7 track) tapes where the drive could-1 just keep going until it picked up the datastream  again.    ' These days you can (a) learn to backup, % (b) pay through the nose or (c) shrugS( your shoulders and accept that your data was not worth that much anyway.    Antonioi     --     ---------------o- Antonio Carlini             arcarlini@iee.org.   ------------------------------  # Date: Sat, 13 Jul 2002 19:05:02 GMTe' From: Stuart Fuller <stufuller@usa.net>o Subject: Re: un-INITing a tape? 0 Message-ID: <7krpga.pko.ln@dadsys1.fuller.local>   Dirk Munk wrote:   > Stuart Fuller wrote: >> Mark Daniel wrote:M >>   >>  > >>>Any way to un-INITIALIZE a (DAT) tape, or recover the data? >>>c >>>OpenVMS Alpha V7.30 >>>o8 >>>(I'm surprised I've never needed to ask this before.) >>>  >>>As always, TIA. >>> I >>>+--------------------------------------------------------------------+ G >>> Mark Daniel                         http://wasd.vsm.com.au/adelaide H >>> mailto:Mark.Daniel@wasd.vsm.com.au (Mark.Daniel@dsto.defence.gov.au)I >>>+--------------------------------------------------------------------+  >> 0 >> . >> No. >> ZG >> At least not with the standard hardware.  Some data recovery vendors H >> claim to be able to recover the data on the tape following the volume, >> labels that you just wrote with the INIT. >> e >>         Stu >  > You can do that yourself.BF > If you mount the tape as foreign, you can copy the contents to disk. > ) > (copy mka500:x dka100:[my-dir]file1.dat . > copy mka500:x dka100:[my-dir]file2.dat etc.) > L > It is possible to skip to the next tapemark too, with a forward command orF > so. Many years ago I used this to copy all kind of data from a tape. >  > Dirk  I Have you actually tried this with a DAT or DLT tape (or any other modern   tape drive)?  J This used to work on 9-track drives, but it does not work on newer drives.           Stu    ------------------------------  % Date: Sat, 13 Jul 2002 19:53:48 +0100 + From: "antonio.carlini" <arcarlini@iee.org>e Subject: Re: un-INITing a tape?~& Message-ID: <3D30773C.2050101@iee.org>   Dirk Munk wrote:   > You can do that yourself. F > If you mount the tape as foreign, you can copy the contents to disk.    0 You can certainly mount it and you can certainly6 copy data. There won't be any, however. Ever since the4 days of the TK50 (or thereabouts) the mechanism used6 to encode data on the tape has meant that the firmware: on the device cannot go past the EOT mark. It's a feature.  0 Both DAT and DLT devices have this as a feature.  0 Back when DEC had a storage division, this was a8 frequent request ... people did not like the engineering5 team's answer ("Nope. We cannot do that") but I never 4 saw anyone ever get anything other than that answer.9 The philosophy was that if your data mattered enough that 3 you were willing to pay, you could go to one of thea3 3rd party data recovery firms that had the time and 0 the kit to (perhaps) get your data back for you.  B > Many years ago I used this to copy all kind of data from a tape.    . Many years ago you had nice simple 9 track (or/ maybe even 7 track) tapes where the drive coulde1 just keep going until it picked up the datastream  again.    ' These days you can (a) learn to backup,i% (b) pay through the nose or (c) shruge( your shoulders and accept that your data was not worth that much anyway.t   Antonio.     -- @   ---------------e- Antonio Carlini             arcarlini@iee.orgh   ------------------------------  % Date: Sat, 13 Jul 2002 20:31:01 +0100-E From: Jamie Stallwood <this.no.work.try.something.else@project76.net> F Subject: Re: VS3100/38: Infinitely looping bugchecks at boot (VMS 552)8 Message-ID: <rrv0ju43fbir3fcglglqceb0eh9e37dc7p@4ax.com>  3 On Sat, 13 Jul 2002 18:26:00 +0100, Jamie Stallwoodd6 <this.no.work.try.something.else@project76.net> wrote:   >Hi, >dG >My vaxstation has just decided to go round and round in a loop when ito" >boots, which is a bit of a pain.  >iF >The screen log is attached if anyone has any clues to help me, and it >would be most appreciated.  >. >Jamie Stallwood >   E I have since discovered this only occurs if I with the network on thetC AUI terminator. If I boot with it on the unterminated 10Base2, it'sm5 fine. Has anyone seen this happen on the 3100 before?    ------------------------------  # Date: Sun, 14 Jul 2002 03:34:55 GMTf, From: "Kenneth Farmer" <kfarmer@openvms.org>- Subject: Who said Carly doesn't like OpenVMS? = Message-ID: <zj6Y8.13824$Sb3.536969@twister.southeast.rr.com>y  ' Who said Carly doesn't like OpenVMS? :)l  9 http://www.openvms.org/stories.php?story=02/07/13/7134900n   --   Kenneth Farmer http://www.OpenVMS.org   ------------------------------   End of INFO-VAX 2002.384 ************************