0 INFO-VAX	Tue, 30 Jan 2001	Volume 2001 : Issue 59      Contents:# Re: All microsoft web sites offline # Re: All microsoft web sites offline  Create a random number Re: Create a random number DCL-E-NOCMDPROC  Re: DCL-E-NOCMDPROC  Re: DCL-E-NOCMDPROC  Re: DCL-E-NOCMDPROC  DEC DBMS Schema Mismatch Errors # Re: DEC DBMS Schema Mismatch Errors # Re: DEC DBMS Schema Mismatch Errors # Re: DEC DBMS Schema Mismatch Errors # Re: DEC DBMS Schema Mismatch Errors  defragment products " FAST I/O - IO$_SETUP / IO$_PERFORM Re: GZIP enhancements  Re: GZIP enhancements  Re: Himalaya / Alpha / VMS/ Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours / Re: It's the end for VMS and other Wild Rumours $ Re: It's the end for VMS get off now MicroVAX 3100 / 40 question ??? # Re: MicroVAX 3100 / 40 question ??? # Re: MicroVAX 3100 / 40 question ??? . Re: Midrange I/O with VMS WAS: DS20 - Slow I/O. Re: Midrange I/O with VMS WAS: DS20 - Slow I/O. Re: Midrange I/O with VMS WAS: DS20 - Slow I/O. RE: Midrange I/O with VMS WAS: DS20 - Slow I/OK Re: Monitoring external temperature (was Re: Monitoring external temprature  Monitoring external temprature4 Re: No technical computing, was: expanding the niche4 RE: No technical computing, was: expanding the niche4 Re: No technical computing, was: expanding the niche Pathworks and Windows ME Re: Pathworks and Windows ME RE: Pathworks and Windows ME& recv causing an application to crash ?- Re: Relative merits of LPD vx TELNET printing  Re: samba 2.0.6 > Re: TQK70 controller in serpentine config, q-bus cfg questions> Re: TQK70 controller in serpentine config, q-bus cfg questions> Re: TQK70 controller in serpentine config, q-bus cfg questionsL VAX 6310, VAX 6320, HSC50 and a couple of SA600's - how to plug'em together?$ very sloooooow Glasfibre performance( Re: very sloooooow Glasfibre performance( RE: very sloooooow Glasfibre performance( Re: very sloooooow Glasfibre performance! Re: VMS actually mentioned but... ! RE: VMS actually mentioned but... ! Re: VMS actually mentioned but... ! Re: VMS actually mentioned but...  Volume Shadowing: 1, fooguy:0 ! Re: Volume Shadowing: 1, fooguy:0 ! Re: Volume Shadowing: 1, fooguy:0 ! Re: Volume Shadowing: 1, fooguy:0 ! Re: Volume Shadowing: 1, fooguy:0 ! Re: Volume Shadowing: 1, fooguy:0 ! Re: Volume Shadowing: 1, fooguy:0 ! Re: Volume Shadowing: 1, fooguy:0 ! Re: Volume Shadowing: 1, fooguy:0 # Re: Volume Shadowing: 131, fooguy:1 # Re: Volume Shadowing: 131, fooguy:1  Re: What succeeds TSM?B Re: Whereabouts of Ranalyzer shareware linked for alpha processor?* Re: [VIM] VMS version doesn't do logicals?  F ----------------------------------------------------------------------  # Date: Tue, 30 Jan 2001 01:29:32 GMT / From: StevenU@POBoxes.com (Steven P. Underwood) , Subject: Re: All microsoft web sites offline2 Message-ID: <3a7617e4.247340179@news.telocity.com>  3 On Mon, 29 Jan 2001 10:52:24 +0000, andrew harrison ! <andrew.nospam@uk.sun.com> wrote:    > 1 >They were only able to respond to 3% of the page 7 >requests coming in instead of the "normal" 97%. As an  5 >avid MS web site surfer (well I have downloaded IE5  6 >a couple of times for UNIX) my fond recollection was 6 >slightly lower than 97% but perhaps they were having  >an abnomal day then as well.  >  >--  >Andrew Harrison >Enterprise IT Architect  F I have always suspected them of "playing" with any non IE web browser.E Anytime I try to go to their site without IE, I tend to lose 5-10% of D the pages I try to access, while simultaneously I have 0% loss using IE.    Steve  Steven P. Underwood,DNRC Whitinsville,MA  StevenU@POBoxes.com    ------------------------------  % Date: Mon, 29 Jan 2001 18:22:47 -0800 ! From: Shane.F.Smith@Healthnet.com , Subject: Re: All microsoft web sites offlineD Message-ID: <OF267706A9.9617F735-ON882569E4.000CDA31@foundation.com>  K I think that's quite likely. It's similar to something I used to encounter.   K It used to be that whenever I had to re-image a machine, I'd have to use IE J to go and download Netscape again. Every time I tried to download from theH Netscape site, IE would blow up near the end. Every time I used a mirrorG that didn't have "netscape" in the domain name, it worked perfectly. It 0 wasn't even intermittant, it was /every/ /time/.   Shane           C StevenU@POBoxes.com (Steven P. Underwood) on 01/29/2001 05:29:32 PM   % Please respond to StevenU@POBoxes.com    To:   Info-VAX@Mvb.Saic.Com  cc:   - Subject:  Re: All microsoft web sites offline     3 On Mon, 29 Jan 2001 10:52:24 +0000, andrew harrison ! <andrew.nospam@uk.sun.com> wrote:    > 1 >They were only able to respond to 3% of the page 6 >requests coming in instead of the "normal" 97%. As an4 >avid MS web site surfer (well I have downloaded IE55 >a couple of times for UNIX) my fond recollection was 5 >slightly lower than 97% but perhaps they were having  >an abnomal day then as well.  >  >--  >Andrew Harrison >Enterprise IT Architect  F I have always suspected them of "playing" with any non IE web browser.E Anytime I try to go to their site without IE, I tend to lose 5-10% of D the pages I try to access, while simultaneously I have 0% loss using IE.    Steve  Steven P. Underwood,DNRC Whitinsville,MA  StevenU@POBoxes.com    ------------------------------  # Date: Tue, 30 Jan 2001 04:18:07 GMT - From: "Dave Pampreen" <davepampreen@home.com>  Subject: Create a random number > Message-ID: <3grd6.264318$hD4.63894754@news1.rdc1.mi.home.com>  I Does anyone have a program which creates random passwords?   I'd prefer a = Fortran example, but I can convert it from most any language.    Dave   ------------------------------  % Date: Tue, 30 Jan 2001 00:05:54 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> # Subject: Re: Create a random number , Message-ID: <3A764BAA.C3B14457@videotron.ca>   Dave Pampreen wrote: > K > Does anyone have a program which creates random passwords?   I'd prefer a ? > Fortran example, but I can convert it from most any language.    $length = 5  $ALLIN1/NOINIT( GENERATE_PASSWORDS(6,CLI$pwd1,CLI$pron1) EXITG $Write sys$output "Password is=>''PWD1' and is pronounced as=>''PRON1'"   I You can have generate_PASSWORDS give you up to 5 choices by adding 4 more / pairs of password-pronounciation arguments (eg: > GENERATE_PASSWORDS(6,CLI$PWD1,CLI$PRON1,CLI$PWD2,CLI$PRO2 etc)  K I beleive that there is probably a system service which ALLIN1 uses to give L that function since the generated passwords are of the same type as those in authorize.    A The first parameter is the desired length of password (up to 16).    ------------------------------  % Date: Mon, 29 Jan 2001 19:54:09 +0100 * From: "Paul Janssen" <intrinsic@skynet.be> Subject: DCL-E-NOCMDPROC* Message-ID: <954e92$8ih$1@news0.skynet.be>  G Is there somebody who can shed some light on the DCL-E-NOCMDPROC, error @ opening captive command procedure - access denied error message.  K This message is part of the interactive login for the tcpip$ftp account, as > I initially could not connect to this system remotely via FTP.  J When removing the RESTRICED flag for this user I can log in and FTP works.K However another system was configured the same and here everything works as * foreseen, even with the RESTRICED flag on.   Strange... or not if you know.   ------------------------------  % Date: Mon, 29 Jan 2001 14:26:03 -0500 - From: "Peter Weaver" <peter.weaver@stelco.ca>  Subject: Re: DCL-E-NOCMDPROC4 Message-ID: <Ctjd6.124571$Z2.1565220@nnrp1.uunet.ca>  5 "Paul Janssen" <intrinsic@skynet.be> wrote in message $ news:954e92$8ih$1@news0.skynet.be...C > Is there somebody who can shed some light on the DCL-E-NOCMDPROC,  error B > opening captive command procedure - access denied error message.  F This means that either the account's LGICMD is missing, or the account5 can not Execute the LGICMD. It could also be that the + SYS$MANAGER:SYLOGIN.COM is not set for W:E.    --   RULES OF THE AIR   ----------------- =  #20. Good judgment comes from experience. Unfortunately, the 1       experience usually comes from bad judgment.    ------------------------------  # Date: Mon, 29 Jan 2001 19:50:35 GMT * From: Alan E. Feldman <alan48@my-deja.com> Subject: Re: DCL-E-NOCMDPROC) Message-ID: <954hi7$boj$1@nnrp1.deja.com>   * In article <954e92$8ih$1@news0.skynet.be>,-   "Paul Janssen" <intrinsic@skynet.be> wrote: C > Is there somebody who can shed some light on the DCL-E-NOCMDPROC,  error B > opening captive command procedure - access denied error message. >    From HELP/MESS NOCMDPROC:   D  NOCMDPROC,  error opening captive command procedure - access denied  7   Facility:     CLI, Command Language Interpreter (DCL)   D   Explanation:  When you attempted to log in, you failed because you have aD                 captive account and DCL received an error during the login.H                 For example, DCL could not find your LOGIN.COM file. YouB                 may also have incorrect protection on the system's SYLOGIN G                 file (SYS$MANAGER:SYLOGIN.COM or /SYSTEM/EXEC logical).  The E                 system's SYLOGIN file must be protected with at least  WORLD:E F                 access to the file and the directory that contains it.  (   User Action:  See your system manager.    A > This message is part of the interactive login for the tcpip$ftp  account, as @ > I initially could not connect to this system remotely via FTP.  F I am not familiar with TCPIP$FTP account. Is this a special account in	 some way?    > E > When removing the RESTRICED flag for this user I can log in and FTP  works.D > However another system was configured the same and here everything works as, > foreseen, even with the RESTRICED flag on. >   , Check the protections on all involved files.    > Strange... or not if you know. >  >    --F NOTE: If you wish to e-mail me, please do NOT use the deja address. ItD is not a valid address. Instead, use the address below, removing the long wrong part first. Thanks.    Disclaimer: JMHO Alan E. Feldman  &-)( afeldman@gfigroup.ButItSaidItPrinted.com     Sent via Deja.com  http://www.deja.com/   ------------------------------  # Date: Mon, 29 Jan 2001 20:27:21 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: DCL-E-NOCMDPROC7 Message-ID: <Jmkd6.178$cu.1398@gazette.loc1.tandem.com>   W In article <954e92$8ih$1@news0.skynet.be>, "Paul Janssen" <intrinsic@skynet.be> writes: H :Is there somebody who can shed some light on the DCL-E-NOCMDPROC, errorA :opening captive command procedure - access denied error message.   K   Probably a protected user- or system-wide login, or a protected directory .   containing the user- or system-wide login...      From HELP/MESSAGE NOCMDPROC...  C NOCMDPROC,  error opening captive command procedure - access denied   7   Facility:     CLI, Command Language Interpreter (DCL)   K   Explanation:  When you attempted to log in, you failed because you have a K                 captive account and DCL received an error during the login. H                 For example, DCL could not find your LOGIN.COM file. YouJ                 may also have incorrect protection on the system's SYLOGINK                 file (SYS$MANAGER:SYLOGIN.COM or /SYSTEM/EXEC logical). The M                 system's SYLOGIN file must be protected with at least WORLD:E F                 access to the file and the directory that contains it.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Mon, 29 Jan 2001 15:45:45 -0700 + From: "Barry Treahy, Jr." <treahy@mmaz.com> ( Subject: DEC DBMS Schema Mismatch Errors( Message-ID: <3A75F299.F3A973EA@mmaz.com>  F I know that this is probably not the best place for this question, butH I'm open to recommendations as we have been running the old DEC DBMS 4.3/ for eons without Compaq and now Oracle support.   F I have a situation where say nine out of ten times a program runs justG fine, but once in a blue moon it complains with a SSVERSION error which F is basically a mismatch of the schemas used to compile the program andF the actual schema in the database.  I'm at a loss as to why it chokes, any ideas??    Barry    --  ? Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO   A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028    ------------------------------  % Date: Mon, 29 Jan 2001 23:25:21 +0000 % From: Alan Greig <a.greig@virgin.net> , Subject: Re: DEC DBMS Schema Mismatch Errors* Message-ID: <3A75FBE0.6D2083CE@virgin.net>   "Barry Treahy, Jr." wrote:  H > I know that this is probably not the best place for this question, butJ > I'm open to recommendations as we have been running the old DEC DBMS 4.31 > for eons without Compaq and now Oracle support.  > H > I have a situation where say nine out of ten times a program runs justI > fine, but once in a blue moon it complains with a SSVERSION error which H > is basically a mismatch of the schemas used to compile the program andH > the actual schema in the database.  I'm at a loss as to why it chokes,
 > any ideas??  >   I I can only suggest the obvious and you've likely already checked but here  goes anyway:  * Does a DBO/VERIFY complete with no errors?J Could there possibly be a spurious logical pointing at a different version" of the database from time to time?H Are you able to do a database load/unload at a convenient time to see if
 things alter? = Is it only one program accessing the database or any program?   I Sorry but it's the best I can think off. I'll try and see if current DBMS  docs provide any clue tomorrow.      > Barry  >  > -- > A > Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO  > C > E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028    --
 Alan Greig   ------------------------------  % Date: Mon, 29 Jan 2001 17:12:12 -0700 + From: "Barry Treahy, Jr." <treahy@mmaz.com> , Subject: Re: DEC DBMS Schema Mismatch Errors( Message-ID: <3A7606DC.290F06D8@mmaz.com>   Alan Greig wrote:    > "Barry Treahy, Jr." wrote: >  > K > I can only suggest the obvious and you've likely already checked but here  > goes anyway: > , > Does a DBO/VERIFY complete with no errors?  ) Clean, other than space management pages.      > L > Could there possibly be a spurious logical pointing at a different version$ > of the database from time to time?  G I'm going back to my build system and double checking everything there,  specifically the CDD.    > J > Are you able to do a database load/unload at a convenient time to see if > things alter?   E I'll do this as a last resort, it is an extremely lengthly process...    > ? > Is it only one program accessing the database or any program?   L Yes, and no... Of all of the MANMAN modules we use, it is only three utilityL commands used for processing and posting cycle count activity but what is so" odd, is that most days it works...     > K > Sorry but it's the best I can think off. I'll try and see if current DBMS ! > docs provide any clue tomorrow.   
 Thanks....   Barry    --  ? Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO   A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028X   ------------------------------  % Date: Tue, 30 Jan 2001 00:59:55 +0000 % From: Alan Greig <a.greig@virgin.net>e, Subject: Re: DEC DBMS Schema Mismatch Errors* Message-ID: <3A76120B.A1C8F3FF@virgin.net>   "Barry Treahy, Jr." wrote:  L Yes, and no... Of all of the MANMAN modules we use, it is only three utility  N > commands used for processing and posting cycle count activity but what is so$ > odd, is that most days it works... >i  L I have actually seen bizarre MANMAN problems that might be related. In a fewM cases there are Fortran array bound violations even within current MANMAN. BytN default MANMAN is not compiled with /check=bounds. For some reason this codingP error seems to occur mostly in Dataport modules. Virtually any sort of crash canO happen with this type of random over-writing but usually it doesn't. We had one O utility that crashed only about once or twice a year then ran fine for the nextWP year or so. I gave up on CA solving it and delved into the Fortran. When I foundJ the problem and suggested to support that they at least compile field test8 MANMAN with /check=bounds they said "what does that do"!  O Not sure of your compiler version but it might be worth a try of a recompile ofWM the modules and common subroutines with /debug and/check switches if all elsedO fails. However it does sound as if you are falling very early in the code if it ; can't even open the database making corruption less likely.t   >d > >VM > > Sorry but it's the best I can think off. I'll try and see if current DBMSe# > > docs provide any clue tomorrow.  >o > Thanks.... >e > BarryV >a > -- >WA > Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO  >sC > E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028t   --
 Alan Greig   ------------------------------  # Date: Tue, 30 Jan 2001 01:26:28 GMTV- From: "Dave Pampreen" <davepampreen@home.com>t, Subject: Re: DEC DBMS Schema Mismatch Errors> Message-ID: <8Lod6.264231$hD4.63784717@news1.rdc1.mi.home.com>  2 Which MANMAN commands and which version of MANMAN?  H As far as CA helping, they lost Alison, Kevin and John late last year soK their DEC support side sucks.  From what I gather, the HP side is trying top pick up the slack.  J Are you part of CAMUS?  If so, send it to the listserv there and you might get some help.   Dave    6 "Barry Treahy, Jr." <treahy@mmaz.com> wrote in message" news:3A75F299.F3A973EA@mmaz.com...H > I know that this is probably not the best place for this question, butJ > I'm open to recommendations as we have been running the old DEC DBMS 4.31 > for eons without Compaq and now Oracle support.n >,H > I have a situation where say nine out of ten times a program runs justI > fine, but once in a blue moon it complains with a SSVERSION error which'H > is basically a mismatch of the schemas used to compile the program andH > the actual schema in the database.  I'm at a loss as to why it chokes,
 > any ideas??e >r > Barry  >  > -- >lA > Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIOc >lC > E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028o >g >,   ------------------------------   Date: 30 Jan 2001 00:12:53 GMT' From: "Jim Strehlow" <jims@data911.com>f Subject: defragment products0 Message-ID: <9550u5$mfk@dispatch.concentric.net>  8 Please let me know which disk defragmenter you are using on OpenVMS Alpha version 7x.  I Did you choose it after comparing it against Compaq/Raxco/Executive/otherg brand?" Does one work better than another:  - on a clusterR  - on shadow setsl  - on stripe setsc  - for organizing free space@  - for moving sections of a highly fragmented file to free space(    thus reducing the number of fragmentsH   (i.e. can you target the software to best optimize one specific file?)   etc.  
 Thank you.  < Larger disks and larger tape drives make the $BACKUP/RESTORE option less appealing.   Jim Strehlow, jims@data911.com System Manager   ------------------------------  # Date: Mon, 29 Jan 2001 22:40:10 GMTo' From: moi_is_me <moi_is_me@my-deja.com>t+ Subject: FAST I/O - IO$_SETUP / IO$_PERFORMb) Message-ID: <954rg9$lko$1@nnrp1.deja.com>   
 Greetings,?    When setting up FAST i/o, you have to call sys$io_setup, and%7    indicate whether you want an AST completion routine.e  B    If you do, but then later decide to invoke SYS$IO_PERFORMW, the<    synchronous version of SYS$IO_PERFORM, the AST completion    routine is STILL invoked.  ?    This seems a little odd to me. I would have called the asyncs2    version if i had wanted notification via an AST  D    Anyone else tried this ... is this a feature/expected behaviour ?      TIA   -Pierre-     Sent via Deja.com  http://www.deja.com/   ------------------------------   Date: 29 Jan 2001 20:00:32 GMT4 From: lewis_nospam@wheels.mitre.org (Keith A. Lewis) Subject: Re: GZIP enhancements( Message-ID: <954i50$h1g$1@top.mitre.org>   malmberg@eisner.decus.org (John E. Malmberg) writes in article <TNV+JF7h+6In@eisner.decus.org> dated 11 Jan 2001 22:33:34 -0500:N >Even though GZIP compresses better than INFO-ZIP, INFO-ZIP's abilty to handleM >directory trees makes it the more popular choice on OpenVMS, and will remainaA >so even if GZIP learns how to preserve some RMS file attributes.e  H We use gzip here on files other than stream_lf.  A co-worker of mine hasI written two utilities (SQUEEZE.COM and UNSQUEEZE.COM) to save and restore G the RMS file attributes in a separate file while gzipping or gunzippingeK them.  The second file is actually a command procedure which can be used tonH decompress the file if UNSQUEEZE.COM is not available on the destination machine.  L Does anybody want the utilities?  Should I post them here?  The total length is about 300 lines.u  3 --Keith Lewis              klewis_no_spam@mitre.orgr PGP key available.         a> The above may not (yet) represent the opinions of my employer.   ------------------------------  # Date: Mon, 29 Jan 2001 20:52:35 GMTc2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: GZIP enhancements7 Message-ID: <nKkd6.180$cu.1390@gazette.loc1.tandem.com>e  _ In article <954i50$h1g$1@top.mitre.org>, lewis_nospam@wheels.mitre.org (Keith A. Lewis) writes:tM :Does anybody want the utilities?  Should I post them here?  The total length- :is about 300 lines.  M   Send them along and I'll add them to the next Freeware -- since I've copies M   of GZIP and INFO-ZIP queued for the next freeware, I can drop these two DCL1.   procedures into the appropriate directory...  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Mon, 29 Jan 2001 21:43:48 +0000t% From: Alan Greig <a.greig@virgin.net>n# Subject: Re: Himalaya / Alpha / VMS7* Message-ID: <3A75E414.AEE5884E@virgin.net>   Hoff Hoffman wrote:o  Z > In article <3A735EB9.9BEE165D@infopuls.com>, Christof Brass <brass@infopuls.com> writes: >m  I >   Himalaya will be moving to Alpha, with the implementation of lockstepI0 >   in the next Alpha microprocessor core (EV7). >s  X What did Winkler mean when he failed to mention this as the future for Himalaya and saidW that Tandem/Himalaya technology was about to be implemented on Intel? Was he just plainPV wrong. If so where is the retraction? I take it you are not denying that he said this?   --
 ALan Greig   ------------------------------  # Date: Mon, 29 Jan 2001 19:01:28 GMTr- From: Terry C Shannon <shannon@world.std.com> 8 Subject: Re: It's the end for VMS and other Wild RumoursD Message-ID: <Pine.SGI.4.21.0101291316370.28218-100000@world.std.com>  * On Sun, 28 Jan 2001, Christof Brass wrote:   > Jordan Henderson wrote:NQ > > >> omission (no references to VMS) than a death warrant. Windoze is not readyh' > > >> for prime time in the VMS space,o > > >nN > > >Bot according to the numbers published. I beleive NSK made over a billionP > > >while VMS just under a billion. (although you'd have to verify this). I canS > > >beleive VMS has improved. But if it did turn around, you'd think that Capellasi% > > >would have proudly mentioned it.e  D NSK revenue for the YEAR might be $1B, VMS certainly is higher on an annualized basis.  > > >E > > M > > Capellas does things for strategic reasons.  There may be joint marketing L > > with Compaq partners at work here (hint: M$), but that doesn't mean that > > VMS is dead. > > K > > As Terry pointed out, CPQ would be crazy not to support a business thatu* > > is so profitable and actually growing. > > K > > Look, you can read whatever you want into it, but I'm still optimistic. I > > There are some here who want to read everything in the negative.  How G > > long has there been whining and nashing of teeth over the fact thatII > > Gartner never has anything good to say about VMS?  Well, Gartner justeI > > issued a really good report on VMS, but now, there's something new toT, > > cry about, so you focus on that instead. > >   I The Death of VMS rumour is pure, unadulterated codswallop. This according D to Michael Capellas and Rich Marcello. If you are concerned, contact richard.marcello@compaq.comn  I Perpetuating a specious and unfounded rumour on comp.os.vms can only help9J competitors spread FUD. Such postings do nothing to incease he credibility" of the posters in the eyes of CPQ.   ------------------------------  % Date: Mon, 29 Jan 2001 16:20:12 -0500R' From: "Bill Todd" <billtodd@foo.mv.com>r8 Subject: Re: It's the end for VMS and other Wild Rumours( Message-ID: <954mi1$s67$1@pyrite.mv.net>  8 Terry C Shannon <shannon@world.std.com> wrote in message> news:Pine.SGI.4.21.0101291316370.28218-100000@world.std.com...   ...u  < > The Death of VMS rumour is pure, unadulterated codswallop.  E The truth or falsehood of that statement depends upon how one defines I 'death', which definition (especially when applied to non-living objects)  has a fair amount of latitude.  E Latin, for example is termed a 'dead' language, even though doubtlessYK millions of people can read (and some even write, and a few even speak) it.fG And then (for living beings) there's the concept of 'brain death':  the 5 vital functions may continue, but the house is empty.A  H By some measures, VMS could be considered to have been dead, or at leastE comatose, for years:   its market significance certainly bears littlenL resemblance to what it was, say, fifteen years ago.  By other measures, someL might consider VMS to be 'alive' until the last machine running it is turned off.  H In the context of the concerns of many (possibly most) of the interestedJ parties here, VMS's viability hinges on sufficient long-term promotion andC development to keep it competitive with other platforms and hence aeG reasonable choice for new as well as existing use.  Actually, to remainoH competitive overall it must maintain a significant lead in at least someD areas, given that it trails significantly in areas like familiarity,L application breadth, and ability to run on 'industry-standard' hardware (no,< I'm not advocating a port, but it *is* a possible drawback).  E By that criterion, VMS is dying and Compaq is not taking any steps to C reverse the situation.  Compaq has relegated VMS to a small list ofLE enumerated niches, but fails to promote it even in those - and has no J intention of promoting use of VMS in any situation that could be addressedH with any of its other platforms (the source of this last is private, butI absolutely authoritative:  paradoxically, the statement was *meant* to besL reassuring, which just shows how out-of-touch Compaq is with the aspirationsC of the VMS community represented here).  And VMS development is notLK advancing at a rate sufficient to keep others from catching up in the areas ( where VMS currently enjoys an advantage.  I Conversely, if you wish to define 'viable' as 'can be purchased and run',oL VMS will certainly be around for a long time (unless Alpha disappears).  ButH systems relegated to near-maintenance-mode tend to see dwindling use andB higher costs for those who hang on - whether they're sold off to a sub-vendor or not.  L So for those of us who believe that VMS *could* be revived if Compaq had theK will, the demonstrable *lack* of such will relegates VMS to an increasinglyeK vegetative state, whether you choose to call that 'death' or not.  And whenyI you compare Compaq's treatment of VMS with IBM's treatment of S/390, it's-H easy to see the potential that is being squandered for no understandable reason.#    This according F > to Michael Capellas and Rich Marcello. If you are concerned, contact > richard.marcello@compaq.comt  J And I'm certain he will reassure you:  it's his job.  But unfortunately heH lacks the authority to follow through on his reassurances - and CapellasE does not appear interested in giving him the authority it would take.n   >hK > Perpetuating a specious and unfounded rumour on comp.os.vms can only helpeL > competitors spread FUD. Such postings do nothing to incease he credibility$ > of the posters in the eyes of CPQ.  K All the evidence indicates that the eyes of CPQ are focused elsewhere.  AndtL strange as it may seem, I suspect that many here don't give a rat's arse howF 'credible' CPQ feels they are:  the issue is how credible *Compaq* is.   - bill   >c   ------------------------------  % Date: Mon, 29 Jan 2001 16:37:30 -05004% From: "John Vottero" <John@mvpsi.com>-8 Subject: Re: It's the end for VMS and other Wild Rumours/ Message-ID: <t7bom0h2d40i69@news.supernews.com>D  : "Terry C Shannon" <shannon@world.std.com> wrote in message> news:Pine.SGI.4.21.0101291316370.28218-100000@world.std.com... >b   [snip]  K > The Death of VMS rumour is pure, unadulterated codswallop. This accordingbF > to Michael Capellas and Rich Marcello. If you are concerned, contact > richard.marcello@compaq.comt >iK > Perpetuating a specious and unfounded rumour on comp.os.vms can only help L > competitors spread FUD. Such postings do nothing to incease he credibility$ > of the posters in the eyes of CPQ. >r  K Please send your compaints about rumours to Michael Capellas.  He's the one,A that started the rumour by pretending that OpenVMS doesn't exist!i   ------------------------------  % Date: Mon, 29 Jan 2001 21:37:44 +0000w% From: Alan Greig <a.greig@virgin.net> 8 Subject: Re: It's the end for VMS and other Wild Rumours* Message-ID: <3A75E2A8.BB303B14@virgin.net>  U Terry C Shannon wrote:Perpetuating a specious and unfounded rumour on comp.os.vms cane	 only helpc  L > competitors spread FUD. Such postings do nothing to incease he credibility$ > of the posters in the eyes of CPQ.  Q Terry, I did what I thought I had to do. I want to save VMS and I'd quite like todU install additional systems. I was prepared to accept that VMS had turned around until4O this week. Maybe I *am* paranoid but Capellas actually said: "Only the paranoid T survive" and "reasons to be cautious". As to my credibility with Compaq. Well CompaqV and DEC before them are the ones who told me that TOPS-20 and Alpha/NT were safe. JustT weeks before both were dropped. And you call into question my credibility with them. Sheesh!a  U It's this nonsense from Compaq that drives customers away to another vendor. You knoweR fine it has done it for years yet you persist in saying we should shut up and thatV every Compaq action that seems out of place is a cock-up. Ok, let's accept it is for aT moment. That's a perfectly acceptable way to run a company is it? And we should make> allowances for our suppliers incompetence. Sorry Terry but no.  W I do not want VMS to die either by ill-will or just incompetence; and I exclude the VMStU team from this as they do appear to be sending positive messages. But then so did theiA Alpha/NT team and the LCG DEC-10/DEC-20 group at DEC before them.N  V If I have a third Compaq/DEC operating system dropped on me I really will need my headR examined. Thus I make a noise when I see similar signs. Maybe I'm wrong but I feelS Compaq hint at dropping VMS every so often and just see how loud the cries are thise time.u  T I know how much it hurts to read this stuff because it hurts me to write it. It willW hurt me even more if it ever happens and I don't intend to let a supplier get away with2V a description of the future in which the supplier tells the customer just exactly whatN they are going to get in future like it or not. And Winkler in particular saidW Intel/Windows is where it's going whether we want it or not because that's where CompaqOW will drive it. In the end this may be the route we go down as well but it should be oursG decision. Very, very clearly some Compaq management do not accept this.   W Compaq bought into the enterprise game but quite clearly a large part of the management U believes all of this non Wintel stuff to be a distraction and are taking it apart bittW by bit. Perhaps Capellas could get some of his Wintel managers and VPs to actually talkeV to the VMS and Unix customers to get a better understanding. How would you feel if youS were a happy Tru-64 customer who had just heard that Compaq will drive W2K into thehU soft underbelly of Unix and then Windows64 Datacentre all the way up driving Unix outn" of even the largest installations?  W Nobody is asking for Compaq to pull away from Microsoft but they have to sell it on its Q own merits. Not attack their own products with it. Remember what happened to DEC.cN Or do you want to change your name to Shannon Knows Microsoft in 5 years time? --
 Alan Greig   ------------------------------  # Date: Mon, 29 Jan 2001 21:43:55 GMTu- From: Terry C Shannon <shannon@world.std.com> 8 Subject: Re: It's the end for VMS and other Wild RumoursD Message-ID: <Pine.SGI.4.21.0101291638110.27090-100000@world.std.com>  % On Mon, 29 Jan 2001, Bill Todd wrote:s   > : > Terry C Shannon <shannon@world.std.com> wrote in message@ > news:Pine.SGI.4.21.0101291316370.28218-100000@world.std.com... >  > ...  > > > > The Death of VMS rumour is pure, unadulterated codswallop. > G > The truth or falsehood of that statement depends upon how one defines K > 'death', which definition (especially when applied to non-living objects)R  > has a fair amount of latitude. >   J I would define "dead" as no longer receiving investment/no longer bringing in revenue.h  M > All the evidence indicates that the eyes of CPQ are focused elsewhere.  AndlN > strange as it may seem, I suspect that many here don't give a rat's arse howH > 'credible' CPQ feels they are:  the issue is how credible *Compaq* is. >   D In whose eyes? In the eyes of customers who shell out the big bucks,I absolutely. And based on CPQ's 4FQ numbers (Strecker's Folly aside) and a.G modest growth in the VMS business, credibility is not one of the firm'so biggest problems right now.n   ------------------------------    Date: 29 Jan 2001 14:43:40 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett)s8 Subject: Re: It's the end for VMS and other Wild Rumours, Message-ID: <bQnXOLZt9t2g@malvm1.mala.bc.ca>  ) In article <954mi1$s67$1@pyrite.mv.net>, l-     "Bill Todd" <billtodd@foo.mv.com> writes:e  E > reverse the situation.  Compaq has relegated VMS to a small list ofhG > enumerated niches, but fails to promote it even in those - and has no L > intention of promoting use of VMS in any situation that could be addressedJ > with any of its other platforms (the source of this last is private, butK > absolutely authoritative:  paradoxically, the statement was *meant* to betN > reassuring, which just shows how out-of-touch Compaq is with the aspirations( > of the VMS community represented here)  K     This pretty much says it all, doesn't it. The number of situations that K can't be addressed by Unix or Windoze is small ( and shrinking ), so CompaqrK has signed the death warrant for VMS if this is truly their attitude - it'stK just a matter of time. It also pretty much rules out the possibility of anyeH 3rd parties porting their non-VMS apps to VMS, why bother if Compaq will5 only be promoting them on the other platforms anyway.   H    How anyone could expect a statement like that to be reassuring to the VMS community is beyond me.    ------------------------------  % Date: Tue, 30 Jan 2001 00:07:07 +0100-9 From: "Michel Herrscher Consultant" <michel@herrscher.fr> 8 Subject: Re: It's the end for VMS and other Wild Rumours% Message-ID: <954tbe$ku1$1@wanadoo.fr>D  A Terry C Shannon <shannon@world.std.com> a crit dans le message :S9 Pine.SGI.4.21.0101291316370.28218-100000@world.std.com...tK > Perpetuating a specious and unfounded rumour on comp.os.vms can only helpdL > competitors spread FUD. Such postings do nothing to incease he credibility$ > of the posters in the eyes of CPQ. >o  + Seeing this debate from Europe is terrific.l+ Terry's opinion seems the right one for me.   P Has all the discussion been started from a VMS supporter or a competitor who has# only this to compete against OVMS ?r   Will we ever known ?   -- Michel HERRSCHER Consultant- mhc@herrscher.fr7 Tl : +33 (0)450 870 912       Fax : +33 (0)450 871 741m Gsm : +33 (0)609 044 711   ------------------------------  % Date: Mon, 29 Jan 2001 18:20:55 -0500r' From: "Bill Todd" <billtodd@foo.mv.com> 8 Subject: Re: It's the end for VMS and other Wild Rumours( Message-ID: <954tkc$5lt$1@pyrite.mv.net>  < Malcolm Dunnett <nothome@spammers.are.scum> wrote in message& news:bQnXOLZt9t2g@malvm1.mala.bc.ca...* > In article <954mi1$s67$1@pyrite.mv.net>,/ >     "Bill Todd" <billtodd@foo.mv.com> writes:  >eG > > reverse the situation.  Compaq has relegated VMS to a small list of I > > enumerated niches, but fails to promote it even in those - and has noCD > > intention of promoting use of VMS in any situation that could be	 addressednL > > with any of its other platforms (the source of this last is private, butJ > > absolutely authoritative:  paradoxically, the statement was *meant* to beD > > reassuring, which just shows how out-of-touch Compaq is with the aspirations * > > of the VMS community represented here) >tH >     This pretty much says it all, doesn't it. The number of situations thatF > can't be addressed by Unix or Windoze is small ( and shrinking ), so CompaqH > has signed the death warrant for VMS if this is truly their attitude - it'sI > just a matter of time. It also pretty much rules out the possibility ofk any0J > 3rd parties porting their non-VMS apps to VMS, why bother if Compaq will7 > only be promoting them on the other platforms anyway.t >DJ >    How anyone could expect a statement like that to be reassuring to the > VMS community is beyond me.d  L Well, I wouldn't want to leave the impression that the author was completelyH 'round the bend:  what I paraphrased was included as what was apparentlyJ intended to be a minor caveat (or attempt to keep the discussion rooted inK reality) in a larger context that was *meant* to be reassuring.  But it didaI kind of leap out at me, and it seemed, aside or not, pretty unambiguous --L though I suspect that the author might re-phrase it differently now if given the opportunity...   - bill   ------------------------------  # Date: Mon, 29 Jan 2001 23:09:53 GMTf- From: Terry C Shannon <shannon@world.std.com>18 Subject: Re: It's the end for VMS and other Wild RumoursD Message-ID: <Pine.SGI.4.21.0101291757140.28563-100000@world.std.com>  & On Mon, 29 Jan 2001, Alan Greig wrote:   >  > W > Terry C Shannon wrote:Perpetuating a specious and unfounded rumour on comp.os.vms canf > only helpn > N > > competitors spread FUD. Such postings do nothing to incease he credibility& > > of the posters in the eyes of CPQ. > S > Terry, I did what I thought I had to do. I want to save VMS and I'd quite like to-W > install additional systems. I was prepared to accept that VMS had turned around untilaQ > this week. Maybe I *am* paranoid but Capellas actually said: "Only the paranoidsV > survive" and "reasons to be cautious". As to my credibility with Compaq. Well CompaqX > and DEC before them are the ones who told me that TOPS-20 and Alpha/NT were safe. JustV > weeks before both were dropped. And you call into question my credibility with them.	 > Sheesh!  >   < I am neither impugning your credibility nor besmirching yourI escutcheon. There have been a lot of postings in this newsgroup which aref> based on incomplete information and unsupportable conclusions.  G We all know that VMS marketing--at the levels above Rich Marcello--dotheH sucketh mightily. And VMS has been damned by faint praise for ages. That= said, the OS actually experienced single-digit growth in 4FQ.s  I Given the downturn in the Wintel market (I can't wait to see how Dell and G Gateway fare) and the fact that CPQ is finally stressing its enterprise 4 bragging rights, why on earth would they "kill" VMS?  E VMS contributes far more revenue, and very similar margins, than that  which is contributed by NSK.  6 A heck of a lot of WildFires go out the door with VMS.  2 CPQ has committed to providing COE support on VMS.  H Pulling the plug now would not only be the height of stupidity, it would be tantamount to suicide.   F The fact that customers continue to buy the OS indicates that CPQ (and& VMS) are still credible in their eyes.  F Many of the financial analysts who attended the conference last FridayI subscribe to my newsletter, and we frequently correspond. I have received G zero inquiries from these folks, or from my large end-user or IT vendorh- subscribers, about the alleged elegy for VMS.i  J Hence I believe an apparent act of omission has been blown way, way out of proportion.g   ------------------------------  % Date: Mon, 29 Jan 2001 18:29:46 -0500 ' From: "Bill Todd" <billtodd@foo.mv.com> 8 Subject: Re: It's the end for VMS and other Wild Rumours( Message-ID: <954u51$606$1@pyrite.mv.net>  8 Terry C Shannon <shannon@world.std.com> wrote in message> news:Pine.SGI.4.21.0101291638110.27090-100000@world.std.com... >m >a' > On Mon, 29 Jan 2001, Bill Todd wrote:8 >0 > >o< > > Terry C Shannon <shannon@world.std.com> wrote in messageB > > news:Pine.SGI.4.21.0101291316370.28218-100000@world.std.com... > >n > > ...o > > @ > > > The Death of VMS rumour is pure, unadulterated codswallop. > >iI > > The truth or falsehood of that statement depends upon how one definesdD > > 'death', which definition (especially when applied to non-living objects)" > > has a fair amount of latitude. > >  > L > I would define "dead" as no longer receiving investment/no longer bringing
 > in revenue.t  J So by that definition, VMS is as alive and well as RSX is.  And I wouldn'tK argue that this won't continue to be true.  But I don't think that's what ac# lot of people here are looking for.o   > J > > All the evidence indicates that the eyes of CPQ are focused elsewhere. AndeL > > strange as it may seem, I suspect that many here don't give a rat's arse how J > > 'credible' CPQ feels they are:  the issue is how credible *Compaq* is. > >  >gF > In whose eyes? In the eyes of customers who shell out the big bucks,K > absolutely. And based on CPQ's 4FQ numbers (Strecker's Folly aside) and aeI > modest growth in the VMS business, credibility is not one of the firm'ss > biggest problems right now.e  L The question is what the numbers would look like without the VMS revenue andK profit (plus some incalculable amount of additional revenue that is tied to E use of VMS but accrues to other Compaq systems):  that, or at least aoA significant percentage of that, is what's on the line for Compaq.   J Customers can be loyal, and customers can be stupid, but at some point theL light dawns, and if that's sooner rather than later, the situation *will* be" a problem, even if it's not today.   - bill   ------------------------------  # Date: Mon, 29 Jan 2001 23:25:13 GMTb- From: Terry C Shannon <shannon@world.std.com>h8 Subject: Re: It's the end for VMS and other Wild RumoursD Message-ID: <Pine.SGI.4.21.0101291818350.25203-100000@world.std.com>  7 On Tue, 30 Jan 2001, Michel Herrscher Consultant wrote:n   >=20E > Terry C Shannon <shannon@world.std.com> a =E9crit dans le message : ; > Pine.SGI.4.21.0101291316370.28218-100000@world.std.com...dL > > Perpetuating a specious and unfounded rumour on comp.os.vms can only he= lpL > > competitors spread FUD. Such postings do nothing to incease he credibil= ityo& > > of the posters in the eyes of CPQ. > >s >=20- > Seeing this debate from Europe is terrific.a- > Terry's opinion seems the right one for me.b >=20L > Has all the discussion been started from a VMS supporter or a competitor = who hasi% > only this to compete against OVMS ?e >=20 > Will we ever known ? >=20  A If you trace the origins of this thread, it's clearly the work oftI enthusiastic VMS supporters who believe VMS is being marginalized. I notetE with interest that our colleague Andrew Harrison has prudently stayed  out of this debate.a  H I doubt that we'll see Compaq issue a retraction. That would be too muchE like responding to "when did you stop beating your wife" allegations.o  J IMHO the best thing Compaq could do is announce a VMS Applications PortingJ Fund designed to attract more ISVs to the platform. Of course, that's just  my opinion and I could be wrong!   terry so   ------------------------------  % Date: Mon, 29 Jan 2001 23:31:32 +0000p% From: Alan Greig <a.greig@virgin.net>i8 Subject: Re: It's the end for VMS and other Wild Rumours* Message-ID: <3A75FD53.B5D65BA5@virgin.net>  " Michel Herrscher Consultant wrote:  C > Terry C Shannon <shannon@world.std.com> a crit dans le message :'; > Pine.SGI.4.21.0101291316370.28218-100000@world.std.com... M > > Perpetuating a specious and unfounded rumour on comp.os.vms can only help7N > > competitors spread FUD. Such postings do nothing to incease he credibility& > > of the posters in the eyes of CPQ. > >b >h- > Seeing this debate from Europe is terrific.n- > Terry's opinion seems the right one for me.s >nR > Has all the discussion been started from a VMS supporter or a competitor who has% > only this to compete against OVMS ?d >l  O Go and watch the teleconference then see who's view you think is right. Neither,T Terry nor yourself has yet seen it but both claim the views of myself and others whoQ have seen it or most of it are wrong. Just follow the link from www.compaq.com. ItO warn you it will take around three hours before the brief mention of VMS. Enjoye# listening to Winkler in particular.    >e > Will we ever known ? >   K Yes it was started by a competitor who wanted to spread FUD about VMS. That-J competitor, unfortunately, was Compaq at their annual analysts conference.   --
 Alan Greig   ------------------------------  % Date: Mon, 29 Jan 2001 18:51:23 -0500-' From: "Bill Todd" <billtodd@foo.mv.com>r8 Subject: Re: It's the end for VMS and other Wild Rumours( Message-ID: <954vdf$6rj$1@pyrite.mv.net>  8 Terry C Shannon <shannon@world.std.com> wrote in message> news:Pine.SGI.4.21.0101291818350.25203-100000@world.std.com...   ...a  J IMHO the best thing Compaq could do is announce a VMS Applications Porting3 Fund designed to attract more ISVs to the platform.a  L [Damn IE didn't right-arrow the original - one of these days I'll figure out why.]n  H Er, didn't we hear something about ISV financial support 'way back when?F Does your statement mean that it didn't happen (which is getting to beI somewhat of a refrain), or are you proposing something to add to existing  support (whatever it is)?m   - bill    Of course, that's just   my opinion and I could be wrong!   terry ss   ------------------------------  % Date: Mon, 29 Jan 2001 23:46:05 +0000 % From: Alan Greig <a.greig@virgin.net> 8 Subject: Re: It's the end for VMS and other Wild Rumours* Message-ID: <3A7600BC.5FFE61AC@virgin.net>   Terry C Shannon wrote:  ( > On Mon, 29 Jan 2001, Alan Greig wrote:K > Given the downturn in the Wintel market (I can't wait to see how Dell and I > Gateway fare) and the fact that CPQ is finally stressing its enterprisel6 > bragging rights, why on earth would they "kill" VMS? >g  V Because Winkler said that W2K will inevitably win the war against Unix. Let alone VMS.[ Compaq will make sure this happens he stressed. I wouldn't be blowing my top if they hadn'tlY been so direct at saying this. Perhaps Terry you'll have steam coming out of your ears as W well if you actually get to hear it. I know what I say sounds incredulous but it reallyi	 happened.   X Why did they feel confident about Wintel? Because Winkler added that's where Compaq cameZ from, that's where it always would be and that was its most important market. AdditionallyT Compaq had concluded that Wintel would eventually win the war entirely and it was inX Compaq's interest to help this happen as quickly as possible. You shouldn't be surprised\ about this, said Winkler, because Compaq's prime mission is to help W64 Datacentre take overY the enterprise. Compaq's most important partners are Intel, Microsoft and Oracle. Nothingy will be done to upset them.n  K >Hence I believe an apparent act of omission has been blown way, way out ofr  
 > proportion.f  [ And without the context of the rest of the presentation I might be inclined to believe thisp too.   --
 Alan Greig   ------------------------------  % Date: Mon, 29 Jan 2001 19:07:12 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca>n8 Subject: Re: It's the end for VMS and other Wild Rumours, Message-ID: <3A7605A9.11289684@videotron.ca>   Terry C Shannon wrote:F > NSK revenue for the YEAR might be $1B, VMS certainly is higher on an > annualized basis.h  M So, if VMS is bigger than NSK, why did Capellas ignore VMS and focus on otherf
 products ?  K > The Death of VMS rumour is pure, unadulterated codswallop. This accordingm( > to Michael Capellas and Rich Marcello.  M Actions speak louder than words. It's the actions we are not seeing. Marcello I can say all he wants, but if he doesn't have the support of Compaq and ifrM Compaq doesn't include VMS as part of its PR and advertising, then there is a  lack of credibility.  L And why the 4 letter word does Capellas justify giving away the technologiesL which give Compaq's own product a technological edge ? Compaq may say white, but it acts black.  L > competitors spread FUD. Such postings do nothing to incease he credibility$ > of the posters in the eyes of CPQ.  L Is Compaq like Perrier or Tylenol when it comes to PR gaffes ? It may chooseK to dismiss customers telling them that Compaq has a serious PR problem withi< regards to VMS, but that is at Compaq's own risk and perils.  E Also note that letters started to fly to Capellas on the night of therK financial results being presented. So if on the subsequent presentation, he M provided a token VMS presence, it may have been as a result of those letters.nG The fact is that during the financial results presentation, VMS was noth' mentioned, although the other OSs were.e   ------------------------------  % Date: Mon, 29 Jan 2001 19:21:18 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca>i8 Subject: Re: It's the end for VMS and other Wild Rumours, Message-ID: <3A7608F5.52A1E7D2@videotron.ca>   Terry C Shannon wrote:L > I would define "dead" as no longer receiving investment/no longer bringing
 > in revenue.o  ' How would you define "allowed to die" ?   F > In whose eyes? In the eyes of customers who shell out the big bucks,K > absolutely. And based on CPQ's 4FQ numbers (Strecker's Folly aside) and avI > modest growth in the VMS business, credibility is not one of the firm'sm > biggest problems right now.e    N Oh yeas, I forgot, compaq only cares about the few big remaining VMS customersN it has because it really wants everyone else to go to wintel instead. But withK only those few large customers left, they too will have to go away from VMSr because of lack of software.   ------------------------------  % Date: Mon, 29 Jan 2001 19:38:04 -0500i' From: "Bill Todd" <billtodd@foo.mv.com>y8 Subject: Re: It's the end for VMS and other Wild Rumours( Message-ID: <955253$8je$1@pyrite.mv.net>  8 Terry C Shannon <shannon@world.std.com> wrote in message> news:Pine.SGI.4.21.0101291757140.28563-100000@world.std.com...   ...   >  There have been a lot of postings in this newsgroup which are! > based on incomplete information   
 By necessity.n    > and unsupportable conclusions.  ; Incorrect.  Unprovable, perhaps, but eminently supportable.m   >wI > We all know that VMS marketing--at the levels above Rich Marcello--dothoJ > sucketh mightily. And VMS has been damned by faint praise for ages. That? > said, the OS actually experienced single-digit growth in 4FQ.l  I Simply stopping the traditional doses of poison (from the sales force anddF the corporate insistence on VMS anonymity) allowed this growth with noD active support.  Think what might have happened if Compaq had made aJ credible effort to turn around the perception that VMS was already defunctK or close to it, and actually *advertised* its strengths to markets that mayg3 be completely unaware of them (as most likely are).e   > K > Given the downturn in the Wintel market (I can't wait to see how Dell and I > Gateway fare) and the fact that CPQ is finally stressing its enterprise 6 > bragging rights, why on earth would they "kill" VMS?  J Well, most of the 'enterprise' bragging rights Compaq is stressing seem to@ be Wintel-based.  And virtually all of the rest are Himalaya- orL Tru64-based.  So it's not clear what relevance Compaq assigns to VMS in this area.e  F That said, I don't think too many people here think Compaq is activelyF trying to 'kill' VMS:  it seems much more likely that they're just notJ interested in taking substantial action to grow it, especially in any areaK where they might be able to sell one of their other platforms instead.  AndmK I'm not sure how much difference it may make to people here if VMS dies (ore; becomes unviable for them) from neglect rather than poison.m   >iG > VMS contributes far more revenue, and very similar margins, than thatd > which is contributed by NSK. >g8 > A heck of a lot of WildFires go out the door with VMS. >i4 > CPQ has committed to providing COE support on VMS. > J > Pulling the plug now would not only be the height of stupidity, it would > be tantamount to suicide.i  H See above.  It isn't that Compaq wants to pull the plug, but that CompaqJ wants to retain its existing VMS customer base (and perhaps even grow it aL bit) without increasing its commitment to something that would be reassuringH (in terms of on-going development, competitiveness, volume, advertising,K ...) to that base.  Nice for Compaq.  Not so nice for customers:  no wonderm they're worried.   >rH > The fact that customers continue to buy the OS indicates that CPQ (and( > VMS) are still credible in their eyes.  I The fact that they buy VMS indicates that VMS is credible.  The fact thatlL they worry as much as they do about that decision (and in many cases migrate= elsewhere when that's feasible) indicates that Compaq is not.t   >tH > Many of the financial analysts who attended the conference last FridayK > subscribe to my newsletter, and we frequently correspond. I have receivedcI > zero inquiries from these folks, or from my large end-user or IT vendoru/ > subscribers, about the alleged elegy for VMS.t  I That's because the analysts, unlike people in this forum, have likely notrH been fed stories about the VMS 'Renaissance' and rosy future:  they justK believe it's still dying off from neglect at the usual rate (and seem to ber right), so no news there.n   >dL > Hence I believe an apparent act of omission has been blown way, way out of
 > proportion..  I Only if one discounts the rosy predictions for VMS's future - based on aniC alleged change in corporate attitude toward VMS - and was thereforenI expecting the presentation to be the same old, same old stance - which ofp course it was.  L Conversely, if one *believed* the optimistic assertions in OpenVMS Times andL private communications, then the return to the same old, same old was indeed a shock.  I Capellas has had no lack of indication of how important external exposureuC (outside the existing base) is to the base's perception of Compaq'scG attitude.  He can't be unaware of it at this point - and still chose to.L ignore VMS in his presentation (save for a single late reference, it seems),I making it the lone omission among the systems Compaq provides directly orbF indirectly (well, I didn't listen to it, so I suppose it's possible heF omitted Linux as well - haven't seen a reference to it here).  And theD Winkler presentation appears to have been a rabid attack on anything' non-Wintel, which is hardly reassuring.t  J The reason people find the words (or lack of them) so important is becauseD they're all that's available to hang onto (except that they *aren't*J available, once again).  I personally agree with you:  it's *actions* thatE matter, and the lack of words simply reinforces the fact that there'scK nothing happening, in word *or* deed, to make one optimistic.  But the lackyF of even words arguably makes it far more likely that people can't look( forward to any positive actions, either.   - bill   ------------------------------   Date: 30 Jan 2001 00:47:41 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)8 Subject: Re: It's the end for VMS and other Wild Rumours, Message-ID: <9552vd$8dt@gap.cco.caltech.edu>  t In article <Pine.SGI.4.21.0101291316370.28218-100000@world.std.com>, Terry C Shannon <shannon@world.std.com> writes: >a< >The Death of VMS rumour is pure, unadulterated codswallop.   C I disagree.  (Maybe it depends upon your definition of dead.)  MC's J conference clearly described a Compaq future that did not include OpenVMS.G You'd have to read volumes between the lines to see anthing else in hisoH statements.  If he thought OpenVMS was important for Compaq for the longI run he would have said so.  He did not.  That tells everyone who's paying-I attention to get the heck off OpenVMS, and probably, to get the heck awaydK from Compaq.  I mean, he must appreciate how skittish the OpenVMS user basesK is, and his little talk probably just drove hundreds of millions of dollarsmD of marginal OpenVMS sales to IBM, HP, and maybe Sun.  Never mind theG technical advantages of OpenVMS, when the vendor is that tepid in their C support it's a good sign that you'd best look elsewhere.  Moreover,sI Compaq's whole Alpha strategy appears so hopelessly muddied that I'm none D too disposed to jump from OpenVMS onto Tru64, which would make senseH otherwise, because I have no desire to get nailed two years later by yetE another Compaq management decision to dump that as well.   I saw what-J happened to the WNT/Alpha, I'm living through the OpenVMS limbo, so why onH earth would I give Compaq management another opportunity to make my life miserable? b   >This accordingVE >to Michael Capellas and Rich Marcello. If you are concerned, contact' >richard.marcello@compaq.com  K MC has spoken, and "slow death" (at least) was his implicit message.  If he D did not intend to send that message then he has no grasp of VMS userL psychology.  RM is between a rock and a hard place - he appears to be tryingE hard but there's little evidence that MC is giving him any support.   G Which isn't to say that people shouldn't write to RM, but I suspect thelH answer you get back from him should be given as much weight as a similarK response from the head of the WNT/Alpha group, immediately pre-termination. I He can only say what he wants to do, not what MC is going to let him do. h- And MC is the only person whose vote counts. r  J >Perpetuating a specious and unfounded rumour on comp.os.vms can only helpK >competitors spread FUD. Such postings do nothing to incease he credibilitys# >of the posters in the eyes of CPQ.   M Nothing any competitor could say or do would speak as loudly as MC's silence.e   David Mathog mathog@seqaxp.bio.caltech.edud? Manager, sequence analysis facility, biology division, Caltech rJ **************************************************************************J *                                RIP VMS                                 *J **************************************************************************   ------------------------------  % Date: Mon, 29 Jan 2001 19:31:02 -060007 From: "David J. Dachtera" <djesys.nospam@earthlink.net> 8 Subject: Re: It's the end for VMS and other Wild Rumours- Message-ID: <3A761956.4AF5BE93@earthlink.net>-   Terry C Shannon wrote:K > The Death of VMS rumour is pure, unadulterated codswallop. This according F > to Michael Capellas and Rich Marcello. If you are concerned, contact > richard.marcello@compaq.come  % He doesn't answer my e-mails anymore.i  tK > Perpetuating a specious and unfounded rumour on comp.os.vms can only helpoL > competitors spread FUD. Such postings do nothing to incease he credibility$ > of the posters in the eyes of CPQ.  B Y'know, I saw in another thread where Bill Hymen repeated his post. (again) about writing to key people at Compaq.  . I read through it again and noticed something:  ! The key VMS people are in Nashua.t   Capellas is in Texas.c  F Now, I don't know how many of you have this problem, but my higher-upsG at work are in the next town over, less than 10 minutes away. They tendiH to live in a world very distant from the everyday operational aspects of@ IT. (They have also been assimilated and are deeply enamoured of Micro$hit.)d  G Somehow, we need to gain access to the Texas people so we can bring theiH woes of OpenVMS to their attention. If we don't, they'll just keep theirE Gatesan blinders on, and proceed merrily forward believing that M$ isg/ all that matters, just like what I see at work.t  ? The end of VMS may be much nearer for me than it is for Compaq.h  D I got into computers after working in a factory for some years in myE twenties and deciding that I needed more from my working career. I am0G once again upon that threshold, and am preparing to move on, perhaps to & my long time "love" - public speaking.   -- e David J. Dachterag dba DJE Systemsn http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.e   ------------------------------  % Date: Mon, 29 Jan 2001 20:40:53 -0500n- From: JF Mezei <jfmezei.spamnot@videotron.ca>o8 Subject: Re: It's the end for VMS and other Wild Rumours, Message-ID: <3A761B97.A4DB0905@videotron.ca>   Terry C Shannon wrote:I > We all know that VMS marketing--at the levels above Rich Marcello--doth J > sucketh mightily. And VMS has been damned by faint praise for ages. That? > said, the OS actually experienced single-digit growth in 4FQ.c  K And all the complaints made since the flawed financial conference call have S been exactly about how the flawed marketing is sending the wrong message about VMS.   K Ever since the ice storm in Montreal in 1998, the mere mention of "freezing>M rain" sends shivers down our spines, and people react quite strongly to power19 failures for fear that it might last a month like before.O  L Ever since Palmer, VMS customers get shivers down their spines whenever theyH hear mention of VMS technologies being given away to Microsoft and reactL strongly when VMS is omitted from official presentations where all the other OSs are featured.I  I Capellas has been made aware of those sensitivities, yet, he continues to F treat VMS as a ghost product. Are they are surprised when people reactN strongly because they forgot to include VMS in their corporate presentations ?  N Either they are incompetent, or their have some intentions which are different( from what Marcello is trying to acheive.  K > Given the downturn in the Wintel market (I can't wait to see how Dell andhI > Gateway fare) and the fact that CPQ is finally stressing its enterprise.6 > bragging rights, why on earth would they "kill" VMS?  I Downturn ? ther wintel servers *grew* by over 20% (not sure, but I recalloL 27%). In fact , they grew more than the niched business critical stuff (Unix and Himalaya).   G > VMS contributes far more revenue, and very similar margins, than thate > which is contributed by NSK.  K Can you corroborate this with facts ? We we all told to expect that VMS was N the driver for Compaq's profits and that with a downturn in PCs, VMS was bound to stand out. But it didn't. e  8 > A heck of a lot of WildFires go out the door with VMS.  > Why then was it not mentioned during the analysts conference ?  M And if Wildfires had done as good as predicted, how come they haven't reachedt? their 1 billion target yet and it won't be reached untl March ?h  4 > CPQ has committed to providing COE support on VMS.  L woopty doo. Making VMS better does nothing to increase its sales when CompaqJ continues to ignore its product which is already technologically superior.  J > Pulling the plug now would not only be the height of stupidity, it would > be tantamount to suicide.h  J Nobody said they'd be pulling the plug. They're just continuing the PalmerS policies of controlling VMS's downsizing to extinction. That is what it looks like.i  H > The fact that customers continue to buy the OS indicates that CPQ (and( > VMS) are still credible in their eyes.  L That is because last year, there were an attempt at marketing with the ballsL and posters. But now that Compaq has reverted to the treatment of silence, I. am not sure that the growth will be sustained.  H > Many of the financial analysts who attended the conference last FridayK > subscribe to my newsletter, and we frequently correspond. I have receivedaI > zero inquiries from these folks, or from my large end-user or IT vendor,/ > subscribers, about the alleged elegy for VMS.   N Because they have already written off VMS so they don't notice it missing from that PC company called Compaq.   ------------------------------  % Date: Mon, 29 Jan 2001 20:54:12 -0500a% From: "John Vottero" <John@mvpsi.com>e8 Subject: Re: It's the end for VMS and other Wild Rumours/ Message-ID: <t7c7nin3slab69@news.supernews.com>a  : "Terry C Shannon" <shannon@world.std.com> wrote in message> news:Pine.SGI.4.21.0101291757140.28563-100000@world.std.com... >o   [snip]   >eK > Given the downturn in the Wintel market (I can't wait to see how Dell andtI > Gateway fare) and the fact that CPQ is finally stressing its enterprisee6 > bragging rights, why on earth would they "kill" VMS? >m  I How can you say that Compaq is "stressing its enterprise bragging rights"sK when they don't even mention OpenVMS?  Isn't OpenVMS the biggest enterprisedK bragging right that they have?  Tru64 has the same soft underbelly as everyeI other Unix right?  They can brag about NSK but there are still only a few I people that really need NSK.  In a few years the only O/S's left standingpG will be Linux, Windows NT and proprietary.  Proprietary will consist oftJ Solaris, OS400, MVS, NSK and, hopefully, OpenVMS.  Compaq seems to realizeL this except they seem to think the important part is the same part that Dell and Gateway can do.    ------------------------------  # Date: Tue, 30 Jan 2001 06:23:52 GMT - From: Terry C Shannon <shannon@world.std.com>78 Subject: Re: It's the end for VMS and other Wild RumoursD Message-ID: <Pine.SGI.4.21.0101300122050.29821-100000@world.std.com>  % On Mon, 29 Jan 2001, Bill Todd wrote:i   > : > Terry C Shannon <shannon@world.std.com> wrote in message@ > news:Pine.SGI.4.21.0101291818350.25203-100000@world.std.com... >  > ...( > L > IMHO the best thing Compaq could do is announce a VMS Applications Porting5 > Fund designed to attract more ISVs to the platform.e > N > [Damn IE didn't right-arrow the original - one of these days I'll figure out > why.]b > J > Er, didn't we hear something about ISV financial support 'way back when?H > Does your statement mean that it didn't happen (which is getting to beK > somewhat of a refrain), or are you proposing something to add to existinga > support (whatever it is)?  >   J Yep, I seem to recall such a statement a year or two back, but it paled inI comparison to the $100M fund for Tru64. Whatever, it would be nice to seea. additional emphasis placed on ISV recruitment.   ------------------------------  # Date: Tue, 30 Jan 2001 06:34:36 GMTv- From: Terry C Shannon <shannon@world.std.com>h8 Subject: Re: It's the end for VMS and other Wild RumoursD Message-ID: <Pine.SGI.4.21.0101300132400.29821-100000@world.std.com>  ( On Mon, 29 Jan 2001, John Vottero wrote:   > < > "Terry C Shannon" <shannon@world.std.com> wrote in message@ > news:Pine.SGI.4.21.0101291757140.28563-100000@world.std.com... > >  >  > [snip] >  > >tM > > Given the downturn in the Wintel market (I can't wait to see how Dell andaK > > Gateway fare) and the fact that CPQ is finally stressing its enterprise 8 > > bragging rights, why on earth would they "kill" VMS? > >w > K > How can you say that Compaq is "stressing its enterprise bragging rights"p) > when they don't even mention OpenVMS?  c  G Easy. Read the 4FQ earnings announcement and listen to the webcast. TheeG focus was on enterprise (half the firm's revenues and 90 percent of then profit), not on peecees.   ------------------------------   Date: 29 Jan 2001 19:44:12 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)- Subject: Re: It's the end for VMS get off now , Message-ID: <954h6c$pib@gap.cco.caltech.edu>  R In article <94vvl3$8ur$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes: >aJ >If people wanted to do something that *might* be constructive, they couldM >band together and present Capellas with a list of what actions it would takeiL >to keep them from bailing out:  at least that would force Compaq to ante upK >or suffer immediate consequences, and help clear the air at the same time.rK >But the time for that kind of effort to be effective may be long gone now.n  J The "People" who need to present this list to Capellas are the big guns atK E-Trade, NASDAQ and the like.  I strongly suggest that anybody reading this H message who works for such an organization (you know who you are!) bringJ the current crisis ( and I think it does merit that name) to the attentionK of their CIO/CTO.  The thing that has me most worried is that this strategyuG of "keep the customers but don't talk up the product" seems designed to H increase or at least maintain the value of a product that is about to beH sold, without simultaneously bringing it to the attention of the outsideH world.  Ie, let us complete the deal in peace.  Unfortunately, since theE most likely buyer is CA, if these companies want to save themselves acG fortune and a tremendous amount of pain and suffering they'd best get a.E commitment out of Capellas that they can at least have first crack at I buying OpenVMS as some sort of consortium, or do something, ANYthing elseaB to keep it out of CA's clutches.  Because if they think OpenVMS isE expensive now, it's nothing compared to the bloodletting that CA willd impose.   H And if these bigwigs do have a chat about OpenVMS's future with CapellasH maybe they can suggest that Compaq spare a few doucats to correct the IOH performance issues.  I find it hard to believe that all the problems I'mH seeing aren't also slowing down their big/expensive systems as well, and/ for them, performance has a real dollar value. b   Regards,   David Mathog mathog@seqaxp.bio.caltech.edue? Manager, sequence analysis facility, biology division, Caltech d   ------------------------------  % Date: Mon, 29 Jan 2001 15:34:50 -0600t5 From: "Greg Christianson" <gchristianson@hotmail.com>s( Subject: MicroVAX 3100 / 40 question ???- Message-ID: <mkld6.66$cd1.4129@nnrp1.sbc.net>o  J     I need to low-level format a drive on a MicroVAX 3100 model 40.  The "J TEST 75  " which works on the model 10's and 20's does not work on a model6 40.  Any help would be appreciated.  Thanks in advance   Greg   ------------------------------  % Date: Mon, 29 Jan 2001 19:48:28 -0500o. From: "Brian Walton" <brimus@gw.total-web.net>, Subject: Re: MicroVAX 3100 / 40 question ???/ Message-ID: <t7c3k72vpg5nd0@corp.supernews.com>a  C Try a "test /util" that is what works on my MicroVAX 3100 model 30.o   Briane  @ "Greg Christianson" <gchristianson@hotmail.com> wrote in message' news:mkld6.66$cd1.4129@nnrp1.sbc.net...uL >     I need to low-level format a drive on a MicroVAX 3100 model 40.  The "L > TEST 75  " which works on the model 10's and 20's does not work on a model8 > 40.  Any help would be appreciated.  Thanks in advance >e > Greg >t >s   ------------------------------  % Date: Mon, 29 Jan 2001 21:46:04 -0600e5 From: "Greg Christianson" <gchristianson@hotmail.com>i, Subject: Re: MicroVAX 3100 / 40 question ???. Message-ID: <fMqd6.175$cd1.8366@nnrp1.sbc.net>  9 "Brian Walton" <brimus@gw.total-web.net> wrote in messaget) news:t7c3k72vpg5nd0@corp.supernews.com...eE > Try a "test /util" that is what works on my MicroVAX 3100 model 30.r >a > Brian  >h  K That was it.  "test/util scsi" and I got a menu of options.  Thanks for the  help.a   Greg   ------------------------------  # Date: Mon, 29 Jan 2001 19:30:25 GMTe% From: Uwe Zessin <zessin@my-deja.com> 7 Subject: Re: Midrange I/O with VMS WAS: DS20 - Slow I/Os) Message-ID: <954gci$ai8$1@nnrp1.deja.com>-  + In article <Op416Gz1ZpNR@eisner.decus.org>,4-   young_r@eisner.decus.org (Rob Young) wrote:D> > In article <87g0i2l56k.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:a( > > fooguy <jweisen@my-deja.com> writes: >2G > > And what do you get? Not much... A big load of 'industry standard',lA > > yeah right, stuff that is limited to about 100Mb/sec. Is it aa? > > replacment for the CIPCA and HSJs, well, in someones dreamsnC > > perhaps. Or one day in the future, maybe. But if you are hopingtE > > for a multi GB/sec resiliant *VMS quality* IO system, you are outnA > > of luck. Do they expect to run EV9 big systems on this stuff?w >iB > 	Yes.  HSG80 has WHL (write history logging) now (1).  Hooks for; > 	mini-merge coming soon.  Perhaps someone can chime in one? > 	deliverables.  The PCI to fibre cards are reasonably priced,a: > 	drop out to switched fabric... at 100 MB/sec the CIPCAs0 > 	and CIXCDs of the world are left in the dust. >u	 > 				Robi >o6 > (1)  http://www5.compaq.com/info/SP7090/SP7090PF.PDF >a > 	See page 3 on WHL infoe   Ahem.  **POOF**,  I think.  ? That is the SPD of the DRM (Data Replication Manager) firmware.t; I'd bet that most OpenVMS sites will not use DRM, but HBVS.p  1 As the paragraph at the bottom of that page says:nA That write log is only used when the links to the target site ares@ down. If the link comes up again and the log hasn't overflown it@ just prevents a full re-synchronization (a copy of all data that is in the remote copy sets).  @ WHL and Mini-merge on the HSG-80 have nothing to do with OpenVMS@ host-based volume-shadowing (HBVS), I am afraid. OpenVMS doesn't; even see the target volumes (if it does, the target site isa improperly set up!).   --
 Uwe Zessin3 (If you want to send mail, please use user "zessin"e/ who lives at "decus.decus.de", not my-deja.com)e     Sent via Deja.com  http://www.deja.com/   ------------------------------    Date: 29 Jan 2001 15:25:04 -0500* From: young_r@eisner.decus.org (Rob Young)7 Subject: Re: Midrange I/O with VMS WAS: DS20 - Slow I/Om+ Message-ID: <p9xnuDTWnYno@eisner.decus.org>p  Q In article <954gci$ai8$1@nnrp1.deja.com>, Uwe Zessin <zessin@my-deja.com> writes:g- > In article <Op416Gz1ZpNR@eisner.decus.org>, / >   young_r@eisner.decus.org (Rob Young) wrote:t? >> In article <87g0i2l56k.fsf@prep.synonet.com>, Paul Repacholi ! > <prep@prep.synonet.com> writes:h) >> > fooguy <jweisen@my-deja.com> writes:e >>H >> > And what do you get? Not much... A big load of 'industry standard',B >> > yeah right, stuff that is limited to about 100Mb/sec. Is it a@ >> > replacment for the CIPCA and HSJs, well, in someones dreamsD >> > perhaps. Or one day in the future, maybe. But if you are hopingF >> > for a multi GB/sec resiliant *VMS quality* IO system, you are outB >> > of luck. Do they expect to run EV9 big systems on this stuff? >>C >> 	Yes.  HSG80 has WHL (write history logging) now (1).  Hooks forV< >> 	mini-merge coming soon.  Perhaps someone can chime in on@ >> 	deliverables.  The PCI to fibre cards are reasonably priced,; >> 	drop out to switched fabric... at 100 MB/sec the CIPCAs 1 >> 	and CIXCDs of the world are left in the dust.n >>
 >> 				Rob >>7 >> (1)  http://www5.compaq.com/info/SP7090/SP7090PF.PDF3 >> >> 	See page 3 on WHL info >  > Ahem.  **POOF**,  I think. > A > That is the SPD of the DRM (Data Replication Manager) firmware. = > I'd bet that most OpenVMS sites will not use DRM, but HBVS.c > 3 > As the paragraph at the bottom of that page says:eC > That write log is only used when the links to the target site areuB > down. If the link comes up again and the log hasn't overflown itB > just prevents a full re-synchronization (a copy of all data that > is in the remote copy sets). > B > WHL and Mini-merge on the HSG-80 have nothing to do with OpenVMSB > host-based volume-shadowing (HBVS), I am afraid. OpenVMS doesn't= > even see the target volumes (if it does, the target site is  > improperly set up!). >   = 	But it does.  If a node crashes in an HSJ based environment, > 	the HSJ's write log ensures that a mini-merge can take place.: 	Several folks are reluctant to go to HSG until mini-merge= 	capabilities show up.  Currently, that is all that write loge; 	is used for, but again .. that write log is very necessaryc& 	if mini-merge is to show up in HSG.    / 	Also, a snippet from an article laying around:'   Q:      What is a mini-merge?a  A         A mini-merge is the same basic operation as a full merge, B         that is, a compare operation of the shadow members to makeB         sure all data is consistent on all members.  However, in aB         mini-merge, controllers that support write-logging containC         specific information on which LBNs were targeted for writes A         by the failing node in the cluster.  Instead of having toiF         compare the entire contents of the disks in a shadow set, onlyD         the LBNs that were targeted for writes are compared.  Due to@         this specific information on LBNs, mini-merges typicallyF         complete in less than 5 seconds.  Usually the only record that;         a mini-merge occurred is the OPCOM message from thea5         SHADOW_SERVER that one started and completed.o  A         The following controllers/disks support the write-logging-         feature:                HSC40              HSC6X              HSC70              HSC9X              HSJ30              HSJ40              HSJ50   ---s  G 	So .. just what are you trying to point out about the controller write  	logs?   Naming conventions?   				Robv   ------------------------------  # Date: Tue, 30 Jan 2001 03:36:49 GMTe% From: Uwe Zessin <zessin@my-deja.com>d7 Subject: Re: Midrange I/O with VMS WAS: DS20 - Slow I/Oc) Message-ID: <955csg$4ig$1@nnrp1.deja.com>l  + In article <p9xnuDTWnYno@eisner.decus.org>,t-   young_r@eisner.decus.org (Rob Young) wrote:e [...]m >aB > So .. just what are you trying to point out about the controller
 > write logs?   B I do understand what the write log history in MSCP controllers and7 certain DSSI Disks is for. Thank's for the reminder ;-)t  A My point is that the write log as implemented in ACS V8.5P is forsB the DRM only. I don't have the SPDs for ACS V8.5F or S handy and IC don't have time to look them up, yet - (I'm on my way to a customerf; in a couple of minutes) do they talk about write logs, too?r   > Naming conventions?   $ The same name for a different thing.   --
 Uwe Zessin3 (If you want to send mail, please use user "zessin"w/ who lives at "decus.decus.de", not my-deja.com)      Sent via Deja.comr http://www.deja.com/   ------------------------------  % Date: Mon, 29 Jan 2001 21:59:10 -0600f+ From: "Main, Kerry" <Kerry.Main@compaq.com>v7 Subject: RE: Midrange I/O with VMS WAS: DS20 - Slow I/OwN Message-ID: <910612C07BCAD1119AF40000F86AF0D805284CE7@kaoexc3.kao.cpqcorp.net>   Uwe,  : The latest FC and DRM support information can be found at:8 <http://www.openvms.compaq.com/openvms/fibre/index.html>K <http://www.openvms.compaq.com/openvms/fibre/fibrechannel.htm> presentationsL (link at bottom of the page) was  updated this month - Jan 2001. It includes OpenVMS 2001 FC plans.   Regards,  
 Kerry Main Senior Consultant  Compaq Canada Inc. Professional Services  Voice: 613-592-4660x Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----, From: Uwe Zessin [mailto:zessin@my-deja.com] Sent: January 29, 2001 2:30 PM To: Info-VAX@Mvb.Saic.Com.7 Subject: Re: Midrange I/O with VMS WAS: DS20 - Slow I/Of    + In article <Op416Gz1ZpNR@eisner.decus.org>,p-   young_r@eisner.decus.org (Rob Young) wrote: > > In article <87g0i2l56k.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:E( > > fooguy <jweisen@my-deja.com> writes: >tG > > And what do you get? Not much... A big load of 'industry standard',tA > > yeah right, stuff that is limited to about 100Mb/sec. Is it ah? > > replacment for the CIPCA and HSJs, well, in someones dreams C > > perhaps. Or one day in the future, maybe. But if you are hoping E > > for a multi GB/sec resiliant *VMS quality* IO system, you are outiA > > of luck. Do they expect to run EV9 big systems on this stuff?  >tB > 	Yes.  HSG80 has WHL (write history logging) now (1).  Hooks for; > 	mini-merge coming soon.  Perhaps someone can chime in one? > 	deliverables.  The PCI to fibre cards are reasonably priced,w: > 	drop out to switched fabric... at 100 MB/sec the CIPCAs0 > 	and CIXCDs of the world are left in the dust. >y	 > 				Roba >a6 > (1)  http://www5.compaq.com/info/SP7090/SP7090PF.PDF >r > 	See page 3 on WHL infoe   Ahem.  **POOF**,  I think.  ? That is the SPD of the DRM (Data Replication Manager) firmware.-; I'd bet that most OpenVMS sites will not use DRM, but HBVS.r  1 As the paragraph at the bottom of that page says:oA That write log is only used when the links to the target site are @ down. If the link comes up again and the log hasn't overflown it@ just prevents a full re-synchronization (a copy of all data that is in the remote copy sets).  @ WHL and Mini-merge on the HSG-80 have nothing to do with OpenVMS@ host-based volume-shadowing (HBVS), I am afraid. OpenVMS doesn't; even see the target volumes (if it does, the target site iss improperly set up!).   --
 Uwe Zessin3 (If you want to send mail, please use user "zessin"a/ who lives at "decus.decus.de", not my-deja.com)s     Sent via Deja.com  http://www.deja.com/   ------------------------------  # Date: Mon, 29 Jan 2001 23:44:37 GMTe2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)T Subject: Re: Monitoring external temperature (was Re: Monitoring external temprature7 Message-ID: <Ffnd6.188$cu.1416@gazette.loc1.tandem.com>   d In article <0kld6.3146$wb7.82172@news1.mts.net>, "Mark-Simon Pope" <mpope@NOSPAM@bristol.ca> writes:G :I've got a rack with a couple of DS20's, and some fiber attached disk,  :running VMS 7.1.r  J   V7.1-2 for the base system, and likely V7.2-1 for Fibre Channel widgets.  M :I'm looking for a product that can monitor the rooms temperature so I can bee' :alerted when the cooling systems fail.   0   Various air conditioning products offers this.  J   ServerWorks performs this, and various Alpha systems can shut themselvesG   down.  (See the shutdown commands and related platform details in the.*   SRM console and platform documentation.)  L   SYI$_THERMAL_VECTOR and the ServerWorks stuff is the mechanism, for those :   systems that can report the temperature.  (See below...)  H   Some pointers on sending pages from OpenVMS are referenced in the FAQ.  F   The attached program uses an undocumented API (for the Server SystemC   MIB), works best on V7.1-2 and later, and requires ServerWorks...o  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com     $ RUN SHOW_POWER    O                   0   1   2   3   4   5   6   7   8   9  10  11  12  13  14  15tO                 _______________________________________________________________lO Power status      +   .   +   .   .   .   .   .   .   .   .   .   .   .   .   .uO Fan status        +   +   .   .   .   .   .   .   .   .   .   .   .   .   .   .SO Thermal status    +   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .3O Temperature (F)  67   .   .   .   .   .   .   .   .   .   .   .   .   .   .   . O Temperature (C)  19   .   .   .   .   .   .   .   .   .   .   .   .   .   .   .0    P --------------------------------------------------------------------------------3 #pragma module SHOW_POWER "X-1"                    P /*      * Copyright (C) 1998 by       o9  * Digital Equipment Corporation, Maynard, Massachusetts. )  * All rights reserved.                     *                            H  * This software is furnished under a license and may be used and copiedH  * only  in  accordance  of  the  terms  of  such  license  and with theH  * inclusion of the above copyright notice. This software or  any  otherH  * copies thereof may not be provided or otherwise made available to anyH  * other person.  No title to and  ownership of the  software is  hereby  * transferred.   *H  * The information in this software is  subject to change without noticeH  * and  should  not  be  construed  as a commitment by Digital Equipment  * Corporation.j  *H  * Digital assumes no responsibility for the use  or  reliability of its:  * software on equipment which is not supplied by Digital.  *  * Abstract:  *E  *	This user program tests each of the new environmental GETSYI item o:  *      codes and display the results in a pretty table.  ;  *                                                         m  * Authors:   *   *	Paul A. Jacobi, November 1998  *  * Modified by:m  *  */n   #include	<stdio> l #include	<stdlib>s #include	<string>e   #include	<hwrpbdef>e #include	<iledef>v #include	<ssdef> #include	<syidef>l #include	<vms_macros>t   /* External routines */e   extern int sys$getsyiw();  extern void sys$exit();    /* External global data */   /* forward routines */   static void show_header(); static void show_power();  static void show_fan();i static void show_thermal();o static void show_temperature(); K static int getsyi_vector(unsigned char *getsyi_buffer, int *buffer_length, o(                          int item_code);  0                                                 /*++  * main - main routine)  *                                          * FUNCTIONAL DESCRIPTION:  *                          J  *	This routine is the main test routine.  It calls a routine to test each/  *	of the new environmental GETSYI item codes. o  *
  * INPUTS:.  *	Standard C command line arguments - ignored  * .1  * OUTPUTS:                                         *	Nonet4  *                                                    *--  */t  & static int main(int argc, char **argv) {c       show_header();       show_power();        show_fan();        show_thermal();      l       show_temperature();e   }     0                                                 /*++%  * show_header - Show instance headera)  *                                          * FUNCTIONAL DESCRIPTION:  *                          *  *	This routine shows the instance header.  *
  * INPUTS:  *	Nonev  * t1  * OUTPUTS:                                      d  *	Noneu4  *                                                    *--  */  static void show_header()f {l       printf("\n");,     printf("               ");/     printf("  00  01  02  03  04  05  06  07");0/     printf("  08  09  10  11  12  13  14  15");t     printf("\n");:       printf("                ");e/     printf("________________________________"); .     printf("_______________________________");     printf("\n");T       return;" }i           /*++!  * show_power - Show power vectora)  *                                          * FUNCTIONAL DESCRIPTION:  *                          F  *	This routine shows power vector.  If $GETSYI returns an error, this(  *	routines show the decimal error code.  *
  * INPUTS:  *	None   * g1  * OUTPUTS:                                         *	Nonei4  *                                                    *--  */ # static void show_power()           t {l9     unsigned char getsyi_buffer[SYI$K_ENV_VECTOR_LENGTH];t0     int buffer_length = SYI$K_ENV_VECTOR_LENGTH;
     int i;     unsigned char *bp;     int status;w  N     status = getsyi_vector(getsyi_buffer, &buffer_length, SYI$_POWER_VECTOR);      if bad_status(status) {nJ         printf("ERROR: SYI_POWER_VECTOR returned status of %d\n", status);         return;0     }r       printf("Power Status   ");       bp = getsyi_buffer;    :'     for (i=0; i < buffer_length; i++) {-         switch (*bp++) {)             case SYI$K_ENV_STATUS_FAILED:o                 printf("   -");e                 break;%             case SYI$K_ENV_STATUS_OK:t                 printf("   +");3                 break;.             case SYI$K_ENV_STATUS_NOT_PRESENT:                 printf("   .");o                 break;*             case SYI$K_ENV_STATUS_UNKNOWN:             default:                 printf("   ?");m                 break;	         }      }h       printf("\n");        return;D }9     i /*++  * show_fan - Show fan vectorp)  *                                       i  * FUNCTIONAL DESCRIPTION:  *                          D  *	This routine shows fan vector.  If $GETSYI returns an error, this(  *	routines show the decimal error code.  *
  * INPUTS:  *	None   * e1  * OUTPUTS:                                         *	Noner4  *                                                    *--  */l static void show_fan() {J9     unsigned char getsyi_buffer[SYI$K_ENV_VECTOR_LENGTH];e+     int i;                                 o0     int buffer_length = SYI$K_ENV_VECTOR_LENGTH;     unsigned char *bp;     int status;c  L     status = getsyi_vector(getsyi_buffer, &buffer_length, SYI$_FAN_VECTOR);      if bad_status(status) {fH         printf("ERROR: SYI_FAN_VECTOR returned status of %d\n", status);         return;      }e       printf("Fan Status     ");       bp = getsyi_buffer;     '     for (i=0; i < buffer_length; i++) {f         switch (*bp++) {)             case SYI$K_ENV_STATUS_FAILED:w                 printf("   -");F                 break;%             case SYI$K_ENV_STATUS_OK:n                 printf("   +");(                 break;.             case SYI$K_ENV_STATUS_NOT_PRESENT:                 printf("   .");t                 break;*             case SYI$K_ENV_STATUS_UNKNOWN:             default:                 printf("   ?");f                 break;	         }g     }        printf("\n");e       return;- }c      o /*++%  * show_thermal - Show thermal vectorc)  *                                       -  * FUNCTIONAL DESCRIPTION:  *                          H  *	This routine shows thermal vector.  If $GETSYI returns an error, this(  *	routines show the decimal error code.  *
  * INPUTS:  *	Nones  * I1  * OUTPUTS:                                         *	None 4  *                                                    *--  */t static void show_thermal() {-9     unsigned char getsyi_buffer[SYI$K_ENV_VECTOR_LENGTH];r
     int i;0     int buffer_length = SYI$K_ENV_VECTOR_LENGTH;     unsigned char *bp;     int status;s  P     status = getsyi_vector(getsyi_buffer, &buffer_length, SYI$_THERMAL_VECTOR);             <     if bad_status(status) { L         printf("ERROR: SYI_THERMAL_VECTOR returned status of %d\n", status);         return;t     }        printf("Thermal Status ");       bp = getsyi_buffer;    e'     for (i=0; i < buffer_length; i++) {          switch (*bp++) {)             case SYI$K_ENV_STATUS_FAILED:h                 printf("   -");a                 break;%             case SYI$K_ENV_STATUS_OK:.                 printf("   +");                  break;.             case SYI$K_ENV_STATUS_NOT_PRESENT:                 printf("   .");o                 break;*             case SYI$K_ENV_STATUS_UNKNOWN:             default:                 printf("   ?");y                 break;	         }      }a       printf("\n");l       return;. }o     V /*++G  * show_temperature - Show temperature vector in Fahrenheit and Celsiuss)  *                                       i  * FUNCTIONAL DESCRIPTION:  *                          G  *	This routine shows temperature vector.  If $GETSYI returns an error,R2  *      this routines show the decimal error code.  *
  * INPUTS:  *	None   * 31  * OUTPUTS:                                         *	None 4  *                                                    *--  */e static void show_temperature() {u9     unsigned char getsyi_buffer[SYI$K_ENV_VECTOR_LENGTH]; 
     int i;0     int buffer_length = SYI$K_ENV_VECTOR_LENGTH;     unsigned char *bp;     int status;l     int temp_f;n  :     status = getsyi_vector(getsyi_buffer, &buffer_length, 5                            SYI$_TEMPERATURE_VECTOR); o     if bad_status(status) { P         printf("ERROR: SYI_TEMPERATURE_VECTOR returned status of %d\n", status);         return;m     }.  0     /* Show temperature in degrees Fahrenheit */       printf("Temperature (F)");                            t     bp = getsyi_buffer;    i'     for (i=0; i < buffer_length; i++) {w         switch (*bp) {.             case SYI$K_ENV_STATUS_NOT_PRESENT:                 printf("   .");m                 break;*             case SYI$K_ENV_STATUS_UNKNOWN:                 printf("   ?");                  break;             default:B                 /* For better accuracy use 90/5 instead of 9/5 and:                  * 325 for 32.5, then divide result by 10.                  */c3                 temp_f = (((*bp * 90)/5) + 325)/10;a%                 printf("%4d",temp_f);a                 break;  	         }n
         bp++;s     }v       printf("\n");i  -     /* Show temperature in degrees Celsius */n       printf("Temperature (C)");                            f     bp = getsyi_buffer;    i'     for (i=0; i < buffer_length; i++) {i         switch (*bp) {.             case SYI$K_ENV_STATUS_NOT_PRESENT:                 printf("   .");e                 break;*             case SYI$K_ENV_STATUS_UNKNOWN:                 printf("   ?");p                 break;             default:"                 printf("%4d",*bp);                 break; e	         }C
         bp++;      }        printf("\n");h       return;a   }m     o /*++A  * getsyi_vector - Call GETSYI with supplied buffer and item code )  *                                          * FUNCTIONAL DESCRIPTION:  *                            *	 N  *      This routines creates and item list from the supplied buffer and item (  *	code and calls GETSYI system service.  *
  * INPUTS:5  *	getsyi_buffer	- Pointer to buffer to return vector-G  *	buffer_length	- Pointer to buffer length, updated with actual length<  *	item_code	- GETSYI item code:  * a1  * OUTPUTS:                                         *	Nonei4  *                                                    *--  */>K static int getsyi_vector(unsigned char *getsyi_buffer, int *buffer_length, w'                          int item_code)  {t     unsigned short retlen;     ILE3 item_list[2];     int status;i  )     item_list[0].ile3$w_code = item_code; 0     item_list[0].ile3$w_length = *buffer_length;1     item_list[0].ile3$ps_bufaddr = getsyi_buffer;o/     item_list[0].ile3$ps_retlen_addr = &retlen;8!     item_list[1].ile3$w_code = 0;m#     item_list[1].ile3$w_length = 0;a    6     status = sys$getsyiw(0, 0, 0, item_list, 0, 0, 0);     if bad_status(status)i         retlen = 0;h  "     *buffer_length = (int) retlen;     return(status);o   }t  P --------------------------------------------------------------------------------* $ DEFINE SLB SYS$LIBRARY:SYS$STARLET_C.TLB& $ DEFINE LLB SYS$LIBRARY:SYS$LIB_C.TLB8 $ CC/LIS/DEFINE=__NEW_STARLET SHOW_POWER+SLB/LIB+LLB/LIB $ LINK/SYSEXE SHOW_POWER   ------------------------------  % Date: Mon, 29 Jan 2001 15:23:52 -0600D1 From: "Mark-Simon Pope" <mpope@NOSPAM@bristol.ca>.' Subject: Monitoring external temprature 0 Message-ID: <0kld6.3146$wb7.82172@news1.mts.net>  F I've got a rack with a couple of DS20's, and some fiber attached disk,D running VMS 7.1.  This is in an air conditioned room with some otherH equipment.  I'm capable of sending alerts to my alpha numeric pager from these boxes.  L I'm looking for a product that can monitor the rooms temperature so I can be& alerted when the cooling systems fail.   Ever heard of anything?y  I I walked into a steaming hot computer room this morning, the air had beensJ down most of the weekend. Previously, we had been able to rely on securityH patrols to alert us.  They still walk through the room, guess that's all they do now.   thanks,u Mark-S.    ------------------------------   Date: 29 Jan 2001 20:34:06 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)= Subject: Re: No technical computing, was: expanding the nichee, Message-ID: <954k3u$pib@gap.cco.caltech.edu>  R In article <94io73$jpd$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes: >sB >Unfortunately, you chose to extrapolate and make statements aboutJ >enterprise-level systems which bear almost no resemblance to the sorts of >programs you run. >B  J Well, I have asked people with such systems to run the benchmarks and see F if those bigger systems somehow or other bypass the problems I see on D smaller systems.  Some of the biggest performance problems are with I respect to really low level operations like creating or extending a file.sL Those operations should be pretty much the same there as on my little DS10s.  I >Oh, yes:  a 2- or 3-to-1 difference (ignoring the apples/oranges issue), L >while significant, is far from the 10-to-1 figure you were throwing around.  I 2-3:1 is the best I've been able to do for "disk IO" on VMS, in this casepB using a RAMdisk.  And that's compared to the default file caching I performance on Linux.  The 10:1 figure is a comparison of VMS (optimized)oD writing to a physical disk versus Linux through file cache.  Without6 optimization the latter ratio is much closer to 100:1.   > E >The bottom line is that you're still confusing default behavior withs >attainable behavior.   F Attainable, but at a very high cost.  Compared to "runs very fast withE unmodified C code" on other OS's.  This is not a winning combination.e2 Fixing this would make VMS much more competitive.   C Not that that matters one iota since Compaq clearly doesn't care toeE to make VMS competitive again outside of its chosen niche.  (It's notsD even all that clear that they're spending very much money to keep itH competitive within those niches).   Which is why all of this performanceD discussion is a moot point.  If they won't fix any of these problems* there's hardly any point discussing them.    Regards,   David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech pJ **************************************************************************J *                                RIP VMS                                 *J **************************************************************************   ------------------------------  % Date: Mon, 29 Jan 2001 16:51:34 -0500e; From: "Everhart, Glenn (FUSA)" <GlennEverhart@FirstUSA.com>o= Subject: RE: No technical computing, was: expanding the nicheAN Message-ID: <4B279CF3578CD211B945009027178017046D0360@swilnts809.wil.fusa.com>  : Anyone for toning down the whining a few tens of decibels?6 We've gotten confirmation Capellas did NOT mean Compaq? is leaving VMS, but the contrary (as made sense). We know theree> is work going on in Spit Brook to speed VMS I/O up. This takesE time though, and VMS engineering is not in the habit of preannouncingn@ stuff. Don't expect to see 200 beta versions of code, as happens? with stuff like reiserfs on Linux as it is built. But the folks-A need some time & patience to design, build and test. If you don'tfB like that, VMS can support other filesystems; go and write one you@ like better. (yeah, nontrivial, but doable.) One legacy Spiralog@ left was an API that is easier than the old one, but others have? written filesystems for VMS using the interfaces that have beene there for ages. D   Same with advertising I fear. It takes some time to build internal
 consensus.<    If you feel it's a waste, OK. Maybe it will prove so. The> complaining here is however harmful and I suspect reduces the ; chance that good might come to VMS. IMO the product doesn't  deserve ill words. Glenn Everhart     -----Original Message-----# From: mathog@seqaxp.bio.caltech.edu & [mailto:mathog@seqaxp.bio.caltech.edu]& Sent: Monday, January 29, 2001 3:34 PM To: Info-VAX@Mvb.Saic.Como= Subject: Re: No technical computing, was: expanding the nichet    J In article <94io73$jpd$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes:r >tB >Unfortunately, you chose to extrapolate and make statements aboutJ >enterprise-level systems which bear almost no resemblance to the sorts of >programs you run. >t  J Well, I have asked people with such systems to run the benchmarks and see F if those bigger systems somehow or other bypass the problems I see on D smaller systems.  Some of the biggest performance problems are with I respect to really low level operations like creating or extending a file.iL Those operations should be pretty much the same there as on my little DS10s.  I >Oh, yes:  a 2- or 3-to-1 difference (ignoring the apples/oranges issue),jL >while significant, is far from the 10-to-1 figure you were throwing around.  I 2-3:1 is the best I've been able to do for "disk IO" on VMS, in this case:B using a RAMdisk.  And that's compared to the default file caching I performance on Linux.  The 10:1 figure is a comparison of VMS (optimized) D writing to a physical disk versus Linux through file cache.  Without6 optimization the latter ratio is much closer to 100:1.   >*E >The bottom line is that you're still confusing default behavior with  >attainable behavior.   F Attainable, but at a very high cost.  Compared to "runs very fast withE unmodified C code" on other OS's.  This is not a winning combination.v2 Fixing this would make VMS much more competitive.   C Not that that matters one iota since Compaq clearly doesn't care towE to make VMS competitive again outside of its chosen niche.  (It's notvD even all that clear that they're spending very much money to keep itH competitive within those niches).   Which is why all of this performanceD discussion is a moot point.  If they won't fix any of these problems* there's hardly any point discussing them.    Regards,   David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech  J **************************************************************************J *                                RIP VMS                                 *J **************************************************************************   ------------------------------  % Date: Tue, 30 Jan 2001 04:54:17 +0100*  From: Paul Sture <paul@sture.ch>= Subject: Re: No technical computing, was: expanding the niche + Message-ID: <VA.0000026a.0e3f0178@sture.ch>   O In article <4B279CF3578CD211B945009027178017046D0360@swilnts809.wil.fusa.com>, n Glenn (FUSA) Everhart wrote:= > From: "Everhart, Glenn (FUSA)" <GlennEverhart@FirstUSA.com>  > Newsgroups: comp.os.vms ? > Subject: RE: No technical computing, was: expanding the nichet' > Date: Mon, 29 Jan 2001 16:51:34 -0500  > < > Anyone for toning down the whining a few tens of decibels?  I Hear hear! Much of it is beginning to sound like a stuck record, and not   adding anything constructive.   O Yes, I know emotions are running high at the moment, but let's remember we are 3N professionals. Constant whining is guaranteed to switch the audience off, and  sooner rather than later._  L Logical, rational arguments stand a much better chance of being listened to  and acted upon.i  8 > We've gotten confirmation Capellas did NOT mean CompaqA > is leaving VMS, but the contrary (as made sense). We know there @ > is work going on in Spit Brook to speed VMS I/O up. This takesG > time though, and VMS engineering is not in the habit of preannouncingdB > stuff. Don't expect to see 200 beta versions of code, as happensA > with stuff like reiserfs on Linux as it is built. But the folks C > need some time & patience to design, build and test. If you don't D > like that, VMS can support other filesystems; go and write one youB > like better. (yeah, nontrivial, but doable.) One legacy SpiralogB > left was an API that is easier than the old one, but others haveA > written filesystems for VMS using the interfaces that have beenn > there for ages. F >   Same with advertising I fear. It takes some time to build internal > consensus.> >    If you feel it's a waste, OK. Maybe it will prove so. The@ > complaining here is however harmful and I suspect reduces the = > chance that good might come to VMS. IMO the product doesn'tr > deserve ill words.  
 Precisely.   > Glenn Everhart >  >  > -----Original Message-----% > From: mathog@seqaxp.bio.caltech.edua( > [mailto:mathog@seqaxp.bio.caltech.edu]( > Sent: Monday, January 29, 2001 3:34 PM > To: Info-VAX@Mvb.Saic.Com ? > Subject: Re: No technical computing, was: expanding the niche  >  > L > In article <94io73$jpd$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com>	 > writes:  > >eD > >Unfortunately, you chose to extrapolate and make statements aboutL > >enterprise-level systems which bear almost no resemblance to the sorts of > >programs you run. > >  > L > Well, I have asked people with such systems to run the benchmarks and see H > if those bigger systems somehow or other bypass the problems I see on F > smaller systems.  Some of the biggest performance problems are with K > respect to really low level operations like creating or extending a file.sN > Those operations should be pretty much the same there as on my little DS10s. > K > >Oh, yes:  a 2- or 3-to-1 difference (ignoring the apples/oranges issue), N > >while significant, is far from the 10-to-1 figure you were throwing around. > K > 2-3:1 is the best I've been able to do for "disk IO" on VMS, in this case D > using a RAMdisk.  And that's compared to the default file caching K > performance on Linux.  The 10:1 figure is a comparison of VMS (optimized)uF > writing to a physical disk versus Linux through file cache.  Without8 > optimization the latter ratio is much closer to 100:1. >  > > G > >The bottom line is that you're still confusing default behavior with" > >attainable behavior.  > H > Attainable, but at a very high cost.  Compared to "runs very fast withG > unmodified C code" on other OS's.  This is not a winning combination.N4 > Fixing this would make VMS much more competitive.  > E > Not that that matters one iota since Compaq clearly doesn't care to G > to make VMS competitive again outside of its chosen niche.  (It's not F > even all that clear that they're spending very much money to keep itJ > competitive within those niches).   Which is why all of this performanceF > discussion is a moot point.  If they won't fix any of these problems, > there's hardly any point discussing them.  > 
 > Regards, >  > David Mathog > mathog@seqaxp.bio.caltech.edu A > Manager, sequence analysis facility, biology division, Caltech )L > **************************************************************************L > *                                RIP VMS                                 *L > ************************************************************************** >E   ___C
 Paul Sture Switzerland    ------------------------------  % Date: Mon, 29 Jan 2001 15:31:43 -0500d6 From: "Dominic Olivastro" <DOlivastro@ChiResearch.com>! Subject: Pathworks and Windows ME % Message-ID: <Tkkd6.31$Cu6.669@client>0  H This is really a question for the pathworks group, but no one answers me over there._  I We have Pathworks client V. 7.1A, and one PC now has Windows ME.  Are the 3 two compatible, or do I need a newer version of PW?"   Dom      -- Dominic Olivastro  CHI Research, IncK 10 White Horse PikeN Haddon Heights, NJ 08035   Phone:  1-856-546-0600 Fax:       1-856-546-9633S% mailto:    DOlivastro@ChiResearch.come   ------------------------------    Date: 29 Jan 2001 22:48:31 +0100* From: eplan@kapsch.net (Peter LANGSTOEGER)% Subject: Re: Pathworks and Windows MEp* Message-ID: <3a75e52f$1@news.kapsch.co.at>  ^ In article <Tkkd6.31$Cu6.669@client>, "Dominic Olivastro" <DOlivastro@ChiResearch.com> writes:I >This is really a question for the pathworks group, but no one answers me  >over there.   Better ask your service rep...  J >We have Pathworks client V. 7.1A, and one PC now has Windows ME.  Are the4 >two compatible, or do I need a newer version of PW?  0 Check	http://www.openvms.compaq.com/pathworks32/  K IIRC, you need PATHWORKS32 V7.2 for NT5 (aka Win2k) but there is no mentionEL of Win00 (aka Windows ME) at this site at all (and SPD link is broken, too).  H I would assume, that with PATHWORKS32 V7.2 you have a better chance (and0 you also get euro currency symbol support ;-)...  E Maybe PATHWORKS Engineering jumps in here (but better chance for thist3 is of course in the pathworks group). Be patient...    -- d< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888p< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Mon, 29 Jan 2001 16:35:16 -0600f+ From: "Main, Kerry" <Kerry.Main@compaq.com> % Subject: RE: Pathworks and Windows ME N Message-ID: <910612C07BCAD1119AF40000F86AF0D805284CE0@kaoexc3.kao.cpqcorp.net>   Peter,  & The new SPD web pages can be found at:" <http://www5.compaq.com/info/spd/>  L Additional information and contacts can be found on the "Contact us" link on the PW32 pages ie.: <http://www.openvms.compaq.com/pathworks32/contactus.html>  ; Dom, did you try this contact (and support email) page yet?f   Regards,  
 Kerry Main Senior Consultants Compaq Canada Inc. Professional Services" Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----0 From: eplan@kapsch.net [mailto:eplan@kapsch.net] Sent: January 29, 2001 4:49 PM To: Info-VAX@Mvb.Saic.Com % Subject: Re: Pathworks and Windows MEI    9 In article <Tkkd6.31$Cu6.669@client>, "Dominic Olivastro" $ <DOlivastro@ChiResearch.com> writes:I >This is really a question for the pathworks group, but no one answers me  >over there.   Better ask your service rep...  J >We have Pathworks client V. 7.1A, and one PC now has Windows ME.  Are the4 >two compatible, or do I need a newer version of PW?  0 Check	http://www.openvms.compaq.com/pathworks32/  K IIRC, you need PATHWORKS32 V7.2 for NT5 (aka Win2k) but there is no mentionsL of Win00 (aka Windows ME) at this site at all (and SPD link is broken, too).  H I would assume, that with PATHWORKS32 V7.2 you have a better chance (and0 you also get euro currency symbol support ;-)...  E Maybe PATHWORKS Engineering jumps in here (but better chance for this 3 is of course in the pathworks group). Be patient...    --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888 < <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  # Date: Mon, 29 Jan 2001 22:11:31 GMT * From: Eugene Zharkov <zharkov@my-deja.com>/ Subject: recv causing an application to crash ?l) Message-ID: <954pqa$k0h$1@nnrp1.deja.com>   + I have an application that is crashing fromt. time to time. I am geeting ACCVIO, ASTFLT, and/ other errors. My feeling is that the stack gets , corrupted. I am trying to reduce the size of- the application, so that I can either post it-, here or I will find the problem myself. This- may take time some time, though. So I thoughtC0 I would ask here, may someone has seen something. that, may be this is a known problem and there is a patch.S  ( I am on Alpha VMS 7.1, UCX V4.2 - ECO 2.& The program seem to be crashing inside& recv or on return from recv. I believe+ that the stack is large enough. The programe* is build with /reentrancy=multithread, but) no extra threads are actually created andi+ the recv call is made from the main thread.d0 The program makes multiple recv calls. Sometimes is does not crash at all.        --       Sent via Deja.coma http://www.deja.com/   ------------------------------  % Date: Mon, 29 Jan 2001 23:41:48 +0100t+ From: "Fred A G" <nospam@allowed.localhost>n6 Subject: Re: Relative merits of LPD vx TELNET printing5 Message-ID: <_kmd6.1187$pq6.3175@nntpserver.swip.net>n  H "Mark Garrett" <Mark.Garrett@wedontwantyourspam.com.au> wrote in message= news:B699F23B.10404%Mark.Garrett@wedontwantyourspam.com.au...o@ > in article 94q6ig$ruh$1@bob.news.rcn.net, Carmine Castiglia at3 > ccastiglia@engsint.com wrote on 26/01/2001 08:41:0 >0C > > Sorry if this is a widely known subject but it's unclear to me:c > >:C > > Given OpenVMS v 7.1-1H2 and DECNet Phase IV (UCX), which is the3 better/most.H > > efficient/most desirable/etc form of printer management: UCX$LPD_SMB or > > UCX$TELNETSYM? > >  > >rE > Just watch out the telnet comms is 7-bit so just don't try anythings! > containing 8bit data, eg fonts.o  F Hm, the telnetsym can handle both "raw TCP" and Telnet printing, so itD should be up to the receiver, e.g. a print server, to handle Reverse) Telnet ("un/padding" IAC in data stream).b   Regardso /Fad   ------------------------------    Date: 29 Jan 2001 19:06:34 -05002 From: malmberg@eisner.decus.org (John E. Malmberg) Subject: Re: samba 2.0.6+ Message-ID: <YiVOjt3ZGg8d@eisner.decus.org>h  0 In article <Oozvf8elmJpy-pn2-l3kuYITIn1NX@Tom2>,9 Thomas.Hahnemann@nospam_s-t.de (Thomas Hahnemann) writes:e  A > Now it runs. My mistake was to specify "interface =" instead of.D > "interfaces =" and to look only into  the text files assuming they3 > contain then same information like the html docs.e  5 > But the problem with large directory still remains.i  D My suspicions are that it is a result of internally SMBD effectively: calling stat() several times for the display of each file.  @ This requires translating the filename from DOS format through aB codepage with a small loop and then scanning it again to translate> any "\" found to be "/"., and then the resulting filename must> be translated to VMS converting any funny characters to either ODS-2 limitations or ODS-5.i  = Then there is also a section of code that tries to simulate a ? case insensitive filesystem on UNIX.  This part can be disabledt: by a setting in smb.conf.  In my tests it seemed to help a9 little bit.  There was a report posted recently that made = the client have to use lower case when retrieving files on ansA earlier version of SAMBA.  I do not recall seeing that, but again D I was looking at directory listing times and may have missed testing for that side effect.g  B Plus it must try to create a unique 8.3 filename.  This is suppose0 to be able to be disabled by a smb.conf setting.  D Now if you are displaying a directory in a Microsoft Window ExplorerD window on the client, it is searching each file of certain types for an icon to display.*  8 With Oplocks enabled, this could slow things down a bit.  $ > Now I stop the processes by typing >   stop "NMBD 2.0.6"h > ande >   stop "SMB_    myhostnnn" >A > Q: How can I shutdown samba ?   + You have got the NMBD shutdown down pat :-)D  A The "SMB_ *" names will eventually go away after a time when they = discover that the client has not transferred any data.  Sincee> the SMB or Lanman protocol is mostly "connectionless" if there> are no files actively open, the "SMB_ *" processes can go away: in idle periods even if you have a drive mapped.  The next8 activity will cause your TCPIP program to start up a new& "SMB_ myhostnnn" process or processes.  @ Shutting the SMB programs down is basically issuing the commands> specific to the TCPIP program you have to disable the starting of the "SMB_ *" processes.  > For SAMBA 2.0.6 on OpenVMS, when you have disabled your TCP/IPD program from starting the SMBD process, you can run the SMBD process5 interactively and redirect the output to SYS$OUTPUT:.,  B Due to either a bug or a feature in the UNIX code, the debug level7 must be 4 or higher for this redirection to take place.   B The fun part in running SMBD interactively in debug is finding theF places where the client will time out, and where it will wait for everC for the server to respond.  Also occasionally one of the systems onoF your network will happen to try to browse your server just as you wereE trying to connect up a client, thus hijacking your debugging session.t  	 Have fun,x -Johna wb8tyw@qsl.network Personal Opinion Only.  F You may want to consider joining the SAMBA-VMS (at) SAMBA.ORG mailing.A list.  It does have a lot of overlap with the people here, but istB usually not too active.  There may be some people that are on that4 list that could be interested in what you are doing.  C To sign up to for the SAMBA-VMS mailing list, see your local mirrorf of http://www.samba.org    ------------------------------  % Date: Mon, 29 Jan 2001 14:03:57 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca>sG Subject: Re: TQK70 controller in serpentine config, q-bus cfg questionsc, Message-ID: <3A75BE9A.E2788129@videotron.ca>   Bob Koehler wrote:' >    1  2  3  4  5  6  7  8  9 10 11 12t' >    ----------  ----  ----  ----  ---- ' >             |  |  |  |  |  |  |  |  | ' >             |  |  |  |  |  |  |  |  |/' >             ----  ----  ----  ----  -f  ( I think your graphic is reversed :-) :-)   The order should be: 1 L-R' 2 L-R  3 L-R  4 Lr 4 Rk 5 Ro 5 L, 6 Lh 6 Rs 7 Rs 7 L  8 L  8 Rn  9 for a total of 13 available devices in the first backpanen   For the second backpane 8 1 L (gets the cable that links it to the first backpane) 2 Lr 3 Li 4 L  4 R  etc (same as above).  I >    About the slots:  I recall only two memory boards in my MicroVAX II,eH >    but my old MicroVAX Systems handbook claims "up to three ... memoryH >    expansion modules", so I'm not sure if there are 3 Q/CD slots or 4.  J There are 3. 1 goes to CPU, and 2 and 3 go for memory expansion to 16 meg.   ------------------------------  % Date: Mon, 29 Jan 2001 19:02:16 +0000 + From: "antonio.carlini" <arcarlini@iee.org>wG Subject: Re: TQK70 controller in serpentine config, q-bus cfg questions ' Message-ID: <3A75BE38.12B7AF8E@iee.org>    NS Gaffin wrote: > 7 > Also, I have seen what a flip chip is, and where theyo8 > were in my TK70 boot problem. I presume these are also > known as grant cards?b  - I think they are *only* known as grant cards. ' Flip-chips are something else and older.. (unless you've run into Intel's latest Pentia)   >  > My current situation is this:* > 5 >                         A-B                     C-D* > & > Slot 1              M7625-BA (KA655)$ > Slot 2              CMX-1651 (mem)+ > Slot 3          M759 (TQK70)    Flip Chip*( > Slot 4          M7516 (DELQA)   EMPTY!( > Slot 5          CQD220          EMPTY!# > Slot 6              ICP1622S COMMl0 > Slot 7-15               EMPTY.          EMPTY. >   0 From the picture it looks like you have a BA123.3 This is a 12 slot backplane (the 13th isn't Q-bus),C0 as opposed to the smaller BA23 8 slot backplane./ The easiest way to distinguish them is that thee1 BA23 is really quite heavy whereas with the BA123 . you'd have to be pretty stupid to even try :-)  1 You can however comfortably sit on the BA123 and, 0 if it still has its wheels, you can scoot around on it.    / It matters because the grant continuity differs! slightly between the two.    In the BA123 it goes ,  - 1->2->3->4->5U->5L->6L->6U->7U->7L->8L->8U..., etc.    ( where 5U means slot five, upper position" (A/B) and 5L means slot five lower position (C/D).n  # So the first four slots do not neede any grants even if you put a l* half height card (like you TQK70 or DELQA)" in them ... but such cards must go in the upper (A/B) position.  $ Then you fill slot 5 A/B then slot 5# C/D then slot 6 C/D then slot 6 A/B " etc. If you draw this out it looks like a snake, hence the name.   & Going by your diagram, if the ICP card* needs to work (and needs either interrupts% or DMA) and the CMD card in slot five. is a half-height card (which iss what it looks like going by thes picture) then you really needi( a grant card in 5 C/D. On the other hand! the grant card in  the lower halft) of slot 3 (C/D) is unnecessary. (In fact,p if you had a quad card in slot. 4 this grant card might actually be a very bad idea). So move the grant card.  3 > My question at this point is, have I followed thet6 > proper serpentine configuration, or is it considered > discontiguous?  2 OK up to the CMD220 but still move the grant card.  6 > I wanted to ask this before I delved deeper into the- > issue of setting CSRs and vectors at Hoff'si2 > suggestion. Also- Hoff if you are out there, you6 > referred to setting "The CSR and vector on one Q-bus4 > module and add it into the first slot." How do you7 > check what these are set to? When all of the ICP1622SO  4 The bad thing your friend did was break your system.- The good thing is that (unless you are really 2 unlucky and have a very early version) your KA650 + will understand a few extra commands at thea >>> prompt:w   >>> SHOW DEVICEi  ' will show which devices the CPU can seer' and identify (and hence the chances are.% good that the CSRs are set correctly)y  
 >>> SHOW QBUSt  % will show which devices respond wheret( in the QBUS address space ... useful for+ spotting where things are when you have not  got things set quite right.e  
 >>> CONFIG  ! Useful for working out what your k  CSRs should be. In your case you need something like:  
 >>> CONFIG QNA,1s TK50,1 EXIT  ' and it will list the settings you need.c& (I don't have a KA65x box here to test& so this is from memory ... it might be& control-Z to terminate the list and it might not be TK50 ... it might be TQK50 ... CONFIG isn't that forgiving if you give it the wrong name).  & If your CMD220 emulates an MSCP device' (I have no idea whether it does or not)u! then you will need to tell CONFIG & that you have an MSCP disk controller: UDA50,1I
 should do.   IIRC, DELQA, TQK70 and 1st MSCPn" device do not tread on each others! toes anyway so it is quite likely ) that the default config you have is fine.t  $ I have no idea what the ISP card is.  4 > are in, they represent (if I have read the jumpers2 > correctly) CSRs 0-7 (000 through 111). How can I7 > verify what they actually read when they are set intol > the slots?   >>> SHOW DEVICEe and  >>> SHOW QBUS  as described above.i   > 8 > Is it a command you can issue at the terminal once the1 > firmware is powered up? I can issue a configureN > command with the KA655 CCL.m >    Yup.    2 > I know this is a lot to digest. The power supply- > and tape drive have been verified not to beo/ > the problem. Any and all suggestions welcome.'  " The output from those two commands  should help pinpoint whether the QBUS config is OK.  3 Beyond that it usually helps to have the controllero1 cable plugged in at both ends, a bootable tape int0 the drive and the drive powered up. Don't laugh,. I've done all of these before ... and no doubt I'll do them all again :-)   Antonio.   --     ---------------n- Antonio Carlini             arcarlini@iee.org    ------------------------------  # Date: Mon, 29 Jan 2001 20:25:32 GMT*2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)G Subject: Re: TQK70 controller in serpentine config, q-bus cfg questions*7 Message-ID: <0lkd6.177$cu.1398@gazette.loc1.tandem.com>w  Z In article <G7xJxJ.H3t@freenet.buffalo.edu>, ag079@freenet.buffalo.edu (NS Gaffin) writes:5 :I have looked at the "Ask the Wizard" section in thec4 :openvms.compaq site- searching for "serpentine". At2 :the moment,5 documents come up: 3064, 3232, 3264,4 :5219, and 6136pro_009. Am I missing any recommended
 :reading?   (   Yep.  1149 is one you want to look at.  + :What exactly does serpentine suggest? Thate6 :the bus goes from Slot1 A-B to Slot1 C-D to Slot2 C-D' :to Slot2 A-B to Slot3 A-B, etc.... ?? e  D   You have Q22 slots and you have CD slots.  Q22/CD slots are wired A   straight across, while Q22/Q22 slots are wired in a serpentine -
   fashion.    ( :But here is a little background anyway:6 :Currently, I am trying to boot a system using a KA6555 :CPU using a tape drive (TK70) that contains VMS 5.4.s6 :As far as I know, this was successful two months ago.7 :(The CPU was KA630, and then one of my colleagues came . :in and unceremoniously swapped the CPU cards-1 :nevertheless, the same problem occurs- that of aa& :CTRLERR, after a successful power up.  H   This would imply that your colleague messed up the Q-bus, in addition I   to replacing the KA630 processor and MS630 memory with KA655 and MS650 o&   memory, and swapping the CPU cables.  6 :Also, I have seen what a flip chip is, and where they7 :were in my TK70 boot problem. I presume these are also  :known as grant cards?  I   There are several types of grant cards...  Dual-width I/O cards can be oH   used, there are short grant cards (rare, if you find one make SURE it J   is a Q-bus and not a UNIBUS grant), and there are the better full-depth J   purple-handled dual-width Q-bus grant cards -- with M9047 on the handle.?   There are also black air-flow/spacer modules, with no etch...1   :My current situation is this: :s
 :			A-B			C-D  :a" :Slot 1      	    M7625-BA (KA655) :Slot 2		    CMX-1651 (mem)i :Slot 3		M759 (TQK70)	Flip Chip-	          -B   Remove the grant card, that is *PROBABLY* incorrectly installed.   :Slot 4		M7516 (DELQA)	EMPTY!j :Slot 5		CQD220		EMPTY!n :Slot 6		    ICP1622S COMM :Slot 7-15		EMPTY.		EMPTY.  F   Fifteen slots?  This is not a standard enclosure I recognize...  AndG   since I do not see the extender modules listed, this does not appear a'   to be two connected Q-bus enclosures.u   .. :...have I followed the 5 :proper serpentine configuration, or is it consideredm :discontiguous?u  C   I have no idea if this Q-bus is correctly configured, as I do nota?   recognize the enclosure.  This could be most any pattern from"F   Q22/CD across to a combination of Q22/Q22 and Q22/CD to all Q22/Q22.  >   Here is the terse version of the "classic" Q-bus enclosures.  :         BA23:  Q22-CD for 3 slots, then Q22-Q22 serpentine:         BA123: Q22-CD for 4 slots, then Q22-Q22 serpentine,         BA213: Q22-CD for all 12 Q-bus slots+         BA215: Q22-CD for all 6 Q-bus slotsw,         BA430: Q22-CD for all 10 Q-bus slots+         BA440: Q22-CD for all 7 Q-bus slotsh  J   But since I do not recognize the box, all bets on the pattern are off...  8 :...I wanted to ask this before I delved deeper into the, :issue of setting CSRs and vectors at Hoff's1 :suggestion. Also- Hoff if you are out there, you 5 :referred to setting "The CSR and vector on one Q-busn3 :module and add it into the first slot." How do you  :check what these are set to?   D   With the manuals for the module.  No, AFAIK, there is no site withG   all of the various Q-bus module manuals or settings available on-lineKD   anywhere.  (There are scattered listings, but I know of know largeG   listings -- if somebody does know of such a site, please let me know E#   and I'll add the URL to the FAQ.)    :When all of the ICP1622Sn3 :are in, they represent (if I have read the jumperse1 :correctly) CSRs 0-7 (000 through 111). How can IN6 :verify what they actually read when they are set into :the slots?-  8   I have no idea what an ICP1622S is, nor how to set it.  +   I will assume it is a third-party module.>  7 :Is it a command you can issue at the terminal once the-0 :firmware is powered up? I can issue a configure :command with the KA655 CCL.  3   See topic 1149 for the last (lengthy) discussion.e   :I have placed pictures at:e/ :http://freenet.buffalo.edu/~ag079/chassis.htmle ..  C   I can't readily identify the chassis from the picture.   With thenH   reported fifteen Q-bus slots, it is not a Q-bus enclosure I recognize.I   (It is also apparently not a UNIBUS enclosure, given the lack of extra k6   space next to what is apparently the CPU module. :-)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------    Date: 30 Jan 2001 05:52:34 +0200+ From: Erno Palonheimo <esp@theremin.hut.fi> U Subject: VAX 6310, VAX 6320, HSC50 and a couple of SA600's - how to plug'em together? - Message-ID: <uth3de14ry5.fsf@theremin.hut.fi>d  
 Greetings,  D As a hardcore VAX hobbyist I've got a CI-based VAXcluster consistingD of a VAX 6310, a VAX 6320 and a HSC50.  Now I'm about to get hold ofC some SA600 disk racks - which would be extremely nice as my clustersA doesn't have any of its original disks left. :-( Now, which stuff.E plugs where?  I've understood that the VAX machines and the HSC50 canu@ be connected together using the BNC cables, but do I need a star= coupler?  Should the SA600 racks be connected to the HSC50 or0B someplace else?  I'm only used to DSSI VAXclusters, CI seems to goA beyond my comprehension (I haven't got any documentation...).  If A someone could give me a pointer to good documentation, or explainc> things briefly, I'd appreciate that a lot.  Thanks in advance.   --  A Erno Palonheimo ; esp@iki.fi ; http://iki.fi/esp/ ; +358505604765a   ------------------------------  # Date: Mon, 29 Jan 2001 23:38:00 GMTd From: Dirk Munk <munk@home.nl>- Subject: very sloooooow Glasfibre performance.' Message-ID: <3A75FED6.11C8BF81@home.nl>t  F We are experiencing a very disturbing performance problem with our new HSG80 setup.  D We have a ES40 connected to a HSG80 cabinet (by means of a switch of course).> The ES40, the switch and the HSG80 are not in use by any other program/system etc.a  ? When we restore a DBMS 7 (codasyl) database, the performance isaF miserable. We get about 30 - 40 IO's /second and see a queue length of about 5 on the disks.i? The whole procedure takes appr. 3 - 4 hr.. In contrast the samesE procedure on another (slower) ES40 takes appr. 30 minutes (to a Mylex  raid controller).i  G Of course writeback cache is enabled, and the max. cached transfer size  is 128 or higher.e  E We see the same problem on 2 ES40's, with their own switch and HSG80.   E We use VMS 7.2-1 with all applicable patches for fibre and shadowing.e   Has anyone seen this before ??   Regards,   Dirk   ------------------------------  % Date: Tue, 30 Jan 2001 00:01:20 +0000a% From: Alan Greig <a.greig@virgin.net>r1 Subject: Re: very sloooooow Glasfibre performanceh* Message-ID: <3A760450.4A8F3430@virgin.net>   Dirk Munk wrote:  H > We are experiencing a very disturbing performance problem with our new > HSG80 setup. >yF > We have a ES40 connected to a HSG80 cabinet (by means of a switch of
 > course).@ > The ES40, the switch and the HSG80 are not in use by any other > program/system etc.M > A > When we restore a DBMS 7 (codasyl) database, the performance isyH > miserable. We get about 30 - 40 IO's /second and see a queue length of > about 5 on the disks.uA > The whole procedure takes appr. 3 - 4 hr.. In contrast the sameaG > procedure on another (slower) ES40 takes appr. 30 minutes (to a Mylext > raid controller).d >o  K I can tell you that on an ES40 /HSZ80 combination  DBMS 7 is lightning fastyJ at backups and restores We looked at the HSG80 but went for the Ultra-SCSIK variant. Partly because of possible teething problems and also that even in K a mirrored triple stripe set combination the disks would tend to be more oftJ a limitation than the dual busses for our application.  I hope you can get a quick solution.v   >aI > Of course writeback cache is enabled, and the max. cached transfer sizec > is 128 or higher.i >eG > We see the same problem on 2 ES40's, with their own switch and HSG80.w > G > We use VMS 7.2-1 with all applicable patches for fibre and shadowing.  >s  > Has anyone seen this before ?? >w
 > Regards, >n > Dirk   --
 Alan Greig   ------------------------------  % Date: Mon, 29 Jan 2001 21:47:34 -0600c+ From: "Main, Kerry" <Kerry.Main@compaq.com> 1 Subject: RE: very sloooooow Glasfibre performance.N Message-ID: <910612C07BCAD1119AF40000F86AF0D805284CE6@kaoexc3.kao.cpqcorp.net>   Dirk,s  C >>> When we restore a DBMS 7 (codasyl) database, the performance isoL miserable. We get about 30 - 40 IO's /second and see a queue length of aboutJ 5 on the disks. The whole procedure takes appr. 3 - 4 hr.. In contrast theJ same procedure on another (slower) ES40 takes appr. 30 minutes (to a Mylex raid controller). <<<a  D Fyi, 30-40 IO's per second is a pretty small number and something isD definately wrong if you are seeing a queue length of 5 on individual$ spindles with this number of IO's ..  K One thought does come to mind - are you restoring to a newly inited disk or H to (or from?) one that is low in space and perhaps extremely fragmented?  8 Are the same symptoms seen when restoring to a new disk?  6 Is the restore from disk to disk or from tape to disk?   Regards,  
 Kerry Main Senior Consultantp Compaq Canada Inc. Professional Servicesi Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----% From: Dirk Munk [mailto:munk@home.nl]- Sent: January 29, 2001 6:38 PM To: Info-VAX@Mvb.Saic.Com - Subject: very sloooooow Glasfibre performances    F We are experiencing a very disturbing performance problem with our new HSG80 setup.  D We have a ES40 connected to a HSG80 cabinet (by means of a switch of course).> The ES40, the switch and the HSG80 are not in use by any other program/system etc.|  ? When we restore a DBMS 7 (codasyl) database, the performance is F miserable. We get about 30 - 40 IO's /second and see a queue length of about 5 on the disks. ? The whole procedure takes appr. 3 - 4 hr.. In contrast the same E procedure on another (slower) ES40 takes appr. 30 minutes (to a Mylexc raid controller).e  G Of course writeback cache is enabled, and the max. cached transfer size  is 128 or higher.   E We see the same problem on 2 ES40's, with their own switch and HSG80.   E We use VMS 7.2-1 with all applicable patches for fibre and shadowing.i   Has anyone seen this before ??   Regards,   Dirk   ------------------------------  # Date: Tue, 30 Jan 2001 05:47:58 GMTC From: Dirk Munk <munk@home.nl>1 Subject: Re: very sloooooow Glasfibre performance-' Message-ID: <3A76558D.8E1C1CFA@home.nl>    "Main, Kerry" wrote:   > Dirk,i >eE > >>> When we restore a DBMS 7 (codasyl) database, the performance isuN > miserable. We get about 30 - 40 IO's /second and see a queue length of aboutL > 5 on the disks. The whole procedure takes appr. 3 - 4 hr.. In contrast theL > same procedure on another (slower) ES40 takes appr. 30 minutes (to a Mylex > raid controller). <<<r > F > Fyi, 30-40 IO's per second is a pretty small number and something isF > definately wrong if you are seeing a queue length of 5 on individual& > spindles with this number of IO's ..  O It is not to 1 spindle, but to a stripe set of 2 mirrorsets (4 spindles total).(   > M > One thought does come to mind - are you restoring to a newly inited disk or J > to (or from?) one that is low in space and perhaps extremely fragmented? >S: > Are the same symptoms seen when restoring to a new disk?  # The disks are new and almost empty.h   > 8 > Is the restore from disk to disk or from tape to disk?  % From internal SCSI disk to fibredisk.    >l >.
 > Regards, >a > Kerry Main > Senior Consultant  > Compaq Canada Inc. > Professional Services2 > Voice: 613-592-4660e > Fax  :  819-772-7036 > Email: Kerry.Main@Compaq.com >i > -----Original Message-----' > From: Dirk Munk [mailto:munk@home.nl]   > Sent: January 29, 2001 6:38 PM > To: Info-VAX@Mvb.Saic.Coml/ > Subject: very sloooooow Glasfibre performance, > H > We are experiencing a very disturbing performance problem with our new > HSG80 setup. >iF > We have a ES40 connected to a HSG80 cabinet (by means of a switch of
 > course).@ > The ES40, the switch and the HSG80 are not in use by any other > program/system etc.) >nA > When we restore a DBMS 7 (codasyl) database, the performance ishH > miserable. We get about 30 - 40 IO's /second and see a queue length of > about 5 on the disks. A > The whole procedure takes appr. 3 - 4 hr.. In contrast the sameiG > procedure on another (slower) ES40 takes appr. 30 minutes (to a Mylexh > raid controller).v > I > Of course writeback cache is enabled, and the max. cached transfer size) > is 128 or higher.i > G > We see the same problem on 2 ES40's, with their own switch and HSG80.t >fG > We use VMS 7.2-1 with all applicable patches for fibre and shadowing.  >h  > Has anyone seen this before ?? >y
 > Regards, >  > Dirk   ------------------------------  % Date: Mon, 29 Jan 2001 23:44:12 +0000p) From: Christof Brass <brass@infopuls.com>d* Subject: Re: VMS actually mentioned but..., Message-ID: <3A76004C.B2A9D78D@infopuls.com>   Jordan Henderson wrote:d > , > In article <3A74ABA4.8D15AC94@virgin.net>,) > Alan Greig  <a.greig@virgin.net> wrote:  > >rM > >Ok, I've watched the second Capellas speech now. Took about three hours ofuL > >video drops before I got there and he does mention VMS but...What he saidK > >was that OpenVMS had a large installed base in healthcare and they wouldsG > >especially work to keep those accounts. They might try to expand thevH > >market but "there are reasons to be cautious. Especially in the first > >quarter"t > > I > >He pointedly did not, as I hear it, commit to anything solid about the(J > >future of VMS. He specifically said "there are reasons to be cautious".K > >Perhaps lower than expected VMS sales are causing problems but I thought = > >they were up? Sounded like he was implying something more.o > >nJ > >The VMS mention was the very last item in his speech. Tagged on?  And IG > >suspect he knew we were looking for a message. He told us "there aresL > >reasons to be cautious". If VMS sales have plummeted then the decision to > >retire it becomes easier. > >oI > >It does sound as if Capellas might be convinced VMS could grow but the M > >majority of the rest of management want rid of it. In any case I think now J > >is the time to make it as clear as possible to Compaq that we are happyK > >with VMS and will continue to buy it but only if Compaq give us some farnM > >more positive news. I really can't recommend a new installation right now.nF > >It reminds me of just before Alpha/NT was dropped. Doesn't mean theM > >outcome will be the same but I think now *is* the time to make some noise.d > > K > >Last year, based on strong assurances, I pushed for VMS on Alpha instead D > >of just running the VAXes into the ground while moving to anotherL > >platform. Several European sites have migrated without problem and we areJ > >now looking at a much larger migration in the US. However our corporateM > >head office (in Houston funnily enough) seems almost immune to any pro-VMS6M > >literature saying that we "don't see the big picture" and "that VMS really J > >does not have a future" Now there are also plenty of people arguing forK > >the switch but right at the moment I'm not confident enough to be one oft > >them. > >B > E > If I've said anything to make people feel that I'm unsympathetic toaE > their problems selling VMS internally, let me make it clear tht I'mv
 > not at all.o > C > I've worked with management who disparage OpenVMS for the reasonsaC > given above ("big picture", "VMS has no future").  I'm not asking  > anyone to be unrealistic.  > A > I'm hoping someone inside Compaq will bring Mr. Greig's post toMC > the attention of Compaq management at the highest levels.  If hisTE > organization abandons VMS, there is absolutely no reason to believe + > that this could be a positive for Compaq.  > C > One of the worst problems I see for OpenVMS, as I've said before,(F > is the brain drain, good people finally being convinced that OpenVMSB > has no future and moving on to work other fields.  Compaq may beD > spurring this along with their inattentiveness.  What Compaq mightF > not recognize that the people they most need to keep involved in VMS@ > are the people who can read-between-the-lines most accurately. > @ > As Bill Todd points out, IBM has no problem selling Wintel ANDD > AS/400, MVS, AIX aggressively.  Compaq should want to be more like% > IBM and less like Dell and Gateway., >  > >--Q
 > >Alan Greigr > >  >  > -Jordan Henderson. > jordan@greenapple.coml  ; Well spoken, but don't trust IBM! Never forget what they tou9 OS/2! It's very similar to what ashole Palmer did to VMS.V  + Ceterum censeo Robert Palmer esse delendam.l   ------------------------------  % Date: Mon, 29 Jan 2001 21:17:27 -0600 + From: "Main, Kerry" <Kerry.Main@compaq.com>-* Subject: RE: VMS actually mentioned but...N Message-ID: <910612C07BCAD1119AF40000F86AF0D805284CE4@kaoexc3.kao.cpqcorp.net>   Alan,e  5 First let me don my hard hat and armour .. Ok.. done.6  G These threads have been interesting to watch - similar to a small matchgH accidentally starting a grass fire and watching it grow so quickly .. oh well...n  @ >>> He specifically said "there are reasons to be cautious". <<<  < MC was talking about the industry in general - not OpenVMS.   $ Here is another external reference: 5 <http://news.cnet.com/news/0-1003-201-4614848-0.html> E Quote: "Capellas also used the meeting to reiterate expectations of a - slowdown in the first six month of the year.    F "There's absolutely no surprise there will be continued price pressure% throughout the first half," he said. u  F European sales and most of Compaq's large global accounts will be less2 affected than U.S. sales, he predicted." End quote  H So, while I will not say the earnings report could not have been handledI better, or that current OpenVMS marketing could not be improved, but likenC Terry has said, a number of MC misquotes are being blown way out of  proportion.   K Gartner has recently given the best review and new oulook for OpenVMS in 10e years.  : Terry mentioned OpenVMS had single digit growth last year.  8 Recheck the big Compaq press releases from last October:; <http://www5.compaq.com/newsroom/pr/2000/pr2000100301.html> A <http://www.openvms.compaq.com/ebusiness_without_compromise.html>a  H Also, Compaq had many NEW OpenVMS Customer wins last year. No, all theirK names are not public, although Eurex, Sydney Exchange, ISE went live in Mayh# etc are public new OpenVMS wins... s  cK So, while I can appreciate the long simmering distrust and frustration that G might be still purcolating within certain area's of the OpenVMS base, ICJ would suggest that folks not read more into these recent events than there
 really is.   Regards,  
 Kerry Main Senior ConsultantQ Compaq Canada Inc. Professional Services  Voice: 613-592-4660e Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----, From: Alan Greig [mailto:a.greig@virgin.net] Sent: January 28, 2001 6:31 PM To: Info-VAX@Mvb.Saic.Come& Subject: VMS actually mentioned but...      J Ok, I've watched the second Capellas speech now. Took about three hours ofI video drops before I got there and he does mention VMS but...What he saidIH was that OpenVMS had a large installed base in healthcare and they wouldD especially work to keep those accounts. They might try to expand theE market but "there are reasons to be cautious. Especially in the first  quarter"  F He pointedly did not, as I hear it, commit to anything solid about theG future of VMS. He specifically said "there are reasons to be cautious".tH Perhaps lower than expected VMS sales are causing problems but I thought: they were up? Sounded like he was implying something more.  G The VMS mention was the very last item in his speech. Tagged on?  And ItD suspect he knew we were looking for a message. He told us "there areI reasons to be cautious". If VMS sales have plummeted then the decision to  retire it becomes easier.r  F It does sound as if Capellas might be convinced VMS could grow but theJ majority of the rest of management want rid of it. In any case I think nowG is the time to make it as clear as possible to Compaq that we are happy H with VMS and will continue to buy it but only if Compaq give us some farJ more positive news. I really can't recommend a new installation right now.C It reminds me of just before Alpha/NT was dropped. Doesn't mean theoJ outcome will be the same but I think now *is* the time to make some noise.  H Last year, based on strong assurances, I pushed for VMS on Alpha insteadA of just running the VAXes into the ground while moving to another.I platform. Several European sites have migrated without problem and we aretG now looking at a much larger migration in the US. However our corporaterJ head office (in Houston funnily enough) seems almost immune to any pro-VMSJ literature saying that we "don't see the big picture" and "that VMS reallyG does not have a future" Now there are also plenty of people arguing forsH the switch but right at the moment I'm not confident enough to be one of them.e   --
 Alan Greig   ------------------------------  % Date: Mon, 29 Jan 2001 23:55:36 -0500l- From: JF Mezei <jfmezei.spamnot@videotron.ca>S* Subject: Re: VMS actually mentioned but..., Message-ID: <3A764942.984DFAB0@videotron.ca>   "Main, Kerry" wrote:E > Terry has said, a number of MC misquotes are being blown way out of 
 > proportion.   ? It isn't so much what he said, but rather what he hasn't said. r  K WHY ?, WHY ? I ask, would a company who has decided to give a product whichhH was neglected under a previous administration a new life, not take every+ possibility to brag about the turn around ?k  N We're not talking about 30 second VMS ads during superbowl here. We're talkingE about free advertising during interviews with the press and financiale  statement telephone conferences.  K I doubt that Compaq's top management is so stupid as to "forget" to mention N VMS. Do they delegate the speech writing to some critters who are not aware ofM the need to mention VMS and by the time they get the speech, it is too late ?n  M Or perhaps Microsoft's WORD has a special worm in it to delete all referencesb to VMS in documents ?h    M > Gartner has recently given the best review and new oulook for OpenVMS in 10r > years.  L That was in June of last year when Compaq had started to make noises that itL might give VMS a new chance. If Gartner were to rewrite that report today, IM think that they would be forced to add some caution about VMS since Compaq isu again silent about VMS.l  < > Terry mentioned OpenVMS had single digit growth last year.  N Again, if VMS turned around and started to grow again, you'd think that CompaqH would have been bragging about turning around one of the more profitable& products. But no. Nada, no word. WHY ?   ------------------------------  # Date: Tue, 30 Jan 2001 06:29:47 GMTe- From: Terry C Shannon <shannon@world.std.com> * Subject: Re: VMS actually mentioned but...D Message-ID: <Pine.SGI.4.21.0101300125280.29821-100000@world.std.com>  * On Mon, 29 Jan 2001, Christof Brass wrote: <snip> > = > Well spoken, but don't trust IBM! Never forget what they tor; > OS/2! It's very similar to what ashole Palmer did to VMS.l >   F I seem to recall that Microsoft wrote OS/2 for IBM. IBM wanted to makeJ OS/2 a multiplatform OS but neglected to specify this in the contract withI Microsoft. Microsoft then delivered an IA-32 specific OS. IBM complained,tI Microsoft's lawyers pointed out that there was no multiplatform caveat int
 the contract.l  E And so we ended up with the first instantiation of Microsoft Businesso+ Partner as Organ Donor. But not the last...p   ------------------------------  # Date: Mon, 29 Jan 2001 21:30:45 GMTe" From: fooguy <jweisen@my-deja.com>& Subject: Volume Shadowing: 1, fooguy:0) Message-ID: <954ne6$hnj$1@nnrp1.deja.com>u  " 1) Install license PAK from Compaq3 2) Modify SYSGEN params for SHADOWING (set it to 2)c	 3) Rebootr/ 4) MOUNT DSA0:/SHADOW=(DKB200:,DKC0:) SHADOWVOLm  H And I get an error that it can't make either disk a member of the shadow set.  C Both disks are brand spanking new and freshly init'ed. Am I missings  something, or is it just Monday?   --- *********************************************-( "All I every wanted from life was to see, Larry Wall give Bill Gates a Perl Necklace."     Sent via Deja.com: http://www.deja.com/   ------------------------------    Date: 29 Jan 2001 16:48:21 -0500* From: young_r@eisner.decus.org (Rob Young)* Subject: Re: Volume Shadowing: 1, fooguy:0+ Message-ID: <qDnA4dTOQHHT@eisner.decus.org>   N In article <954ne6$hnj$1@nnrp1.deja.com>, fooguy <jweisen@my-deja.com> writes:$ > 1) Install license PAK from Compaq5 > 2) Modify SYSGEN params for SHADOWING (set it to 2)  > 3) Reboott1 > 4) MOUNT DSA0:/SHADOW=(DKB200:,DKC0:) SHADOWVOL0 > J > And I get an error that it can't make either disk a member of the shadow > set. > E > Both disks are brand spanking new and freshly init'ed. Am I missing " > something, or is it just Monday? >   8 What is the error message?  Here is what my script does:  M $       mount_command = "MOUNT/NOREBUILD/NOASSIST ''mount_type' ''shadow_set' 0 /SHADOW=(''shadow_members') ''label' ''logical'") $               show symbol mount_command  $               'mount_command'C  ' local symbols to support the above are:l   mount_type = "/SYSTEM" shadow_set = "DSA232:"( shadow_members = "$7$DUA740:,$7$DUA741:" label = "DSK_LBL1" logical = labelM  F Where the above are pulled from a flat text file and massaged into the6 symbols.  Not necessary but trying to give background.  2 VMSNOD   | DSA232:,$7$DUA740:,$7$DUA741:| DSK_LBL1   ---r  ; 	A guess might be that the labels don't match.  But if they  	do, show us the error message.i   				Robr   ------------------------------  % Date: Mon, 29 Jan 2001 16:53:11 -0500a- From: "Peter Weaver" <peter.weaver@stelco.ca>a* Subject: Re: Volume Shadowing: 1, fooguy:04 Message-ID: <zDld6.124612$Z2.1566494@nnrp1.uunet.ca>  / "fooguy" <jweisen@my-deja.com> wrote in messager# news:954ne6$hnj$1@nnrp1.deja.com...t$ > 1) Install license PAK from Compaq5 > 2) Modify SYSGEN params for SHADOWING (set it to 2)l > 3) Rebootc1 > 4) MOUNT DSA0:/SHADOW=(DKB200:,DKC0:) SHADOWVOLe >r  F Did you also set the ALLOCLASS in MODPARAMS? I think you also need the  ALLOCLASS in your mount command;  9      $ MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) SHADOWVOLs     --   RULES OF THE AIR   -----------------e:  #21. It's always a good idea to keep the pointy end going"       forward as much as possible.   ------------------------------  # Date: Mon, 29 Jan 2001 22:30:50 GMTc" From: fooguy <jweisen@my-deja.com>* Subject: Re: Volume Shadowing: 1, fooguy:0) Message-ID: <954quq$l33$1@nnrp1.deja.com>y  - $MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) FOO   ? %MOUNT-F-NOMASTER, no master member found for this virtual unit E %MOUNT-I-SHDMEMFAIL, _$1$DKB200: failed as a member of the shadow sete- -SYSTEM-W-NOSUCHDEV, no such device availablee? %MOUNT-F-NOMASTER, no master member found for this virtual unityC %MOUNT-I-SHDMEMFAIL, _$1$DKC0: failed as a member of the shadow set - -SYSTEM-W-NOSUCHDEV, no such device availablet  4 In article <zDld6.124612$Z2.1566494@nnrp1.uunet.ca>,0   "Peter Weaver" <peter.weaver@stelco.ca> wrote:1 > "fooguy" <jweisen@my-deja.com> wrote in message % > news:954ne6$hnj$1@nnrp1.deja.com...t& > > 1) Install license PAK from Compaq7 > > 2) Modify SYSGEN params for SHADOWING (set it to 2)f
 > > 3) Reboot 3 > > 4) MOUNT DSA0:/SHADOW=(DKB200:,DKC0:) SHADOWVOLd > >  >hH > Did you also set the ALLOCLASS in MODPARAMS? I think you also need the" > ALLOCLASS in your mount command; >a; >      $ MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) SHADOWVOL  >n > -- >   RULES OF THE AIR >   -----------------f< >  #21. It's always a good idea to keep the pointy end going$ >       forward as much as possible. >  >-   --- *********************************************u( "All I every wanted from life was to see, Larry Wall give Bill Gates a Perl Necklace."     Sent via Deja.comr http://www.deja.com/   ------------------------------  % Date: Mon, 29 Jan 2001 18:20:51 -0500S% From: "John Vottero" <John@mvpsi.com>t* Subject: Re: Volume Shadowing: 1, fooguy:0/ Message-ID: <t7buntmd9dbl20@news.supernews.com>G  K If these are newly inited disks then I think VMS is trying to tell you that G it doesn't know which of the disks should be the master (should it copy * DKB200 to DKC0 or vis-a-versa).  Try this:  % $ MOUNT DSA0:/SHADOW=($1$DKB200:) FOO   # Then add the second disk like this:t  - $MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) FOOS    / "fooguy" <jweisen@my-deja.com> wrote in messagel# news:954quq$l33$1@nnrp1.deja.com...h/ > $MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) FOO  >sA > %MOUNT-F-NOMASTER, no master member found for this virtual unithG > %MOUNT-I-SHDMEMFAIL, _$1$DKB200: failed as a member of the shadow sett/ > -SYSTEM-W-NOSUCHDEV, no such device availablesA > %MOUNT-F-NOMASTER, no master member found for this virtual unit-E > %MOUNT-I-SHDMEMFAIL, _$1$DKC0: failed as a member of the shadow setB/ > -SYSTEM-W-NOSUCHDEV, no such device availablea > 6 > In article <zDld6.124612$Z2.1566494@nnrp1.uunet.ca>,2 >   "Peter Weaver" <peter.weaver@stelco.ca> wrote:3 > > "fooguy" <jweisen@my-deja.com> wrote in messagee' > > news:954ne6$hnj$1@nnrp1.deja.com... ( > > > 1) Install license PAK from Compaq9 > > > 2) Modify SYSGEN params for SHADOWING (set it to 2)  > > > 3) Rebooto5 > > > 4) MOUNT DSA0:/SHADOW=(DKB200:,DKC0:) SHADOWVOLg > > >l > >aJ > > Did you also set the ALLOCLASS in MODPARAMS? I think you also need the$ > > ALLOCLASS in your mount command; > >u= > >      $ MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) SHADOWVOL  > >t > > -- > >   RULES OF THE AIR > >   ----------------- > > >  #21. It's always a good idea to keep the pointy end going& > >       forward as much as possible. > >P > >s >w > --/ > *********************************************h* > "All I every wanted from life was to see. > Larry Wall give Bill Gates a Perl Necklace." >n >s > Sent via Deja.comn > http://www.deja.com/   ------------------------------  # Date: Mon, 29 Jan 2001 23:52:36 GMTr$ From: Ryan Moore <rmoore@san.rr.com>* Subject: Re: Volume Shadowing: 1, fooguy:0; Message-ID: <Pine.LNX.4.31.0101291547370.416-100000@jaipur>f  " On Mon, 29 Jan 2001, fooguy wrote:$ > 1) Install license PAK from Compaq5 > 2) Modify SYSGEN params for SHADOWING (set it to 2)  > 3) Rebooti1 > 4) MOUNT DSA0:/SHADOW=(DKB200:,DKC0:) SHADOWVOLe  I As mentioned earlier, make sure you system's allocation class isn't zero.sE Here's a log for me to set up a shadowset.  Are you sure both devicesl3 report the same number of disk blocks and geometry?l  % Neither devicse has been used before:o   $ show dev dga4310  E Device                  Device           Error    Volume         Free F  Name                   Status           Count     Label        Blocks. $1$DGA4310:   (xxxxxx)  Online               0   $ show dev dga5310  E Device                  Device           Error    Volume         FreenF  Name                   Status           Count     Label        Blocks. $1$DGA5310:   (xxxxxx)  Online               0  # $ init /system $1$dga4310: testshadr; $ mount /sys dsa90: /shad=($1$dga4310:,$1$dga5310) testshadS- %MOUNT-I-MOUNTED, TESTSHAD mounted on _DSA90:aH %MOUNT-I-SHDWMEMSUCC, _$1$DGA4310: (xxxxxx) is now a valid member of the
 shadow setJ %MOUNT-I-SHDWMEMCOPY, _$1$DGA5310: (xxxxxx) added to the shadow set with a copy operation $ sho dev dsa90:  E Device                  Device           Error    Volume         Free F  Name                   Status           Count     Label        BlocksF DSA90:                  Mounted              0  TESTSHAD      17768664B $1$DGA4310:   (xxxxxx)  ShadowSetMember      0  (member of DSA90:)F $1$DGA5310:   (xxxxxx)  ShadowCopying        0  (copy trgt DSA90:   0% copied)m   $ mcr sysgen SYSGEN>  SHOW SHADOWG Parameter Name           Current    Default     Min.      Max.     UniteG --------------           -------    -------    -------   -------   ----rL SHADOWING                       2          0         0          3 Coded-valuJ SHADOW_SYS_DISK                 0          0         0         -1 Bit_CodeF SHADOW_SYS_UNIT                 0          0         0       9999 UnitI SHADOW_MAX_COPY                 4          4         0        200 ThreadsmI SHADOW_MBR_TMO                120        120         1      65535 Seconds-I SHADOW_SYS_TMO                120        120         1      65535 SecondseI SHADOW_SYS_WAIT               256        256         1      65535 SecondseG SHADOW_REMOVE_1 "                "    "    "    "    "     "ZZZZ" AsciirG SHADOW_REMOVE_2 "                "    "    "    "    "     "ZZZZ" Asciio   -Ryand   ------------------------------  # Date: Mon, 29 Jan 2001 23:56:01 GMTa" From: fooguy <jweisen@my-deja.com>* Subject: Re: Volume Shadowing: 1, fooguy:0) Message-ID: <954vud$plm$1@nnrp1.deja.com>,  9 I tried to mount one member, and got the exact same error@  / In article <t7buntmd9dbl20@news.supernews.com>,u(   "John Vottero" <John@mvpsi.com> wrote:H > If these are newly inited disks then I think VMS is trying to tell you thatD > it doesn't know which of the disks should be the master (should it copy, > DKB200 to DKC0 or vis-a-versa).  Try this: >t' > $ MOUNT DSA0:/SHADOW=($1$DKB200:) FOOq >k% > Then add the second disk like this:l >a/ > $MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) FOOw >t1 > "fooguy" <jweisen@my-deja.com> wrote in message % > news:954quq$l33$1@nnrp1.deja.com... 1 > > $MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) FOOt > > C > > %MOUNT-F-NOMASTER, no master member found for this virtual unit E > > %MOUNT-I-SHDMEMFAIL, _$1$DKB200: failed as a member of the shadow  setr1 > > -SYSTEM-W-NOSUCHDEV, no such device availablegC > > %MOUNT-F-NOMASTER, no master member found for this virtual unit'G > > %MOUNT-I-SHDMEMFAIL, _$1$DKC0: failed as a member of the shadow sett1 > > -SYSTEM-W-NOSUCHDEV, no such device availablei > >e8 > > In article <zDld6.124612$Z2.1566494@nnrp1.uunet.ca>,4 > >   "Peter Weaver" <peter.weaver@stelco.ca> wrote:5 > > > "fooguy" <jweisen@my-deja.com> wrote in messagew) > > > news:954ne6$hnj$1@nnrp1.deja.com...e* > > > > 1) Install license PAK from Compaq; > > > > 2) Modify SYSGEN params for SHADOWING (set it to 2)e > > > > 3) Rebooty7 > > > > 4) MOUNT DSA0:/SHADOW=(DKB200:,DKC0:) SHADOWVOLc > > > >h > > >wH > > > Did you also set the ALLOCLASS in MODPARAMS? I think you also need thec& > > > ALLOCLASS in your mount command; > > > ? > > >      $ MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) SHADOWVOLr > > >  > > > -- > > >   RULES OF THE AIR > > >   -----------------i@ > > >  #21. It's always a good idea to keep the pointy end going( > > >       forward as much as possible. > > >a > > >a > >i > > --1 > > *********************************************M, > > "All I every wanted from life was to see0 > > Larry Wall give Bill Gates a Perl Necklace." > >  > >d > > Sent via Deja.comr > > http://www.deja.com/ >n >    --- ********************************************* ( "All I every wanted from life was to see, Larry Wall give Bill Gates a Perl Necklace."     Sent via Deja.comr http://www.deja.com/   ------------------------------  % Date: Mon, 29 Jan 2001 19:46:39 -0500o- From: JF Mezei <jfmezei.spamnot@videotron.ca>a* Subject: Re: Volume Shadowing: 1, fooguy:0, Message-ID: <3A760EE5.6C558357@videotron.ca>  
 fooguy wrote:W > / > $MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) FOOb > A > %MOUNT-F-NOMASTER, no master member found for this virtual units   Have you tried:   / MOUNT/SYSTEM DSA0: /SHADOW=($1$DKB200:) FOO   ?t   and then6 MOUNT/SYSTEM DSA0: /SHADOW=($1$DKB200:,$1$DKC0:) FOO ?  N I was under the impression that in the case of a new shadowset, if you provideM multiple drives, it is the first one which is used as the master. But perhapsrL these drives were in previous shadow sets in the past ? Forcing the buildingN of a shadow set with just one member would force that member to be the master.  M Also, are you able to mount each drive individually, outside of the shadowing ? software ? This resets the flag that it used to be a shadowset.i   ------------------------------  % Date: Mon, 29 Jan 2001 19:53:55 -0600b7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>o* Subject: Re: Volume Shadowing: 1, fooguy:0- Message-ID: <3A761EB3.3D34B3C2@earthlink.net>s  
 fooguy wrote:h > / > $MOUNT DSA0:/SHADOW=($1$DKB200:,$1$DKC0:) FOOt > A > %MOUNT-F-NOMASTER, no master member found for this virtual unittG > %MOUNT-I-SHDMEMFAIL, _$1$DKB200: failed as a member of the shadow setp/ > -SYSTEM-W-NOSUCHDEV, no such device availablelA > %MOUNT-F-NOMASTER, no master member found for this virtual unitaE > %MOUNT-I-SHDMEMFAIL, _$1$DKC0: failed as a member of the shadow set / > -SYSTEM-W-NOSUCHDEV, no such device availableo   Ok. Try:   $ SHO DEV DK  D If the disks you're trying to MOUNT don't show up in that list, then> there's your problem (post the results, if need be). Also, ...  * $ WRITE SYS$OUTPUT F$GETSYI( "ALLOCLASS" )  D If it doesn't return a value of "1", you've still got to set this in- MODPARAMS, then run AUTOGEN again and reboot.h  E Also, as John V. suggests, the only way I've ever seen this work withhF newly INITIALIZEd disks is to MOUNT the first disk into the shadow-setG to establish it, then add the second member (you can have up to three).0, It'll copy in until the members are in sync.   -- i David J. Dachterac dba DJE Systemsl http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.i   ------------------------------  # Date: Tue, 30 Jan 2001 00:31:16 GMT " From: fooguy <jweisen@my-deja.com>, Subject: Re: Volume Shadowing: 131, fooguy:1) Message-ID: <95520l$rhl$1@nnrp1.deja.com>r  
 YEAHHHH!!!  E Ok, we got the shadow set up and running. We're in the full copy now.e   Soon, the hurting begins.e   Thanks Everyone, John    ; In article <Pine.LNX.4.31.0101291547370.416-100000@jaipur>,    <rmoore2@san.rr.com> wrote: $ > On Mon, 29 Jan 2001, fooguy wrote:& > > 1) Install license PAK from Compaq7 > > 2) Modify SYSGEN params for SHADOWING (set it to 2)o
 > > 3) Reboott3 > > 4) MOUNT DSA0:/SHADOW=(DKB200:,DKC0:) SHADOWVOLw >dE > As mentioned earlier, make sure you system's allocation class isn'ta zero.CG > Here's a log for me to set up a shadowset.  Are you sure both devices 5 > report the same number of disk blocks and geometry?t >e' > Neither devicse has been used before:u >M > $ show dev dga4310 > G > Device                  Device           Error    Volume         FreeaH >  Name                   Status           Count     Label        Blocks0 > $1$DGA4310:   (xxxxxx)  Online               0 >i > $ show dev dga5310 > G > Device                  Device           Error    Volume         Free-H >  Name                   Status           Count     Label        Blocks0 > $1$DGA5310:   (xxxxxx)  Online               0 >e% > $ init /system $1$dga4310: testshad.= > $ mount /sys dsa90: /shad=($1$dga4310:,$1$dga5310) testshad2/ > %MOUNT-I-MOUNTED, TESTSHAD mounted on _DSA90:oF > %MOUNT-I-SHDWMEMSUCC, _$1$DGA4310: (xxxxxx) is now a valid member of thea > shadow setE > %MOUNT-I-SHDWMEMCOPY, _$1$DGA5310: (xxxxxx) added to the shadow set2 with a > copy operation > $ sho dev dsa90: >mG > Device                  Device           Error    Volume         FreeoH >  Name                   Status           Count     Label        BlocksH > DSA90:                  Mounted              0  TESTSHAD      17768664D > $1$DGA4310:   (xxxxxx)  ShadowSetMember      0  (member of DSA90:)H > $1$DGA5310:   (xxxxxx)  ShadowCopying        0  (copy trgt DSA90:   0%	 > copied)  >a > $ mcr sysgen > SYSGEN>  SHOW SHADOW@ > Parameter Name           Current    Default     Min.      Max. UnitB > --------------           -------    -------    -------   ------- ----C > SHADOWING                       2          0         0          3o
 Coded-valuC > SHADOW_SYS_DISK                 0          0         0         -1t Bit_CodeH > SHADOW_SYS_UNIT                 0          0         0       9999 UnitC > SHADOW_MAX_COPY                 4          4         0        200n Threads-C > SHADOW_MBR_TMO                120        120         1      65535  Seconds C > SHADOW_SYS_TMO                120        120         1      65535a Seconds:C > SHADOW_SYS_WAIT               256        256         1      65535J Seconds6C > SHADOW_REMOVE_1 "                "    "    "    "    "     "ZZZZ": Ascii C > SHADOW_REMOVE_2 "                "    "    "    "    "     "ZZZZ"  Asciit >e > -Ryanh >1 >r   --- ********************************************* ( "All I every wanted from life was to see, Larry Wall give Bill Gates a Perl Necklace."     Sent via Deja.comM http://www.deja.com/   ------------------------------  # Date: Tue, 30 Jan 2001 03:58:47 GMTr$ From: Ed Wilts <ewilts@mediaone.net>, Subject: Re: Volume Shadowing: 131, fooguy:1< Message-ID: <XZqd6.45304$lh.2751293@typhoon.mn.mediaone.net>  
 fooguy wrote:   G > Ok, we got the shadow set up and running. We're in the full copy now.   F The hurting begins when you forget to monitor your OPCOM messages for I shadow transitions.  If a shadow member fails, it will be dropped out of  F the shadow set and you'll get copious messages in your operator log.  I However, many people forget to monitor these and find themselves running 7H with a single-member set for days on end.  Then, when the second member   fails, you'll hurt really badly.   MONITOR YOUR SHADOW SETS!b           .../Ed     ------------------------------  # Date: Tue, 30 Jan 2001 00:39:55 GMT:" From: ltx_tech_support@my-deja.com Subject: Re: What succeeds TSM? ) Message-ID: <9552gr$rv3$1@nnrp1.deja.com>e  F Lantronix actually makes devices that were designed for implementationC such as yours.  They have terminal servers that support LAT and RASjG Servers that you can use to dial into your network with.  You can check4? them out at http://www.lantronix.com/products/mds/ets16pr_32pr/s     Sent via Deja.com@ http://www.deja.com/   ------------------------------  # Date: Tue, 30 Jan 2001 02:04:59 GMTN  From: snead_george_w@my-deja.comK Subject: Re: Whereabouts of Ranalyzer shareware linked for alpha processor?A( Message-ID: <9557g6$48$1@nnrp1.deja.com>  < Now that I've experienced the great usefulness of navigating; a large application by deriving html from the source files,E> I'd like to refine it a step further.  "Ranalyzer" has its own: parser with a limited capability of finding functions in a= source file.  Why not turn directly to the real authority  --06 the compiler itself as expressed through its listing!!  E A DecC compiler listing "cc/list/show=all" is a marvel of completnessmC about the code just compiled.  Such a listing can easily run 50,000S blocks.a  : Does anyone know of any awk,perl, ... tools for processing; a DecC compiler listing to generate HTML derivative listing-< for ease of navigation and tracking of function and variable definitions and references?r     Sent via Deja.comD http://www.deja.com/   ------------------------------  # Date: Mon, 29 Jan 2001 20:50:15 GMTv2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)3 Subject: Re: [VIM] VMS version doesn't do logicals?a7 Message-ID: <bIkd6.179$cu.1390@gazette.loc1.tandem.com>g   In article <rdeininger-2901011156110001@user-2ivec43.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) writes: < :I've offered once or twice to kit it up for the freeware.  I :I received no response, so I haven't made in progress in that direction.t  J   Apologies on my clear lack of responsiveness.  I had previously snagged G   a copy of VIM-SAVESET.ZIP dated 7-Feb-2000 for inclusion on the next EH   Freeware (V5), containing VIM.SAVESET also dated 7-Feb-2000.  Is that    the current/correct kit?  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   End of INFO-VAX 2001.059 ************************