1 INFO-VAX	Mon, 20 Aug 2001	Volume 2001 : Issue 461       Contents:E Re: Compaq destroys Storageworks (was Re: 7.3 system disk corruption) F Re: Could VMS be a desktop OS? (was: The Alpha Systems Customer Update Re: DEC-C float & double Re: DEC-C float & double	 DecWrite?  hello < How is the cache memory in an HSG80 divided and/or utilized?@ Re: How is the cache memory in an HSG80 divided and/or utilized?# Re: OT: TOPS-20 and TOPS-10 live on # Re: OT: TOPS-20 and TOPS-10 live on  Re: Run 3 phase PS on 2 phases?  Re: some SFF questions$ Re: Terry Shannon Tech Talk on Tru64 Re: The Final Knell Has Sounded  Re: The Final Knell Has Sounded  Re: The Final Knell Has Sounded  Re: The Final Knell Has Sounded  Re: The Final Knell Has Sounded  RE: tpcip upgrade  Re: tpcip upgrade ! Re: VAX/ALPHA FORTRAN and me! :-)  VSXXX-JA headphones  Re: VSXXX-JA headphones > Re: Wailing and moaning.... (was: The Final Knell Has Sounded)> Re: Wailing and moaning.... (was: The Final Knell Has Sounded)> Re: Wailing and moaning.... (was: The Final Knell Has Sounded) where is 7.3 ? Re: where is 7.3 ? Re: where is 7.3 ? Re: where is 7.3 ? Re: where is 7.3 ? RE: where is 7.3 ? Re: where is 7.3 ?  F ----------------------------------------------------------------------  # Date: Sun, 19 Aug 2001 12:01:56 GMT # From: Mark Hemker <hemker@home.com> N Subject: Re: Compaq destroys Storageworks (was Re: 7.3 system disk corruption)8 Message-ID: <sravnt0icvmenrl7np2sd0si6v3u8e9nk6@4ax.com>  E I upgraded a pair of HSG80s two weeks ago from V8.5F to V8.6F without F stopping anything.  VMS just failed over the paths from one controllerC to the other and then failed them all back to the first controller. + The entire procedure took all of 5 minutes.    Mark  A On Sun, 19 Aug 2001 07:44:23 GMT, Dirk Munk <munk@home.nl> wrote:    >  >  >Bart Zorn wrote:  > : >> "Simon Dodsley" <simon@sdinfotech.com> wrote in message7 >> news:3b7efb56_1@mk-nntp-1.news.uk.worldonline.com... / >> > Yeah, we all want to upgrade to HSG80's !!  >> >J >> > That wonderful bit of kit that, currently, can only have its firmwareN >> > upgraded by shutting down every system connected to it.  What a wonderfulL >> > way to keep your Wildfires running on 24x7 - Sorry Mr Business CriticalI >> > Customer, your 24x7 solution you paid an arm and a leg for has to be N >> > shutdown to upgrade the disk arrays. What's that you say, can't you do it >> on L >> > the fly like the old controllers? No, sorry, not yet, Compaq screwed up
 >> > again...  >> > >> >N >> This must be a new feature of the HSG80's. I have upgraded pairs of HSG80'sL >> without the application knowing it. Note: OpenVMS DID notice: some of theH >> disks failed over to another path. But the applications didn't blink. > J >We did an upgrade too abouth 2 months ago, without having to shutdown theK >systems. But we didn't do the upgrade to 8.6 yet, so I can't comment about  >that. >  >  >> >> >> Bart Zorn   ------------------------------  % Date: Mon, 20 Aug 2001 00:40:46 +0200 , From: Didier Morandi <Didier.Morandi@gmx.ch>O Subject: Re: Could VMS be a desktop OS? (was: The Alpha Systems Customer Update & Message-ID: <3B80406D.C31C9531@gmx.ch>   Fred Kleinsorge wrote: > D > Click on options from the login panel, and select the old desktop.   'twas not the question. ; I was talking about the configuration described in the FAQ.   E Actually I rebooted the station, which restarted DECW, and now I have 7 the normal welcome window with the huge |d|i|g|i|t|a|l|    D.   ------------------------------  # Date: Sun, 19 Aug 2001 18:54:40 GMT # From: ualski <ualski@earthlink.net> ! Subject: Re: DEC-C float & double - Message-ID: <3B800B54.CE9D34A1@earthlink.net>    JF Mezei wrote:  >  > Jesper Naur wrote:L > Now, how does the compiler know it should convert a float to a double in aM > routine such as printf with undefined arguments ? Does "printf" get special C > treatment because it is so closely associated with the language ?  > N > Or doers the compiler automatically convert any float to double when passing+ > to a routine whose argument are unknown ?   M It's the second way you mentioned.  The standard (at least used to) calls for L converting floats to double before passing their value to the subroutine.  IQ suppose that means they all get special treatment.  You can avoid the performance L hit associated with the conversion by making all your floats doubles or keep, them as floats and pass the address instead.   -- Aaron Sliwinski   ------------------------------  % Date: Sun, 19 Aug 2001 21:34:30 +0200 . From: "Jesper Naur" <jesper.naur@post.tele.dk>! Subject: Re: DEC-C float & double ; Message-ID: <3b8013cf$0$296$edfadb0f@dspool01.news.tele.dk>   8 JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message& news:3B7FF074.7C4CA77E@videotron.ca... > L > Now, how does the compiler know it should convert a float to a double in aE > routine such as printf with undefined arguments ? Does "printf" get  special C > treatment because it is so closely associated with the language ?  > F > Or doers the compiler automatically convert any float to double when passing + > to a routine whose argument are unknown ?   C I've done a couple of more tests, and it does seem, that floats are L converted to doubles, whenever the called routine arguments are unknown. TryG to compile the testprogram below with cc/lis/mac, and you'll see what I  mean.      #include <stdio.h>  $ void test( float sky, double water ) {     printf(" sky=%f ", sky );     printf(" water=%f ", water ); }   0 void test2( char *fmt, float sky, double water ) {     printf( fmt, sky );    printf( fmt, water ); }   
 void test4();   % void test3( float sky, double water )  {     test4( sky, water );  }   & void test5( float sky, double water );  % void test6( float sky, double water )  {     test5( sky, water );  }          Best regards     Jesper Naur    ------------------------------  # Date: Mon, 20 Aug 2001 03:04:00 GMT 0 From: William Barnett-Lewis <wlewis@mailbag.com> Subject: DecWrite?* Message-ID: <3B807E1C.47ED77A@mailbag.com>  D While setting up the C compiler on my new to me VLC, that one of theC licensed items in the list is DecWrite. Would anyone out there have F media for this product available? I can only do cdrom on this machine.   Thanks in advance,   William  --  * You better watch out    What you wish for;+ It better be worth it   So much to die for. -                                 Courtney Love    ------------------------------  # Date: Sun, 19 Aug 2001 20:05:22 GMT , From: "Garry Wang" <garrywang00@hotmail.com> Subject: hello< Message-ID: <6_Uf7.26527$Y6.14451900@news3.rdc2.on.home.com>   hello    ------------------------------  # Date: Sun, 19 Aug 2001 18:12:50 GMT $ From: Scott Vieth <svieth@wi.rr.com>E Subject: How is the cache memory in an HSG80 divided and/or utilized? ) Message-ID: <3B80025F.FC85BBCE@wi.rr.com>    Hi:   2 I've got a question regarding the cache on HSG80s.  ? Let's say I've got a pair of HSG80s with max cache running in a - redundant config with mirrored cache enabled.   C We have 512MB on each HSG80, half is devoted to mirroring the other ) HSG80.  That gives us 256MB to work with.   A If I have four units on the controller, does each get 25% of that > 256MB?  Or do all the units utilize all of the 256MB together?  G If the units are D10, D20, D30 and D40 AND D40 is very active I/O-wise, H will its blocks monopolize the cache and result in the cached blocks for# D10, D20 and D30 being flushed out?   E Can anyone shed any light on the behavior of the cache in the HSG80s?    Thanks,   
 -Scott :^)   ------------------------------    Date: 19 Aug 2001 21:33:59 -0500+ From: young_r@encompasserve.org (Rob Young) I Subject: Re: How is the cache memory in an HSG80 divided and/or utilized? 3 Message-ID: <LBMAfI3xIJ85@eisner.encompasserve.org>   P In article <3B80025F.FC85BBCE@wi.rr.com>, Scott Vieth <svieth@wi.rr.com> writes: > Hi:  > 4 > I've got a question regarding the cache on HSG80s. > A > Let's say I've got a pair of HSG80s with max cache running in a / > redundant config with mirrored cache enabled.  > E > We have 512MB on each HSG80, half is devoted to mirroring the other + > HSG80.  That gives us 256MB to work with.  > C > If I have four units on the controller, does each get 25% of that @ > 256MB?  Or do all the units utilize all of the 256MB together? >   ? 	Without able to find info on the cache algorithms, it was told H 	to me that is a shared pool.  Related, the Evil Machine at one time hadA 	a weakness whereby their cache was divided by OS.  That has gone * 	away.  One big happy pool there now also.  D 	But then for each unit online, does the HSG80 controller set aside @ 	a portion per unit?  Highly unlikely.  Otherwise , totally idleC 	units would be monopolizing cache.  Everyone does LRU or a version F 	thereof out of their shared cache pool.   It doesn't get much better  	than that.   I > If the units are D10, D20, D30 and D40 AND D40 is very active I/O-wise, J > will its blocks monopolize the cache and result in the cached blocks for% > D10, D20 and D30 being flushed out?  >   C 	Are those random or sequential I/Os coming out of D40?  If random, A 	one could imagine that D40 would be churning a certain amount of > 	the cache, whereby if D20 and D30 are re-hitting their HSG80  	controller for I/Os, those @ 	blocks aren't getting kicked out (not LRU).  If sequential, youG 	are doing pre-fetch there and hopefully the controller is smart enough " 	to chew up a portion and no more.  G > Can anyone shed any light on the behavior of the cache in the HSG80s?   E 	But suppose your Alphas have plentuous memory.  Perhaps so much that > 	you have dedicated 1 Gigabyte to VIOC (you aren't running 7.3; 	yet ;-) , VIOC is doing well , kicking LRU blocks out when D 	the need arises.   When you get down at a controller level and run F 	VTDPY , you are surprised at how poor your read caching is.  But all I 	your indices are cached at the server, so you are traipsing out to disk  > 	to read in less (rare) accessed data.  Your HSG80s are a win ? 	becuase backups are aided by the pre-fectching and the ability 8 	to cache the writes speeds writing up.  Your poor read G 	caching is to be expected as those random reads really do have to hit  C 	disk and most likely aren't that active.  If they are, they are in > 	your server cache!!!  Where they should be!  The Evil MachineA 	has attempted to turn a few of us around and point out how large E 	and wonderful their storage cache is.... that is the wrong place to   	cache!   A 	You can further verify how well your VIOC is doing (backups will 5 	skewer the stats) by looking at read hits at 8 a.m.  I 	( $ show mem /cache/full ) , making a note of it and then about 11 a.m.  ( 	do some math on $ show mem/cache/full .   				Rob    ------------------------------  # Date: Mon, 20 Aug 2001 00:33:37 GMT ' From: bad bob <sfmc68@bellatlantic.net> , Subject: Re: OT: TOPS-20 and TOPS-10 live on0 Message-ID: <3B805EA2.10CE9655@bellatlantic.net>   jmfbahciv@aol.com wrote: > > > In article <howard-52B209.08051417082001@enews.newsguy.com>,- >    Howard S Shubs <howard@shubs.net> wrote: ; > >In article <mpupntkgmu19llkp0eekvto7h7qfcrihio@4ax.com>, * > > Alan Greig <a.greig@virgin.net> wrote: > > J > >> Just in the last week or so pre-built binaies have been made availbleF > >> for WIndows and the emulator claims to be able to compile and run, > >> under VMS but I haven't tried that yet. > > = > >How fast does the emulator run versus the actual hardware?  >  > Who the fuck cares? ? Ahh, perfect! I just get back from vacation, going thru my mail : and news groups after 10 days away, and I read a perfectly8 salient and succient comment.  Of course it is from BAH! GREAT!!  thanks bad bob  >  > /BAH > ) > Subtract a hundred and four for e-mail.    ------------------------------  % Date: Sun, 19 Aug 2001 22:13:59 -0400 ' From: Howard S Shubs <howard@shubs.net> , Subject: Re: OT: TOPS-20 and TOPS-10 live on< Message-ID: <howard-83E18E.22135919082001@enews.newsguy.com>  0 In article <3B805EA2.10CE9655@bellatlantic.net>,)  bad bob <sfmc68@bellatlantic.net> wrote:   A > Ahh, perfect! I just get back from vacation, going thru my mail < > and news groups after 10 days away, and I read a perfectly: > salient and succient comment.  Of course it is from BAH!	 > GREAT!!   J I think she's annoyed with me for some reason, but I dunno why.  Possibly L relates to something in alt.folklore.computers, but if so she needs to calm  down.  --   Howard S ShubsD "Run in circles, scream and shout!"  "I hope you have good backups!"   ------------------------------  % Date: Mon, 20 Aug 2001 19:00:35 -0500  From: eccm <eccm@swbell.net>( Subject: Re: Run 3 phase PS on 2 phases?* Message-ID: <3B81A4A3.5316549A@swbell.net>  B if the fat cable goes to the holybox with lots of regular outlets,D then just getcha outlet strips and fire it up. also if the power boxD is an 871 or similar, see http://www.montagar.com/~patj/dec/hcps.htmE and go to DEC Manuals and Technical Data and see if your power box is ( listed. you can rewire for single phase.   "David J. Dachtera" wrote: >  > Dean Woodward wrote: > > K > > I've inherited a DEC rack-mount 3-phase power supply (P/N 70-30101-08). I > > Since I live in a stereotypical residential neighborhood, I only have F > > 2-phase power available.  I've opened it up (oh no! the warranty'sL > > voided!) and it appears at a casual glance that each of the three phases9 > > is simply used to power 1/3 of the available outlets.  > > K > > So, can I run this on 2-phase and just use the outlets that work?  What G > > if I wire two of the phases together inside so I can use all of 'em H > > (don't need 'em all, but it'd save hunting down which are powered byJ > > which...)  It'd be nice to simply wire this to my UPSes and distributeE > > power from it (not to mention the geek factor- I've already found C > > conduit that fits where the 3-phase supply came in originally.)  > > H > > Other equipment in the rack consists primarily of BA-356 shelves and8 > > other 110 powered equipment; no 2- or 3-phase stuff. > > G > > Photos of the PS (internal & external) can be posted to webspace on  > > request. > F > Go to http://www.montagar.com/users.html and look for Pat Jankowiak,E > then follow one if his e-mail links. He has some info. you may find 	 > useful.  > -- > David J. Dachtera  > dba DJE Systems  > http://www.djesys.com/ > * > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sun, 19 Aug 2001 22:36:21 +0200 < From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <noone@home.com> Subject: Re: some SFF questions ' Message-ID: <3B802345.DB3C63B@home.com>   ? To send a MIME formatted mail with some binary (e.g ZIP or PDF) 1 file as an attachement you can do the following :   4 $ mpack -s some_subject -o mime.tmp some_binary_file% $ mail mime.tmp "nbl%""user@domain"""   = The mail arrives at user@domain with the subject given in the < mpack command and with "some_binary_file" as an attachement.  B In just 3-4 DCL lines I run a TEX file throught LaTeX (giving a PS file),F convert to PDF with Ghostscript, convert to MIME format with MPACK and send the file with NBL.    No 3GL hacking ;-)   Of course, YMMV, as usual.   Regards  Jan-Erik Sderholm.   *  Dirk Munk <munk@home.nl> wrote in message# news:<3B7CE926.BC1238D8@home.nl>... I > We are currently developing a application for bulk e-mails, and we want  > to use SFF for this. > [snip...] H > Furthermore the readme states that MIME mail should work to, but thereB > are no examples how to accomplish this with callable interfaces.   ------------------------------  % Date: Mon, 20 Aug 2001 21:02:41 -0500  From: eccm <eccm@swbell.net>- Subject: Re: Terry Shannon Tech Talk on Tru64 * Message-ID: <3B81C141.62F3ECF2@swbell.net>  F Haha! a bletcherous personal flame, born of /dev/null and drooled fromE the lips of a known entity, perched upon a pinnacle of its own cruft, > er.. from a yahoo-over-goolge account! What are we coming to?   E Shannon's on the user society board of directors, so maybe he gets to B go for free, so what if he does?. There was a time not so long agoE when that was not the case. Shannon Knows Anaheim Hotel Prices, yes..qF but that's immaterial. Those that are of any knowledge in professionalE matters don't compute "shannon" with "lackey", as they are aware that 3 the two are not same type variable. get knowledge.    > BTW- there aren't really any good hotel rooms in Anaheim, just overpriced ones.  ! Damn -this is just too much fun. /  8 Newbie JrSysAdmin@yahoo.com wrote via groups.google.com: >  > "># > > Copyright 2001 Terry C. ShannonS8 > > Not affiliated with ... Compaq Computer Corporation. > F > lackey, do you even know what a good hotel room in anaheim costs, or3 > did your "non-affiliate" take care of it for you?    ------------------------------  % Date: Mon, 20 Aug 2001 20:22:48 -0500d From: eccm <eccm@swbell.net>( Subject: Re: The Final Knell Has Sounded* Message-ID: <3B81B7E8.E57D4933@swbell.net>  D David, you are mistaken in that the the sessions which the DFWCUG is? scheduled to present are still on. Please verify details beforeo spreading FUD. h  D This might not be widely known but the only two boxes to survive theF onslaught of over 5000 hackers at DEFCON9 were the DFWCUG's VMS Alpha, and a Sun box. ?  E I won't speculate about the merits of the Sun, but I will say that it-F had a sysadmin present 24 hours a day for the entire event, as well asE a secondary system which did nothing but monitor attacks on the box. i  F As for the VMS system, we turned it on, basically left it alone exceptD when we wanted to look at it, and slept and drank through the event, without a care.    "David J. Dachtera" wrote: >  > <exasperation> >r* > Well, it's tough to argue at this point: > H > Alpha is dead with no viable replacement currently on the market or on > the foreseeable horizon.  C Oh yeah, like the pdp11, vax, and a host of other not-necessesarilyeD DECompaq machines? These systems have traditionally gone on in theirE useful and critical tasks for years. There's a chip fab I can't name,lD that was until last week, running a microvax II which controlled theF wafer-level testing of the chips. The machine died, and what happened?D a compadre and I took the RD54's and stuck them one at a time into aE cluster-booted diskless microvax II and backed them up to 4mm DAT foroE the nice man. So, VAX isn't dead after all. How can Alpha be? It willsC be years before Alpha is no longer used. There will be transitions, E and speedbumps on the road, but the situation is not as gloomy as theC image you present. > H > No more Alpha newsletter. Oh, there'll be mention of it in the OpenVMS> > Times for a while, but that'll fade, too. History repeats...  A Change is required to survive. "if a man doesn't change, he dies"t (from some western..)c   >  > No more CI clusters.  ? That's an issue of modernization. CI is very cool and holy, butt@ somewhat dated and not necessary for new systems due to improvedC methods available from Compaq's modern system architecture. How can-D you kow there won't be a solution available for the legacy systems?  >  > No more SCSI clusters.  C What proof do you have of this? SCSI clustering has more to do withu7 the glue chips and O/S on the system than with the CPU.    > I > The last non-government VMS holdouts are abandoning in droves, with theh8 > possible exceptions of Cerner and who ever bought SMS.  D Can you cite numbers and facts? Sure alot of customers are migrating% to alternative systems, but not all. w > J > No VMS newbie sessions at CETS-2001. I hear even Wayne Sauer's venerableB > performance and management sessions got bumped. No DFW stuff was% > accepted, either, from what I hear.u   already answered.t > 4 > No more platforms + no more newbies = no more VMS. > E OpenVMS ported to IVMS (to coin a term) on Itanium + quality hardwarer + newbies = VMS. v  A Maybe you're unhappy about things, but taking always publicly the D negative side is not constructive. How do you think I feel? They areA making me take unix training! (actually, unix is kind of easy but D quite bizzarre) I work in what is mostly a Sun and Intel engineeringD environment, but all the stuff in my company, that, were it to breakC would cost millions and cause damage, runs on VMS. They run the sunl@ and intel in the cheap seats, and the VMS in the space where theF degree of departure from desired reality is counted in hard currency.   J > Congratulations Compaq - you won. VMS is now officially dead, outside ofJ > the government markets you chose for it. (The business schools will haveB > to rewrite their text books - instead of the market choosing theJ > product, the product chose the market. Wonder what Clem Stone or old man$ > Rockefeller would say about that?) > C Congratulations Compaq - you won. VMS is now officially going to berF ported to a new archtecture, just with the VAX, for the benefit of all! the markets which will choose it.v     > </exasperation>  > A > Gee - maybe it's time for an INDEPENDENT OpenVMS User Group and F > Symposium - followed soon by an Independent OpenVMS, in the hands ofI > profit-minded entrepreneurs who know something about marketing to boot.o  D Maybe I can get a user group started at DEFCON10? The DEFCON_VMSUG.   > As far as an independent VMS - write it, find a sugar daddy, a. mouthpiece, and a marketroid, and you go girl!   > F > (Geez! I've *REALLY* gone off the deep end! What's the number of the7 > employment office at the scavenger company again...?).  F Just forget it. Take two xanax, get a hobbyist license, and call me inD the morning. I'm being a smart ass because I'm just sick of all this< FUD. I had to change careers because what I did was becomingF obsolesced. If you think it's over, get a head start on the rest of usD blind fools who can't see it coming. You are in a unique position toD take advantage of this Alpha-Intel situation, being a consultant and all. t       >  > -- > David J. Dachtera  > dba DJE Systems- > http://www.djesys.com/ > * > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/a   ------------------------------  % Date: Sun, 19 Aug 2001 22:47:58 -0400h' From: "Bill Todd" <billtodd@foo.mv.com>o( Subject: Re: The Final Knell Has Sounded( Message-ID: <9lptoe$2pq$1@pyrite.mv.net>  ) "eccm" <eccm@swbell.net> wrote in message $ news:3B81B7E8.E57D4933@swbell.net...   ...s   > "David J. Dachtera" wrote: > >e > > <exasperation> > >u, > > Well, it's tough to argue at this point: > >fJ > > Alpha is dead with no viable replacement currently on the market or on > > the foreseeable horizon. >0E > Oh yeah, like the pdp11, vax, and a host of other not-necessesarily: > DECompaq machines?  L *Exactly* like PDP-11s and VAXes:  no longer a competitive platform, and nowF suited only to 'legacy' use where advancing along with the rest of the industry is not important.  F Some people will be perfectly happy with that, and more power to them.K Others had hoped to see Alpha vigorously developed (to maintain or increaseaI its performance lead) and equally vigorously promoted (to keep its profitgG healthy and its future secure) - and are now royally pissed at Compaq'snL unilateral and arbitrary reversal of its frequent and unambiguous statementsI that Alpha's future viability and performance were things they should bet  their businesses on.  J I don't know why it's so hard for people who (for whatever reasons) aren'tJ upset at Compaq to understand why others *are* very upset:  just because aL platform change may not cause *you* all that much grief doesn't mean it will: be equally convenient for everyone else (e.g., people withB impossible-to-port critical applications or to whom performance isL important).  And *any* port, even if handled optimally, is more work than noK port at all, and thus an annoyance if it was totally unnecessary (which thedG evidence strongly suggests is the case, Compaq's lies notwithstanding).    ...t  J > > No more Alpha newsletter. Oh, there'll be mention of it in the OpenVMS@ > > Times for a while, but that'll fade, too. History repeats... > C > Change is required to survive. "if a man doesn't change, he dies"e > (from some western..)r  L Where you look for words of enlightenment makes your own superficiality moreE understandable.  Sometimes change is required, sometimes stability isIK required.  Change for change's sake is usually (though not always) a ratherd	 bad idea.n   ...l  K > > The last non-government VMS holdouts are abandoning in droves, with thel: > > possible exceptions of Cerner and who ever bought SMS. > F > Can you cite numbers and facts? Sure alot of customers are migrating& > to alternative systems, but not all.  L Of course not all:  some people will put up with anything to keep using VMS.F But even if just 'a lot of customers' migrate there is major cause forL concern, because there is without any shadow of a doubt some threshold levelH below which Compaq will cease to continue to develop and support VMS (atH least at any reasonable price) - and the casual manner in which it threwJ away a very profitable Alpha business suggests that this threshold may not= be all the much lower than the current level of VMS business.t   ...   6 > > No more platforms + no more newbies = no more VMS. > >aG > OpenVMS ported to IVMS (to coin a term) on Itanium + quality hardware  > + newbies = VMS.  D Or, VMS sales tank during the *3 years* while it's only available onF lame-duck hardware with the prospect of having to port the system whenL hardware with more of a future is finally usable:  no newbies, fewer oldies,/ Compaq finds it unprofitable, and good-bye VMS.   C > Maybe you're unhappy about things, but taking always publicly they$ > negative side is not constructive.  H It can be constructive if it warns off people who might otherwise becomeK lulled (yet again) into one more round of complacency by Compaq's lies.  IteI could also be constructive if it resulted in sufficient financial pain onaL Compaq's part to cause them to rethink their priorities.  At the very least,K it provides a sorely-needed balance to the bullshit Compaq spread around ineG attempts to justify its decision on technical and financial grounds andhJ paper over the uncompromisingly definitive nature of the commitments it so cavalierly broke.c   ...   L > > Congratulations Compaq - you won. VMS is now officially dead, outside ofL > > the government markets you chose for it. (The business schools will haveD > > to rewrite their text books - instead of the market choosing theL > > product, the product chose the market. Wonder what Clem Stone or old man& > > Rockefeller would say about that?) > >uE > Congratulations Compaq - you won. VMS is now officially going to be  > ported to a new archtecture,  L Would that be just as 'officially' as the Alpha commitments up to June 25th?# Somehow that's not very reassuring.e   .../  8  I'm being a smart ass because I'm just sick of all this > FUD.  J Hey, you sound just like Kerry!  And about as convincing.  While there hasK been some wholly-understandable but likely excessive paranoia caused by theiB Alpha upheaval, most criticism of Compaq has centered on issues of
 substance.  8  I had to change careers because what I did was becomingH > obsolesced. If you think it's over, get a head start on the rest of us& > blind fools who can't see it coming.  H I can't speak for David, but I couldn't care less about you blind fools:J you have enough information to make judgements, and if you choose unwiselyH that's your problem.  I do, however, think that every statement you makeH that might suck *other* people into your foolishness deserves a suitableH rejoinder:  if you're sick of the discussion, get the pro-Compaq side to' shut up, and most of it will evaporate.a   - bill   ------------------------------    Date: 19 Aug 2001 20:54:42 -0700- From: afeldman@gfigroup.com (Alan E. Feldman)h( Subject: Re: The Final Knell Has Sounded= Message-ID: <af1e4ce6.0108191954.274d0a96@posting.google.com>   W "Bill Todd" <billtodd@foo.mv.com> wrote in message news:<9ljlfl$rtt$1@pyrite.mv.net>...s: > "Rob Young" <young_r@encompasserve.org> wrote in message/ > news:9RQpvsMYVJcs@eisner.encompasserve.org... G > > In article <2jaqnt0g526377h39lllhe1mirdvin0rpt@4ax.com>, Alan Greigb >  <a.greig@virgin.net> writes:m9 > > > On Fri, 17 Aug 2001 10:10:45 -0400, "Sue Skonetski"r) > > > <susan.skonetski@compaq.com> wrote:  > > >i > > >>Here we go again.  > > >>L > > >>The sky is not falling and I wish that I had a dime for every supposed >  death > > >>knoll. > > >> [snip] > > D > > Not too many examples of highly profitable segments getting axed( > > in downturns?   Gee... I wonder why?  E Well, NBC axed Jeopardy! when it was at its peak! (I don't know if it1E was during a "downturn".) The Jeopardy! book claims that the networksPE (at the time, anyway) just like to re-decorate (I forgot just how thetF book put it) their daytime line-ups once in a while. And the book alsoF claims that NBC never topped the ratings the Jeopardy! got in its timeF slot. And, of course, NBC also axed Star Trek. (Okay, Star Trek wasn't2 very profitable then, but it sure has been since!)  D But Star Trek was saved once by a letter writing campaign. If peopleD write to Compaq in a civil tone, politely explaining what VMS can do@ for the company, maybe that would help (one can dream); but thenB again, maybe not. Of course, VMS may "beat the esimates" yet againA (see below). Please, rude letters will *not* help. Would you takedD advice from a rude, profanity-filled, insulting letter? I think not! [snip]J > Why not take a sample of the number of comp.os.vms participants who haveN > voiced nervousness compared with the number who have voiced confidence?  I'mD > sure the ratio exceeds 2:1, and possibly by quite a bit.  Hardly a> > scientific sample, but certainly sufficient to justify 'we'.  A Yes, hardly scientific. I would guess that those who are the most D worried are the ones who are screaming the loudest and most often. IC mean, you don't see too many people going around with signs saying,   > "NOOOOO!!!!!!. THE WORLD IS ***NOT*** ABOUT TO END. THE SKY ISD ***NOT*** FALLING. EVERYTHING IS GOING TO BE OKAY!!!!! STOP WORRYING! NOW!!! DON'T WORRY, BE HAPPY!!! "n  F Most of the time things are not as bad as the most-worried proclaim it< to be. And the same goes for the most optimistic viewpoints.   Long live VMS!   Dislcaimer: JMHO   &-) Alan E. Feldmanm afeldman@gfigroup.com    ------------------------------  % Date: Mon, 20 Aug 2001 01:06:03 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>t( Subject: Re: The Final Knell Has Sounded, Message-ID: <3B809AB9.3B9D952E@videotron.ca>   "Alan E. Feldman" wrote:F > write to Compaq in a civil tone, politely explaining what VMS can doB > for the company, maybe that would help (one can dream); but then > again, maybe not.   K Been there, done that. You know what happens ? You get a nice reply tellingwM you about efforts underway to put VMS in a "renaissance" mode, and that "very N shortly" Compaq will be starting an advertising campaign that will turn Compaq> into an enterprise company. So you shut up and remain patient.  J Then you see those famous ads that showcase wintel products as "enteprise"L products. Then you listen to financial statements where one would expect VMSL to be mentioned because it is supposedly so profitable at a time when wintelI is not, but it is not mentioned, its profits probably accounted under thes< "wintel server" group since both are "enterprise" to compaq.  J Compaq has made its decision. And there is truly very little you can do toL change its direction. Remember that Compaq is a slave to Microsoft and IntelJ and it does not have the freedom to do as its VMS customers wish it to do.  J Compaq, like its Digital predecessor will recat when the complains get tooM loud. But its reaction is just enough to force the complainers to shut up and N wait, and meanwhile Compaq returns to its old wintel self until the complaints surface again.  M When Compaq bought Digital, there were hopes that VMS's fortunes would changesN under new management that was seamingly interested in expanding beyond wintel.I However, Compaq has had VMS for long enough now that it is pretty safe tofK conclude that Compaq does not have intentions to expand much beyond Wintel, N and its true goal is to retain those non-wintel customers until wintel is able7 to serve them well and then sell them wintel solutions.   N Look at the iPAQ. Compaq got a lot of milage and brownie points from MicrosoftN because Compaq was able to bring to Microsoft tons of customers for a platform: which had been considered a failure until now (WindowsCE).  L Imagine how many brownie points Compaq will get when it gives Microsoft some real enterprise customers.  L In one of those financial announcements, one tidbit got my ear: Capellas hadL repeated at least twice the fact that they had sold a large number of wintelM servers and desktops to the health industry. Of course no mention of VMS in a-2 market where it sopposedly has large market share.  M It is of my opinion that the current management at Compaq didn't buy VMS fromBL digital, they bough the VMS customers from Digital and Compaq will work hardE to find Compaq-approved solutions for them (aka: wintel). Until thosetE solutions are found, VMS is being kept on basic life support with thee appearance of being healthy.    K VMS used to be a key platform for international electronic funds transfers.uK But alas, SWIFT (probably at the urging of Palmer) decided to drop VMS as atJ blessed platform and VMS will shortly no longer be able to be part of thatJ industry. The Compaq person in charge of dealing with SWIFT had a reactionE such as "we are working with SWIFT to ensure that our remaining ST400 K customers will have the time to migrate and potentially delaying the cutoff J date when ST400 no longer works to allow for easier migration, and we hope/ that they will all migrate to PROLIANT servers.d  M Now, the decision to drop VMS was done under the palmer era when it was clearoJ Palmer was out to kill VMS. Last year, while Compaq seemed for a moment toM give VMS a chance, you would think that that same person would have gone backlK to SWIFT and asked them to re-think their position and perhaps Compaq couldtM have funded the porting of the new SWIFT software to VMS in order not only tofM retain the customers but also grow the customer base now that Compaq *seemed*4N serious about VMS. Instead, they were hoping to give customers as much time asC possible so that they could migrate to the NT version on Proliants.r  K Tandem was built to a large part by Tandem funding some of the software andeN companies that wrote for NSK and this is how Tandem got to have some strategicK software/applications that were available only on NSK, making for some very'; captive customers. Compaq could do the same to rebuild VMS.t    K Remember that international funds transfer was one of those "niche" marketsdD that VMS had and was mentioned whenever Compaq tried to reassure VMSH customers. And their reaction is that they hope that  the customers willN migrate to Proliants (instead of Sun or HP - the True64 version was abandonned some time ago).a    N So, one of the remaining niche markets for VMS is allowed to go to the waysideK because Compaq sees the opportunity to sell Proliant servers now that SWIFT L sanctioned software will run on NT. And guess what, in a couple of years, itK is Compaq and NT that will brag about doing x% of the world's internationala* funds transfers, a title VMS used to have.  N Ask yourself: as software on NT becomes available to compete against the otherE remaining niches VMS still has, are you confident that Compaq will be M agressive to protect the VMS niche, or do you beleive that Compaq will see annA opportunity to sell Proliant servers with NT to those customers ?u   ------------------------------  % Date: Mon, 20 Aug 2001 01:07:56 -0400 ( From: Hamlyn Mootoo <univms@bigfoot.com>( Subject: Re: The Final Knell Has Sounded+ Message-ID: <3B809B2C.D0E2D5B5@bigfoot.com>    "Alan E. Feldman" wrote: > Would you takeF > advice from a rude, profanity-filled, insulting letter? I think not!D It all depends on how smart you are, and whether you're own emotionsG cloud your judgement.  What kills businesses is the problems they DON'ToH here about from customers who will never buy from them again.  If I wereF running a business and could only choose whether to have only negativeF feeback given to me or only positive, but not both, I would choose theE negative feedback every time.  What a business does right, is usuallyeF fairly obvious - what it does wrong, is not always so obvious.  So theH level of profanity in a letter might serve as an excellent gauge of whatB your're doing wrong in business.  One might even develop a scoringG system to produce a negativity index by counting the "D", "A", "F", andnD "B" words used in the letter to illustrate the deficiency of service) experienced by the author of the letter. s  H By the way, are you any distant relation to Marty of the same last name?! He was one of my favorite actors.t   HM >  > Dislcaimer: JMHO   &-) > Alan E. Feldman  > afeldman@gfigroup.com    ------------------------------  % Date: Sun, 19 Aug 2001 16:25:55 -0400n+ From: "Main, Kerry" <Kerry.Main@compaq.com>m Subject: RE: tpcip upgradeR Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4D495B2@kaoexc01.americas.cpqcorp.net>   Mike,   H >>> Please, really excuse me again for such matter, but I just found the0 Install docs regarding OS and TCPIP Services,<<<  E As a fyi, the OpenVMS and TCPIP documentation can be found online at: E http://www.openvms.compaq.com:8000/index.html (OpenVMS Doc Home Page)tE http://www.openvms.compaq.com:8000/index.html#tcpip (TCPIP V5.1/V5.0A  documentation)G http://www.openvms.compaq.com:8000/73FINAL/6524/6524pro.html#backup_secrL (TCPIP install doc's which shows VMS V7.1 and V7.2-1 are supported - V7.3 as	 well now)t  H Only additional note would be to ensure that after the upgrade, you also@ install the latest maint kit for TCPIP V5.1. It is available at:L http://ftp.support.compaq.com/patches/public/Readmes/vms/dec-axpvms-tcpip_ec# o-v0501-151-4.README (one long url)    Regards,    
 Kerry Main Senior Consultantp Compaq Canada Corp./ Professional Services- Voice: 613-592-4660a Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----@ From: techwebsite@netscape.net [mailto:techwebsite@netscape.net] Sent: August 18, 2001 4:12 PMb To: Info-VAX@Mvb.Saic.Com  Subject: tpcip upgrade    
 Excuse me,  E Is it possible to upgrade from TCP/IP Services V5.0 (AXP) to a higher E version (i.e. V5.1) without upgrading the OS (currently using V7.2)?,n? and if so, how may I accomplish that task, looking what kind ofe information?  C I'd like to upgrade those services because I had some problems withr
 the systemD due to It crashes sometimes due to INCONSTATE, Inconsistent I/O data base.pC I was looking for issues about it on DejaNews, and accordinf what Ii* read address me to upgrade such a service.  D Please, really excuse me again for such matter, but I just found theE Install docs regarinf OS and TCPIP Services, but I did not see such ar= scenery or any comments about it are appreciated in advanced.e     Thanks   Mike   ------------------------------  % Date: Sun, 19 Aug 2001 16:50:57 -0400n- From: JF Mezei <jfmezei.spamnot@videotron.ca>u Subject: Re: tpcip upgrade, Message-ID: <3B80269E.8543C91A@videotron.ca>  L which TCPIP version is supposed to have a DHCP *CLIENT* in it, and will that one run on VMS 7.2 ?  8 I was under the impression that it would only run on 7.3   ------------------------------  % Date: Sun, 19 Aug 2001 16:43:25 -0400s2 From: rdeininger@mindspring.com (Robert Deininger)* Subject: Re: VAX/ALPHA FORTRAN and me! :-)L Message-ID: <rdeininger-1908011643260001@user-2iveak4.dialup.mindspring.com>  / In article <3B7A93BF.26274BD5@cableinet.co.uk>,t$ tim.llewellyn@cableinet.co.uk wrote:  ,B > yup, that is the function I meant, sorry I posted the wrong name
 > earlier.G > Couldn't find the Fotran Manual on www.openvms.compaq.com (I know its  > lurking there somewhere).   # http://www.compaq.com/fortran/docs/S   -- i Robert Deininger rdeininger@mindspring.comr   ------------------------------   Date: 19 Aug 2001 22:08:15 GMT# From: system@niuhep.physics.niu.edue Subject: VSXXX-JA headphones+ Message-ID: <9lpdcf$6qg$1@husk.cso.niu.edu>i  M I believe that VSXXX-JA headsets were designed for use with VaxStation 4000s.IM They have what looks like but is a bit smaller than a phone jack for plugging M into the VS.  Are these things of any use without a vaxstation? or adaptable n for other uses?l  " Thanks for any help you can give,  Robert Morphis   ------------------------------  % Date: Mon, 20 Aug 2001 01:53:14 -0400g2 From: rdeininger@mindspring.com (Robert Deininger)  Subject: Re: VSXXX-JA headphonesL Message-ID: <rdeininger-2008010153150001@user-2ivecgp.dialup.mindspring.com>  P In article <9lpdcf$6qg$1@husk.cso.niu.edu>, system@niuhep.physics.niu.edu wrote:  O > I believe that VSXXX-JA headsets were designed for use with VaxStation 4000s.aO > They have what looks like but is a bit smaller than a phone jack for pluggingcO > into the VS.  Are these things of any use without a vaxstation? or adaptable y > for other uses?c  I This headset is also supposed to be compatible with the audio port on DECoH 3000 (turbochannel) alpha systems.  The audio port on these systems fits* the cable from the handset on a telephone.  6 If you have a surplus VSXXX-JA, I might be interested.   -- l Robert Deininger rdeininger@mindspring.comw   ------------------------------  % Date: Sun, 19 Aug 2001 19:52:22 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>dG Subject: Re: Wailing and moaning.... (was: The Final Knell Has Sounded)u, Message-ID: <3B805136.C73F0830@videotron.ca>   Larry Kilgallen wrote:F > about work out.  But certainly bothering the rest of us with puttingI > doom-and-gloom posts into comp.os.vms does not reach Compaq management,c  5 How do you know it does not reach Compaq management ?t  F Besides, it is good to grieve with others who share the same feelings.   ------------------------------  % Date: Sun, 19 Aug 2001 20:29:10 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>tG Subject: Re: Wailing and moaning.... (was: The Final Knell Has Sounded)a' Message-ID: <3B8067E6.29CCBC51@fsi.net>n   Larry Kilgallen wrote: > b > In article <v9eqntob529t0ukb4rngs8s2erebp20hqt@4ax.com>, Alan Greig <a.greig@virgin.net> writes:J > > On Fri, 17 Aug 2001 15:57:36 +0100, Nic Clews <sendspamhere@127.0.0.1>
 > > wrote: > >o > >>E > >>That way they'd be given something more 'productive', walking thetL > >>streets proclaiming doom and gloom while the rest of us just get on with > >>the job in hand. > > G > > Carry on sleep-walking Nic. The purpose (or at least my purpose) of H > > many of the doom profits is to try and *force* Compaq to take strongI > > re-assuring measures right now which we can hold it to later. Leavinge > E > Well please let us know how those activities we don't know anything.F > about work out.  But certainly bothering the rest of us with puttingI > doom-and-gloom posts into comp.os.vms does not reach Compaq management,dI > so please cut out _that_ part of your activity and get to the real workf > you have in mind.d  F I regularly forward noteworthy posts to Marcello, Gorham and Capellas.F If they don't read their e-mail, that is beyond my ability to control.  F I suppose certified mail would not be out of the question, but it does get a bit costly after a while.o   -- o David J. Dachterag dba DJE Systemsu http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/t   ------------------------------  % Date: Mon, 20 Aug 2001 01:55:30 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)G Subject: Re: Wailing and moaning.... (was: The Final Knell Has Sounded)-L Message-ID: <rdeininger-2008010155310001@user-2ivecgp.dialup.mindspring.com>  5 In article <3B805136.C73F0830@videotron.ca>, JF MezeiH% <jfmezei.spamnot@videotron.ca> wrote:9   > Larry Kilgallen wrote:H > > about work out.  But certainly bothering the rest of us with puttingK > > doom-and-gloom posts into comp.os.vms does not reach Compaq management,m > 7 > How do you know it does not reach Compaq management ?   B Our friends in VMS engineering have said, more than once, that theB newsgroup is not a good way to reach the upper echelons at Compaq.  H > Besides, it is good to grieve with others who share the same feelings.  $ I think it is possible to overdo it.   -- k Robert Deininger rdeininger@mindspring.com    ------------------------------  # Date: Sun, 19 Aug 2001 23:46:19 GMT ! From: Beyonder <beyonder@vrx.net>b Subject: where is 7.3 ? 8 Message-ID: <lhk0otks4ij6v011u1vc16bukdsirktpbe@4ax.com>  A I haven't seen it anywhere. montagar hsan't had an update in agesa either.n   sigh...p   B.   ------------------------------  % Date: Sun, 19 Aug 2001 20:42:53 -0400a' From: Howard S Shubs <howard@shubs.net>t Subject: Re: where is 7.3 ? < Message-ID: <howard-735D43.20425319082001@enews.newsguy.com>  8 In article <lhk0otks4ij6v011u1vc16bukdsirktpbe@4ax.com>,#  Beyonder <beyonder@vrx.net> wrote:T  C > I haven't seen it anywhere. montagar hsan't had an update in agest	 > either.   K You could order it from Compaq, but I wouldn't if I were you.  I've yet to t$ hear or see anything good about 7.3. -- o Howard S ShubsD "Run in circles, scream and shout!"  "I hope you have good backups!"   ------------------------------    Date: 19 Aug 2001 19:33:56 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)u Subject: Re: where is 7.3 ?n3 Message-ID: <6P4aSh7c0VhR@eisner.encompasserve.org>l  \ In article <lhk0otks4ij6v011u1vc16bukdsirktpbe@4ax.com>, Beyonder <beyonder@vrx.net> writes: > I haven't seen it anywhere.   6 Subscription copies were delivered several months ago.  N http://eisner.decus.org/DECUServe/DECnotes/LATEST_RELEASE_INFORMATION/275.HTML   ------------------------------  % Date: Sun, 19 Aug 2001 21:35:05 -0400e  From: John Santos <JOHN@egh.com> Subject: Re: where is 7.3 ? 6 Message-ID: <1010819212111.14122B-100000@Ives.egh.com>  * On Sun, 19 Aug 2001, Howard S Shubs wrote:  : > In article <lhk0otks4ij6v011u1vc16bukdsirktpbe@4ax.com>,% >  Beyonder <beyonder@vrx.net> wrote:G > E > > I haven't seen it anywhere. montagar hsan't had an update in agesb > > either.t > M > You could order it from Compaq, but I wouldn't if I were you.  I've yet to  & > hear or see anything good about 7.3.  E I'm running it on two systems right now and haven't had any problems.-I One VAX hobbyist system and one Alpha test system (not used much recentlym but it will be soon...)c  E However there have been reports of data corruption and there were tworD ECO's released a few days ago that address this.  One is for the XFC@ (extended file cache - Alpha only) and the other is for the SCSIA drivers (VMS73_FIBRE_SCSI).  There have been a bunch of ECO's fornB Alpha, but none yet for VAX except for a DECNET-PLUS ECO.  Most ofB the ECO's pertain to alpha-only things (such as XFC) but I hope us8 VAX die-hards aren't left out for things that affect us!  F Whether to upgrade yet depends on the usual issues.  For a developmentC or hobbyist system, I would go ahead now.  For a production system,o@ I would wait a couple more months for the problems to shake out,D unless I really needed some new feature.  Of course, the more peopleB upgrade, the quicker problems will be found and fixed, so I hereby@ encourage everyone to upgrade immediately and find all the bugs!  ? I am probably going to upgrade my main development cluster (twop? Alphas and a VAX) over the Labor Day weekend, Sep 3 for all thed non-US readers.d   -- n John Santosa Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Sun, 19 Aug 2001 22:12:50 -0400q' From: Howard S Shubs <howard@shubs.net>  Subject: Re: where is 7.3 ?o< Message-ID: <howard-6412FA.22125019082001@enews.newsguy.com>  6 In article <1010819212111.14122B-100000@Ives.egh.com>,"  John Santos <JOHN@egh.com> wrote:  5 > Whether to upgrade yet depends on the usual issues.e  N Yes.  All I can say is that if I can't upgrade, the update is useless to me.  M USELESS.  TO.  ME.  And I've tried to upgrade a 7.2-1 system with NO patches oN (other than the MUP) and the upgrade failed every time.  Let me say it again: K USELESS!  And it has -file corruption-??  When was the last time -DEC- had nM this kind of problem?  I figure that Compaq's QA and field tests have failed lO fatally.  What a nasty fall.  File corruption!  Inablility to upgrade from the rA most recent (shy of 7.2-1H1) version of the system!  Inexcusable.  -- m Howard S ShubsD "Run in circles, scream and shout!"  "I hope you have good backups!"   ------------------------------  % Date: Mon, 20 Aug 2001 01:36:45 -0400e+ From: "Main, Kerry" <Kerry.Main@compaq.com>i Subject: RE: where is 7.3 ?dR Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4D495B4@kaoexc01.americas.cpqcorp.net>   Howard,r  A There are a number of Customers using OpenVMS V7.3 in prod today.   
 Reference:L http://www.openvms.compaq.com/openvms/brochures/Bank-Austria/ (and they have, been running VMS V7.3 for quite awhile now).  K Having stated this, there are patches for VMS V7.3 that resolve some issues  under specific conditions - 8 http://ftp.support.compaq.com/patches/.new/openvms.shtml  E The decision to upgrade is typically made on specific new features orpI additional performance requirements. When an upgrade is planned, applyingoJ the latest patches that the system manager deems to be applicable to their: environment are also usually part of that upgrade process.  L >>> And I've tried to upgrade a 7.2-1 system with NO patches (other than the) MUP) and the upgrade failed every time.<<i  J Just out of curiosity, but there is an upgrade issue that is documented inL the cover letter that came with OpenVMS V7.3. Is this the same issue you had or a different one?   K Fwiw, I've personally upgraded 2 systems from VMS V7.2-1 (and they both hadt6 a number of 7.2-1 patches) to VMS V7.3 with no issues.   Regards,  
 Kerry Main Senior Consultanto Compaq Canada Corp.i Professional Servicese Voice: 613-592-4660m Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----. From: Howard S Shubs [mailto:howard@shubs.net] Sent: August 19, 2001 10:13 PM To: Info-VAX@Mvb.Saic.Com  Subject: Re: where is 7.3 ?V    6 In article <1010819212111.14122B-100000@Ives.egh.com>,"  John Santos <JOHN@egh.com> wrote:  5 > Whether to upgrade yet depends on the usual issues.   L Yes.  All I can say is that if I can't upgrade, the update is useless to me.  L USELESS.  TO.  ME.  And I've tried to upgrade a 7.2-1 system with NO patches  F (other than the MUP) and the upgrade failed every time.  Let me say it again: vK USELESS!  And it has -file corruption-??  When was the last time -DEC- had nL this kind of problem?  I figure that Compaq's QA and field tests have failed  J fatally.  What a nasty fall.  File corruption!  Inablility to upgrade from the A most recent (shy of 7.2-1H1) version of the system!  Inexcusable.S -- t Howard S ShubsD "Run in circles, scream and shout!"  "I hope you have good backups!"   ------------------------------  % Date: Mon, 20 Aug 2001 01:58:41 -0400g  From: John Santos <JOHN@egh.com> Subject: Re: where is 7.3 ?o6 Message-ID: <1010820011433.14122A-100000@Ives.egh.com>  * On Sun, 19 Aug 2001, Howard S Shubs wrote:  8 > In article <1010819212111.14122B-100000@Ives.egh.com>,$ >  John Santos <JOHN@egh.com> wrote: > 7 > > Whether to upgrade yet depends on the usual issues.u > P > Yes.  All I can say is that if I can't upgrade, the update is useless to me.  O > USELESS.  TO.  ME.  And I've tried to upgrade a 7.2-1 system with NO patches fP > (other than the MUP) and the upgrade failed every time.  Let me say it again: M > USELESS!  And it has -file corruption-??  When was the last time -DEC- had  O > this kind of problem?  I figure that Compaq's QA and field tests have failed  Q > fatally.  What a nasty fall.  File corruption!  Inablility to upgrade from the SC > most recent (shy of 7.2-1H1) version of the system!  Inexcusable.m  E What happened with the upgrade?  There is an issue, documented in theuC cover letter with 7.3 that if you've installed certain 7.2-x ECO's,wE the changed files aren't recorded correctly in some PCSI file, so thesC new V7.3 files don't replace them correctly.  The work-around is torF edit the PCSI file to fix it, and there is now an ECO kit that will doE the edit for you.  But if you haven't installed any 7.2-1 ECO's, that  shouldn't affect you.h  F There have also been battles waged (forever, it seems) between certainE layered products, mostly compilers, and VMS for control of the object B libraries and shareable images.  There was definitely a problem atC about 7.0-7.1 having to do with some DEC C routines moving from onepA module to another which caused problems when the upgrade tried to"B replace the modules.  I didn't hit anything like this in V7.3, butC it is possible you did.  So that was probably the last time DEC hadt  such a problem with upgrades....  3 As for the last time they had disk file corruption,v  4      $ search sys$help:vax*.release_notes corruption  7 on my V7.1 VAX finds 3 different ECOs that address disk-3 corruption issues, VAXMOUN05_071, VAXRMS02_071, and0= VAXSHAD08_071.  So I think is is more common and has a longert history than you suspect.B  B The V7.3 file corruption isn't endemic.  (It never happened to me,B as far as I can tell.)  I think the two new ECO's address problems; reported previously in this newsgroup.  Two of the problemssA involved bound volume sets and the other is described as follows:o  H       o  Exception  conditions,  such  as  those  resulting  from  powerH          surges,  disk  errors or AUTOGEN operations, can produce hangs,:          crashes or data corruption in an SMP environment.            Images Affected:_  '           -  [SYS$LDR]SYS$PKWDRIVER.EXEV  D According to help, PKWDRIVER is the SCSI port driver for Symbios 875B and follow-on chipsets.  (I don't seem to have any of those, but IA think it is used in KZPCM's.  Anyway, if you install the ECO, youp should be okay.)  F The XFC problems seem to be more serious, since from the descriptions,B they can happen during normal operations, but only on bound volumeC sets.  Maybe not many people use bound volume sets on Alphas, which-- is why they didn't show up during field test?-  @ I'm not saying that V7.3 is perfect (i.e. it inherits the V7.2-1D directory modification date mis-feature and the broken purge/confirmG from V7.2, but no one except me seems to be annoyed by these problems.)p  ? I don't think it is any worse than any other major O/S upgrade,5H probably much cleaner than most.  (You can't go by numbering - sometimesG a Vx.0 is only a minor upgrade from the previous version, and sometimestF a .n+1 version is major.  V7.3 has about 2 years of changes from V7.2,H and though it looks mostly the same on the outside has a lot of internalH changes.  The new XFC cache is a big change.  There have also been majorC changes in the lock manager, volume shadowing, SMP support, clustere internals, and other places.  A Like I said, I think it is okay for hobbyist and test/developmentdA systems right now, but I would hold off on production systems foriD another few months, unless I really, really needed some new feature.   > -- , > Howard S ShubsF > "Run in circles, scream and shout!"  "I hope you have good backups!"  - Always have good backups before upgrading ;-)d   -- c John SantosC Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------   End of INFO-VAX 2001.461 ************************