1 INFO-VAX	Thu, 29 Jun 2000	Volume 2000 : Issue 360       Contents: Re: Balanced Viewpoint?  Re: Compaq advertises  RE: Compaq advertizes % Re: Does VMS require a SYSTEM account % Re: Does VMS require a SYSTEM account % Re: Does VMS require a SYSTEM account % Re: Does VMS require a SYSTEM account . DS90 issues a %xD1 when ending a print session2 Re: DS90 issues a %xD1 when ending a print session& Re: good news (for me,  I think) . . .& Re: good news (for me,  I think) . . .& Re: good news (for me,  I think) . . .& Re: good news (for me,  I think) . . .& Re: good news (for me,  I think) . . .& RE: good news (for me,  I think) . . . Re: Locate the hotfiles ! LPR from VMS v6.2 to Linux Redhat 7 Re: Northern Light vs. Google (and the winner is . . .) 7 Re: Northern Light vs. Google (and the winner is . . .) 7 Re: Northern Light vs. Google (and the winner is . . .) 7 Re: Northern Light vs. Google (and the winner is . . .) 7 Re: Northern Light vs. Google (and the winner is . . .) 7 Re: Northern Light vs. Google (and the winner is . . .) 7 Re: Northern Light vs. Google (and the winner is . . .) 7 RE: Northern Light vs. Google (and the winner is . . .) 7 Re: Northern Light vs. Google (and the winner is . . .) . Re: OpenVMS clusters vs other systems clusters. RE: OpenVMS clusters vs other systems clusters. Re: OpenVMS clusters vs other systems clusters. Re: OpenVMS clusters vs other systems clusters. OpenVMS loses big, was:  RE: Compaq advertises2 RE: OpenVMS loses big, was:  RE: Compaq advertises( Oracle Press Release, incl. OpenVMS info, Re: Oracle Press Release, incl. OpenVMS info, RE: Oracle Press Release, incl. OpenVMS info PARITY ERROR. Re: Remote access programs from Windows to VMS See how a device is spooled 
 Re: SMTP Mail 
 Re: SMTP Mail 
 Re: SMTP Mail 
 Re: SMTP Mail  SSH on vms?  Re: SSH on vms?  Re: SSH on vms?  Re: SSH on vms? K Re: Sun Porting Clearinghouse for Staroffice, anybody going to port to VMS? % Re: VAX based 1032 database to Oracle % Re: VAX based 1032 database to Oracle % Re: VAX based 1032 database to Oracle  Re: VAX on Intel?  Re: VAX on Intel?   F ----------------------------------------------------------------------  % Date: Wed, 28 Jun 2000 15:29:27 -0400 ' From: "Bill Todd" <billtodd@foo.mv.com>   Subject: Re: Balanced Viewpoint?( Message-ID: <8jdjl3$b65$1@pyrite.mv.net>  8 SysAdmin <djesys.nospam@fsi.net.mapson> wrote in message( news:395A2E6B.CBB90531@fsi.net.mapson...F > Of recent, much has been made of the importance of taking a balancedI > view of the computing world and the various options, but mostly focused G > on operating systems. I'd like to offer some perspectives from my own F > experience. Again, and as always, your mileage may vary considerably1 > from mine; so, please do not flame or derogate.   I Speaking only for myself, my inclination to flame or derogate tends to be L strongly influenced by the inflamatory/derogatory content of the material toJ which I'm reacting.  Though I've been known to lose patience after lengthyH interchanges in which the other party seems uninclined to assimilate and* respond to the issues I've tried to raise.    These discussions are9 > not meant to be exhaustive. Please review them as such.  >  > Multimedia > ---------- > I > BeOS is probably better suited to multimedia applications than OpenVMS, E > Linux or Windows. MacOS is currently deeply entrenched here, with a ? > large, loyal user base and a considerable stable of available H > applications. BeOS is modestly priced and runs on "commodity" hardware > (Intel CPUs only). >  > Multi-Tasking  > -------------  > I > Either OpenVMS or Linux are better suited to multitasking than Windows. H > Given the multimedia focus of BeOS, it is possible that it, too, is as > strong a player as MacOS.   G I'd suggest that NT may be at least as strong a multi-tasking player as @ MacOS, and perhaps at least as strong as BeOS and Linux as well.   >  > Business Servers > ---------------- > B > While Windows/NT is deeply entrenched here along with Novell and8 > commercial UN*X, W/NT is not as strong in multitasking  L There it is again:  could you be specific as to what's missing compared with Novell and Unixes?    or as stable as? > some of the alternatives may be under similar application and G > environmental conditions. It has the advantage of a large support and F > user base and a large stable of available applications. Hardware andG > software are both available at prices which are greatly attractive to I > all but the smallest, most cash-strapped businesses. Linux and *BSD are G > closing most of those gaps, while OpenVMS languishes, a distant "also J > ran" due to high cost to acquire and support both hardware and software,. > as well as a dearth of application software. >  > Business Desktops  > -----------------  > I > Windows/NT, and its successor W2K-Pro, remain the business desktop O.S. H > of choice with W/9x running closely along or slightly ahead. Again, weH > find the advantages to be a large, familar support and user base and a) > large stable of available applications.   J I can't resist pointing out this rare appropriate association of the words$ 'stable' and 'available' with Win9x.    Hardware and software are both G > available at prices which are greatly attractive to even the smallest F > business with the lowest (frequently non-existant) budgets. MacOS isG > also sometimes found here where the necessary application software is J > available. The various UN*X flavors have yet to make significant inroadsC > here. Budget constraints and other pressures may afford them some 3 > leverage to garner small portions of this market.  >  > Consumer Desktops  > -----------------  > H > W/9x remains the consumer desktop O.S. of choice. Here also, we find aG > large, familiar user base and large stable of available applications. F > Hardware and software are both available at prices which are greatlyE > attractive to the masses. The various UN*X flavors have yet to make D > significant inroads here; however, anti-Microsoft sentiment in theF > market may give such alternative platforms an advantage which can beC > exploited enough to capture at least another small portion of the C > market. Garnering user acceptance remains the greatest challenge.  > - > So, Why All The "Affordable OpenVMS" Noise? - > -------------------------------------------  > = > For those markets to which OpenVMS remains well suited, ...  > 	 > <MUSIC> / > "All we are say--ing, is give 'VMS a chance!" 
 > </MUSIC>  H Nothing to disagree with there.  But I still believe that VMS needs someH cosmetic alterations to make it more 'familiar', plus at least a nascentL (and visible) resurgence to garner interest, before it will be 'well suited'B to the very low end of the market (which I'll arbitrarily peg, forD discussion purposes, as the under-$2500-full-system segment, vs. theH under-$5000-full-system - including monitor, etc. - segment that I think: Compaq could place it in immediately without much effort).   >  > ---- > In closing, let me reiterate:  > G > Again, and as always, your mileage may (and quite probably will) vary > > considerably from mine; so, please do not flame or derogate.  ' Actually, I thought it was pretty good.    - bill   >  >  > David J. Dachtera    ------------------------------  % Date: Wed, 28 Jun 2000 17:52:20 +0000 - From: SysAdmin <djesys.nospam@fsi.net.mapson>  Subject: Re: Compaq advertises. Message-ID: <395A3B54.44F8CD33@fsi.net.mapson>   Rob Young wrote: >   > Just spotted what Kerry wrote: > 9 > >> From:        Main, Kerry[SMTP:Kerry.Main@compaq.com] 0 > >> Sent:        Tuesday, June 27, 2000 9:23 PM > >> To:  Info-VAX@Mvb.Saic.Com ' > >> Subject:     RE: Compaq advertizes  > >>' > >> Also, big ad ran in Boston Globe..  > >> > >> Note - Forbes article: 8 > >> http://www.forbes.com/tool/html/00/Jun/0626/mu7.htm > >>' > >> <<< Will they ever market VMS? >>>  > >>2 > >> Wait until tomorrows (June 28) announcements. > >>O > >> Folks on this list should like what they hear .. well, ok maybe one or two  > >> folks might not ..  > >> > E >         Let me guess.. .. one of them might call soccer , football? E >         And perchance have eaten kippers on occasion?  Surely eaten  >         fish and chips.  >  >         Spot on! > % >                                 Rob   @ As of 12:50 CDT (US, 17:50 Zulu), a search of news on Forbes.com returned only a single link:  5 http://www.forbes.com/tool/html/98/jul/0716/side1.htm   F ...and THAT is about Sega dreamcst, not a commercial operating system!   David J. Dachtera    ------------------------------  # Date: Wed, 28 Jun 2000 18:19:39 GMT = From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)  Subject: RE: Compaq advertizes0 Message-ID: <009EC494.5BEAB368@SendSpamHere.ORG>  x In article <910612C07BCAD1119AF40000F86AF0D80528449A@kaoexc4.kao.dec.com>, "Main, Kerry" <Kerry.Main@compaq.com> writes:/ >Wait until tomorrows (June 28) announcements.    J June 28 is here and more than 1/2 gone.  If there's an announcement, where was it made?   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM    ------------------------------    Date: 28 Jun 2000 14:51:07 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) . Subject: Re: Does VMS require a SYSTEM account+ Message-ID: <rXso9bHT7nk3@eisner.decus.org>   ^ In article <slkbod98jev84@corp.supernews.com>, "Peter Weaver" <peter.weaver@stelco.ca> writes:E > I am on the phone with a support person from the OpenVMS Management A > Station support group and she is telling me that the Management B > Station software *HAS* to run under the SYSTEM account. I am 99%H > confident that VMS does not require a SYSTEM account. I would considerE > any product that *has* to run from the SYSTEM account to be broken. F > Does anyone know (I am still on the phone so it is difficult to lookF > up right now) if running VMS without a SYSTEM account is a supported > configuration.  D Running without a SYSTEM account is not supported, because any batchA submissions from SYSTARTUP_VMS.COM have to have someplace to run. @ Running with a SYSTEM account that is disabled for all but BatchD is supported, and in fact is required in order to get accountabilityG of individual privileged system managers (that is, no shared accounts).   C If the OpenVMS Management Station does not function correctly other < than via an interactive login to SYSTEM, it is quite broken.   ------------------------------  % Date: Wed, 28 Jun 2000 15:11:40 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> . Subject: Re: Does VMS require a SYSTEM account, Message-ID: <395A4DE7.F2860DC6@videotron.ca>   Peter Weaver wrote: F > Does anyone know (I am still on the phone so it is difficult to lookF > up right now) if running VMS without a SYSTEM account is a supported > configuration.  M For a long time, I had a DISUSER floag on SYSTEM and experienced no problems. L I r-enabled it so that I could submit  batch jobs during startup (although I3 could have used SUBMIT/USER in the systartup file).   I However, I had another cryptic account name with the same UIC as SYSTEM.    < What happens to a file owned by a UIC which does not exist ?   ------------------------------   Date: 28 Jun 00 12:37:47 PST From: mckinneyj@cpva.saic.com . Subject: Re: Does VMS require a SYSTEM account( Message-ID: <3cmfnU6MZoNb@cpva.saic.com>  , In article <395A4DE7.F2860DC6@videotron.ca>,0  JF Mezei <jfmezei.spamnot@videotron.ca> writes:
 >[...snip...]  > > > What happens to a file owned by a UIC which does not exist ?   Nothing, try...     $ CREATE/OWNER=[12345,12345] t.t ^Z   ------------------------------   Date: 28 Jun 2000 19:51:00 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman). Subject: Re: Does VMS require a SYSTEM account6 Message-ID: <8jdkv4$edp$1@mailint03.im.hou.compaq.com>  ^ In article <slkbod98jev84@corp.supernews.com>, "Peter Weaver" <peter.weaver@stelco.ca> writes:D :I am on the phone with a support person from the OpenVMS Management@ :Station support group and she is telling me that the ManagementA :Station software *HAS* to run under the SYSTEM account. I am 99% G :confident that VMS does not require a SYSTEM account. I would consider D :any product that *has* to run from the SYSTEM account to be broken.E :Does anyone know (I am still on the phone so it is difficult to look E :up right now) if running VMS without a SYSTEM account is a supported  :configuration.   ?   Well, that 1% has it -- OpenVMS must have a SYSTEM username.    F   I tried an experiment that is a variant of this question a number ofH   years back -- simply disusering SYSTEM -- and a number of things went ?   flagrantly weird, and (worse) a few things went subtly weird.   C   There are a few things that are tied a little more tightly to the &   SYSTEM username than I might prefer.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 28 Jun 2000 20:04:57 +0200 > From: "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr>7 Subject: DS90 issues a %xD1 when ending a print session 3 Message-ID: <8jdeor$1td5$1@s2.feed.news.oleane.net>   5 We've got a printer behind a DS90M with UXC$TELNETSYM   C After any printing, we get an extra D1(hex) after the file content. @ Same thing when we logout the port from an administrator session  
 Any Idea ?   Jean-Franois Marchal  X9000 - LYON   ------------------------------  % Date: Wed, 28 Jun 2000 20:49:43 -0500 + From: Shael Richmond <ksrich@bellsouth.net> ; Subject: Re: DS90 issues a %xD1 when ending a print session - Message-ID: <395AAB37.FD3F4F50@bellsouth.net>    "Jean-Franois Marchal" wrote: > 7 > We've got a printer behind a DS90M with UXC$TELNETSYM  > E > After any printing, we get an extra D1(hex) after the file content. B > Same thing when we logout the port from an administrator session >   H Disable input flow control.  We saw this when DNAS first came out.  Most; of our printers ignored it, but a few spaced an extra line.    Shael    ------------------------------  % Date: Wed, 28 Jun 2000 14:54:30 -0400 ' From: "Bill Todd" <billtodd@foo.mv.com> / Subject: Re: good news (for me,  I think) . . . ( Message-ID: <8jdhjh$8mc$1@pyrite.mv.net>  8 John Nixon <jorlnixon@worldnet.att.net> wrote in message@ news:58n65.6038$AM4.353365@bgtnsc04-news.ops.worldnet.att.net...   ...   D > On our VAX VMS 6.X systems, when we put 3 or 4 thousand files in a > directory, we would haveD > all sorts of problems.   Now,  on hearing that VMS 7.2 fixed these	 problems,  > I have heard aD > proposal to allow the application to put over  60,000 files in one$ > directory.   I think this is nuts,> > but I have not had the opportunity to test the improvements. >  > Can anyone comment on this?   L 'Fixed' is a relative concept:  VMS directories are still 'flat' (as opposedK to, e.g., b-tree-structured) - it's only their caching behavior that's been 	 improved.   I Having 60K files in a single directory is 'nuts' only if the directory is D not efficiently randomly-accessible (e.g., b-tree-structured) by anyH criterion (e.g., name) normally used to access it - or if full-directoryK scan/list operations frequently occur.  Several file systems developed over C the past 5 - 10 years can efficiently handle normal (non-scan/list) 9 operations in directories with far more than 60K entries.    - bill   > 8 > Thanks for all your help in the past (and the future). >  >    ------------------------------   Date: 28 Jun 2000 20:03:27 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)/ Subject: Re: good news (for me,  I think) . . . , Message-ID: <8jdlmf$c44@gap.cco.caltech.edu>  R In article <8jdhjh$8mc$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes: > 9 >John Nixon <jorlnixon@worldnet.att.net> wrote in message A >news:58n65.6038$AM4.353365@bgtnsc04-news.ops.worldnet.att.net...  >  >....  > E >> On our VAX VMS 6.X systems, when we put 3 or 4 thousand files in a  >> directory, we would have E >> all sorts of problems.   Now,  on hearing that VMS 7.2 fixed these 
 >problems, >> I have heard a E >> proposal to allow the application to put over  60,000 files in oneu% >> directory.   I think this is nuts,e? >> but I have not had the opportunity to test the improvements.i >> >> Can anyone comment on this? >oM >'Fixed' is a relative concept:  VMS directories are still 'flat' (as opposedoL >to, e.g., b-tree-structured) - it's only their caching behavior that's been
 >improved. > J >Having 60K files in a single directory is 'nuts' only if the directory isE >not efficiently randomly-accessible (e.g., b-tree-structured) by anyfI >criterion (e.g., name) normally used to access it - or if full-directorywL >scan/list operations frequently occur.  Several file systems developed overD >the past 5 - 10 years can efficiently handle normal (non-scan/list): >operations in directories with far more than 60K entries.  H Putting 60k files in a directory on ANY OS is really asking for trouble,J even if that OS nominally supports that many files.  The only question is , how the OS will bite you for your impudence.  I For instance, let's say on Unix that 20k of the 60k files start with "a",o. and you only want to look at them, so you do:    % ls -al a*n  L and it barfs because the shell can't handle the resulting list of 20k files.F Or maybe your shell has been tricked out to handle that many, and the I result merely scrolls through 1000 pages really quickly, most of which goe' right off the top of the scroll buffer.v  K Or consider NT.  It doesn't take kindly to this type of abuse either.  WhenpC you try to open a directory with too many files in it using WindowsnL Explorer it will sit there thinking for a very, very long time.  I've got a J directory on my Unix/NT systems that has 830 subdirectories (it's the PDB H protein structure database, which is "hashed" by a two letter field thatK is part of the names of the entries) and even that takes a longish time to  E open up.  Good luck if you need to view this file list if it is beinge constantly updated!r  F On OpenVMS you could probably put all the files in, but even with the H improvements in directory handling it would take a long time to do many  delete operations.  K Or imagine that you have this 60k entry directory mounted over the network, I or that some operation has to stat() every file in a directory to figure c- out what the names of the subdirectories are.-  L A couple of hundred files in a directory is more than enough, a thousand is 0 pushing it, and 60k is just begging for trouble.   Regards,   David Mathog mathog@seqaxp.bio.caltech.edut? Manager, sequence analysis facility, biology division, Caltech p   ------------------------------    Date: 28 Jun 2000 13:49:33 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett),/ Subject: Re: good news (for me,  I think) . . .y, Message-ID: <0IwjBdV9t7tW@malvm2.mala.bc.ca>  ) In article <8jdhjh$8mc$1@pyrite.mv.net>, a.      "Bill Todd" <billtodd@foo.mv.com> writes: >  >> I have heard aaE >> proposal to allow the application to put over  60,000 files in oneg% >> directory.   I think this is nuts,r? >> but I have not had the opportunity to test the improvements.t >> >> Can anyone comment on this? > N > 'Fixed' is a relative concept:  VMS directories are still 'flat' (as opposedM > to, e.g., b-tree-structured) - it's only their caching behavior that's beenb > improved.t > K > Having 60K files in a single directory is 'nuts' only if the directory istF > not efficiently randomly-accessible (e.g., b-tree-structured) by anyJ > criterion (e.g., name) normally used to access it - or if full-directoryM > scan/list operations frequently occur.  Several file systems developed over-E > the past 5 - 10 years can efficiently handle normal (non-scan/list)n; > operations in directories with far more than 60K entries.  >   I     IIRC they're flat, but maintained in sorted order ( by filename ), so K non-wildcard searches can be (and are) done using a binary search. It wouldeG seem that in the special case of a stable set of files which are always-J accessed by fully qualified name 60,000 files might not be a problem. ThisF is of course of no help for full scan operations or for wildcards. TheG requirement to maintain the directory in sorted order is what is at thehJ heart of the slowness one encounters when creating or deleting files which$ are not at the end of the directory.  E     Having said that, I'd still agree that 60K files in one directoryr
 is 'nuts'.   ------------------------------  % Date: Wed, 28 Jun 2000 20:57:10 -0400v' From: "Bill Todd" <billtodd@foo.mv.com>c/ Subject: Re: good news (for me,  I think) . . .p( Message-ID: <8je6re$s7v$1@pyrite.mv.net>  = David Mathog <mathog@seqaxp.bio.caltech.edu> wrote in message & news:8jdlmf$c44@gap.cco.caltech.edu...L > In article <8jdhjh$8mc$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes:    ...i  G > >'Fixed' is a relative concept:  VMS directories are still 'flat' (as2 opposedmI > >to, e.g., b-tree-structured) - it's only their caching behavior that's  been > >improved. > > L > >Having 60K files in a single directory is 'nuts' only if the directory isG > >not efficiently randomly-accessible (e.g., b-tree-structured) by any K > >criterion (e.g., name) normally used to access it - or if full-directory I > >scan/list operations frequently occur.  Several file systems developed  overF > >the past 5 - 10 years can efficiently handle normal (non-scan/list)< > >operations in directories with far more than 60K entries. >hJ > Putting 60k files in a directory on ANY OS is really asking for trouble,K > even if that OS nominally supports that many files.  The only question ish. > how the OS will bite you for your impudence. >aK > For instance, let's say on Unix that 20k of the 60k files start with "a",-/ > and you only want to look at them, so you do:. >o
 > % ls -al a*a >,G > and it barfs because the shell can't handle the resulting list of 20kl files.  L Sounds like the exception I noted, extended only slightly (instead of 'full'J directory scan/list operations, perhaps I should have said 'large portion'K scan/list operations - but I gave people credit for a bit of common sense).v  G > Or maybe your shell has been tricked out to handle that many, and thetK > result merely scrolls through 1000 pages really quickly, most of which goc) > right off the top of the scroll buffer.    Same observation.A   >aG > Or consider NT.  It doesn't take kindly to this type of abuse either.  WhenE > you try to open a directory with too many files in it using Windows A > Explorer it will sit there thinking for a very, very long time.o   Same observation (yawn).     I've got aK > directory on my Unix/NT systems that has 830 subdirectories (it's the PDBpJ > protein structure database, which is "hashed" by a two letter field thatL > is part of the names of the entries) and even that takes a longish time toG > open up.  Good luck if you need to view this file list if it is beingn > constantly updated!-   Need I say it again?   >oG > On OpenVMS you could probably put all the files in, but even with theoI > improvements in directory handling it would take a long time to do manyi > delete operations.  5 Well, yuh - that's why a tree structure is desirable.b   >oD > Or imagine that you have this 60k entry directory mounted over the network,J > or that some operation has to stat() every file in a directory to figure/ > out what the names of the subdirectories are.   L I can imagine a great many things that you might not want to do with a largeG directory, at least using common applications.  Just as I can imagine a H great many things that you might not want to do with a large file, usingI common applications.  That does not mean that either is unreasonable when ! the situation makes it desirable.>   >e? > A couple of hundred files in a directory is more than enough,s   For most purposes.    a thousand is
 > pushing it,h  - On systems not designed to handle such sizes.m  %  and 60k is just begging for trouble.S  E Only if you use it in places likely to be encountered by programs not K designed to handle it gracefully.  Conversely, for special situations where J it's useful, the alternative (creating an artificial sub-tree to limit theK number of entries in any given level) is a pain, and most operating systemsoD provide mechanisms ('hidden' or 'system' files that are not normallyI displayed in Explorer, for example, or access-control limitations to keep,? users from accessing it save for the few who need to) to keep ag5 special-purpose large directory from being a problem./   - bill   >.
 > Regards, >o > David Mathog > mathog@seqaxp.bio.caltech.edue@ > Manager, sequence analysis facility, biology division, Caltech   ------------------------------  % Date: Thu, 29 Jun 2000 11:44:58 +1000n/ From: "Phil Howell" <howellp@snowyhydro.com.au>z/ Subject: Re: good news (for me,  I think) . . .R2 Message-ID: <xSx65.25059$N4.980623@ozemail.com.au>  8 John Nixon <jorlnixon@worldnet.att.net> wrote in message@ news:58n65.6038$AM4.353365@bgtnsc04-news.ops.worldnet.att.net...L > After several false starts, and millions of dollars spent, we have finally > decided that it is better toI > migrate our VAX applications to Alpha VMS than to rewrite them on Unix.- >-I > Well, at least we have decided that we need to move the applications to@ > Alpha VMS for nowaH > because we don't know how long it will take us to move to unix, and we don't  > want to be > stuck on VAX indefinitely. ><G > SOooooo,  I am probably going to be asking a lot of VAX to Alpha type> > questions. >o > Here is a starter question:o >rD > I am intimately aware of the performance problems related to large > directories on pre VMS7.2.I > I am also aware of the improvements to large directory performance with  VMSV > 7.2, >  >u- > But,  HOW MUCH    improvement can I expect?r >cD > On our VAX VMS 6.X systems, when we put 3 or 4 thousand files in a > directory, we would haveD > all sorts of problems.   Now,  on hearing that VMS 7.2 fixed these	 problems,  > I have heard aD > proposal to allow the application to put over  60,000 files in one$ > directory.   I think this is nuts,> > but I have not had the opportunity to test the improvements. >h > Can anyone comment on this?l >fF This behaviour is fairly common in unix applications that use the fileF system as a sort of distributed lock manager and create files based onJ pid/date/time/user etc. if you can isolate the routines that do this, thenG when you port it to vms you will have different options for locking and. inter-process communication- Phil   ------------------------------  % Date: Wed, 28 Jun 2000 23:35:51 -0400i+ From: "Main, Kerry" <Kerry.Main@compaq.com> / Subject: RE: good news (for me,  I think) . . .dJ Message-ID: <910612C07BCAD1119AF40000F86AF0D8052844A8@kaoexc4.kao.dec.com>   John,   J You (and others doing OpenVMS VAX to Alpha migrations) might be interested in the following urls:8 <http://www3.sympatico.ca/n.rieck/docs/alpha_diary.html>? <http://WWW.OPENVMS.DIGITAL.COM:8000/72final/6459/6459PRO.HTML>s? <http://WWW.OPENVMS.DIGITAL.COM:8000/72final/5601/5601pro.html>M   Regards,  
 Kerry Main Senior Consultant,
 Compaq Canaday Professional Servicest Voice : 613-592-4660 FAX   : 819-772-7036 Email : kerry.main@compaq.coms       -----Original Message-----4 From: John Nixon [mailto:jorlnixon@worldnet.att.net]& Sent: Wednesday, June 28, 2000 9:33 AM To: Info-VAX@Mvb.Saic.Comu* Subject: good news (for me, I think) . . .    J After several false starts, and millions of dollars spent, we have finally decided that it is better toG migrate our VAX applications to Alpha VMS than to rewrite them on Unix.   G Well, at least we have decided that we need to move the applications to- Alpha VMS for now L because we don't know how long it will take us to move to unix, and we don't
 want to be stuck on VAX indefinitely.  E SOooooo,  I am probably going to be asking a lot of VAX to Alpha typen
 questions.   Here is a starter question:,  B I am intimately aware of the performance problems related to large directories on pre VMS7.2.K I am also aware of the improvements to large directory performance with VMSa 7.2,    + But,  HOW MUCH    improvement can I expect?E  B On our VAX VMS 6.X systems, when we put 3 or 4 thousand files in a directory, we would haveL all sorts of problems.   Now,  on hearing that VMS 7.2 fixed these problems, I have heard aB proposal to allow the application to put over  60,000 files in one" directory.   I think this is nuts,< but I have not had the opportunity to test the improvements.   Can anyone comment on this?n  6 Thanks for all your help in the past (and the future).   ------------------------------  # Date: Wed, 28 Jun 2000 23:36:23 GMT  From: kparris@my-deja.coml  Subject: Re: Locate the hotfiles) Message-ID: <8je25l$5hh$1@nnrp1.deja.com>    twong@livingstonintl.com wrote:oG > One of the disk has much higher i/o rate in comparing with others.  I-@ > desire to enable some global buffer counts on those hot files. > ; > Is there any way to locate the files with high i/o rate ?a  B Do SET FILE/STATISTICS on the files you suspect might be the majorG ones, and use MONITOR RMS/FILE=xxx/ITEM=ALL to watch them while they'rexF busy and see which ones are the most active.  MONITOR RMS will also beB handy in helping you size your global buffer counts, as it reportsC (under /ITEM=CACHING) what the global buffer hit rates are, and hitaE rates are a good indication of how much good global buffers are doing F for you -- low hit rates may indicate too few global buffers, but thenD at some point, if you raise the global buffer count and the hit rate> doesn't increase any more, you've probably passed the point ofE diminishing returns and were probably doing about as well as could bes expected with the last count.s  G Note that a file unfortunately has to be closed before you can do a SETiB FILE/STATISTICS or SET FILE/GLOBAL_BUFFER=n command on it, so thisD generally implies at least a short application outage while you turn" things on or change buffer counts.  E If you really have no idea what files are the likely suspects, or you.E don't know enough about your application and fear you might miss some F very important ones, consider taking advantage of some statistics thatG VMS has been gathering since dynamic lock remastering was introduced ineF V5.5, but which (to my knowledge) are not made visible yet in any 3rd-E party performance product or VMS tool (save SDA) -- the lock activity6B counters in root resource blocks.  In my experience, lock activity? rates for RMS files tend to correlate well with actual file I/O C activity, so files with high lock activity rates tend to be the hotnE files in I/O terms as well.  I did a presentation at DFWdays covering 	 this; seepG http://www.geocities.com/~kparris/decus_presentations/s2000_lockmgr.pptsC for details and a pointer to a sample procedure (LOCK_ACTV.COM) and.G program (LCKACT.MAR) which can display the lock activity statistics forM@ you.  The files listed first in its output are the hottest ones.G -----------------------------------------------------------------------iG Keith Parris|Integrity Computing,Inc.|parris@decuserve.decus.org-nospamoF VMS Consulting: Clusters, Perf., Alpha porting, Storage&I/O, Internals    & Sent via Deja.com http://www.deja.com/ Before you buy.r   ------------------------------  # Date: Thu, 29 Jun 2000 00:00:02 GMTl From: sboulay@my-deja.com * Subject: LPR from VMS v6.2 to Linux Redhat) Message-ID: <8je3hq$6dg$1@nnrp1.deja.com>r   Help!n  F We are desperately trying to print from VMS to a Linux box via LPR. WeE send the job but all we get is a "Retained on error". Here's the fulln message:  $ BASE> show queue UCX$LPD_QUEUE /full@ Server queue UCX$LPD_QUEUE, idle, on XYZ::, mounted form DEFAULTA   /BASE_PRIORITY=4 /DEFAULT=(FEED,FORM=DEFAULT) /NOENABLE_GENERICjF   /OWNER=[SYSTEM] /PROCESSOR=UCX$LPD_SMB /PROTECTION=(S:M,O:D,G:R,W:S)   /RETAIN=ERRORr  4   Entry  Jobname         Username     Blocks  Status4   -----  -------         --------     ------  ------?      34  LOGIN           SBOULAY           3  Retained on error ,        %UCX-F-NOMSG, Message number 030AEB74F          Submitted 28-JUN-2000 18:53:08.77 /FORM=DEFAULT /PRIORITY=1007          File: _XYZ$DKA0:[VMS$COMMON.SYSMGR]LOGIN.COM;3lA          Completed 28-JUN-2000 18:53:08.85 on queue UCX$LPD_QUEUE   "  and here's our printcap.dat file: #o # LOCAL PRINTERS #  UCX$LPD_QUEUE:\v         :lp=UCX$LPD_QUEUE:\o         :sd=UCX$LPD_SPOOL: #  LOCAL1|local1:\t.         :lf=/SYS$SPECIFIC/UCX_LPD/LOCAL1.LOG:\         :lp=LOCAL1:\         :rm=10.0.0.98:\i         :rp=lp0:\t)         :sd=/SYS$SPECIFIC/UCX_LPD/LOCAL1:e #n      F What are we doing wrong? The Linux box at 10.0.0.98 is quite alive andG pinging back. We also tried printing via lpr from Win2000 and it works.    Please respond! Thanks!-  
    Sylvain    & Sent via Deja.com http://www.deja.com/ Before you buy.l   ------------------------------  % Date: Wed, 28 Jun 2000 15:03:16 -0400m- From: JF Mezei <jfmezei.spamnot@videotron.ca>w@ Subject: Re: Northern Light vs. Google (and the winner is . . .), Message-ID: <395A4BEF.6DD25E3A@videotron.ca>   Nigel Arnot wrote:G > Seems to me that this is a good case for the "if it ain't broke don'tuE > fix it" school of system management. In other words, never upgrade, E > just throw out the originally installed software with the hardware!o  L Once worked at a bank which *had* the policy. They knew that upgrades causedL many headaches and unscheduled outages/problems so they wanted to reduce it.  N But when the time came to bring in new hardware that required a new version ofM the OS, which in turn required a new version of many of the middleware (CICS, N IMS, DB2 etc) and a recompile of all apps, that simple hardware upgrade turned into a huge project.  K After that project, it was quickly decided that in order to keep a smoothereN flow, the bank should keep its software "current"  so that any new software orG hardware that requires recent software will not cause a major upheaval, A Continuous upgrades ended up costing less and provided for bettersN time/resource planning because it doesn't monopolise all your staff for months on a single project.   L Look at the Y2K problem on large IBM mainframes. If companies had kept theirI software up-to-date with regular updates to them, one of the big problemsnE would not have occured: having to recompile a very old source code ont compilers which had evolved.  K Another good example is the VAX-C to DEC-C. Regular software updates on theuN VAX would have eventually had much of the softwrae "migrated" to  DEC-C on theK vax which means that when it was moved to an alpha, it was not a big issue.gM But keep your software stale, and when forced to migrate to Alpha, it becomesa a big project.   ------------------------------  % Date: Wed, 28 Jun 2000 20:36:08 -0400-' From: "Bill Todd" <billtodd@foo.mv.com> @ Subject: Re: Northern Light vs. Google (and the winner is . . .)( Message-ID: <8je5k5$r9p$1@pyrite.mv.net>  5 Rob Young <young_r@eisner.decus.org> wrote in messageh% news:GGnph8Yw4phY@eisner.decus.org...TE > In article <009EC486.C7B2CB7A.50@maxwell.ph.kcl.ac.uk>, Nigel Arnot % <sysmgr@maxwell.ph.kcl.ac.uk> writes:0   ...l  I > > Imagine that the next-generation server-farm box is using a Transmeta L > > Crusoe processor or similar. That means it dynamically adjusts its clockG > > speed to match the workload and minimize the power drain. The poweroH > > consumption of the server farm will therefore tend to scale with theJ > > level of activity. This gets even better if spin-down of inactive diskE > > drives can also be incorporated. If they have multiple nodes with G > > identical disk data mirrored on each for redundancy and peak loads,tK > > then at quiet times letting one out of N take all the strain might work I > > well. Or one of the non-rotating storage technologies may replace thel disk2 > > drive making the idle power become negligible. > >i >f> > But the difficulty here is that none of them are idle enough$ > to even spin down the disk drives.  F I'm not sure you understood what Nigel was saying.  If the servers areK indeed mirroring each other (and index modifications aren't constant), thenaL there will certainly be plenty of times when one server of a mirror pair canL easily handle the load and the other can spin down its disks and put its CPU> into stand-by mode without noticeably affecting response time.  H Unless host-mirrored large servers are also able to monitor activity andI spin down unneeded disks in a similar manner, this could make the low-end E server farm *more* energy efficient than the centralized alternative.-     Couple that with the= > fact they have to spin up the disk drives and you see where@@ > their edge of a very fast search suddenly goes in the opposite> > direction and they become one of the slower search services.   ...i  B > But what if Google is only 5% efficient?  Sure, they are getting< > the job done and done well.. but they are wasting valuable= > resources and only a matter of time before the tree huggerse > catch on.S  L I'm not sure what you're suggesting.  Since Google is using reasonably largeK disk drives, their disk power consumption seems likely to about the same as G any alternative.  Since they're likely using less than state-of-the-arttL close-to-1-GHz CPUs, that portion of the load shouldn't be *too* much worse.) So where is the comparative inefficiency?-   - bill   >e > Rob  >s   ------------------------------  % Date: Wed, 28 Jun 2000 19:36:18 -0500r* From: Keith Brown <kbrown780@usfamily.net>@ Subject: Re: Northern Light vs. Google (and the winner is . . .), Message-ID: <395A9A02.9BE71E23@usfamily.net>   JF Mezei wrote:o >  > Keith Brown wrote:D > > Good Grief! What about the cost of the army of support people to? > > keep it running. Running 10,000 of anything would be a realc
 > > bitch. > I > If this is funded by Intel, then it doesn't matter how much it costs to L > maintain or how difficult such a 10000 billybox network is to manage. WhatL > matters is that Intel can claim its 8086 is fully capable to process large- > amounts of data just like the big guys can.i > N > Think of Northen Light which was somewhat a poster boy for VMS, proving thatO > VMS could be a "me too" in the search engine business. So Intel also wanted am! > "me too" and now they have one.l > J > And if they can show that they have mamangement tools to manage 1000s ofL > billyboxes, then it becomes great advertising to companies who may have toK > manage 100s of billyboxes since they are confident that this software cana! > handle their much smaller load.h  @ More likely they really do have an army but probably won't admit it.  -- w Keith Brownr kbrown780@usfamily.net   ------------------------------  % Date: Wed, 28 Jun 2000 19:43:03 -0500a* From: Keith Brown <kbrown780@usfamily.net>@ Subject: Re: Northern Light vs. Google (and the winner is . . .), Message-ID: <395A9B97.1EFFFEE5@usfamily.net>   Rob Young wrote: > T > In article <8jbdp0$efp$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes: > >v8 > > Main, Kerry <Kerry.Main@compaq.com> wrote in messageH > > news:910612C07BCAD1119AF40000F86AF0D805284497@kaoexc4.kao.dec.com...
 > >> Bill, > >> > F >         Even though David J. thinks Bill needs a vacation, I believeE >         he is spot on here.  As Bill points out, this is replicateduF >         hither and yon, backups are not necessary.. automated serverG >         setup is a must and as mentioned in the Intel eBusiness piece)C >         they "Googal-ize" the servers (envisioning plugging it indG >         and answering a few questions in automated scripts, i.e. IP).w > A >         My initial criticisms/cynicisms were along the lines of C >         employee head count (7000 drives, with great MTBF someoneoH >         is changing out several drives a week, etc.) power consumptionI >         (Environmentally very harmful.  If nothing else , could rile up D >         a few tree huggers and point out how many resources GoogleD >         is wasting... ironically this could be one of the greatestC >         criticisms).  Not to be overlooked is the whole manner ofSH >         efficiency which points a finger at the power consumption too. > D >         It seems they want it both ways, one argument for multiple4 >         PCs is because of computational intensity: > 4 > http://www.techweb.com/wire/story/TWB20000530S0011 > D >         Yeah, right.  Bet those 4000 CPUs are really humming doingI >         13 million searches per day... at 8 million searches peak hourspJ >         i.e. 8 hour period, that's 278 searches per second.  Yeah, I bet/ >         those 4000 CPUs are really straining.  > A >         But more realistically, from the Intel eBusiness piece:  > O > http://www.intel.com/ebusiness/estrategies/casestudies/snapshots/google_p.htmt > K > "In Google's environment, disk I/O performance is an overriding factor  "u > C >         In fairness, maybe each PC has data and index such that a J >         portion of the index is cachable (i.e. hot)... unscientific testE >         of "monica clinton" takes .24 seconds and a retry takes .03tI >         seconds.  Smart enough to redirect to same cluster and the term.H >         is still in cache at 10:35 p.m. EDT.  Again, "hillary chelsey"F >         takes .27 seconds, retry .05 seconds.  A term that is likelyF >         in cache?  How about genome?  Yep.  .06 seconds.  Now, let'sL >         see how "genome soup" does .. .37 seconds.  Terms to be in memory:B >         "genome dna" , so it does do quite well on cached index. > H >         My point is that not all index is cached and when it has to goC >         to disk, real-world results.  Again, a single term not inoH >         cache:  "zugswang" took .31 seconds.  No "and" involved so theH >         overhead of .06 seconds is saved.. thankfully not taxing thoseF >         very busy CPUs.  But that's a good data point.. depending onL >         the term it takes .24 to .31 seconds (if not in cache I suppose)..* >         i.e. 240 to 310 milliseconds ;-) > K >         But not much can be cached apparently as 3 minutes later "clintoneM >         soup" takes .54 seconds to resolve [oops, not true, "soup" isn't ino> >         cache, see below].  Hillary is still there, "clintonI >         hillary" takes .06 seconds.  "dna soup" takes .49 seconds.  Ah,iJ >         soup is finally cached on "genome soup" so perhaps it takes 5 orM >         6 go rounds to trick up the caching algorithm.  Yep, "chelsey soup"eL >         takes a while (chelsey isn't cached) but "dna soup" is now flying.N >         How about "doj"?   Yep.. .06, "reno"?  .31 seconds.  "clinton reno"?K >         Slow...  Footnote:  Took about 10 go rounds to force chelsey into-G >         cache.. so maybe someone else had searched for "soup" lately.  >  >         Enough.i > H >         For now trying to come up with a rebuttal to Bill's valid caseJ >         in his initial reply.  Didn't want to be too flippant and I wantI >         to slightly tweek the initial scenario.  Still believe Northern G >         Light will come out on top in all this but it will take aboutyA >         a year or a year and a half for Marvel (Terabyte MemoryT% >         machine) to arrive (guess).0 > G >         Best campaign Northern Light could come up with then would be M >         the real-time indexing aspect and a shot of their parking lot and a0M >         shot of their lights-out operation with the staff reading magazinesnO >         and cut to a shot to a nameless Internet Search Service abuser of theMJ >         environment showing shots of their parking lots and the acres ofN >         servers and a quick pan to the power meters out back that have smokeJ >         coming off of them because they are whizzing around so fast.  ToP >         really stick it to them, perhaps a few fish floating on the surface ofC >         a pristine stream; boiled alive by nuclear heat effluent.g > . >         More about real-time indexing later. > % >                                 Robo    : I don't believe that replication is a good replacement for> backup. Suppose your data is deleted or corrupted, should this be replicated too?   -- o Keith Brown  kbrown780@usfamily.net   ------------------------------  % Date: Wed, 28 Jun 2000 21:14:49 -0400m' From: "Bill Todd" <billtodd@foo.mv.com> @ Subject: Re: Northern Light vs. Google (and the winner is . . .)( Message-ID: <8je7sg$t23$1@pyrite.mv.net>  5 Keith Brown <kbrown780@usfamily.net> wrote in messagen& news:395A9B97.1EFFFEE5@usfamily.net...   ...a  < > I don't believe that replication is a good replacement for@ > backup. Suppose your data is deleted or corrupted, should this > be replicated too?  I In this particular situation, replication is an excellent replacement forpK backup, since it protects you from media failure, and in the unlikely event,J (since the process is almost fully automated) of corruption the portion ofE the database that was lost can be reconstructed from the original Webg; material - a process which is continually occurring anyway.a  L And, as I said before, if that's not sufficiently comforting, you can alwaysJ capture and save the incoming updates and use them to reconstruct the lost? portion of the data, to get it back on line a bit more quickly.   D Indeed, for this large and volatile an environment, backup is prettyK unattractive.  For that matter, it's convincingly arguable that traditionallE backup is in general on the way out, given the decreasing gap in costpH between on-line and off-line storage and the increasing relative cost ofA creating, organizing, and storing said off-line storage:  keepingoL differential (to minimize space requirements; perhaps partial as well, sinceB a lot of data is easily recoverable by other means) redundant (andL separated, to guard against both media and whole-site failures) snapshots onI line is one promising alternative, and particularly attractive in that ithL can allow users to recover from their own mistakes without invoking the Gods of the Data Center.A   - bill   >F > --
 > Keith Browno > kbrown780@usfamily.net   ------------------------------    Date: 28 Jun 2000 22:28:48 -0500* From: young_r@eisner.decus.org (Rob Young)@ Subject: Re: Northern Light vs. Google (and the winner is . . .)+ Message-ID: <eZLwVoBSmsRd@eisner.decus.org>   Y In article <395A9B97.1EFFFEE5@usfamily.net>, Keith Brown <kbrown780@usfamily.net> writes:    > < > I don't believe that replication is a good replacement for@ > backup. Suppose your data is deleted or corrupted, should this > be replicated too? >   : 	There must be nice journalling techniques here if you are; 	committed to it.  i.e. critical stuff is journalled.  Hardy8 	to corrupt flat files.  Be very careful doing copies orG 	synchronizations.  A program/interface that won't allow you to destroy  	things would be convenient.  > 	Besides, see the other thread.  How does one go about backing> 	up 20, 30, 80 Terabytes over 100 MBit pipes?  Very carefully,
 	I suppose.     @ 	If the data gets stomped, crawl the web from scratch and suffer: 	the consequences?  Maybe.  Or take a few weeks to restore@ 	the 12 Terabytes?  Maybe you only lost or corrupted 5000 files?   	Crawl the web for them.   				Rob-   ------------------------------    Date: 28 Jun 2000 22:22:45 -0500* From: young_r@eisner.decus.org (Rob Young)@ Subject: Re: Northern Light vs. Google (and the winner is . . .)+ Message-ID: <ZOWe2atIF5jK@eisner.decus.org>/  R In article <8je5k5$r9p$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes: > 7 > Rob Young <young_r@eisner.decus.org> wrote in message2' > news:GGnph8Yw4phY@eisner.decus.org...$F >> In article <009EC486.C7B2CB7A.50@maxwell.ph.kcl.ac.uk>, Nigel Arnot' > <sysmgr@maxwell.ph.kcl.ac.uk> writes:j >  > ...w > J >> > Imagine that the next-generation server-farm box is using a TransmetaM >> > Crusoe processor or similar. That means it dynamically adjusts its clockFH >> > speed to match the workload and minimize the power drain. The powerI >> > consumption of the server farm will therefore tend to scale with theRK >> > level of activity. This gets even better if spin-down of inactive disk F >> > drives can also be incorporated. If they have multiple nodes withH >> > identical disk data mirrored on each for redundancy and peak loads,L >> > then at quiet times letting one out of N take all the strain might workJ >> > well. Or one of the non-rotating storage technologies may replace the > disk3 >> > drive making the idle power become negligible.d >> > >>? >> But the difficulty here is that none of them are idle enough % >> to even spin down the disk drives.s > H > I'm not sure you understood what Nigel was saying.  If the servers areM > indeed mirroring each other (and index modifications aren't constant), then N > there will certainly be plenty of times when one server of a mirror pair canN > easily handle the load and the other can spin down its disks and put its CPU@ > into stand-by mode without noticeably affecting response time. >     < 	Oh... seems that isn't the way it works though... the index9 	is split amongst chunks of 40 servers and load balancers ? 	fire to an idle server.  I may be misinterpreting this though:a  - http://www.internetwk.com/lead/lead060100.htmh  D "Within the data centers, Google uses its own traffic management andN load-balancing software to direct traffic to the best server. The index of theN Web is broken down into parts, with each section of the index distributed to a9 cluster of about 40 servers for redundancy and failover."h  < 	If they have many more servers than they need, that is very< 	wasteful.  So the 4000 must come into play (certainly isn't= 	for the storage needs).  Maybe they have 2 extra servers perm= 	nodule?  In other words, those 40 per grouping are probably  / 	tickled periodically during peak search times.a  - 	Perhaps you can give us your interpretation?   7 	I'm not a network guy, but maybe that 40 server number 7 	wasn't just arbitrary.  How many ports in a large hub?y  ? 	So far, that is the best overview of what they have that I caneC 	find.  The actual thesis itself was particularly troubling to find ' 	and I have already lost the link :-( .     J > Unless host-mirrored large servers are also able to monitor activity andK > spin down unneeded disks in a similar manner, this could make the low-end G > server farm *more* energy efficient than the centralized alternative.i >   < 	Don't know about that.  Each server has 2 disks... year end; 	they have 2 * 10000 disks or 20000 disks.  I want to say ao? 	large search service disk farm has a 1000 18 gig drives, maybe  	more but how much more?     >   Couple that with the> >> fact they have to spin up the disk drives and you see whereA >> their edge of a very fast search suddenly goes in the oppositec? >> direction and they become one of the slower search services.e >  > ...p > C >> But what if Google is only 5% efficient?  Sure, they are getting = >> the job done and done well.. but they are wasting valuable > >> resources and only a matter of time before the tree huggers >> catch on. > N > I'm not sure what you're suggesting.  Since Google is using reasonably largeM > disk drives, their disk power consumption seems likely to about the same asnI > any alternative.  Since they're likely using less than state-of-the-artpN > close-to-1-GHz CPUs, that portion of the load shouldn't be *too* much worse.+ > So where is the comparative inefficiency?m >   9 	From an earlier post... if they are now up to 18 million C 	searches a day and within an 8 hour period they perform 12 millionaD 	of them, they are doing 417 searches per second.  Since each search? 	appears to take .5 second (actually less), those 4000 CPUs arev4 	mostly idle.  Even less than 5% loaded I would say.  B 	Each server has 80 Gigs of storage from the above article... from> 	another article (in May I believe) they talked about using 80D 	Terabytes of storage out of the 320 Terabytes in the servers.  OverF 	20% in use there.  But now they are up to 500 million articles, maybe@ 	12 Terabytes for actual data.  I'll have to dig to see if I canC 	estimate that or if that info is readily available.  If 12 or evenoC 	20 .. 20 Terabytes means they are actually using 6% of the storageoB 	available for data with over 1 Terabyte for index (from the aboveH 	article).  The rest replicated hither and yon.  But storage efficiency C 	is one of those things hard to pin down.. wasn't really moaning a  E 	whole lot about that.. but using less than 5% of the CPU horsepower aD 	available does seem to be quite a waste of resources.  But they do I 	point out their index phase is very CPU intensive.  Whatever that means.i  F 	Let's talk next about the 20000 fans they will have spinning year-endE 	(2 per server, right?).  That seems like an interesting topic.  ThentG 	off to the waste heat a power supply uses -- 10000 of those, year-end.k   				Rob    ------------------------------  % Date: Wed, 28 Jun 2000 22:10:05 -0400o+ From: "Main, Kerry" <Kerry.Main@compaq.com>s@ Subject: RE: Northern Light vs. Google (and the winner is . . .)J Message-ID: <910612C07BCAD1119AF40000F86AF0D8052844A4@kaoexc4.kao.dec.com>   re: last few notes in thread ..n  J I suspect this debate could go on forever in the sense that is essentiallyJ the argument for massively distributed computing vs centralized computing.  H While the massively distributed model does have a place in the technicalL computing realm, it very much depends on the applications being written suchG that all of the processors and storage devices can be utilized. This iseI almost always home grown, heavily customized applications. This does have-F advantages in the some technical computing arenas as the recent CeleraL Genomics project has shown. The SETI project is likely another good example.  I However, these massively distributed compute and storage environments arezK often based on the premise of minimal backups as the data can be re-enteredt8 or re-built or restarted from the last check-point etc.   J This is certainly not the case when dealing with mission critical businessK or other types of applications that deal with critical data that either canlJ not be re-entered at all, ot at best without a great deal of effort and/orK political embarrasement. The recent Hotmail issues of losing end user emaillK information is a good example (many other examples available today) of whatt4 happens in todays world when you lose Customer data.  F In this mission critical environment, data replication is certainly noK replacement for backups. A file created yesterday, but accidentally deletedlL or corrupted today typically must be restorable from the last nights backup.    L Data replication in most mission critical applications is usually defined asL an availability feature - not a data backup feature. It is why shadowing (orI mirroring) is recommended as an availability feature that complements then  data integrity backup processes.  ? Yes, big storage arrays are pushing the backup windows and tapeeJ technologies, but this does not negate the need for backups, it just meansF that additional creative technologies need to be adopted to handle it.   Regards,  
 Kerry Main Senior Consultant,
 Compaq Canadan Professional Serviceso Voice : 613-592-4660 FAX   : 819-772-7036 Email : kerry.main@compaq.com        -----Original Message-----@ From: young_r@eisner.decus.org [mailto:young_r@eisner.decus.org]' Sent: Wednesday, June 28, 2000 11:29 PMi To: Info-VAX@Mvb.Saic.ComI@ Subject: Re: Northern Light vs. Google (and the winner is . . .)    8 In article <395A9B97.1EFFFEE5@usfamily.net>, Keith Brown  <kbrown780@usfamily.net> writes:   > < > I don't believe that replication is a good replacement for@ > backup. Suppose your data is deleted or corrupted, should this > be replicated too? >   : 	There must be nice journalling techniques here if you are; 	committed to it.  i.e. critical stuff is journalled.  Hard 8 	to corrupt flat files.  Be very careful doing copies or? 	synchronizations.  A program/interface that won't allow you top destroyu 	things would be convenient.  > 	Besides, see the other thread.  How does one go about backing> 	up 20, 30, 80 Terabytes over 100 MBit pipes?  Very carefully,
 	I suppose.  .  @ 	If the data gets stomped, crawl the web from scratch and suffer: 	the consequences?  Maybe.  Or take a few weeks to restore@ 	the 12 Terabytes?  Maybe you only lost or corrupted 5000 files?   	Crawl the web for them.   				Robw   ------------------------------  % Date: Wed, 28 Jun 2000 21:41:12 -0500n* From: Keith Brown <kbrown780@usfamily.net>@ Subject: Re: Northern Light vs. Google (and the winner is . . .), Message-ID: <395AB748.1C62A193@usfamily.net>   Bill Todd wrote: > 7 > Keith Brown <kbrown780@usfamily.net> wrote in messageu( > news:395A9B97.1EFFFEE5@usfamily.net... >  > ...t > > > > I don't believe that replication is a good replacement forB > > backup. Suppose your data is deleted or corrupted, should this > > be replicated too? > K > In this particular situation, replication is an excellent replacement for M > backup, since it protects you from media failure, and in the unlikely eventyL > (since the process is almost fully automated) of corruption the portion ofG > the database that was lost can be reconstructed from the original Weba= > material - a process which is continually occurring anyway.r > N > And, as I said before, if that's not sufficiently comforting, you can alwaysL > capture and save the incoming updates and use them to reconstruct the lostA > portion of the data, to get it back on line a bit more quickly.g > F > Indeed, for this large and volatile an environment, backup is prettyM > unattractive.  For that matter, it's convincingly arguable that traditionaloG > backup is in general on the way out, given the decreasing gap in costxJ > between on-line and off-line storage and the increasing relative cost ofC > creating, organizing, and storing said off-line storage:  keepingoN > differential (to minimize space requirements; perhaps partial as well, sinceD > a lot of data is easily recoverable by other means) redundant (andN > separated, to guard against both media and whole-site failures) snapshots onK > line is one promising alternative, and particularly attractive in that it N > can allow users to recover from their own mistakes without invoking the Gods > of the Data Center.o >  > - bill >  > >o > > -- > > Keith Brown  > > kbrown780@usfamily.net  = I agree with you that snapshots will probably be way to go ass@ the advantages are readily apparent. Your statement that backups= will soon be obsolete is crap. This is the usual spin vendors ; put out when they don't have the functionality so lets telle% everyone they don't need it anymore. s   -- t Keith Brown" kbrown780@usfamily.net   ------------------------------    Date: 28 Jun 2000 19:00:36 -0500* From: kuhrt@eisner.decus.org (Marty Kuhrt)7 Subject: Re: OpenVMS clusters vs other systems clustersa+ Message-ID: <apSUYnalQnA9@eisner.decus.org>"  n In article <395A26EF.6A4F1D3@uk.sun.com>, Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> writes: > jlsue wrote: > H >> On Wed, 21 Jun 2000 01:01:26 -0400, "Bill Todd" <billtodd@foo.mv.com>	 >> wrote:e >>H >> To be completely fair, many of the reasons that VMS lost what hold itF >> did have on the market was because of all the marketing lies (snake4 >> oil) that the Unix segment was putting out there. >>5 > Just for a moment just consider packaging, all partt2 > of marketing. How out of touch do you heve to be4 > to try to market an OS in this day and age without: > including an IP stack and all the accompanying utilities7 > with it. What sort of message do you think this sends 5 > to people building internet/intranet based systems.o  ; So the mouthpiece for the wanna-be solid OS company refutest> unix marketing lies by telling an outright lie?  Is it irony, < a complete detachment from reality, or just plain stupidity?   ------------------------------  % Date: Wed, 28 Jun 2000 18:50:43 -0400 + From: "Main, Kerry" <Kerry.Main@compaq.com>u7 Subject: RE: OpenVMS clusters vs other systems clusterscJ Message-ID: <910612C07BCAD1119AF40000F86AF0D8052844A2@kaoexc4.kao.dec.com>   Andrew,     Thank you for the great setup !!  I >>> because DEC totally missed the move from people making apps to peoplerK buying apps and most crucially it lost its position because DEC and laterly L Compaq though that you didn't need to market the OS and its capabilities. <<  ' Applications, marketing .. absolutely !o  - Check it out: (remember, you asked for it ..)   L <http://www.oracle.com/cgi-bin/press/printpr.cgi?file=200006281130.25961.htmL l&mode=corp&td=1&tm=7&fd=1&fm=5&status=Search&ty=2000&limit=50&fy=2000> (one url)  H "Compaq Adopts the Oracle Internet Platform Across Server Product Lines"  L The Oracle Internet Platform will be optimized for Compaq servers--includingH ProLiant(TM) running Windows NT(R)/Windows 2000(TM), AlphaServer runningE Tru64 UNIX(TM) or OpenVMS(TM), and the companies plan to optimize fori
 Himalaya(TM) eJ  running NonStop(TM) Kernel (NSK)--and StorageWorks(TM) storage offerings.H This Compaq and Oracle E-Business Platform will be available from CompaqL during calendar 4Q00. With this agreement, Compaq will also adopt internally the Oracle Internet Platform.t  K  "Today's announcement represents a major step in our strategy to work withpE partners to build and deliver the best Internet infrastructure to our I customers," said Michael Capellas, Compaq's President and CEO. "Customers B want e-business solutions that are reliable, scalable, and easy toH deploy--with the security of knowing they are backed by vendors they canB trust. With the tight integration of our Internet technologies andI professional services, no two vendors are better positioned to meet theset/ customer requirements than Compaq and Oracle." i  9 <http://www.oracle.com/tellmemore/?228177> (Tell me more)s  L <http://www1.compaq.com/pressrelease/0,1494,wp%7E14583_2%21ob%7E31866_1_1,00 .html>  G [another thread has been started already, so I'll add my comments there  ....]n   Regards,  
 Kerry Main Senior Consultant,
 Compaq Canadal Professional Serviceso Voice : 613-592-4660 FAX   : 819-772-7036 Email : kerry.main@compaq.comi       -----Original Message-----' From: Andrew Harrison SUNUK Consultancy-! [mailto:andrew.nospam@uk.sun.com] ' Sent: Wednesday, June 28, 2000 12:25 PMs To: Info-VAX@Mvb.Saic.ComR7 Subject: Re: OpenVMS clusters vs other systems clusterse     jlsue wrote:  G > On Wed, 21 Jun 2000 01:01:26 -0400, "Bill Todd" <billtodd@foo.mv.com>i > wrote: >oJ > >VMS has no lock on high availability any more - well, it never did, but nowrJ > >its competition includes 'standard' Unix and even Windows environments. The G > >eagerness with which customers are accepting such solutions makes itr clearuL > >that VMS's advantages, real though they may be, are not decisive compared toJ > >other considerations (in other words, the competition offers *adequate*9 > >availability plus many attractions that VMS does not).c > >i >.G > To be completely fair, many of the reasons that VMS lost what hold it0E > did have on the market was because of all the marketing lies (snake 3 > oil) that the Unix segment was putting out there.a >   3 Ahh to old conspiracy theory raises its head again.e  6 OpenVMS didn't lose its position in the market because9 of lies by UNIX competitors. It lost its position becauseo; it cost too much at the low end, because DEC totally missed26 the move from people making apps to people buying apps7 and most crucially it lost its position because DEC and 8 laterly Compaq though that you didn't need to market the OS and its capabilities.  3 And by market I don't just mean demand creation and 2 brand awareness creation through advertising which7 rather quaintly some people in this group semm to think  will solve the OpenVMS problem.w  5 I mean sales programs, loan equipment for developers,a4 special deals for developer systems. Co marketing of7 the OS and selected applications into that applications>7 vendors customer base. Cooperative deals to rationalised6 software licensing charges (parity with NT application6 licensing for example). Marketing programs to selected9 vertical markets backed by things like the willingness to>3 pay for ports of apps ot in some cases the purchasen8 of the ISV itself that are key to sucess in that market.  ; This is all marketing and virtually none of it is addressedo> by the rather sad excercise so beloved of this group of trying0 to spot OpenVMS references in Compaq literature.  8 Given the lack of any of the above in Digitals marketing, of OpenVMS and market development of OpenVMS2 one has to conclude that Digital/Compaq handed the& market in a plate to the UNIX vendors.  3 Just for a moment just consider packaging, all partr0 of marketing. How out of touch do you heve to be2 to try to market an OS in this day and age without8 including an IP stack and all the accompanying utilities5 with it. What sort of message do you think this sendse3 to people building internet/intranet based systems.d   Regardsh Andrew Harrisoni Enterprise IT Architect    ------------------------------  % Date: Wed, 28 Jun 2000 20:19:19 -0400O' From: "Bill Todd" <billtodd@foo.mv.com>o7 Subject: Re: OpenVMS clusters vs other systems clustersa( Message-ID: <8je4ke$q9m$1@pyrite.mv.net>  5 Marty Kuhrt <kuhrt@eisner.decus.org> wrote in messagei% news:apSUYnalQnA9@eisner.decus.org...WA > In article <395A26EF.6A4F1D3@uk.sun.com>, Andrew Harrison SUNUK . Consultancy <andrew.nospam@uk.sun.com> writes: > > jlsue wrote: > > J > >> On Wed, 21 Jun 2000 01:01:26 -0400, "Bill Todd" <billtodd@foo.mv.com> > >> wrote:l > >>J > >> To be completely fair, many of the reasons that VMS lost what hold itH > >> did have on the market was because of all the marketing lies (snake6 > >> oil) that the Unix segment was putting out there. > >>7 > > Just for a moment just consider packaging, all parth4 > > of marketing. How out of touch do you heve to be6 > > to try to market an OS in this day and age without< > > including an IP stack and all the accompanying utilities9 > > with it. What sort of message do you think this sendso7 > > to people building internet/intranet based systems.w > = > So the mouthpiece for the wanna-be solid OS company refutess? > unix marketing lies by telling an outright lie?  Is it irony,n> > a complete detachment from reality, or just plain stupidity?  L Could be he's just not up-to-the-minute on VMS packaging, which would hardly seem to justify such invective.f  G Only a couple of months ago, a friend who investigated the situation (ItL suspect thoroughly, since low-end VMS issues are of considerable interest toH him) reported that the base VMS license available in the low-end pricingL tier (DS10, DS20, ES40) indeed did not include the TCP/IP stack and friends,K and cost around $1200.  Now the least expensive package listed for the DS10tI reportedly *does* include the TCP/IP stack and friends, and costs severalnH times that amount (haven't heard whether the old, $1200 base package wasL actually discontinued, or is just something you have to ask for specifically to find out about it).   - bill   ------------------------------  % Date: Wed, 28 Jun 2000 21:06:03 -0700g5 From: "Larry D Bohan, Jr" <LBohan@dbc.spam_less..com>e7 Subject: Re: OpenVMS clusters vs other systems clustersi2 Message-ID: <+8laObUQ8qczfUZ8LgLzAdY8lpRM@4ax.com>  E On Wed, 28 Jun 2000 20:19:19 -0400, "Bill Todd" <billtodd@foo.mv.com>n wrote:H >Only a couple of months ago, a friend who investigated the situation (IM >suspect thoroughly, since low-end VMS issues are of considerable interest toeI >him) reported that the base VMS license available in the low-end pricingnM >tier (DS10, DS20, ES40) indeed did not include the TCP/IP stack and friends,eL >and cost around $1200.  Now the least expensive package listed for the DS10J >reportedly *does* include the TCP/IP stack and friends, and costs severalI >times that amount (haven't heard whether the old, $1200 base package washM >actually discontinued, or is just something you have to ask for specifically  >to find out about it).i   ???e  ) I can't speak for the DS10 (workstation) y; but DS20's (and likely, ES40's) come with the EIP bundle.  u ie,t  % DIGITAL Enterprise Integration Servert  ? DIGITAL Enterprise Integration Server for OpenVMS, Release 2.0,2F license packages are included with most OpenVMS AlphaServer systems.    E Some OpenVMS AlphaServer systems include Release 1.1 of this package..C Please consult the description of a specific AlphaServer system forn= information about exactly which package is included.  DIGITAL ? Enterprise Integration Server for OpenVMS, Release 2.0 License w   Packages include:   7 License Part Number	Quantity	Description and SPD Numbert  9 QL-OLXA-*AA	1	DIGITAL TCP/IP Services for OpenVMS (46.46)m6 QL-MTFA*-AA	1	DECnet-Plus for OpenVMS Alpha End System (50.45)i8 QL-MV4A*-AA	1	DECwindows Motif for OpenVMS Alpha (42.19)9 QL-09MA*-AA	1	DECprint Supervisor for OpenVMS Alpha, Plusr (44.15)s9 QL-09KA*-AA	1	DECprint Supervisor for OpenVMS Alpha, Open- (44.15)-: QL-5LQA*-AA	1	Archive/Backup System for OpenVMS Management
 Tools (60.72)r9 QL-5LSA9-3B	1	Archive/Backup Agent for Windows NT (60.72)e: QL-5M3A*-AA	1	Datametrics ViewPoint Data Collector (60.72): QL-5TA9A-2B	1	DIGITAL OpenVMS Disk Services for Windows NT (60.80)h7 QM-5TB9A-AA	1	DIGITAL Office Server for OpenVMS (61.49)t9 QM-5TD9A-A*	2	DIGITAL Office Server Client Access Licenses (61.49)c* QM-5LKAA-AB	3	DIGITAL PATHWORKS 32 (56.33)5 QM-5SUAA-AK	2	PATHWORKS for OpenVMS (Advanced Server)e (30.50)l   ------------------------------   Date: 28 Jun 2000 18:31:30 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)7 Subject: OpenVMS loses big, was:  RE: Compaq advertisesn, Message-ID: <8jdga2$3uk@gap.cco.caltech.edu>  g In article <001301bfe10e$74e4b2c0$14b324a6@CJ4733A>, arturo saavedra <arturo.saavedra@wcom.com> writes:h >Compaq A Winner In Gene Racer  J And OpenVMS is the big loser - because none of these genomics folks chose B OpenVMS to run on a single box, going instead with Tru64 or Linux.  I OpenVMS's poor performance for general computing (anything except massiveeK transaction processing), arbitrary and poorly handled 32k and 64k limits oni> various operations, and rampant (and now completely pointless)K incompatibility with source code developed on Unix systems (which is the de B facto standard server OS now), made OpenVMS a noncontender in thisE lucrative new market.  Just as it will in all other new markets.  ButxC that's just losing.  What makes it the _BIG_ loser is that the vastsI majority of the money that went into the development of the Alpha, Tru64,iF and even parts of Linux/Alpha was derived from OpenVMS related sales. 4 So Compaq as a whole wins, but OpenVMS gets screwed.  G Moreover, it isn't just what OpenVMS is now that caused it to lose thismI competition.   This is also the result of neglecting markets because they H weren't viewed as important or sufficiently lucrative.  Over the last 10E years the types of software Celera and the others run have been underbK constant development on _small_ machines almost entirely by people working dE in academica.  Too bad for OpenVMS that it has been a no show in this G market, and consequently, when the time came to run this code on largeraE machines, it was all Unix based.  I doubt any of these companies even G seriously considerd OpenVMS, since the guys doing the deciding have run E nothing but Unix for the last decade.  And even if they had been open_I minded enough to look at OpenVMS, it really would have been illogical for  them to choose it. o  H How did Tru64 get the job over another Unix, for instance, Solaris? It'sE really quite simple.  The code was all developed on Unix systems, so:B  H 1.  Tru64 is a Unix, so it can run that code without major modification.> 2.  Tru64 runs on Alpha, and that chip is faster than a Sparc.1 3.  Tru64 can make full use of the Alpha's speed.o  E Linux made quite an inroad as well, because, thanks now to the nativem/ Compaq compilers, it passes all 3 criteria too.   / OpenVMS washes out, failing criteria 1 and 3.  i  K And guys, these first few genomics sales are just the beginning.  There are I going to be some huge sales to the pharmaceutical industry.  They will beHG building/using tools to manipulate and analyze the vast amounts of datasF present in the genomes of various organisms, gene expression patterns,K protein strucures and every other type of biological information.  But thiskG isn't transaction processing, and it isn't even clear to me that OracleiE style relational databases are going to play a large role.  So unless-H OpenVMS gets head to head with Tru64 in performance, Unix compatibility,D and price, I don't see any future for it in this field.  Or for that< matter, in any other new field that may come down the road.    Regards,   David Mathog mathog@seqaxp.bio.caltech.edue? Manager, sequence analysis facility, biology division, Caltech     ------------------------------  % Date: Wed, 28 Jun 2000 23:16:20 -0400r+ From: "Main, Kerry" <Kerry.Main@compaq.com> ; Subject: RE: OpenVMS loses big, was:  RE: Compaq advertises J Message-ID: <910612C07BCAD1119AF40000F86AF0D8052844A6@kaoexc4.kao.dec.com>   David,  J No one will argue that OpenVMS marketing suffered under the latter portionH of the old Digital regime or even the early Compaq days when things were just settling in.n  	 However -c  J Does the COE project discussed in Terry Shannons articles not address someJ of the issues you raise in the attached ie. combining the good features ofC OpenVMS with some of the core functions associated with UNIX OS's?    H Would this not seem like the best of both worlds? The RASS (reliability,L availability, scalability and security) and unique features (like Galaxy) of= OpenVMS combined with the Linux/UNIX applications available. i  K In terms of marketing and applications, is not todays announcements for newnG Oracle IAS (Internet Application Server) middleware software being madet> available on OpenVMS a good start to making up for lost time?   L Are not the recent endorsements from major Customers like E*Trade on the newG Alpha GS Series a good sign that Customers (and ISV's) are re-examiningl! their strategies around OpenVMS?    
 Reference:9 <http://www.compaq.com/alphaserver/gs/quotes/etrade.html> 5 <http://www.openvms.digital.com/gsseries/quotes.html>l2 <http://www.openvms.compaq.com/openvms/brochures/>  - I agree there is still much work to be done. o  K However, while perhaps not fast enough for readers of this list, surely the J events of the last few months can be seen as steps in the right direction?  " [now donning my suit of armour ..]   :-)p   Regards,  
 Kerry Main Senior Consultant,
 Compaq Canadac Professional Servicess Voice : 613-592-4660 FAX   : 819-772-7036 Email : kerry.main@compaq.com        -----Original Message-----# From: mathog@seqaxp.bio.caltech.edu@& [mailto:mathog@seqaxp.bio.caltech.edu]& Sent: Wednesday, June 28, 2000 2:32 PM To: Info-VAX@Mvb.Saic.Coms6 Subject: OpenVMS loses big, was: RE: Compaq advertises    D In article <001301bfe10e$74e4b2c0$14b324a6@CJ4733A>, arturo saavedra" <arturo.saavedra@wcom.com> writes: >Compaq A Winner In Gene Racea  J And OpenVMS is the big loser - because none of these genomics folks chose B OpenVMS to run on a single box, going instead with Tru64 or Linux.  I OpenVMS's poor performance for general computing (anything except massive K transaction processing), arbitrary and poorly handled 32k and 64k limits onl> various operations, and rampant (and now completely pointless)K incompatibility with source code developed on Unix systems (which is the dedB facto standard server OS now), made OpenVMS a noncontender in thisE lucrative new market.  Just as it will in all other new markets.  ButiC that's just losing.  What makes it the _BIG_ loser is that the vastgI majority of the money that went into the development of the Alpha, Tru64,lF and even parts of Linux/Alpha was derived from OpenVMS related sales. 4 So Compaq as a whole wins, but OpenVMS gets screwed.  G Moreover, it isn't just what OpenVMS is now that caused it to lose thisuI competition.   This is also the result of neglecting markets because theycH weren't viewed as important or sufficiently lucrative.  Over the last 10E years the types of software Celera and the others run have been undernK constant development on _small_ machines almost entirely by people working pE in academica.  Too bad for OpenVMS that it has been a no show in thissG market, and consequently, when the time came to run this code on larger E machines, it was all Unix based.  I doubt any of these companies evensG seriously considerd OpenVMS, since the guys doing the deciding have runwE nothing but Unix for the last decade.  And even if they had been openoI minded enough to look at OpenVMS, it really would have been illogical forr them to choose it. t  H How did Tru64 get the job over another Unix, for instance, Solaris? It'sE really quite simple.  The code was all developed on Unix systems, so:T  H 1.  Tru64 is a Unix, so it can run that code without major modification.> 2.  Tru64 runs on Alpha, and that chip is faster than a Sparc.1 3.  Tru64 can make full use of the Alpha's speed.S  E Linux made quite an inroad as well, because, thanks now to the nativet/ Compaq compilers, it passes all 3 criteria too.a  / OpenVMS washes out, failing criteria 1 and 3.     K And guys, these first few genomics sales are just the beginning.  There are I going to be some huge sales to the pharmaceutical industry.  They will behG building/using tools to manipulate and analyze the vast amounts of data,F present in the genomes of various organisms, gene expression patterns,K protein strucures and every other type of biological information.  But thiseG isn't transaction processing, and it isn't even clear to me that OraclerE style relational databases are going to play a large role.  So unlesseH OpenVMS gets head to head with Tru64 in performance, Unix compatibility,D and price, I don't see any future for it in this field.  Or for that< matter, in any other new field that may come down the road.    Regards,   David Mathog mathog@seqaxp.bio.caltech.edud? Manager, sequence analysis facility, biology division, Caltech e   ------------------------------   Date: 28 Jun 2000 21:02:23 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)1 Subject: Oracle Press Release, incl. OpenVMS infog6 Message-ID: <8jdp4v$fv6$1@mailint03.im.hou.compaq.com>  D   Please see the attached Oracle press release -- this press releaseD   is available at the Oracle website, and I'd expect to see related D   information posted at the OpenVMS website in the very near future.  0   Note: Follow-ups have been set to comp.os.vms.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com     	------------------------- 	Oracle Press Releaseu   	Contact(s): Amy Mackl
 	Oracle Corp.M
 	650/607-2552 & 	amy.mack@oracle.com  Michelle Spolver 	Compaq Computer Corporation e
 	408/285-0159e" 	michelle.spolver@compaq.com          A 	Compaq Adopts the Oracle Internet Platform Across Server Producti Lines   ; 	Compaq and Oracle align to offer customers of all sizes ano0 integrated mission critical e-business platform   4 	HOUSTON AND REDWOOD SHORES, Calif., Jun. 28, 2000 -G (http://www.oracle.com/tellmemore/?228177)  Compaq Computer Corporation F (NYSE: CPQ) and Oracle Corporation (Nasdaq: ORCL) today announced thatF Compaq will adopt the Oracle Internet Platform across its full line ofE server and storage offerings to provide customers of all sizes with amK reliable, scalable, and complete mission critical e-business platform. WithaL this agreement, the companies will jointly optimize and test this Compaq andH Oracle E-Business Platform, and will deliver this offering through their% professional services organizations. r  : 	The Oracle Internet Platform will be optimized for CompaqK servers--including ProLiant* running Windows NT*/Windows 2000*, AlphaServeriG running Tru64 UNIX* or OpenVMS*, and the companies plan to optimize for B Himalaya* running NonStop* Kernel (NSK)--and StorageWorks* storageL offerings. This Compaq and Oracle E-Business Platform will be available fromH Compaq during calendar 4Q00. With this agreement, Compaq will also adopt) internally the Oracle Internet Platform. g  A 	"Today's announcement represents a major step in our strategy toeK work with partners to build and deliver the best Internet infrastructure toiB our customers," said Michael Capellas, Compaq's President and CEO.J "Customers want e-business solutions that are reliable, scalable, and easyK to deploy--with the security of knowing they are backed by vendors they canrB trust. With the tight integration of our Internet technologies andI professional services, no two vendors are better positioned to meet these./ customer requirements than Compaq and Oracle." l  = 	Optimization of the Oracle Internet Platform, which includestL Oracle8i, Oracle Internet Application Server 8i (Oracle iAS), and the OracleD Internet Developer Suite, across Compaq server platforms will enableK customers of all sizes to implement comprehensive e-business solutions withnA flexibility, efficiency, and deployment ease. Such cross-platform-L integration will allow customers to write Java* applications once and deployK them across all Compaq platforms. Additionally, the scalability of Compaq'sgL servers and the Oracle Internet Platform software offer customers the widestE possible range of deployment options--from a single server to a large9K cluster, allowing them to grow their infrastructure while maintaining theirm technology investments.   D 	"This agreement between Oracle and Compaq represents a new level ofF commitment to provide an exceptionally reliable, scalable and completeJ e-business infrastructure to the widest range of customers possible," saidF Gary Bloom, executive vice president, Oracle. "The breadth of Compaq'sE server platforms and storage offerings, coupled with the power of the E complete Oracle Internet Platform, will provide our customers with an A unbeatable infrastructure for developing and deploying e-businessl solutions."    	Expert, Integrated Services  B 	Compaq Professional Services is teaming with Oracle Consulting to> create service offerings to optimize for customers the design,K implementation and management of the Compaq and Oracle E-Business Platform.lH Compaq and Oracle are plan to jointly develop service offerings, such asJ architecture analysis and design consulting, and implement global trainingI programs for their consulting organizations to jointly deliver Compaq and I Oracle E-Business Platform-based products running across Compaq servers.    @ 	Additionally, the two companies intend two jointly establish anE E-Business Reference Lab for ISVs to provide hands-on assistance witheI optimizing, testing and tuning the Oracle Internet Platform across Compaq F servers and storage offerings. Configuration and sizing information onK tested Compaq and Oracle Internet Platform solutions will be made availablenE to ISVs, solution providers and consultants via Compaq ActiveAnswers*0I (www.compaq.com/activeanswers), an online repository of tools, forums and E information to help plan, deploy and operate enterprise solutions and-# accelerate time-to-implementation.    
 	About Oracle0  D 	Oracle Corporation provides the software that powers the Internet.   = 	For more information about Oracle, please call 650/506-7000.U    
 	About CompaqR  B 	Compaq Computer Corporation, a Fortune Global 100 company, is theC largest supplier of computing systems in the world. Compaq designs,vE develops, manufactures and markets hardware, software, solutions, andeD services, including industry-leading enterprise computing solutions,H fault-tolerant business-critical solutions, and communications products,< commercial desktop and portable products, and consumer PCs.   A 	Compaq products and services are sold in more than 200 countriesrH directly to businesses, through a network of authorized Compaq marketingC partners, and directly to businesses and consumers through Compaq'shI e-commerce Web site at http://www.compaq.com. Compaq markets its productsaL and services primarily to customers from the business, home, government, andH education sectors. Customer support and information about Compaq and its> products and services are available at http://www.compaq.com.   
 	# # # # #    C 	This press release may contain forward-looking statements based on J current expectations that involve a number of risks and uncertainties. TheK potential risks and uncertainties that could cause actual results to differpG materially include: delays in the implementation of changes in delivery.J models, increased competitive environment and pricing pressures, delays inL product rollout schedules, component shortages, slow acceptance for new formJ factors, employee retention, disruptions related to restructuring actions,K the financial condition of resellers, delays in new systems implementation,nE operational inefficiencies related to sales cycles, equity investment J volatility, and emerging market political or economic instability. FurtherJ information on the factors that could affect Compaq's financial results isF included in Compaq's Securities and Exchange Commission (SEC) filings,H including the latest Annual Report on form 10-K and the latest Quarterly Report on Form 10-Q. A  , 	For further editorial information, contact:     	Compaq Computer Corporation e   	Michelle Spolver    	408-285-0159    	michelle.spolver@compaq.com     	  s   ------------------------------    Date: 28 Jun 2000 14:40:01 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett) 5 Subject: Re: Oracle Press Release, incl. OpenVMS info8, Message-ID: <NMU3VZcW6zff@malvm2.mala.bc.ca>  7 In article <8jdp4v$fv6$1@mailint03.im.hou.compaq.com>, p7    hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes:d   > F >   Please see the attached Oracle press release -- this press releaseF >   is available at the Oracle website, and I'd expect to see related F >   information posted at the OpenVMS website in the very near future. >   H     Is this the "big announcement" that folks were talking about earlierE today? My marketspeak filter is having trouble understanding what (ifa= any) meaningful change this really represents for a VMS site.i   > < > 	The Oracle Internet Platform will be optimized for CompaqM > servers--including ProLiant* running Windows NT*/Windows 2000*, AlphaServer I > running Tru64 UNIX* or OpenVMS*, and the companies plan to optimize fortD > Himalaya* running NonStop* Kernel (NSK)--and StorageWorks* storageN > offerings. This Compaq and Oracle E-Business Platform will be available from > Compaq during calendar 4Q00.  I     Does this mean they've made some deal to sell more Proliants with lipOJ service being paid to VMS? Anybody want to bet the Win2K version will comeD out in 4Q00, followed by Tru64 in 1Q01, followed ( maybe ) by VMS in? 1Q02 [this estimate based on their track record with Oracle 8i]e  - > With this agreement, Compaq will also adopts+ > internally the Oracle Internet Platform. r >   F     What are they using now? Will this mean they'll be using less Rdb?  L > "Customers want e-business solutions that are reliable, scalable, and easyM > to deploy--with the security of knowing they are backed by vendors they caneD > trust. With the tight integration of our Internet technologies andK > professional services, no two vendors are better positioned to meet thesep1 > customer requirements than Compaq and Oracle."   > J      When Joe NewCustomer calls up Compaq looking for a system to run thisO software does anyone really think that selling him VMS will even enter the minde of the average Compaq salesman?o  C > flexibility, efficiency, and deployment ease. Such cross-platform N > integration will allow customers to write Java* applications once and deploy# > them across all Compaq platforms.c  6     Isn't that supposed to be the whole point of Java?  + > Additionally, the scalability of Compaq'seN > servers and the Oracle Internet Platform software offer customers the widestG > possible range of deployment options--from a single server to a largeaM > cluster, allowing them to grow their infrastructure while maintaining theirl > technology investments.  >   =    Yep, everything from the smallest to the largest Proliant.y   > E > 	This press release may contain forward-looking statements based onuI > current expectations that involve a number of risks and uncertainties. p  H     I'd say that's the understatement of the year in regards to anything relating to Oracle and VMS :-)  E      Sorry to sound so cynical, but I'd hoped a "big announcement" retG Oracle and VMS would be a real commitment to get the VMS version out athD the *same* time as the "tier 1" OSes (not the 1-2 months later which$ turns into 6-9 months and counting).  M      It seems to me this announcement is actually bad news for VMS. If thingsa4 roll out as I predict it may go something like this:  B    Customer: I want to mount a new application on VMS system usingH              these products. I'll also need to get a larger Alphaserver.  F    Oracle/Compaq: The VMS version isn't ready yet, but it will be in 3H                   months. In the meantime why don't you get started withE                   the project on Windows 2K, by the time you're readyeD                   to deploy it the VMS code will be done and you can$                   just port it over.       [3 months later]  2     Customer:  I'm ready, where's the VMS version?  F     Oracle/Compaq: not quite ready yet, just give us another 3 months.    D     Customer: Well OK, if I have to I can survive that long. Maybe I.               can do a pilot rollout on Win2k.       [3 months later]  #     Customer: Still no VMS version?e  H     Oracle/Compaq: It's almost ready just another 3 months or so. But ifL                    you can't wait why don't we just convert that AlphaserverI                    you ordered into a Proliant, then you can deploy righte                    away.   ------------------------------  % Date: Wed, 28 Jun 2000 19:07:17 -0400g+ From: "Main, Kerry" <Kerry.Main@compaq.com> 5 Subject: RE: Oracle Press Release, incl. OpenVMS info*J Message-ID: <910612C07BCAD1119AF40000F86AF0D8052844A3@kaoexc4.kao.dec.com>   As a followup to Steve's note -o  I A few pointers for those folks who have beed asking for more applicationsr and marketing of OpenVMS:n  L <http://www.oracle.com/corporate/press/index.html?228177.html> (Oracle Press release)  9 <http://www.oracle.com/tellmemore/?228177> (Tell me more),  L <http://www1.compaq.com/pressrelease/0,1494,wp%7E14583_2%21ob%7E31866_1_1,00 .html> (Compaq Press release)   G And in case, some folks missed the other posting (and I like the splash-	 stuff :-)d- <http://www.openvms.digital.com/e-postcard1/>n- <http://www.openvms.digital.com/e-postcard2/>6H (under each url is a pointer that you can save and run directly on a PC)   Regards,  
 Kerry Main Senior Consultant,
 Compaq Canadat Professional Servicest Voice : 613-592-4660 FAX   : 819-772-7036 Email : kerry.main@compaq.coms       -----Original Message-----# From: hoffman@xdelta.zko.dec.nospame& [mailto:hoffman@xdelta.zko.dec.nospam]& Sent: Wednesday, June 28, 2000 5:02 PM To: Info-VAX@Mvb.Saic.Comi1 Subject: Oracle Press Release, incl. OpenVMS info-        D   Please see the attached Oracle press release -- this press releaseD   is available at the Oracle website, and I'd expect to see related D   information posted at the OpenVMS website in the very near future.  0   Note: Follow-ups have been set to comp.os.vms.  2  --------------------------- pure personal opinion ---------------------------hL    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com     	------------------------- 	Oracle Press Releasev   	Contact(s): Amy Mack 
 	Oracle Corp.e
 	650/607-2552 & 	amy.mack@oracle.com  Michelle Spolver 	Compaq Computer Corporation s
 	408/285-0159 " 	michelle.spolver@compaq.com          A 	Compaq Adopts the Oracle Internet Platform Across Server Productr Lines   ; 	Compaq and Oracle align to offer customers of all sizes an 0 integrated mission critical e-business platform   4 	HOUSTON AND REDWOOD SHORES, Calif., Jun. 28, 2000 -G (http://www.oracle.com/tellmemore/?228177)  Compaq Computer Corporation F (NYSE: CPQ) and Oracle Corporation (Nasdaq: ORCL) today announced thatF Compaq will adopt the Oracle Internet Platform across its full line ofE server and storage offerings to provide customers of all sizes with aiK reliable, scalable, and complete mission critical e-business platform. WithsL this agreement, the companies will jointly optimize and test this Compaq andH Oracle E-Business Platform, and will deliver this offering through their% professional services organizations. g  : 	The Oracle Internet Platform will be optimized for CompaqK servers--including ProLiant* running Windows NT*/Windows 2000*, AlphaServerbG running Tru64 UNIX* or OpenVMS*, and the companies plan to optimize forrB Himalaya* running NonStop* Kernel (NSK)--and StorageWorks* storageL offerings. This Compaq and Oracle E-Business Platform will be available fromH Compaq during calendar 4Q00. With this agreement, Compaq will also adopt) internally the Oracle Internet Platform. 6  A 	"Today's announcement represents a major step in our strategy tooK work with partners to build and deliver the best Internet infrastructure tolB our customers," said Michael Capellas, Compaq's President and CEO.J "Customers want e-business solutions that are reliable, scalable, and easyK to deploy--with the security of knowing they are backed by vendors they canlB trust. With the tight integration of our Internet technologies andI professional services, no two vendors are better positioned to meet theset/ customer requirements than Compaq and Oracle."    = 	Optimization of the Oracle Internet Platform, which includes L Oracle8i, Oracle Internet Application Server 8i (Oracle iAS), and the OracleD Internet Developer Suite, across Compaq server platforms will enableK customers of all sizes to implement comprehensive e-business solutions withoA flexibility, efficiency, and deployment ease. Such cross-platformaL integration will allow customers to write Java* applications once and deployK them across all Compaq platforms. Additionally, the scalability of Compaq's L servers and the Oracle Internet Platform software offer customers the widestE possible range of deployment options--from a single server to a largeiK cluster, allowing them to grow their infrastructure while maintaining their  technology investments.   D 	"This agreement between Oracle and Compaq represents a new level ofF commitment to provide an exceptionally reliable, scalable and completeJ e-business infrastructure to the widest range of customers possible," saidF Gary Bloom, executive vice president, Oracle. "The breadth of Compaq'sE server platforms and storage offerings, coupled with the power of thenE complete Oracle Internet Platform, will provide our customers with an A unbeatable infrastructure for developing and deploying e-business2 solutions."    	Expert, Integrated Services  B 	Compaq Professional Services is teaming with Oracle Consulting to> create service offerings to optimize for customers the design,K implementation and management of the Compaq and Oracle E-Business Platform.fH Compaq and Oracle are plan to jointly develop service offerings, such asJ architecture analysis and design consulting, and implement global trainingI programs for their consulting organizations to jointly deliver Compaq andaI Oracle E-Business Platform-based products running across Compaq servers. r  @ 	Additionally, the two companies intend two jointly establish anE E-Business Reference Lab for ISVs to provide hands-on assistance witheI optimizing, testing and tuning the Oracle Internet Platform across CompaqeF servers and storage offerings. Configuration and sizing information onK tested Compaq and Oracle Internet Platform solutions will be made availableaE to ISVs, solution providers and consultants via Compaq ActiveAnswers*pI (www.compaq.com/activeanswers), an online repository of tools, forums andrE information to help plan, deploy and operate enterprise solutions andz# accelerate time-to-implementation. r  
 	About Oracleo  D 	Oracle Corporation provides the software that powers the Internet.   = 	For more information about Oracle, please call 650/506-7000.a    
 	About Compaql  B 	Compaq Computer Corporation, a Fortune Global 100 company, is theC largest supplier of computing systems in the world. Compaq designs,aE develops, manufactures and markets hardware, software, solutions, anduD services, including industry-leading enterprise computing solutions,H fault-tolerant business-critical solutions, and communications products,< commercial desktop and portable products, and consumer PCs.   A 	Compaq products and services are sold in more than 200 countriesiH directly to businesses, through a network of authorized Compaq marketingC partners, and directly to businesses and consumers through Compaq'szI e-commerce Web site at http://www.compaq.com. Compaq markets its productstL and services primarily to customers from the business, home, government, andH education sectors. Customer support and information about Compaq and its> products and services are available at http://www.compaq.com.   
 	# # # # #    C 	This press release may contain forward-looking statements based oneJ current expectations that involve a number of risks and uncertainties. TheK potential risks and uncertainties that could cause actual results to differrG materially include: delays in the implementation of changes in deliveryyJ models, increased competitive environment and pricing pressures, delays inL product rollout schedules, component shortages, slow acceptance for new formJ factors, employee retention, disruptions related to restructuring actions,K the financial condition of resellers, delays in new systems implementation,iE operational inefficiencies related to sales cycles, equity investmentoJ volatility, and emerging market political or economic instability. FurtherJ information on the factors that could affect Compaq's financial results isF included in Compaq's Securities and Exchange Commission (SEC) filings,H including the latest Annual Report on form 10-K and the latest Quarterly Report on Form 10-Q.    , 	For further editorial information, contact:     	Compaq Computer Corporation e   	Michelle Spolver    	408-285-0159    	michelle.spolver@compaq.com g   	  f   ------------------------------  % Date: Thu, 29 Jun 2000 12:25:55 +1200 . From: Nivlesh Chandra <NChandra001@itc.gov.fj> Subject: PARITY ERROReM Message-ID: <791C2856E8FDD211BAFB0008C759919591F9C9@exchange01.govnet.gov.fj>r   Need help.. please respondL I have a tape drive that is used for backups. I have a script that runs at aK specified time, get the operators to load the tapes and then it checks thatnI the tapes are accessible. During this it initialises the tapes so that itpI starts fresh. Now this is where my problem lies. Sometimes the tapes thateK are used result in a parity error.. this is sometimes due to the tape mediaCL and sometimes it is just the tape drive that is playing up. I need to detectK that a parity error has occurred and then take steps to find out what it isoL due to .. I have thought of this ... when a parity error occurs.. the scriptL will ask the operators to load another tape that passed the testing and thenH this will be loaded. If this tape passes the test then it means that theK previous tape was faulty and thus it needs to be replaced (a new tape needscL to be initialised with the same label as that tape and the faulty tape needsE to be trashed). If the second tape that is loaded also gives a paritynJ error.. then there is a high chance that the drive has given up on us thusI the technical people need to be informed (well there will need to be someo2 more tests done before this conclusion is reached)  L Anyway when I get the parity error.. I do not know how I could capture it inH my script so that I can detect that a parity error has occurred and thus react on it appropriately..t  A Can someone please tell me as to how I could go about doing this?e# Your help will be most appreciated.n   Thanks NivC   ------------------------------  % Date: Wed, 28 Jun 2000 20:57:44 -0800p) From: prymmer/kahn <pvhp@nospam.best.com>r7 Subject: Re: Remote access programs from Windows to VMSt/ Message-ID: <395AD748.7EDD36E0@nospam.best.com>q   Tomer Cohen wrote:   > Hello,L > Does anyone know about a good remote-access program (emulator?) from PC to+ > VAX/Alpha running VMS (except eXcursion)?o  G For running an X display on a PC or a Mac you might try the MI/X serverTF which is $25.00 shareware for Windows and free for MacOS and available from:p  5     http://tnt.microimages.com/www/html/freestuf/mix/n  F for a Windows app that is not an X display server you could try any of severaldH terminal emulators such as QVT Term available as nominal shareware from:     http://www.qpc.com/p  
 Good luck.   Peter P.   ------------------------------  % Date: Wed, 28 Jun 2000 20:54:21 -0500 + From: Shael Richmond <ksrich@bellsouth.net>t$ Subject: See how a device is spooled- Message-ID: <395AAC4D.E6860825@bellsouth.net>r  3 Is there a way to see where a device is spooled to?   ? We have been creating LAT ports spooling to LPD/TelnetSym printuB queues.  But doing a sho dev/full or sho term won't show the queue name - just that it is spooled.A   Thanks,t   Shael1   ------------------------------  % Date: Wed, 28 Jun 2000 14:52:45 -0400i- From: JF Mezei <jfmezei.spamnot@videotron.ca>C Subject: Re: SMTP Mail, Message-ID: <395A4979.3B95745B@videotron.ca>   Jerry Leslie wrote:sD > The "smtp%" is no longer necessary as of VMS 6.2, per the VMS FAQ: > < >     http://www.openvms.digital.com/wizard/openvms_faq.html  H Ok, so when MAIL sees an @ and some dots in a email address (is that theN actual logic used ?), it calls the SMTP foreign transport automatically. (just0 tested it on a system without dec's SMTP on it).  L Question: it is possible to configure mail to use a different transport when& it thinks it has an internet address ?   ------------------------------  # Date: Wed, 28 Jun 2000 19:56:34 GMTsL From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr") Subject: Re: SMTP Mail8 Message-ID: <009EC488.C086C269@SSRL04.SLAC.STANFORD.EDU>  \ In article <395A4979.3B95745B@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: >Jerry Leslie wrote:E >> The "smtp%" is no longer necessary as of VMS 6.2, per the VMS FAQ:i >> n= >>     http://www.openvms.digital.com/wizard/openvms_faq.htmlO >VI >Ok, so when MAIL sees an @ and some dots in a email address (is that theKO >actual logic used ?), it calls the SMTP foreign transport automatically. (justi1 >tested it on a system without dec's SMTP on it).n >aM >Question: it is possible to configure mail to use a different transport whens' >it thinks it has an internet address ?y  N Answer: The adroit use of logical names can fool mail into using any transport you like.  (   -- Alan   O ===============================================================================t0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056sM  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210 O ===============================================================================a   ------------------------------  % Date: Wed, 28 Jun 2000 16:17:09 -0400I- From: JF Mezei <jfmezei.spamnot@videotron.ca>e Subject: Re: SMTP Mail, Message-ID: <395A5D3C.C8DA48C8@videotron.ca>  * Alan Winston - SSRL Admin Cmptg Mgr wrote:O > >Question: it is possible to configure mail to use a different transport whend) > >it thinks it has an internet address ?S > P > Answer: The adroit use of logical names can fool mail into using any transport > you like.o  F Yes, I know that. But I was hoping for some configurable equivalent ofK "SMTP%". For instance, if the internet gateway resides on another node, you G might want to prefix such mail messages with "NODE2::IN%" for instance.y  H It would have been since if the action taken when an internet address isG detected were user configurable. I realise it has implications with thes quoting of adresses though.a   ------------------------------   Date: 28 Jun 2000 20:56:09 GMT) From: leslie@clio.rice.edu (Jerry Leslie)  Subject: Re: SMTP Mail' Message-ID: <8jdop9$2k8$1@joe.rice.edu>a  . JF Mezei (jfmezei.spamnot@videotron.ca) wrote: : Jerry Leslie wrote:bF : > The "smtp%" is no longer necessary as of VMS 6.2, per the VMS FAQ: : > > : >     http://www.openvms.digital.com/wizard/openvms_faq.html  J : Ok, so when MAIL sees an @ and some dots in a email address (is that theK : actual logic used ?), it calls the SMTP foreign transport automatically. l8 : (just tested it on a system without dec's SMTP on it).  J : Question: it is possible to configure mail to use a different transport - : when it thinks it has an internet address ?u    M The protocol VMS Mail uses can be set by the MAIL$INTERNET_TRANSPORT logical o< name, per the attached excerpt of the UCX 4.2 release notes.  4 --Jerry Leslie     (my opinions are strictly my own)N ==============================================================================  #      2.10.1 SMTP Addressing Changes   ;            Beginning with OpenVMS Version 6.2, OpenVMS Mailr?            automatically interprets a destination address as aneB            Internet address. You are no longer required to add theC            SMTP% prefix and quotes around the address. In addition,h?            OpenVMS Mail automatically translates the following:   B            o  If the node component (username@node) of the addressA               is fully qualified (it contains a period (.)), suchMA               as markson@billows.bldg2.com), OpenVMS Mail assumesnB               the address is an Internet address and uses the SMTPA               protocol by default. If you want to use a different =               Internet protocol, you can define the alternateeE               protocol with the MAIL$INTERNET_TRANSPORT logical name.o  ?            o  If the node component of the address is not fully-A               qualified (does not contain a period), OpenVMS Mailr>               translates the address for DECnet transport. For>               example, if you specified markson@billows as the?               destination address, the utility interprets it as ?               BILLOWS::MARKSON and uses DECnet instead of SMTP.a  <            Beginning with OpenVMS Version 7.0, you can forceB            OpenVMS Mail to use a specific protocol by defining theD            MAIL$INTERNET_MODE logical name. This condition is usefulD            in cases where a mail address such as markson@billows canD            be interpreted as valid for either the Internet or DECnet            protocol.  <            You can assign one of the following values to theB            MAIL$INTERNET_MODE logical name in your LOGIN.COM file:              o  SMTP  =               Mail always interprets the node component of an A               unqualified address as an Internet address specifi- @               cation. (SMTP is the default transport used unlessA               you define an alternate Internet transport with thes4               MAIL$INTERNET_TRANSPORT logical name.)              o  DECNET  @                  Mail always interprets the node component of anD                  unqualified address as a DECnet node specification.  %               o  HYBRID (the default)e  H                  Mail uses an Internet protocol if the node component ofG                  the address contains a period. If there are no periodsvF                  in the node component, Mail uses the DECnet protocol.  C               For example, the following definition causes the MailnF               utility to interpret any address that does not include aE               period in the node component of the specification as an                Internet address:c  .               $ DEFINE MAIL$INTERNET_MODE SMTP  @               Therefore, with OpenVMS Version 7.0 and subsequentB               versions, if you want to avert having an unqualifiedC               Internet address translated for DECnet transport, your!               have three choices:a  @               o  Fully qualify the address (for example, specifyF                  markson@billows.bldg2.com instead of markson@billows)  I               o  Include the SMTP% prefix (specify SMTP%"markson@billows)w  C               o  Define the MAIL$INTERNET_TRANSPORT logical name asa                  SMTP.  H               For more information on OpenVMS Mail and how it interpretsC               addresses, see the appropriate OpenVMS documentation.e   ------------------------------  % Date: Wed, 28 Jun 2000 13:14:37 -0700m2 From: ken nightingale <knightingale@tnrdlib.bc.ca> Subject: SSH on vms?- Message-ID: <395A5CAD.CE1A1125@tnrdlib.bc.ca>i   Hi,pA    Does anyone know of a Secure Shell server on something like it* that runs on VMS?h thanks,  ken nightingaleX -- O   knightingale@tnrdlib.bc.ca   ------------------------------   Date: 28 Jun 2000 20:21:02 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: SSH on vms?6 Message-ID: <8jdmne$etf$1@mailint03.im.hou.compaq.com>  b In article <395A5CAD.CE1A1125@tnrdlib.bc.ca>, ken nightingale <knightingale@tnrdlib.bc.ca> writes:B :   Does anyone know of a Secure Shell server on something like it :that runs on VMS?  B   Please search for "SSH" in the OpenVMS FAQ.  The OpenVMS FAQ is 7   available via the URL http://www.openvms.compaq.com/.u  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 28 Jun 2000 13:30:08 PDTT From: Fairfield@SLAC.Stanford.EDU (Ken Fairfield; SLAC: 650-926-2924; FAX: 926-3515) Subject: Re: SSH on vms?3 Message-ID: <UtQ00$M8mQsf@mccdev.slac.stanford.edu>   . In article <395A5CAD.CE1A1125@tnrdlib.bc.ca>, 9     	ken nightingale <knightingale@tnrdlib.bc.ca> writes:sC >    Does anyone know of a Secure Shell server on something like itm > that runs on VMS?u  H         Yes.  David Jones of  OSU  HTTP  Server  fame has written an SSHH     Server  for VMS, which, BTW, we've been running for the better  part     of a year.  See:  4         http://er6s1.eng.ohio-state.edu/~jonesd/ssh/  H     for  information  and  downloading  of  the  source  for  the latestH     version.   You  will  also  need some version  of  OpenSSL  for  theH     encryption stuff.  I believe  URLs  are  present  in  the  readme's/     release notes.  H         Also, there is a mailing list  for discussing the VMS SSH ServerE     (and the VMS SSH client called "FISH").  Send a one-line message,l       	SUBSCRIBE VMS-SSH        to the list server,  "         MXserver@alpha.sggw.waw.pl  H     to subscribe to the VMS-SSH  list  where  you can ask more questions$     and/or search the archives, etc.    H         Finally, Process  Software  has  ported  the  Data  Fellow's SSHH     Server  _and_  Client  to VMS for inclusion in the next  release  ofH     Multinet, currently in external field test.  If you can wait, and ifH     you use Multinet, this version will be released before  the  end  of
     summer...r           -Ken -- cM  Kenneth H. Fairfield            |  Internet: Fairfield@SLC.Slac.Stanford.EduB:  SLAC, 2575 Sand Hill Rd, MS 46  |  Voice:    650-926-2924:  Menlo Park, CA  94025           |  FAX:      650-926-3515N  -----------------------------------------------------------------------------B  These opinions are mine, not SLAC's, Stanford's, nor the DOE's...   ------------------------------  % Date: Wed, 28 Jun 2000 14:55:01 -0600t% From: Dan O'Reilly <dano@process.com>s Subject: Re: SSH on vms?C Message-ID: <4.3.2.7.2.20000628145434.00b2d1c0@pop.clsp.uswest.net>n  C The upcoming version of both MultiNet and TCPware will support SSH.t  - At 02:14 PM 6/28/2000, ken nightingale wrote:t >Hi,C >    Does anyone know of a Secure Shell server on something like itv >that runs on VMS? >thanks, >ken nightingale >--  >  >knightingale@tnrdlib.bc.cac   ------I +-------------------------------+---------------------------------------+hI | Dan O'Reilly                  |                                       |iI | Principal Engineer            |  "Time flies like an arrow.  Fruit    |wI | Process Software Corporation  |   flies like a banana."               |eI | http://www.process.com        |                    -- Groucho Marx    | I +-------------------------------+---------------------------------------+n   ------------------------------   Date: 28 Jun 2000 23:01:19 CDT= From: wayne@tachysoft.xxx.183943.killspam.00bb (Wayne Sewell)lT Subject: Re: Sun Porting Clearinghouse for Staroffice, anybody going to port to VMS?. Message-ID: <FgKRMb2YD+xu@tachxxsoftxxconsult>  ^ In article <slk71d3pjev66@corp.supernews.com>, "Peter Weaver" <peter.weaver@stelco.ca> writes:H > If you check out http://soldc.sun.com/staroffice/#porting you will seeE > that Sun is looking for people who want to port StarOffice to otherhB > platforms. I know several people in the past have said they wereD > interested in getting StarOffice ported to VMS, is anyone going toG > attempt it? I do not have the C skills to even think about it myself,m# > but maybe someone out there does.a >   L Wasn't the licensing scheme for this rather unfavorable?  I remember hearingO about it a few months ago.  Something about going to the effort of porting withaO the possibility that Sun could restrict/terminate distribution of the port on at whim.t   -- cO ===============================================================================oM Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxxr: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-)sO ===============================================================================iO Otter, on dining with Bluto:"It's perfectly safe if you keep your arms and legss 			away from his mouth."   ------------------------------  % Date: Wed, 28 Jun 2000 14:40:09 -0400o' From: "Bill Todd" <billtodd@foo.mv.com>C. Subject: Re: VAX based 1032 database to Oracle( Message-ID: <8jdgop$7tk$1@pyrite.mv.net>  L Though I'm speaking with no personal acquaintance with the situation, Hoff'sH response below raises what seems to be an obvious question:  would it beF feasible (in terms of the required port time-frame) and less expensiveL (given what CCA apparently wants to charge for the license transfer) to portL the application to Oracle on the existing VAXen, and then to port the result	 to Alpha?v   - bill  = Hoff Hoffman <hoffman@xdelta.zko.dec.nospam> wrote in message90 news:8jda1k$a7s$1@mailint03.im.hou.compaq.com... >e5 > In article <8jd6jv$2m0$1@pr0n.openface.ca>, "linux" $ <linuxmtl@-NOSPAM-yahoo.com> writes:L > :I have a client that need to migrate from their old Vaxes (VAX 6420 & VAXK > :7510) to the new DS20 servers and CCA wants them to pay over 200,000.00$fJ > :just to transfer the licenses to the Alpha systems....a real 'rip-off'. >sH >   Check the costs of the port, and you may well find that your effortsH >   to move (quickly and in one jump) from 1032 on OpenVMS VAX to OracleE >   (Rdb or classic) on OpenVMS Alpha will burn through that $200,000oJ >   licensing figure pretty quickly.  And if the port attempt should fail,J >   the costs can be far higher -- depending on the value of the database,F >   a serious failure could obviously put the company out of business. >tJ >   As a recommendation, once you get the current application and databaseI >   environment over onto OpenVMS Alpha, you can then evaluate what wouldeI >   be required to move from the 1032 database over to the Oracle (Rdb ortI >   classic) database, either incrementally (often the prefered approach)hG >   or as another full "port".  This effort will have to be examined insJ >   comparision with the on-going 1032 support costs, once the environment  >   is running on OpenVMS Alpha. >wH >   Of course, if this is a small-scale and simple database environment,I >   a direct "big bang" port might make more sense and might be as likelytH >   to suceed as an incremental port.  (Without details around the scaleE >   and complexity of the current database environment and the use ofeJ >   1032-specific features, any specific discussions around the trade-offsD >   and the costs of the port can obviously only be rather generic.) >5J >   I'm familiar with Oracle Rdb and quite familiar with OpenVMS, and haveK >   experience porting large applications based on Oracle Rdb (embedded SQLlG >   in C) from OpenVMS VAX over to a mixed OpenVMS Cluster environment.oK >   (I'm presently waiting for an embedded-SQL application build to finish.  :-)rG >   As such, I'd look very carefully at the costs involved, and (from ahG >   technical perspective) I would also look to handle this port in twooH >   stages: first from OpenVMS VAX to OpenVMS Alpha (or mixed), and then$ >   a port from 1032 over to Oracle. >wL > :Since thoses systems are already running Oracle, it would be a great save of1 > :money if we could migrate the s1032 to Oracle.o ><K >   Yes, it probably would, though the costs might only be recoverable over5B >   a long(er) interval.  Though as was inferred elseposting, your
 particularH >   questions and your particular phrasing makes it appear that you will haveJ >   some time to get up to speed on OpenVMS VAX and specifically on a portG >   over to OpenVMS Alpha and over to Oracle.  This is also obviously a-G >   non-trivial additional cost to the effort.  (No offense is intendedm here.) >o, >  --------------------------- pure personal# opinion ---------------------------r1 >    Hoff (Stephen) Hoffman   OpenVMS Engineeringa hoffman#xdelta.zko.dec.com >A   ------------------------------  # Date: Wed, 28 Jun 2000 20:53:29 GMTk From: mschaus@hotmail.comd. Subject: Re: VAX based 1032 database to Oracle) Message-ID: <8jdojv$u2a$1@nnrp1.deja.com>   D OK, guys.  I really hate to have to respond to this (look at all theF crap on every other newsgroup), but this database conversion is simplyG a small portion of the total project.  This part will likely be handled-! by a smaller firm working for us.u  - However, I do thank everybody for their help.N   -Mike Schaus    7 In article <200006272303_MC2-AA54-BBEB@compuserve.com>,s5   "Richard B. Gilbert" <DRAGON@compuserve.com> wrote: E >         If I may say so, you, and your company, seem singularly illvF > prepared to offer this customer what he wants;  you don't even speak hiss > language.  >,    & Sent via Deja.com http://www.deja.com/ Before you buy.o   ------------------------------  % Date: Wed, 28 Jun 2000 17:53:37 -0500n7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> . Subject: Re: VAX based 1032 database to Oracle- Message-ID: <395A81F1.A559062A@earthlink.net>T   linux wrote: >  > Just for the record. > 7 > s1032 is now owned by Computer Corporation of Americas > (http://www.cca-int.com) . > K > I have a client that need to migrate from their old Vaxes (VAX 6420 & VAXtJ > 7510) to the new DS20 servers and CCA wants them to pay over 200,000.00$I > just to transfer the licenses to the Alpha systems....a real 'rip-off'.N [snip]  B Just them taking a pricing example from Compaq (and Digital before them).   -- y David J. Dachtera  dba DJE SystemsO" http://home.earthlink.net/~djesys/  : Unofficial Affordable OpenVMS Home Page and Message Board:+ http://home.earthlink.net/~djesys/vms/soho/-   ------------------------------  % Date: Wed, 28 Jun 2000 15:16:17 -0700.5 From: "Larry D Bohan, Jr" <LBohan@dbc.spam_less..com>- Subject: Re: VAX on Intel?2 Message-ID: <CHdaOaiFBFQfBs1s1K3I4ZLH9FXx@4ax.com>  / On Wed, 28 Jun 2000 11:47:33 -0600, Glen Martino& <GLENMARK@utxvms.cc.utexas.edu> wrote:  B >A VMS bigot who has been swamped with NT/Exchange admin duties of >late...G >(Whatever idiot within the bowels of Redmond who decided that it was ahI >good idea to have EVERYBODY'S mailbox in a SINGLE DATABASE should, IMHO,a5 >have evil things done to him. But I'm not bitter...)o  $ My understanding is that the MSEXCH / was built from a SQL engine, modified for speedh0 (prolbably at the expense of..., you guessed it, reliability/recoverability)s  - the white papers I've seen, describing MSEXCHt0 recovery, illustrate that folks best suited  to 0 MSEXCH admin, are those with DBA/admin training.  6 seems like what was part-time duty on VMS (pmdf admin)5 translates into 3 full-time (MSEXCH admin) positions;a) Hidden expenses MSFT never mentions much.    ------------------------------  % Date: Wed, 28 Jun 2000 22:08:07 -0400f6 From: "Michael A. Foley" <mike.foley@technologist.com> Subject: Re: VAX on Intel?8 Message-ID: <0cy65.243577$MB.4280697@news6.giganews.com>  @ "Larry D Bohan, Jr" <LBohan@dbc.spam_less..com> wrote in message, news:CHdaOaiFBFQfBs1s1K3I4ZLH9FXx@4ax.com...% > My understanding is that the MSEXCH'1 > was built from a SQL engine, modified for speedf2 > (prolbably at the expense of..., you guessed it, > reliability/recoverability)s  =     It's better than that, it's based on the JET engine. Yea,N'     the database from Microsoft Access.a  4     I was in Redmond testing the Alpha NT version of;     the first version of Exchange Server. I asked about the=A     single database and the issues of restoring a single mailbox. >     The deer caught in the headlights look was entertaining to<     watch when I told them they were going to piss off every
     sysadmin.n  >     It was that same trip I told a Microsoftie who was busting;     my chops because I worked at DEC that someday Microsofto?     would come up against something similar. That we at DEC hadaB     the same attitude and it bit us. He laughed. I'm laughing now.D     The  only thing they ever innovated was marketing. The rest they     took from DEC and others.    mike   ------------------------------   End of INFO-VAX 2000.360 ************************