1 INFO-VAX	Tue, 17 Jul 2001	Volume 2001 : Issue 394       Contents:1 Re: 3 Reasons why VMS is alive and probably well+ 1 Re: 3 Reasons why VMS is alive and probably well+  Re: Alpha - Intel and VMS ( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Re: Alpha:  an invitation to communicate( Ancient VMS distributions for hobbyists?! Re: Attunity Connect and Oracle 8 ! Re: Basic quesitons from a newbie $ Re: booting diskless satellite fails' Check out the Wall Street Journal today + Re: Check out the Wall Street Journal today + Re: Check out the Wall Street Journal today + Re: Check out the Wall Street Journal today + Re: Check out the Wall Street Journal today + Re: Check out the Wall Street Journal today + Re: Check out the Wall Street Journal today 2 Re: Compaq's Road to IPF: A Twenty-Question Survey2 Re: Compaq's Road to IPF: A Twenty-Question Survey2 Re: Compaq's Road to IPF: A Twenty-Question Survey2 Re: Compaq's Road to IPF: A Twenty-Question Survey Compaq's WebsiteB Re: Computer Rooms and big UPSes: I need some electrical advice...? Re: DCL arrays, DELETE (was: Re: Basic quesitons from a newbie) 2 Re: DCL code will break in Alpha -> IPF migration?2 Re: DCL code will break in Alpha -> IPF migration?2 Re: DCL code will break in Alpha -> IPF migration?2 Re: DCL code will break in Alpha -> IPF migration?" Re: DHCP on different Cluster Node" Re: DHCP on different Cluster Node" Re: DHCP on different Cluster Node" Re: DHCP on different Cluster Node Free HW,  last chance  Re: FUD  Re: FUD  Help me please in one test. 
 Re: Hobbyists  How to shutdown DECevent RE: How to shutdown DECevent Re: IA64 Rocks My World  Re: IA64 Rocks My World  Re: IA64 Rocks My World  Re: IA64 Rocks My World  Re: IA64 Rocks My World  Re: IA64 Rocks My World  Re: IA64 Rocks My World  Re: IA64 Rocks My World  Re: IA64 Rocks My World  Re: IA64 Rocks My World  Re: Is there a WASD forum? Re: ISV's and VMS future Re: ISV's and VMS future Re: ISV's and VMS future& Re: looping to delete a directory tree& Re: looping to delete a directory tree& Re: looping to delete a directory tree1 Need help with data recovery on failed volume set  No chance for OpenVMS  Re: No chance for OpenVMS  Re: No chance for OpenVMS  Re: No chance for OpenVMS  Re: No chance for OpenVMS  Re: No chance for OpenVMS  Re: No chance for OpenVMS  Re: No chance for OpenVMS  Re: No chance for OpenVMS  Re: No chance for OpenVMS  Re: No chance for OpenVMS B Re: OpenVMS on IPF (was re: IPF already needs a face-lift for VMS)B Re: OpenVMS on IPF (was re: IPF already needs a face-lift for VMS)B Re: OpenVMS on IPF (was re: IPF already needs a face-lift for VMS)B Re: OpenVMS on IPF (was re: IPF already needs a face-lift for VMS) Optical Drive for OpenVMS  Re: OT: Dr Who.  Re: OT: Dr Who.  Re: OT: Dr Who.  Re: OT: Dr Who.  Re: OT: Dr Who. . Re: Pathworks 6.0D on OpenVMS/vax 7.3 question Pathworks Question Re: Pathworks Question RE: Pathworks Question Re: PAWZ and ECP - Re: PAWZ and ECP - Penance for Compaq's VMS sins?" Re: Penance for Compaq's VMS sins?" Re: Penance for Compaq's VMS sins?= Possibly a dumb question on TCPIP settings for 2 IP Addresses  Re: RA7x series disks 4 Returning a value from a C program to a DCL variable8 Re: Returning a value from a C program to a DCL variable8 Re: Returning a value from a C program to a DCL variable RSH question.. Re: RSH question.. Re: RSH question.. Re: RSH question.. SNMP monitor decHUB temperature # Re: SNMP monitor decHUB temperature  SS$_GSDFULL  Re: SS$_GSDFULL  Re: SS$_GSDFULL  Re: SS$_GSDFULL ( Re: Sun's take on Compaq's announcement.( Re: Sun's take on Compaq's announcement.( Re: Sun's take on Compaq's announcement.( Re: Sun's take on Compaq's announcement.( Re: Sun's take on Compaq's announcement.$ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-64 Re: Upgrade PAGEFILE.SYS error% Re: VMS Delete lots of files quickly? % Re: VMS Delete lots of files quickly?  Re: Want 21164 Alpha  F ----------------------------------------------------------------------  % Date: Tue, 17 Jul 2001 12:01:43 +0100 % From: Alan Greig <a.greig@virgin.net> : Subject: Re: 3 Reasons why VMS is alive and probably well+8 Message-ID: <dm68lts626jhcovptmucenhp97r1bhtfes@4ax.com>  1 On Sun, 15 Jul 2001 13:44:01 -0400, "Main, Kerry"  <Kerry.Main@compaq.com> wrote:    K >>>> Well, get someone to explain "the whole picture" here then, instead of # >offering meaningless platitudes.<<  > ) >Have you reviewed the presentation yet??   B Even Richard George said several times yesterday that the completeE picture is not yet available and will not be for several months yet.       -- Alan   ------------------------------   Date: 17 Jul 2001 15:08:11 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) : Subject: Re: 3 Reasons why VMS is alive and probably well++ Message-ID: <9j1kcr$120$3@info.cs.uofs.edu>   8 In article <dm68lts626jhcovptmucenhp97r1bhtfes@4ax.com>,(  Alan Greig <a.greig@virgin.net> writes:4 |> On Sun, 15 Jul 2001 13:44:01 -0400, "Main, Kerry"! |> <Kerry.Main@compaq.com> wrote:  |>   |>  N |> >>>> Well, get someone to explain "the whole picture" here then, instead of& |> >offering meaningless platitudes.<< |> >, |> >Have you reviewed the presentation yet?? |>  E |> Even Richard George said several times yesterday that the complete H |> picture is not yet available and will not be for several months yet.   D And Compaq expects that businesses all over the world are just goingD to put off any IT plans they had and sit around to wait for them to  provide concrete information??  D And the fallout continues.  I was waiting to hear something about myE Student Porting Project in the beginning of this month, (it's success H and maybe even it's very existence was dependant on an offered equipmentG grant from outside the University.)  I have had no contact since before G Black Monday and I fear the plug was probably pulled on this project as  well.   8 Even PHB's can read the writting on the wall eventually.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Tue, 17 Jul 2001 12:51:08 +0100 % From: Alan Greig <a.greig@virgin.net> " Subject: Re: Alpha - Intel and VMS8 Message-ID: <5988ltka15m6in33s325cl5mi2sk6gbuqb@4ax.com>  D On Fri, 13 Jul 2001 19:07:08 +0200, John McLean <mcleanj@dplanet.ch> wrote:   >  > C >Just what are these customers being told that makes them extremely E >positive ?  What are they being told that we are not ?  (Or are they . >simply less skeptical of Compaq's promises ?)  @ "Extremely positive" are not words that I would  use to describeD customer response at yesterday's talk to the customers event. At the? end of Richard George's presentation nobody even clapped for an @ eternity until politeness got the better of me and I thought I'dA better start at least a modest clap as I have no bitter feeling s  towards Richard George.   A Most of my harder questions were answered with "we are working on  providing an answer to this"   -- Alan   ------------------------------  # Date: Tue, 17 Jul 2001 13:28:11 GMT = From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) 1 Subject: Re: Alpha:  an invitation to communicate 0 Message-ID: <009FF22B.42FD96BE@SendSpamHere.ORG>  R In article <9ivmno$prf$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes: > G >"Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message . >news:SPVZGzKByHGB@eisner.encompasserve.org...J >> In article <Pine.SOL.4.33.0107162040280.3839-100000@holyrood.ed.ac.uk>,/ >Peter Boyle <pboyle@holyrood.ed.ac.uk> writes:  >  >....  > J >> > My only comment is that the above argument presupposes that IA64 will; >> > attain the volumes of x86, wich may or may not happen. G >> > If, for the sake of argument, IA64 volumes don't take off how much  >> > patience will intel have? >> >
 >> > Peter >>F >> If VMS is ported to IA64 and IA64 is overwhelmed in the marketplaceD >> by IBM's Power CPUs, VMS will have become that much more portableD >> for porting to Power.  Alpha has not won the hearts of the world, >> despite performance.  > L >Alpha won a great many hearts, from what we've seen on Usenet over the pastL >few weeks.  What it lacked were anything like commensurate sales - and thatK >was not Alpha's fault, but its owner's for lackadaisical marketing and (as - >we now know) highly-questionable commitment.   H I have a clipping from a magazine called "Computer World" that was givenI to my by a friend.  It's dated September 19, 1994 and the column title is H "Alpha may not win, but it's the best".  The author of this article back then predicted:   I "Despite its many strengths, Alpha may be the case of superior technology H  that does not win out.  It is carrying the millstone of Digital's busi-H  ness problems as it tries to win new adherents.  But win, lose or draw,H  the principles around which Alpha was built will continue to influence $  computer design for years to come."  G Gee, there was a lot of truth in that one paragraph.  Alpha influenced  H Pentium design via intel's patent rip-offs.  Digital's business problemsH -- which are also Compaq's -- have succeeded in killing off the greatestG technology going.  Alpha, already in 1994, had won the hearts of those  H using it and it was the envy of those not using it or producing it!  So,/ let's stop trying to denegrate it posthumously.   G I am a fan of Alpha -- always have been, always will be.  How much of a F fan?  Wall Street Journal Tuesday, February 25, 1992 pages A9-A11 hangG framed upon my wall.  I embraced it when it was first created and stood F behind it.  I was dragged through the shit and bowels of our sleazy USG legal system because of it and, despite that, I am STILL there to stand ! behind it.  Why?  It is the best.    --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM              J   "And of course, I'm a genius, so people are naturally drawn to my fiery I   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbes    ------------------------------    Date: 17 Jul 2001 10:23:37 -0500+ From: young_r@encompasserve.org (Rob Young) 1 Subject: Re: Alpha:  an invitation to communicate 3 Message-ID: <ViH2TEqf60kh@eisner.encompasserve.org>   p In article <009FF22B.42FD96BE@SendSpamHere.ORG>, system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes:   > J > I have a clipping from a magazine called "Computer World" that was givenK > to my by a friend.  It's dated September 19, 1994 and the column title is J > "Alpha may not win, but it's the best".  The author of this article back > then predicted:  > K > "Despite its many strengths, Alpha may be the case of superior technology J >  that does not win out.  It is carrying the millstone of Digital's busi-J >  ness problems as it tries to win new adherents.  But win, lose or draw,J >  the principles around which Alpha was built will continue to influence & >  computer design for years to come." >   > 	Yep.  Michael Goldberg.  It was one of the better articles he< 	wrote.  Prior to that , he used to be pretty hard on Alpha.= 	One of two things happened.  Somebody took time out of their ? 	day to ring him up and educate him or he gained that education @ 	on his own.  It is a lot of hard work educating the writers andA 	end-users.... most are interested in other things and won't take B 	the time to find out about it on their own.  But short of letting@ 	people know where you are coming from and where you are headed,B 	few will be interested and before you know it, your customer base9 	has slowly leeched away.  Like rust.  Rust Never Sleeps.    				Rob    ------------------------------  % Date: Tue, 17 Jul 2001 11:00:16 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> 1 Subject: Re: Alpha:  an invitation to communicate 2 Message-ID: <XpY47.841$rc5.60732@news.cpqcorp.net>  ? antonio.carlini wrote in message <3B5358A0.B8D7F22E@iee.org>...    > # >I guess you cannot name the source ! >(otherwise you would have done).  >      Correct.  ) >But if you could name him, would we know  >him (at least by reputation)? >    Yes.  % >I guess he won't make the statements  >publically? >    Dunno.   ------------------------------  % Date: Tue, 17 Jul 2001 11:06:05 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> 1 Subject: Re: Alpha:  an invitation to communicate 2 Message-ID: <ovY47.845$rc5.60834@news.cpqcorp.net>  J If I had been part of the Alpha chip design team, I would feel exactly theD same way.  I would be absolutely sure that we could have kept up ourG performance lead.  I guess the question must have come down to "at what  cost".  L Look at how long it took to get the 1GHz CPU's out there.  It wasn't becauseL the Alpha deigners couldn't design it.  Look how long it took to get EV6 outJ there to begin with.  Look how long it's taken to get the initial EV7s outJ there - it's not because Pete and his team couldn't design it.  What's the, cost per-chip going to be for EV7?  For EV8?      : Bill Todd wrote in message <9ivi7e$ljf$1@pyrite.mv.net>... >  ...    ------------------------------  % Date: Tue, 17 Jul 2001 11:11:30 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> 1 Subject: Re: Alpha:  an invitation to communicate 2 Message-ID: <uAY47.847$rc5.60851@news.cpqcorp.net>  H The most dominant CPU architecture in the world is arguably the ugliest,2 least loved architecture ever developed - the x86.  / Sparc sucks, and Sun sells 'em by the boatload.   F If Micro$soft gets behind IPF (the least loved software company in theI world), then I would expect to see IPF succeed.  The day that they decide A POWER (or AMD) is their direction is the day I'll change my mind.    _Fred   : Bill Todd wrote in message <9ivmno$prf$1@pyrite.mv.net>... > G >"Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message . >news:SPVZGzKByHGB@eisner.encompasserve.org...J >> In article <Pine.SOL.4.33.0107162040280.3839-100000@holyrood.ed.ac.uk>,/ >Peter Boyle <pboyle@holyrood.ed.ac.uk> writes:  >  >... > J >> > My only comment is that the above argument presupposes that IA64 will; >> > attain the volumes of x86, wich may or may not happen. G >> > If, for the sake of argument, IA64 volumes don't take off how much  >> > patience will intel have? >> >
 >> > Peter >>F >> If VMS is ported to IA64 and IA64 is overwhelmed in the marketplaceD >> by IBM's Power CPUs, VMS will have become that much more portableD >> for porting to Power.  Alpha has not won the hearts of the world, >> despite performance.  > L >Alpha won a great many hearts, from what we've seen on Usenet over the pastL >few weeks.  What it lacked were anything like commensurate sales - and thatK >was not Alpha's fault, but its owner's for lackadaisical marketing and (as - >we now know) highly-questionable commitment.  > / >  If IA64 fails to win the hearts of the world D >> despite intense marketing, it does not mean that Alpha would have >> been the successor. > I >Alpha wouldn't have had to have been 'the successor' to have made Compaq  far I >greater profit than anything else seems at all likely to:  it would just E >have had to have had a half-decent slice of the market in which it's 9 >currently moderately profitable despite lousy marketing.  > - >  But VMS does need a 64-bit environment forV) >> the future (even though I prefer VAX).  > I >Only because Compaq just killed the perfectly good 64-bit environment its
 >already had.a >e >- billa >  >l >    ------------------------------   Date: 17 Jul 2001 15:41:40 GMT7 From: woodacre@scala.reading.sgi.com (Michael Woodacre)n1 Subject: Re: Alpha:  an invitation to communicateo. Message-ID: <9j1mbk$44q0h$1@fido.engr.sgi.com>  p In article <009FF22B.42FD96BE@SendSpamHere.ORG>, system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes:  |> Organization: TMESIS Software3 |> Message-ID: <009FF22B.42FD96BE@SendSpamHere.ORG>o |> References: <9ion0n$cfd$1@pyrite.mv.net> <OGG47.775$rc5.60253@news.cpqcorp.net> <Pine.SOL.4.33.0107162040280.3839-100000@holyrood.ed.ac.uk> <SPVZGzKByHGB@eisner.encompasserve.org> <9ivmno$prf$1@pyrite.mv.net>o$ |> Reply-To: system@SendSpamHere.ORG+ |> X-Complaints-To: newsabuse@supernews.como |> Lines: 57> |> Xref: fido.engr.sgi.com comp.os.vms:302404 comp.arch:123343 |> oU |> In article <9ivmno$prf$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes:e |> >J |> >"Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message1 |> >news:SPVZGzKByHGB@eisner.encompasserve.org... M |> >> In article <Pine.SOL.4.33.0107162040280.3839-100000@holyrood.ed.ac.uk>, 2 |> >Peter Boyle <pboyle@holyrood.ed.ac.uk> writes: |> > |> >.... |> >M |> >> > My only comment is that the above argument presupposes that IA64 willR> |> >> > attain the volumes of x86, wich may or may not happen.J |> >> > If, for the sake of argument, IA64 volumes don't take off how much! |> >> > patience will intel have?y |> >> > 
 |> >> > Peters |> >>eI |> >> If VMS is ported to IA64 and IA64 is overwhelmed in the marketplaceCG |> >> by IBM's Power CPUs, VMS will have become that much more portablewG |> >> for porting to Power.  Alpha has not won the hearts of the world,  |> >> despite performance. |> >O |> >Alpha won a great many hearts, from what we've seen on Usenet over the pastoO |> >few weeks.  What it lacked were anything like commensurate sales - and thatHN |> >was not Alpha's fault, but its owner's for lackadaisical marketing and (as0 |> >we now know) highly-questionable commitment. |> tK |> I have a clipping from a magazine called "Computer World" that was giveneL |> to my by a friend.  It's dated September 19, 1994 and the column title isK |> "Alpha may not win, but it's the best".  The author of this article backe |> then predicted:   |>  L |> "Despite its many strengths, Alpha may be the case of superior technologyK |>  that does not win out.  It is carrying the millstone of Digital's busi-eK |>  ness problems as it tries to win new adherents.  But win, lose or draw,oK |>  the principles around which Alpha was built will continue to influence i' |>  computer design for years to come."o |> cJ |> Gee, there was a lot of truth in that one paragraph.  Alpha influenced K |> Pentium design via intel's patent rip-offs.  Digital's business problems   H I'm curious, can you name the patents involved? I've never been clear on1 what exactly was involved in this patent skirmish    Thanks Mike  K |> -- which are also Compaq's -- have succeeded in killing off the greatestnJ |> technology going.  Alpha, already in 1994, had won the hearts of those K |> using it and it was the envy of those not using it or producing it!  So, 2 |> let's stop trying to denegrate it posthumously. |> nJ |> I am a fan of Alpha -- always have been, always will be.  How much of aI |> fan?  Wall Street Journal Tuesday, February 25, 1992 pages A9-A11 hangkJ |> framed upon my wall.  I embraced it when it was first created and stoodI |> behind it.  I was dragged through the shit and bowels of our sleazy USpJ |> legal system because of it and, despite that, I am STILL there to stand$ |> behind it.  Why?  It is the best. |>   |> --rR |> VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM |>            M |>   "And of course, I'm a genius, so people are naturally drawn to my fiery eL |>   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbes |> m   -- d    Michael S. Woodacre N woodacre@sgi.com   p Phone: +44 118 925 7846e   ------------------------------    Date: 17 Jul 2001 18:19:41 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)1 Subject: Re: Alpha:  an invitation to communicaten( Message-ID: <3b54659d@news.kapsch.co.at>  j In article <uAY47.847$rc5.60851@news.cpqcorp.net>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes:I >The most dominant CPU architecture in the world is arguably the ugliest, 3 >least loved architecture ever developed - the x86.e >M0 >Sparc sucks, and Sun sells 'em by the boatload. > G >If Micro$soft gets behind IPF (the least loved software company in the-J >world), then I would expect to see IPF succeed.  The day that they decideB >POWER (or AMD) is their direction is the day I'll change my mind.   Agreed.i. But if it happens, then it is too late for Q !7 Migrate VMS again to POWER ? Nobody will wait for it...0   -- :< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888e< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  # Date: Tue, 17 Jul 2001 16:33:00 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>1 Subject: Re: Alpha:  an invitation to communicate0: Message-ID: <0NZ47.391$N21.400942@typhoon.ne.mediaone.net>  J "Brian Schenkenberger, VAXman-" <system@SendSpamHere.ORG> wrote in message* news:009FF22B.42FD96BE@SendSpamHere.ORG... > >tI > >Alpha won a great many hearts, from what we've seen on Usenet over thei pastI > >few weeks.  What it lacked were anything like commensurate sales - and. thatI > >was not Alpha's fault, but its owner's for lackadaisical marketing and" (as / > >we now know) highly-questionable commitment.u >.J > I have a clipping from a magazine called "Computer World" that was givenK > to my by a friend.  It's dated September 19, 1994 and the column title iseJ > "Alpha may not win, but it's the best".  The author of this article back > then predicted:o > K > "Despite its many strengths, Alpha may be the case of superior technologywJ >  that does not win out.  It is carrying the millstone of Digital's busi-J >  ness problems as it tries to win new adherents.  But win, lose or draw,I >  the principles around which Alpha was built will continue to influencee& >  computer design for years to come." >nH > Gee, there was a lot of truth in that one paragraph.  Alpha influencedJ > Pentium design via intel's patent rip-offs.  Digital's business problemsJ > -- which are also Compaq's -- have succeeded in killing off the greatestH > technology going.  Alpha, already in 1994, had won the hearts of thoseJ > using it and it was the envy of those not using it or producing it!  So,1 > let's stop trying to denegrate it posthumously.| >    A fitting obituary indeed!   ------------------------------  % Date: Tue, 17 Jul 2001 12:34:01 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)1 Subject: Re: Alpha:  an invitation to communicatekL Message-ID: <rdeininger-1707011234020001@user-2iveacr.dialup.mindspring.com>  D In article <uAY47.847$rc5.60851@news.cpqcorp.net>, "Fred Kleinsorge"$ <kleinsorge@star.zko.dec.com> wrote:    lH > If Micro$soft gets behind IPF (the least loved software company in theK > world), then I would expect to see IPF succeed.  The day that they decidenC > POWER (or AMD) is their direction is the day I'll change my mind.y  I This is a safe yardstick to use, for the time being.  But microsoft won'tnD stay dominant forever.  Nothing ever does.  A company that practicesC customer-hating as much as microsoft is headed for a fall, the only I question is when.  And there are lots of examples where the fall has comen rather suddenly.  E In a decade or two, some company that's currently tiny and obscure is-J going to stomp microsoft into the dirt, unless microsoft changes its ways.   -- S Robert Deininger rdeininger@mindspring.com    ------------------------------  # Date: Tue, 17 Jul 2001 16:40:21 GMT.4 From: "Terry C. Shannon" <terryshannon@mediaone.net>1 Subject: Re: Alpha:  an invitation to communicatew: Message-ID: <VTZ47.394$N21.404592@typhoon.ne.mediaone.net>  @ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message, news:ovY47.845$rc5.60834@news.cpqcorp.net...L > If I had been part of the Alpha chip design team, I would feel exactly theF > same way.  I would be absolutely sure that we could have kept up ourI > performance lead.  I guess the question must have come down to "at what  > cost". > F > Look at how long it took to get the 1GHz CPU's out there.  It wasn't becausetJ > the Alpha deigners couldn't design it.  Look how long it took to get EV6 out4L > there to begin with.  Look how long it's taken to get the initial EV7s outL > there - it's not because Pete and his team couldn't design it.  What's the. > cost per-chip going to be for EV7?  For EV8? >A  I Given the current run rate of perhaps half a million CPUs per year, "$500fL per chip in overhead expenses alone" was not the right answer to the "What'sK the cost?" question. Whether IPF turns out to be a Better Answer remains tow be seen.  H Complexity is another issue. Piling wider issue and SMT on top of out ofK order execution makes for a daunting architectural exercise. To proceed and1L succeed with EV8, Compaq would have required incremental staffing and, quite! likely, a waiver of Murphy's Law.p   ------------------------------  # Date: Tue, 17 Jul 2001 16:44:18 GMTi4 From: "Terry C. Shannon" <terryshannon@mediaone.net>1 Subject: Re: Alpha:  an invitation to communicatea: Message-ID: <CXZ47.397$N21.406337@typhoon.ne.mediaone.net>  @ "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message, news:uAY47.847$rc5.60851@news.cpqcorp.net...J > The most dominant CPU architecture in the world is arguably the ugliest,4 > least loved architecture ever developed - the x86. >s1 > Sparc sucks, and Sun sells 'em by the boatload.e >d  I Indeed they do. In-your-face marketing can cover a multitude of sins. I'dwC rather be marketing UltraSparc 3 gear (once Sun manages to ship itssC somewhat-competitive 900MHz part) than worrying about designing anduB delivering UltraSparc 4. The addition of out of order execution to. UltraSparc 4 will be an interesting challenge.  G Compaq's decision to go with IPF reflects their philosophy that the CPUiB won't be the key differentiator in future systems. We shall see...   ------------------------------  # Date: Tue, 17 Jul 2001 16:46:32 GMTd4 From: "Terry C. Shannon" <terryshannon@mediaone.net>1 Subject: Re: Alpha:  an invitation to communicate : Message-ID: <IZZ47.399$N21.407456@typhoon.ne.mediaone.net>  7 "Peter LANGSTOEGER" <eplan@kapsch.net> wrote in messager" news:3b54659d@news.kapsch.co.at...F > In article <uAY47.847$rc5.60851@news.cpqcorp.net>, "Fred Kleinsorge"% <kleinsorge@star.zko.dec.com> writes:eK > >The most dominant CPU architecture in the world is arguably the ugliest,u5 > >least loved architecture ever developed - the x86.g > > 2 > >Sparc sucks, and Sun sells 'em by the boatload. > >0I > >If Micro$soft gets behind IPF (the least loved software company in theeL > >world), then I would expect to see IPF succeed.  The day that they decideD > >POWER (or AMD) is their direction is the day I'll change my mind. >,	 > Agreed.h0 > But if it happens, then it is too late for Q !9 > Migrate VMS again to POWER ? Nobody will wait for it...t >   C Compaq certainly considered the POWER option when they made the IPFt architectural decision.n   ------------------------------  % Date: Tue, 17 Jul 2001 12:55:48 -0400e5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>d1 Subject: Re: Alpha:  an invitation to communicatet2 Message-ID: <f6_47.860$rc5.60979@news.cpqcorp.net>  % Robert Deininger wrote in message ...l   >.J >This is a safe yardstick to use, for the time being.  But microsoft won'tE >stay dominant forever.  Nothing ever does.  A company that practicesiD >customer-hating as much as microsoft is headed for a fall, the onlyJ >question is when.  And there are lots of examples where the fall has come >rather suddenly.m    L I disagree with part of this.  I think that Joe Sixpack who buys a Compaq PCG at Best Buy, and who uses it mostly for Quicken, AOL, and a handfull ofrH games - is pretty happy with Windows and what he gets from Microsoft andL from the hardware vendor who bundles all that nice, pre-configured software.  F The people who *hate* Microsoft tend to be those of us in the computerI business who don't work for, or otherwise make a living off of Microsoft. C Or the end user who is computer savvy, maybe even a large business.7J Nonetheless, the PC is nearly the Universal desktop platform.  I'm writingL this reply on a PC, despite the fact I could be using the VMS system sittingB right next to it - which I do development for.  I have no love for- Microsoft, but I use their product every day.a   >?F >In a decade or two, some company that's currently tiny and obscure isK >going to stomp microsoft into the dirt, unless microsoft changes its ways.0 >0  J Maybe.  But IBM was a "Microsoft" in it's day, and while IBM may no longerJ be the 900 lb gorilla it used to be, it's still a good 750.  Microsoft mayK be knocked off it's perch one day, but it will remain a player for a *long*e time.t   ------------------------------    Date: 17 Jul 2001 13:07:12 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)d1 Subject: Re: Alpha:  an invitation to communicatek3 Message-ID: <op4fynH$u3YT@eisner.encompasserve.org>   j In article <f6_47.860$rc5.60979@news.cpqcorp.net>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes:' > Robert Deininger wrote in message ...t >  >>K >>This is a safe yardstick to use, for the time being.  But microsoft won'trF >>stay dominant forever.  Nothing ever does.  A company that practicesE >>customer-hating as much as microsoft is headed for a fall, the only-K >>question is when.  And there are lots of examples where the fall has comeo >>rather suddenly. >  > N > I disagree with part of this.  I think that Joe Sixpack who buys a Compaq PCI > at Best Buy, and who uses it mostly for Quicken, AOL, and a handfull ofgJ > games - is pretty happy with Windows and what he gets from Microsoft andN > from the hardware vendor who bundles all that nice, pre-configured software. > H > The people who *hate* Microsoft tend to be those of us in the computerK > business who don't work for, or otherwise make a living off of Microsoft.tE > Or the end user who is computer savvy, maybe even a large business.i  L I believe that in their naivete, Joe and Jane sixpack do not hate Microsoft,K they hate computers, since they think all computers are like the example in>H their possession.  If they are a little less self-confident than others,K they _fear_ computers, for the same reason.  They know there are people whocG know how those computers work and they feel unworthy.  Nobody ever felt  that way about typewriters.o  E But in certain regards, the computer industry as a whole is to blame. D I write letters on a Macintosh that has 10-year old software, and itB does what I want.  I have not suffered the anguish of upgrades andC incompatibility because I have another Macintosh on which I can runo@ bleeding edge (but not very) software as I see fit.  Some peopleA use DOS for the same reason.  The industry effort to keep sellingeB people newer and faster gear is what the brokerage regulators callF "churn".  I am not going to write letters any faster than the existingA Macintosh can handle it, and if I did I would make more mistakes.    ------------------------------    Date: 17 Jul 2001 19:14:39 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)1 Subject: Re: Alpha:  an invitation to communicate ( Message-ID: <3b54727f@news.kapsch.co.at>  q In article <IZZ47.399$N21.407456@typhoon.ne.mediaone.net>, "Terry C. Shannon" <terryshannon@mediaone.net> writes: [ >"Peter LANGSTOEGER" <eplan@kapsch.net> wrote in message news:3b54659d@news.kapsch.co.at...im >> In article <uAY47.847$rc5.60851@news.cpqcorp.net>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes:mL >> >The most dominant CPU architecture in the world is arguably the ugliest,6 >> >least loved architecture ever developed - the x86. >> >3 >> >Sparc sucks, and Sun sells 'em by the boatload.- >> >J >> >If Micro$soft gets behind IPF (the least loved software company in theM >> >world), then I would expect to see IPF succeed.  The day that they decideeE >> >POWER (or AMD) is their direction is the day I'll change my mind.F >>
 >> Agreed.1 >> But if it happens, then it is too late for Q !g: >> Migrate VMS again to POWER ? Nobody will wait for it... >aD >Compaq certainly considered the POWER option when they made the IPF >architectural decision.  0 Choosing INTEL or IBM shouldn't be the question.! Choosing Alpha or INTEL should bec; and the answer should have been "my own platform of course"r   -- n< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888s< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  # Date: Tue, 17 Jul 2001 17:19:53 GMTn4 From: "Terry C. Shannon" <terryshannon@mediaone.net>1 Subject: Re: Alpha:  an invitation to communicate8: Message-ID: <Zs_47.419$N21.423658@typhoon.ne.mediaone.net>  F "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message- news:op4fynH$u3YT@eisner.encompasserve.org...4F > In article <f6_47.860$rc5.60979@news.cpqcorp.net>, "Fred Kleinsorge"% <kleinsorge@star.zko.dec.com> writes:A) > > Robert Deininger wrote in message ...  > >n > >>G > >>This is a safe yardstick to use, for the time being.  But microsoftt won'taH > >>stay dominant forever.  Nothing ever does.  A company that practicesG > >>customer-hating as much as microsoft is headed for a fall, the onlyAH > >>question is when.  And there are lots of examples where the fall has come > >>rather suddenly. > >a > >fF > > I disagree with part of this.  I think that Joe Sixpack who buys a	 Compaq PCnK > > at Best Buy, and who uses it mostly for Quicken, AOL, and a handfull ofsL > > games - is pretty happy with Windows and what he gets from Microsoft andF > > from the hardware vendor who bundles all that nice, pre-configured	 software." > > J > > The people who *hate* Microsoft tend to be those of us in the computerB > > business who don't work for, or otherwise make a living off of
 Microsoft.G > > Or the end user who is computer savvy, maybe even a large business.l > C > I believe that in their naivete, Joe and Jane sixpack do not hatet
 Microsoft,J > they hate computers, since they think all computers are like the example inJ > their possession.  If they are a little less self-confident than others,I > they _fear_ computers, for the same reason.  They know there are peopleo whocI > know how those computers work and they feel unworthy.  Nobody ever felt  > that way about typewriters.  >eG > But in certain regards, the computer industry as a whole is to blame.hF > I write letters on a Macintosh that has 10-year old software, and itD > does what I want.  I have not suffered the anguish of upgrades andE > incompatibility because I have another Macintosh on which I can run.B > bleeding edge (but not very) software as I see fit.  Some peopleC > use DOS for the same reason.  The industry effort to keep sellinglD > people newer and faster gear is what the brokerage regulators callH > "churn".  I am not going to write letters any faster than the existingC > Macintosh can handle it, and if I did I would make more mistakes.e  K Yup. Although he waxes sensational more than he waxes eloquent, Mark MinasitK offers some interesting insights on the Wintel crowd's ability to transformeK bugfixes and processor speed bumps into a (until recently) generous annuityW< revenue stream. A brief review of Minasi's magnum opus is at8 http://www.acersoft.com/SKD/Acersoft-SKC-BookReviews.htm   ------------------------------   Date: 17 Jul 2001 17:24:06 GMT3 From: anton@mips.complang.tuwien.ac.at (Anton Ertl)m1 Subject: Re: Alpha:  an invitation to communicate , Message-ID: <9j1sbm$of8$1@news.tuwien.ac.at>  2 In article <uAY47.847$rc5.60851@news.cpqcorp.net>,8  "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes:G >If Micro$soft gets behind IPF (the least loved software company in thea0 >world), then I would expect to see IPF succeed.  @ Like MIPS and Alpha, right?  Or do you think that MS did not getB behind MIPS and Alpha?  Then, how do we see, except in retrospect,F that MS does, or does not, get behind IPF?  And if we can tell only inE retrospect, then we cannot use "MS gets behind" for prediction as you  suggest.  F BTW, what's this IPF nonsense?  <wild speculation>Is Intel's marketingF already preparing to abandoning the current IA64 for an enhanced IA32,B and wants to free the IA64 name of current associations? Or is theE IA64 name just too associated with years of delays etc.?  Or does the ? IA64 name come out of engineering, and is not snazzy enough fort2 marketing (not to mention that it's not a TLA)?</>   Followups to comp.arch.w   - antonn -- yK M. Anton Ertl                    Some things have to be seen to be believed-K anton@mips.complang.tuwien.ac.at Most things have to be believed to be seeni0 http://www.complang.tuwien.ac.at/anton/home.html   ------------------------------  # Date: Tue, 17 Jul 2001 17:27:22 GMTr4 From: "Terry C. Shannon" <terryshannon@mediaone.net>1 Subject: Re: Alpha:  an invitation to communicate0: Message-ID: <_z_47.423$N21.426680@typhoon.ne.mediaone.net>  7 "Peter LANGSTOEGER" <eplan@kapsch.net> wrote in message2" news:3b54727f@news.kapsch.co.at...F > In article <IZZ47.399$N21.407456@typhoon.ne.mediaone.net>, "Terry C., Shannon" <terryshannon@mediaone.net> writes:  F > >Compaq certainly considered the POWER option when they made the IPF > >architectural decision. >a2 > Choosing INTEL or IBM shouldn't be the question.# > Choosing Alpha or INTEL should bec= > and the answer should have been "my own platform of course"  >o  L I agree, but the combination of nearly a decade of marketing malfeasance andG strategic blunders led to an economically unsustainable Alpha business.h  E With Alpha off the RISC playing field, the question is, "who's next?"o   ------------------------------  # Date: Tue, 17 Jul 2001 07:05:36 GMTl) From: ethan_dicks@yahoo.com (Ethan Dicks)a1 Subject: Ancient VMS distributions for hobbyists?d; Message-ID: <3b546e2c.83771270@news-server.columbus.rr.com>t  C I've been going through piles of tapes from my old company and havet@ managed to extract upgrade savesets from TU-58 carts for VMS 3.2> and VMS3.5 (someone else overwrote one of the VMS3.4 tapes andB at least one from my VMS3.3 and VMS3.4 sets won't read, even after coaxing and cleaning).  C It's gotten me thinking about ancient VMS.  I got my first taste of ? VMS as a user in 1984, on a pair of VAX-11/750s (and I have onet< of those specific machines! S/N BT000354)  It's been runningD VMS4.x for a long time; before I ever got a priv'ed account.  Before> I finished reading my VMS3.2 tapes, I was hoping that it was aF full distribution, but it's only an upgrade kit.  VMS3.1 required. :-(  F So... are there any projects to preserve and collect VMS distributionsF prior to 5.0?  It's obviously not of any real commercial value at thisA point, but for those of us who have older systems, it'd be fun tof run contemporary stuff.a   -ethan   ------------------------------    Date: 17 Jul 2001 05:08:52 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)i* Subject: Re: Attunity Connect and Oracle 83 Message-ID: <b33jFl5mgWfQ@eisner.encompasserve.org>   e In article <000c01c10e63$17988ff0$0200a8c0@teamrdb.com>, Eric Ebinger <eebinger@telocity.com> writes:n  J > It is possible to a Java application running on a Windoze system talk toC > a Rdb database running on VMS by using a JDBC-ODBC bridge to talk.I > to the ODBC driver for Rdb.  I am not aware of any "out of the box" way C > to have a Java application running on VMS talk to a Rdb database.w   I don't understand.i  4 You can access Rdb through the VMS calling standard.  C Are you saying that Java cannot make outbound calls via the callingh
 standard ?  A I would expect Java should be like the compiled languages in thisc7 regard, although you might have to create the bindings.n  	 > (OracleeI > does not have a native JDBC driver for Rdb.)   My understanding is that5M > the Attunity "On Platform" package makes it possible for a Java applicatione= > running on VMS to access an Oracle 7/8 database but the "OnyM > Platform" package does not include support for Rdb databases.  To make this N > really offensive Attunity Connect does support Rdb as a datasource for their* > full Attunity Connect package (for $$$).  F If vendors don't have a way to make money, there is no reason for them to support VMS.h   ------------------------------    Date: 17 Jul 2001 11:12:32 -0500- From: koehler@encompasserve.org (Bob Koehler)m* Subject: Re: Basic quesitons from a newbie3 Message-ID: <byxraIoF$Wra@eisner.encompasserve.org>   c In article <14ce1c21.0107161159.24e66526@posting.google.com>, paszty@xoma.com (Nick Paszty) writes:w > H > question 2.  i have searched the documentation for the array structure? > in DCL but have found nothing.  i did find some documentation G > indicating that typical array stuctures are not available in DCL.  isoF > there another clever way to be able to establish a list of items andC > then loop through them to apply an operation or some other useful  > task?  >   $ Simulated using symbol substitution:   $i = 5 $j = 6 $account'i'j = 220  G this sets element 5,6 in the simluated array "account" to the value 22.jG Actually it simply makes a symbol called account56 and sets that to 22,1G but it can be followed through as an array.  Just watch out for elemento 15,6 vs. element 1,56.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation = NASA GSFC Flight Software       | Federal Sector, Civil GroupSE                                 | please remove ".aspm" when replyinge   ------------------------------  / Date: Tue, 17 Jul 2001 08:03:17 +0200 (MET DST) & From: Rudolf Wingert <win@fom.fgan.de>- Subject: Re: booting diskless satellite failsn6 Message-ID: <200107170603.IAA19377@sinet1.fom.fgan.de>   Hello,  F I did see some reasons for that problem. The first was, that two nodesG within the cluster did have the same name (different roots). The secondkE was a missing file (I don't remember the right name). In both cases I-; did remove the satellite adn add them newly to the cluster.    Regards Rudolf Wingert   ------------------------------  % Date: Tue, 17 Jul 2001 06:12:14 -0400e2 From: "Sue Skonetski" <susan.skonetski@compaq.com>0 Subject: Check out the Wall Street Journal today2 Message-ID: <NbU47.808$rc5.60633@news.cpqcorp.net>  3 there is a Compaq ad (any spelling errors are mine)r Text   What will Compaq Customers get with The transisation to thei Itanium (tm) Architecture?  
 The Future ofy 64-bit Computing  8 Then four customer quotes  (yes there are VMS customers)   Then a paragraph that says.e  L Our customers consistently say that they prefer the choice,  flexibility andL cost-efficiency of systems based on industry-leading components, rather thanJ proprietary technologies.  Transferring our Alpha technology to Intel (tm)J to enhance the future of the Itanium (tm) processor family is the best wayH to protect our customers' long-term enterprise computing investments andJ keep them at the forefront of technology.  And clearly, we're not the only ones who think that way.  0 To get the facts go to compaq.com/ipf-enterprise   ------------------------------    Date: 17 Jul 2001 07:05:37 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)g4 Subject: Re: Check out the Wall Street Journal today3 Message-ID: <P0hSWtkAFo5g@eisner.encompasserve.org>f  g In article <NbU47.808$rc5.60633@news.cpqcorp.net>, "Sue Skonetski" <susan.skonetski@compaq.com> writes:A" > Date: 17 Jul 2001 06:12:14 -0500                     ^^^^^^^^                     Wow !y    2 > To get the facts go to compaq.com/ipf-enterprise  > Congratulations, this page looks fine from a secured browser !   ------------------------------  % Date: Tue, 17 Jul 2001 07:46:15 -0400w( From: Hamlyn Mootoo <univms@bigfoot.com>4 Subject: Re: Check out the Wall Street Journal today+ Message-ID: <3B542587.B1884BBA@bigfoot.com>r  F I went to the URL mentioned, but I'm a little embarassed to admit, I'mG confused.  You see, on the Web page there are two quotes attributed to sE officials of the London Stock Exchange.  Here it is verbatim from thet page:u  0 London Stock Exchange, David Lester, President: D "The London Stock Exchange believes Compaq's decision to migrate theH Himalaya platform to Intel Itanium is a positive move that will protectG the Exchange's investment in Himalaya-based systems in the long term."    I ------------------------------------------------------------------------ o; London Stock Exchange plc, Jonathan Wittmann, IT Director: iD "The London Stock Exchange believes Compaq's decision to migrate theH Himalaya platform to Intel Itanium is a positive move that will protectC the Exchange's investment in its Himalaya based systems in the long0 term." o  F What I can't figure out is which one is the real person, and which oneB is the wooden dummy.  It is rumored that UK officials can be a bitG stuffy, and I've heard of "yes-men", but identical utterances?  Is thisaG Compaq's new program of fault-tolerant testimonials - in case you don'toG believe the paid testimonial from one guy, they hope you'll believe the@ other?G Can you tell me where to apply so my parrot can get some work? He's lowe
 on bird seed.o   HM   Sue Skonetski wrote: > 5 > there is a Compaq ad (any spelling errors are mine)i > Text >  > What will Compaq > Customers get with > The transisation to thet > Itanium (tm) Architecture? >  > The Future ofd > 64-bit Computing > : > Then four customer quotes  (yes there are VMS customers) >  > Then a paragraph that says.l > N > Our customers consistently say that they prefer the choice,  flexibility andN > cost-efficiency of systems based on industry-leading components, rather thanL > proprietary technologies.  Transferring our Alpha technology to Intel (tm)L > to enhance the future of the Itanium (tm) processor family is the best wayJ > to protect our customers' long-term enterprise computing investments andL > keep them at the forefront of technology.  And clearly, we're not the only > ones who think that way. > 2 > To get the facts go to compaq.com/ipf-enterprise   ------------------------------   Date: 17 Jul 2001 14:28:00 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)p4 Subject: Re: Check out the Wall Street Journal today* Message-ID: <9j1i1g$72$1@info.cs.uofs.edu>  2 In article <NbU47.808$rc5.60633@news.cpqcorp.net>,5  "Sue Skonetski" <susan.skonetski@compaq.com> writes:s |>O |> Our customers consistently say that they prefer the choice,  flexibility and O |> cost-efficiency of systems based on industry-leading components, rather thanr |> proprietary technologies.    0 Ummm....  Isn't OpenVMS proprietary technology??   bill   -- aJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   a   ------------------------------    Date: 17 Jul 2001 10:46:43 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)i4 Subject: Re: Check out the Wall Street Journal today3 Message-ID: <7A7KXLKjHVSl@eisner.encompasserve.org>7  ^ In article <9j1i1g$72$1@info.cs.uofs.edu>, bill@triangle.cs.uofs.edu (Bill Gunshannon) writes:4 > In article <NbU47.808$rc5.60633@news.cpqcorp.net>,7 >  "Sue Skonetski" <susan.skonetski@compaq.com> writes:. > |>Q > |> Our customers consistently say that they prefer the choice,  flexibility andEQ > |> cost-efficiency of systems based on industry-leading components, rather thanr  > |> proprietary technologies.   > 2 > Ummm....  Isn't OpenVMS proprietary technology??  = Yes, just like Tru64, Solaris, HP-UX, AIX, MacOS and Windows.   B But when something talks about "systems" and "components" they are? generally talking hardware.  I doubt that customers are tellingo= Compaq they prefer Mosaic to Navigator, Internet Explorer ands Opera.  @ The idea of being able to run VMS on a wider variety of machines= than produced by Compaq is quite appealing.  Even if concernsoB about support made it unattractive to a lot of serious businesses,A the evangelism prospects for hobbyists are astounding.  While the B education marketplace has always been where people get their firstA biases on operating systems, that is moving upstream to the home.i  @ A properly run hobbyist program does not cost Compaq anything in@ lost sales -- those people were not going to fork over thousands of dollars anyway.   ------------------------------  % Date: Tue, 17 Jul 2001 10:47:13 -0400M5 From: David Beatty <David.Beatty@qwertysasasdfgh.com>A4 Subject: Re: Check out the Wall Street Journal today2 Message-ID: <qU9UO2goqb+xLAeH=ZsDkLTgfyef@4ax.com>  < On 17 Jul 2001 14:28:00 GMT, bill@triangle.cs.uofs.edu (Bill Gunshannon) wrote:  3 >In article <NbU47.808$rc5.60633@news.cpqcorp.net>,.6 > "Sue Skonetski" <susan.skonetski@compaq.com> writes: >|>sP >|> Our customers consistently say that they prefer the choice,  flexibility andP >|> cost-efficiency of systems based on industry-leading components, rather than >|> proprietary technologies.  . >$1 >Ummm....  Isn't OpenVMS proprietary technology??e >o >billr   Yes.  ' The last time I checked, so is Itanium.i   David R. Beattyn   ------------------------------  % Date: Tue, 17 Jul 2001 11:54:00 -0400s- From: JF Mezei <jfmezei.spamnot@videotron.ca>n4 Subject: Re: Check out the Wall Street Journal today, Message-ID: <3B545F96.7F451765@videotron.ca>   Larry Kilgallen wrote:B > The idea of being able to run VMS on a wider variety of machines. > than produced by Compaq is quite appealing.   N The idea that VMS could run on all Alpha machines from small to very large wasN also very appealing. But Multia, and the systems marketed by API don't run VMS' even though they are still Alpha based.t  K Has Compaq made a written commitment that VMS would be bootable on any IA64r server/workstation ?   ------------------------------  % Date: Tue, 17 Jul 2001 11:11:18 +0200  From: "Magnus M" <mmad@tips.se>w; Subject: Re: Compaq's Road to IPF: A Twenty-Question Surveyr3 Message-ID: <rjT47.440$Ta.2022@news3.global-ip.net>g  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3B538E65.5C23D158@videotron.ca... > Comments on the questions: >hG > A question is missing on whether the customer beleives that this movea@ > strenghtens or weakens Compaq's commitment to VMS (and Tru64). >,A > Also, there is a question on whather Compaq has provided enoughw information toL > make purchasing decision. My initial instinct was YES: Theyt have providedE > enough to convince me to quit VMS ASAP. But I ended up answering NOo
 because of0 > the way I think the answers will be tabulated.  G If you do plan to quit VMS ASAP as you state, I hope that might include  VMS-related Usenet
 activities...a   /magnusa   ------------------------------    Date: 17 Jul 2001 05:54:31 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)r; Subject: Re: Compaq's Road to IPF: A Twenty-Question Surveya3 Message-ID: <RedDApzaoFyP@eisner.encompasserve.org>   U In article <rjT47.440$Ta.2022@news3.global-ip.net>, "Magnus M" <mmad@tips.se> writes:  > < > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3B538E65.5C23D158@videotron.ca... >> Comments on the questions:s >>H >> A question is missing on whether the customer beleives that this moveA >> strenghtens or weakens Compaq's commitment to VMS (and Tru64).t >>B >> Also, there is a question on whather Compaq has provided enough > information toM >> make purchasing decision. My initial instinct was YES: Theyt have providedJF >> enough to convince me to quit VMS ASAP. But I ended up answering NO > because of1 >> the way I think the answers will be tabulated.D > I > If you do plan to quit VMS ASAP as you state, I hope that might includeo > VMS-related Usenet > activities...e   Agreed.e   ------------------------------   Date: 17 Jul 2001 13:19:56 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)t; Subject: Re: Compaq's Road to IPF: A Twenty-Question Surveyt, Message-ID: <9j1e1t$214k$1@info.cs.uofs.edu>  , In article <3B538E65.5C23D158@videotron.ca>,0  JF Mezei <jfmezei.spamnot@videotron.ca> writes: |> Comments on the questions:t |> nQ |> Also, there is a question on whather Compaq has provided enough information tosM |> make purchasing decision. My initial instinct was YES: Theyt have providedaQ |> enough to convince me to quit VMS ASAP. But I ended up answering NO because ofc1 |> the way I think the answers will be tabulated.a  J I agree with this.  I selected "No Opinion" because I was sure you weren'tH looking for the an answer like the YES answer above.  It is not so  muchM the amount of information provided as the amount not provided, but the result  is the same.   bill   -- hJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   T   ------------------------------  # Date: Tue, 17 Jul 2001 16:50:29 GMTi4 From: "Terry C. Shannon" <terryshannon@mediaone.net>; Subject: Re: Compaq's Road to IPF: A Twenty-Question Surveyp: Message-ID: <p1_47.401$N21.409390@typhoon.ne.mediaone.net>  > "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message& news:9j1e1t$214k$1@info.cs.uofs.edu.... > In article <3B538E65.5C23D158@videotron.ca>,2 >  JF Mezei <jfmezei.spamnot@videotron.ca> writes: > |> Comments on the questions:I > |>D > |> Also, there is a question on whather Compaq has provided enough information toF > |> make purchasing decision. My initial instinct was YES: Theyt have providedH > |> enough to convince me to quit VMS ASAP. But I ended up answering NO
 because of3 > |> the way I think the answers will be tabulated.o >tL > I agree with this.  I selected "No Opinion" because I was sure you weren'tJ > looking for the an answer like the YES answer above.  It is not so  muchH > the amount of information provided as the amount not provided, but the result > is the same. >0  K Thanks for the feedback. We put this survey together faster than Compaq putyG the IPF launch together, hence some things fell through the cracks. Ther/ fallout will be addressed in a followup survey.e   cheers,    terry ss   ------------------------------  % Date: Tue, 17 Jul 2001 14:15:19 +0200o2 From: "Ren Schelbaum" <rene.schelbaum@datakom.at> Subject: Compaq's WebsiterG Message-ID: <3b542be3$0$25434$6e365a64@newsreader02.highway.telekom.at>    Have you already looked at  ? http://www.compaq.com/hps/ipf-enterprise/customer_quotes.html ?e  L Looks very much like those "I have lost 40 pounds in a week" advertisments - doesn't it ?   regardsu   Ren Schelbaum Datakom Austria GesmbH  8 by the way - this is my personal opinion - not Datakom's   ------------------------------  % Date: Mon, 16 Jul 2001 21:07:12 -0700 4 From: Roger Wiechman <Roger_Wiechman@HMC.nospam.Edu>K Subject: Re: Computer Rooms and big UPSes: I need some electrical advice... . Message-ID: <3B53B9F0.2C7CD335@HMC.nospam.Edu>   Hamlyn Mootoo wrote: > H > Keep in mind that if you plan to run at 80%, that this is steady-stateF > load.  Should you lose power, and drain your batteries to zero, whenJ > power does come back on again (you might not be there when it does), theI > current draw will be higher due to disk drives motors (non-linear load)aG > drawing more power during startup.  Although storageworks controllersoE > sequence startup of the disk drives, if you have enough independentbF > cabinets in the room, the instantaneous draw could trip the Liebert. >   F ** How true, but much more likely you should plan on the input breakerD  to the UPS being sizable, since when power is restored, you get 80% load4  -plus- a possible substantial battery charger load.  @  I'll continue with more comments from the original post.  I had	 forgotten G  much of my EE when it comes to 3 phase, but have managed to relearn itt+  recently due to the CA power situation. :)e   >  > Scott Vieth wrote: > >y > > All: > >pJ > > I'm doing some planning regarding our current UPS capacity and how bigC > > the next one should be.  We currently have a 75kVA Liebert unitcG > > servicing our computer room.  The LCD panel on the front says it istJ > > running at 55%.  I'm trying to determine how much equipment we can addK > > to the computer room before we hit 80% (a ceiling that we determined ono
 > > our own).m > >sK > > We hired an electrician to come in with an ammeter to measure the power-F > > consumption of some of our "typical" equipment racks.  I have someK > > questions on how to interpret these numbers so I need someone from this H > > group with an electrical background to help me out.  The electricianJ > > measured the current by tracing the circuits back to the breaker panelH > > and putting the ammeter lead around the wire going directly into the > > breaker. > >tA > > 1) We had an SW800 full of HSJ50s and disks plugged-in with anK > > three-phase connection.  The electrician measured an average current onsK > > phase 1 (I1) of 3.6A, an average current on phase 2 (I2) of 5.9A and an H > > average current on phase 3 (I3) of 6.7A.  Considering that this is aI > > "three-phase 208V" setup, what is the power consumption of that rack? E > > Is it (3.6A + 5.9A + 6.7A) * 208V?  Or is it the highest of thosesK > > values, 6.7A, multiplied by 208V?  (Everything in the SW800 was plugged  > > into one PDU).  E ** Actually, since this cabinet really only has single phase loads of- the 120vD  variety, the power may be estimated at (3.6A + 5.9A + 6.7A) * 120V.G  the 208 is the line-to-line voltage, and really doesn't come into play  here. H  I'd also comment that the cabinet is not very well balanced, load wise.  0  Note for the record that 208 = 120 * 3**.5  :<)   > >MK > > 2) I've got an ESA1200 (two PDUs inside) which uses a NEMA L6-20P plug.sD > > That means two hot leads and a ground (I think).  The electricanK > > measured 3.2A on I1 and 3.4A on I2.  This is for PDU number one in thattL > > rack.  The other PDU had these figures: 3.4A on I1 and 3.3A on I2.  WhatL > > is the total power consumption for that rack?  Is it (3.2A + 3.4A + 3.4A1 > > + 3.3A) * 208V or is it (3.2A + 3.4A) * 208V?e  ; ** This one is messy, only cause I forget for sure the NEMA- designations, but,F  I believe this to be a (nominal) 240 V 20A connector.  When connected toF  a 3 phase system, it indeed provides 208 volts.  When the electrician measuredG  both lines, he was simply seeing the effective 'return' path.  I'd say  eachA  PDU as drawing about 3.3 A at 208V, therefore the total power is  (3.3+3.3)*208.   > >iG > > 3) When the electrician was trying to explain this stuff to me, the G > > square root of three came up a few times in different calculations.tC > > Where does the square root of three come from when dealing with F > > electrical power?  Is it because the UPS is outputting three-phaseJ > > power?  He said that the max capacity of our UPS is (75000 VA)/(208V *E > > sqrt(3)) or right around 208 Amps at 208 Volts.  Is this correct?   G ** Since this is really 3 phases of 120v., I'd approximate the max loadnF  as 25KW/phase / 120v = 208 amps/phase.  I really wouldn't describe it  as "at 208V".  G  The square root of 3 relates to the 120 degree phase relation, and how B  power is distributed in 3 phase system.  Note my earlier formula.   > >iL > > 4) Bonus round:  If the UPS has a capacity of 75kVA AND the meter on theK > > front says we're at 55% AND we don't want to go over 80%, then how muchgH > > juice do we have left in our current setup?  Is it 208A (number fromK > > previous question) multiplied by (80%-55%)?  That would be (208A * 25%) ( > > = 52 Amps at 208V.  Is that correct?  E ** Simple.  Your current (52) is correct.  It should be "per phase atg 120v.".nD   You might want to watch to make sure that any additional connected loadG   is better balanced.  There might be some specific "per phase" currenti@   maximum that could be reached before the overall unit limit is reached.E   In other words, I doubt the UPS would be happy with all 75 KW being,-   taken from one phase. (some 600 amps worth)f  dI > > 5) Is there a "power distribution for dummies" page on the web that I 9 > > could read to bring myself up to speed on this stuff?t    * I never found one.E  I > > 6) What is the difference between "watts" and "VA"?  Why aren't UPSesq > > rated in watts?e  > ** The main difference is that VA takes into account the phase relationshipE  of the voltage and current.  You may see this mentioned elsewhere as H  "power factor".  At a power factor of 1.0, (a true resistive load) thenD  they are the same, but this is rarely the case with switching power suppliesD  and other nasty loads.  UPS's must be rated worst case, VA.  In the worstoD  case, a purely capacitive or inductive load, apparent power (watts) mightn=  go to zero, but true power is a different number altogether.I  H > > I reeeally wish I would have paid more attention when I was studyingE > > E.E. for three semesters in college before I switched to Computer' > > Science.  Ha ha ha!!  A * I had forgotten much as well, although I stuck it out in EE/EL,a troubleo0  is it was 30 years ago, and CS didn't exist. :)   > >f > > Thanks,t > >o > > -Scott :^)   * Cheers,   Rogero   ------------------------------    Date: 17 Jul 2001 02:35:27 -0700  From: nclews@csc.com (Nic Clews)H Subject: Re: DCL arrays, DELETE (was: Re: Basic quesitons from a newbie)= Message-ID: <a720d610.0107170135.3413aec2@posting.google.com>a   hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) wrote in message  :question 2.  i have searched the documentation for the array structure@ > :in DCL but have found nothing.  i did find some documentationH > :indicating that typical array stuctures are not available in DCL.  isG > :there another clever way to be able to establish a list of items and D > :then loop through them to apply an operation or some other useful > :task? > H >   Through symbol substitution within the symbol itself, yes.  (I coverB >   this and other topics in the DCL book.)  At its very simplest: >  > $ x=1e
 > $ y'x'=2 > $ sho sym y10 >   Y1 = 2   Hex = 00000002  Octal = 00000000002 > $k  B I will shortly be releasing some DCL freeware which is designed to; help you analyse telecomms itemised bills, specifically forG? determining which discount structure is most applicable to yourRF calling pattern, and is architecture-independent in that it should runE on VAX, Alpha, and (future) iTanium systems. Most importantly it uses"B self-modifed array names and indexed file access / update, so this. code serves as a rough example of these tasks.  C I'll announce in this newsgroup (or drop me an email for an emailed- announcement).  ( Regards, Nic Clews CSC Computer Sciences nclews at csc dot com    ------------------------------    Date: 17 Jul 2001 05:04:30 -0700  From: nclews@csc.com (Nic Clews); Subject: Re: DCL code will break in Alpha -> IPF migration? = Message-ID: <a720d610.0107170404.6cc7e887@posting.google.com>t  7 "Neil Rieck" <n.rieck@sympatico.ca> wrote in message > a > $ say :== write sys$output  > $ arch = f$getsyi("ARCH_NAME") > $ test = "Unknown", > $ if arch .eqs. "VAX"   then test = "good". > $ if arch .eqs. "Alpha" then test = "better"- > $ if arch .eqs. "IA-64" then test = "yikes"s- > $ say "Architecture: ", arch, " (",test,")"o  E With immediate effect I have started to port my LOGIN.COM to IA-64 oreB whatever it will be called. I am building in the ability to handleA multiple architectures, so I'm not even limiting it to IA-64, letl alone VAX or Alpha.r  4 (Small passage of time for edit...)... LOGIN.COM;373 ...m $ EXESUF=F$GETSYI("ARCH_NAME") ...s $ XV:==$XVDIR:XV_'EXESUF'e ...t  C There, done it. (In fact, I just went in and verified it was what I . thought it was, which it was. Which was nice.)C I don't have any IA-64 EXE's that I can test, but it's based on thep) existing, functional VAX and Alpha logic.e   Regards, Nic ClewsD PS. 373 does not represent the total number of edits, its been reset to ;1 more than once.h   ------------------------------  # Date: Tue, 17 Jul 2001 13:23:05 GMTu8 From: hammond@not@peek.ppb.cpqcorp.net (Charlie Hammond); Subject: Re: DCL code will break in Alpha -> IPF migration?,2 Message-ID: <Z_W47.817$rc5.60769@news.cpqcorp.net>  , In article <3B52F6F1.7725D7FA@my-deja.com>, * Everett Chong <echong@my-deja.com> writes:G >I just realized something.  There are DCL command procedures out therey >that use some variation oft >s& >$ IF   F$GETSYI("HW_MODEL") .LT. 1024 >$  THEN >$    ! VAX-specific codet >$  ELSE >$    ! Alpha-specific code-	 >$  ENDIF-  1 (1) HW_MODEL may not be the best thing to check;  9     ARCH_TYPE (or ARCH_NAME) is probably a better choice.j  @ (2) Assuming that the Universe is made up of only VAX and ALPHA !     is a coding error (arguably).    I have written code like:b  '     $ ARCH$TYPE = F$GETSYS("ARCH_TYPE")      $ IF ARCH$TYPE .EQ. 17     $ THEN 3     $   ... vax code...S     $   GOTO AFTER_ARCHe     $ ENDIFo     $ IF ARCH$TYPE .EQ. 2u     $ THEN s     $   ... alpha code...      $   GOTO AFTER_ARCHa     $ ENDIFe     $  ... error code ...o     $AFTER_ARCH:     : Extending this for OpenVMS Itanium is left as an exercise.   --  K     Charlie Hammond -- Compaq Computer Corporation -- Pompano Beach  FL USA H        (hammond@not@peek.ppb.cpqcorp.net -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------  % Date: Tue, 17 Jul 2001 09:32:58 -0400 - From: "Richard D. Piccard" <piccard@ohio.edu> ; Subject: Re: DCL code will break in Alpha -> IPF migration?n( Message-ID: <3B543E8A.BF84E20D@ohio.edu>  G Any DCL code that is architecture type dependent is going to have to bepK re-visited by the user, regardless of how it was coded.  The DCL files thatdK check for HW_MODEL being less than 1024 may actually be the ones with fewerbI changes, since probably some of the DCL differences between VAX and AlphanF versions will have the IA-64 version behaving identically to the AlphaM version.  If the code uses ARCH_TYPE, then the DCL code will for sure have to:5 be modified when a new possible return value appears.e  #                                 RDPa     Charlie Hammond wrote:  - > In article <3B52F6F1.7725D7FA@my-deja.com>,s, > Everett Chong <echong@my-deja.com> writes:I > >I just realized something.  There are DCL command procedures out theree > >that use some variation oft > >r( > >$ IF   F$GETSYI("HW_MODEL") .LT. 1024
 > >$  THEN > >$    ! VAX-specific coden
 > >$  ELSE > >$    ! Alpha-specific codem > >$  ENDIFa >p2 > (1) HW_MODEL may not be the best thing to check;; >     ARCH_TYPE (or ARCH_NAME) is probably a better choice.u > A > (2) Assuming that the Universe is made up of only VAX and ALPHAi# >     is a coding error (arguably).  >t > I have written code like:e >a) >     $ ARCH$TYPE = F$GETSYS("ARCH_TYPE")i >     $ IF ARCH$TYPE .EQ. 1- >     $ THEN >     $   ... vax code...2 >     $   GOTO AFTER_ARCHi
 >     $ ENDIFd >     $ IF ARCH$TYPE .EQ. 2l >     $ THEN >     $   ... alpha code...T >     $   GOTO AFTER_ARCHs
 >     $ ENDIFz >     $  ... error code ...g >     $AFTER_ARCH: >.< > Extending this for OpenVMS Itanium is left as an exercise. >s > --M >     Charlie Hammond -- Compaq Computer Corporation -- Pompano Beach  FL USA.J >        (hammond@not@peek.ppb.cpqcorp.net -- remove "@not" when replying)L >       All opinions expressed are my own and not necessarily my employer's.   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------   Date: 17 Jul 2001 09:21:46 CDT= From: wayne@tachysoft.xxx.320117.killspam.015d (Wayne Sewell) ; Subject: Re: DCL code will break in Alpha -> IPF migration?s. Message-ID: <cYdzDmF0uQeX@tachxxsoftxxconsult>  g In article <rPN47.14991$304.1893706@news20.bellglobal.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:a7 > "Everett Chong" <echong@my-deja.com> wrote in message9' > news:3B52F6F1.7725D7FA@my-deja.com...cI >> I just realized something.  There are DCL command procedures out therec >> that use some variation ofq >>( >> $ IF   F$GETSYI("HW_MODEL") .LT. 1024
 >> $  THEN >> $    ! VAX-specific codef
 >> $  ELSE >> $    ! Alpha-specific codeM >> $  ENDIFt >>K >> These will have to be identified and re-written for Itanic.  What valuestG >> of HW_MODEL will be used for IPF based systems?  Now that Compaq has G >> pulled the plug on Alpha, there ought to be a finite upper bound fora >> HW_MODEL for Alphas.n >>
 >> Everett >> >> --------a8 >> "Huh?  Compaq killed the Alpha?  How could you tell?" >  > How about this...n >  > $ say :== write sys$output  > $ arch = f$getsyi("ARCH_NAME") > $ test = "Unknown", > $ if arch .eqs. "VAX"   then test = "good". > $ if arch .eqs. "Alpha" then test = "better"- > $ if arch .eqs. "IA-64" then test = "yikes"|- > $ say "Architecture: ", arch, " (",test,")"s  I The reason for use of the F$GETSYI("HW_MODEL") was that arch_name was not0M defined in earlier versions of vms on vax.  Thus the above test would fail on:E an old version.  Unfortunately, people still insist on running those.7     -- 4O ===============================================================================rM Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxx5: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-) O ===============================================================================i? Flounder: "I can't believe I threw up in front of Dean Wormer." >    Otter: "Face it, Flounder.  You threw up *on* Dean Wormer."   ------------------------------  % Date: Tue, 17 Jul 2001 08:39:49 +0200t2 From: "Dr. Otto Titze" <titze@ikp.tu-darmstadt.de>+ Subject: Re: DHCP on different Cluster Node 3 Message-ID: <3B53DDB5.E2D41CD2@ikp.tu-darmstadt.de>u   Hi,h   concerning the question of Carl    Carl Karcher schrieb:b > B > What version of TCPIP is this? I'm seeing a similar problem with@ > gateways for all but the first subnet being incorrect on TCPIPI > V5.0A-ECO1. I'm about to test this with TCPIP V5.1-ECO1 this week. I'lleE > be reporting the results to my open case via DSNlink and a followup- > here.-  B This is TCPip V5.1 out of the box of the VMS 7.3 distribution kit.
 In particulart  TCPIP$DHCP_SERVER.EXE	is	V5.1-15   Regards- Otto 5,  -------------------------------------------, | Dr. Otto Titze, Kernphysik TUD           |, | Schlossgartenstr. 9, D-64289 Darmstadt   |, | titze@ikp.tu-darmstadt.de                |, | Tel: +49(6151)16-2916,FAX:16-4321        |,  -------------------------------------------   ------------------------------  % Date: Mon, 16 Jul 2001 19:46:21 -0700t4 From: Roger Wiechman <Roger_Wiechman@HMC.nospam.Edu>+ Subject: Re: DHCP on different Cluster Node . Message-ID: <3B53A6FD.E683942A@HMC.nospam.Edu>   "Dr. Otto Titze" wrote:i >  > Hi,n > F > my idea was to run individual DHCP Servers in different subnets just< > serving these paricular segemnts only. But the servers are< > members of a LAVC. Obviously this is not possible, becauseF > if I start the second one he is waiting for the lock and starts only > when I abort the first one.   A *I'm not convinced a second DHCP server couln't be started.  Whatr9  lock are you referring to?  What IP stack are you using?    > I > This might be useful in setting up a cluster failover. But there is note+ > described how to get rid of this feature.s  1 * You should be able to run failover easy enough.i  r. > Any idea how to get rid of the lock-request? > G > (The reason for a server in each subnet was that I wanted to restrict D > the IP addresses of laptops with a known MAC address. Depending on? > which subnet the laptop shows up it gets there a different IPiH > address. I can't see how to do it with one DHCP server for all subnets > in the domain.)l  nD ** I believe that any current (version3?) DHCP server can do exactly$  what you want from a single server.  F  Don't quote me here, since I don't have any documentation in front of9  me, but the config should have the following look to it:u  # subnet x.x.1.0 subnet 255.255.255.0  {   
  host Doof  {  fixed-address x.x.1.10;%  hardware ethernet 00:B1:D0:41:72:14;w  } }q  # subnet x.x.2.0 subnet 255.255.255.0  {m
  host Doof  {  fixed-address x.x.2.100;m%  hardware ethernet 00:B1:D0:41:72:14;h  } }   E * This is of course if you want a particular IP address given to thist mac address.A  There are other options as well.  Like allowing a particular mact address to have =  any of a range of addresses, or a multitude of other things.c  F  Most of these would involve creating a 'group' definition, and making the particularD  mac address a member of the group.  You can then use allow and deny
 statementsE  freely in subnets or offered ranges to allow or deny members of thisl	 group.  I B  have done this exact thing and it works well. (Multinet V4.3 with patches)6  If you need examples of this type of thing, just ask.    Good luck,       Roger    ------------------------------  % Date: Tue, 17 Jul 2001 10:35:58 +0200d2 From: "Dr. Otto Titze" <titze@ikp.tu-darmstadt.de>+ Subject: Re: DHCP on different Cluster Node 3 Message-ID: <3B53F8EE.1FBFC054@ikp.tu-darmstadt.de>h   Roger,   thanks for your reply   H > > my idea was to run individual DHCP Servers in different subnets just> > > serving these paricular segemnts only. But the servers are> > > members of a LAVC. Obviously this is not possible, becauseH > > if I start the second one he is waiting for the lock and starts only > > when I abort the first one.e > C > *I'm not convinced a second DHCP server couln't be started.  Whats; >  lock are you referring to?  What IP stack are you using?MA Ok my wording was not correct, it starts and then aquires a lock,-@ after getting this it read is config params. (see example of the! log file at the end of this mail)n I use TCPIP V5.1    >  > >rK > > This might be useful in setting up a cluster failover. But there is not - > > described how to get rid of this feature.r > 3 > * You should be able to run failover easy enough. I In this configuration I didn't want failover just to get rid of the lock n issuee    F > ** I believe that any current (version3?) DHCP server can do exactly& >  what you want from a single server.D This might be true, but when trying this I had the already mentioned= wrong assigment of gateways to subnets. (Your proposed confige@ example is obviously not TCPIP, maybe Multinet. By I will to see9 what can be done with additional subnet entries in TCPIP)    >  Good luck,       Rogera Thanks.i   Otto  @ PS: Log file from startup of the 2nd DHCP server (with comments)  1 $ !  This command procedure runs the DHCP server.e $ DEFINE TCPIP$DHCP_FROMINETD 1h. $ RUN/NODEBUG SYS$SYSTEM:TCPIP$DHCP_SERVER.EXE7 JOIN Server Release 4.1.0f-DECsrc  for VAX with OpenVMSs< Copyright 1992-1998 Competitive Automation, Inc.  All Rights	 Reserved.o0 JOIN server startup  on Tuesday July 17 10:02:047 JOIN Server Release 4.1.0f-DECsrc  for VAX with OpenVMSa< Copyright 1992-1998 Competitive Automation, Inc.  All Rights	 Reserved.l1 Acquiring lock on resource "TCPIP$DHCP_SERVER"...r  @ ---- Here the 2nd Server is waiting before reading configuration  ; The first server has created in its working directory these 
 file entries.B  4 RWLOCKDBA.;1        RWLOCKDBB.;1        RWLOCKDBC.;1. RWLOCKDBF.;1	 RWLOCKDBG.;1        RWLOCKDBH.;1 RWLOCKDBI.;1	RWLOCKDBJ.;1   5 ---- when stopping the first server these are removeds!      and the 2nd server continues   - Acquired lock on resource "TCPIP$DHCP_SERVER" ) This standby server is now active server.82 server.sock after  socket(TCPIP$C_AUXS, 0, 0) = -1 final server.sock 3u+ server dns domain name           = mydomaina+ server nis/nis+ domain name      = mydomainc. canonical server IP address      = 130.8.24.39) canonical server name            = lings0 $ debug level                      = 0' ignored interfaces               = noneo( default lease duration           = 86400) finite bootp auto extension      = 0 secsa* ttl of provisional lease         = 60 secs, timeout on ICMP echo             = 500 msecs, minimum legal bootp packet       = 300 bytes& name service                     = dns( name service updateable          = false/ naming policy                    = by ipaddresse' accept client name               = truei( ignore name ownership            = false( DNS expiration tracks DHCP lease = false( update NS on every transaction   = false$ free list length                 = 8' bootp support                    = trueh( bootp addr from pool             = false' registered clients only          = true ( listen on ppp interfaces         = false( Microsoft RAS support            = false( ping bootp clients               = false( reply when relay on same net     = false( send options in offer            = false' auto-reread modified dhcpcap     = truea( validate bootp client net#       = false' auto synchronise disk database   = true ( token ring source routing        = false( release ip when client moves     = false( expand BOOTP packet ?            = false( send bootfile in 'file' field    = false( ignore hardware type             = false( force broadcast reply            = false' use MAC addr as client ID        = truee syncing databases:  	 Netmasks:m Net              Maskl 130.8.133.0     255.255.255.0D 130.8.24.0      255.255.255.0  network 130.8.24.0&         IPaddress ranges administered:(                 130.8.24.9 - 130.8.24.20*         first free IP address = 130.8.24.1 network 130.8.133.0e&         IPaddress ranges administered:+                 130.8.133.41 - 130.8.133.41i+                 130.8.133.78 - 130.8.133.78h+                 130.8.133.86 - 130.8.133.87i,         first free IP address = 130.8.133.411 Listening on interfaces (name  ip  mask  bdcast):p=          ZE0  130.8.24.39      255.255.255.0     130.8.24.255s   ------------------------------  % Date: Tue, 17 Jul 2001 11:47:28 +0200c2 From: "Dr. Otto Titze" <titze@ikp.tu-darmstadt.de>+ Subject: Re: DHCP on different Cluster Node23 Message-ID: <3B5409B0.5F166252@ikp.tu-darmstadt.de>=   Hi,0   Carl Karcher schrieb:tB > What version of TCPIP is this? I'm seeing a similar problem with@ > gateways for all but the first subnet being incorrect on TCPIPI > V5.0A-ECO1. I'm about to test this with TCPIP V5.1-ECO1 this week. I'llrE > be reporting the results to my open case via DSNlink and a followupe > here.   A It just looked for available ECOs on TCPIP. The latest I found isxF TCPIPVSX_EC0151 dated 10-may-2001. In the content DHCP components are 9 NOT mentioned at all. Therefore maybe no help from there.e   Regards  Otto   ------------------------------  % Date: Tue, 17 Jul 2001 07:20:57 -0700s! From: Tom Linden <tom@kednos.com>- Subject: Free HW,  last chance9 Message-ID: <CIEJLCMNHNNDLLOOGNJIOEOMCPAA.tom@kednos.com>:    Have the following to give away:  C XL266 96MB, Matrox Millenium II no drives (have the NT cds as well)cL 3100/76  32 MB, no drives  (needs Dallas Real Time clock that I cannabalized for 4000/90)' 3100/VS42S  16MB, CD and TK50 no drives    2 rz26 2 rz25 2 rz24 1 rz23 with VMS5.5-2 1 quantum trailblazer(rz25?) 1 hitachi dk312c (rz24?) 1 RRD42E  CD-ROM  J all worked when last powered down.  Would prefer somebody to take the lot.  G If you want this, you must contact your carrier and arranged to have it4	 picked upv from
 Tom Linden 1050 Vaquero Roadn Pebble Beach, CA 93953   ------------------------------    Date: 17 Jul 2001 09:56:13 -0500- From: koehler@encompasserve.org (Bob Koehler)r Subject: Re: FUD3 Message-ID: <zeQj6sOung3j@eisner.encompasserve.org>   ` In article <9ifa2o$2bi5$1@info.cs.uofs.edu>, bill@triangle.cs.uofs.edu (Bill Gunshannon) writes: > F > Unless you have been shown the documents that make this reference, I+ > would consider this revisionist history. p  D We all considered it revisionist history on the part of Sun.  But ifE Word 2.0 can be followed by Word 6.0, why should Sun be worried about  renaming old products?  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences CorporationD= NASA GSFC Flight Software       | Federal Sector, Civil GroupWE                                 | please remove ".aspm" when replyingd   ------------------------------    Date: 17 Jul 2001 10:37:41 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)  Subject: Re: FUD3 Message-ID: <pNUmesX$2sPh@eisner.encompasserve.org>4  c In article <zeQj6sOung3j@eisner.encompasserve.org>, koehler@encompasserve.org (Bob Koehler) writes:Db > In article <9ifa2o$2bi5$1@info.cs.uofs.edu>, bill@triangle.cs.uofs.edu (Bill Gunshannon) writes: >> 2G >> Unless you have been shown the documents that make this reference, Ic, >> would consider this revisionist history.  > F > We all considered it revisionist history on the part of Sun.  But ifG > Word 2.0 can be followed by Word 6.0, why should Sun be worried about  > renaming old products?  B On Macintosh we have Word 5.1A, the last version before they broke security with Macros.e   ------------------------------  # Date: Tue, 17 Jul 2001 15:50:03 GMTO From: <kodtest99@yahoo.com>e$ Subject: Help me please in one test./ Message-ID: <L8Z47.14916$M6.5426462@news.rt.ru>F  
 This is test.a" Answer me please for test message. Thanks,t Robert.p  #  ----------------------------------e$ | FREE Vacation!  FANTASNIC offer! |$ |                                  |$ |  http://linkresort.fasturl.it    |#  ----------------------------------c   ------------------------------    Date: 17 Jul 2001 06:20:57 -0700) From: P.Young@unsw.EDU.AU (Patrick Young). Subject: Re: Hobbyists= Message-ID: <55f85d77.0107170520.2f59343c@posting.google.com>s  v jason_odonnell@erinet.com (Jason O'Donnell) wrote in message news:<5c8ffd05.0107092238.4ae55a50@posting.google.com>... > All, > E > It seems that there are several OpenVMS Hobbyists in this group.  IwB > was just curious what systems people have in their homes runningE > OpenVMS.  Any GS series?  As soon as I get things organized, I willD) > have a 2100A RM and and a 2000 at home.  >  > JMOD  9 Since you _really_ had to ask under the topic "Hobbyists"   F Started with various PDP-11 systems in my bedroom and the back shed...  D 1985 - PDP 11/04 (RT-11) - My first system (actually the first piece of DECB kit I ever saw as a new high School student in 1981 - *HAD TO HAVEF IT*) bought it from the School in 1985 or so with RX01 (in later yearsD I realized this was the worst implementation of 8 inch floppy drives= I've ever seen - Apple II 5 1/4 inch drives could at least betE formatted), soon after RK05 drives. Got an RK11-C at one stage (wholehA rack of little modules) but it was broken and not easy to use (or C power) so an RK11-D and BA-11 followed (the 11/04 was a half heightnD model so no place to put an RK11-D). The RK05 was a *legend* - stillC got a "new" Nashua 4416-12 cartridge sitting next to me unopened in F sealed plastic wrapping. Could fix these things anytime as long as the right bulb with the.7 right filament was available for the linear positioner.I  B 1986 - PDP 11/34 (RT-11/L7 UNIX) - traded in the 11/04 (now wish I
 still keptF it due to how much it ended up influencing my life/job/etc - as in whyA I am in c.o.v right now and that we currently run OpenVMS at workaC these days). Also got a couple of VT103(?)s or whatever those silly E things with built in QBUS 11/2 CPUs were called - my memory is not so $ good these days. Also found myself a8 TE10(?) - bright orange - tape drive - now that was fun!  E 1987 -> 1990  Various PDP 11/40s (RT-11/L7 UNIX) saved from the scrap E yard. They were used in UNSW up to 1987/1988 when the Cyber mainframea? that used PDP 11/40 powered batch stations was replaced with an-F IBM3090. PDP 11/60 at around the same time. RK06 then a number of RK07 drives.1  E 1989 - VAX 11/730 (VMS) - paid real money for this one ($2K) - one ofaB the small rackmount ones. RM80 with the M stripped out - ie: couldA take out the MASSBUS H/W and hook it to the 11/730 IDC. Number ofiD years later a couple of RA81s on a UDA-50. Think I used the RK07 andA BA11 combo with this too when I first got it. Couple of TS11 tape  drivesC at this point (remember a whole of CMOS type logic in the module at  thev? top with the controls and trying to fix it each time it broke).   F 1991 or so MicroVAX 3100 Mobo. Ended up using an Apple II power supply for thisD with some small transformer and regulator to supply some voltage the Apple IIE supply didn't that it needed. At some stage also a MicroVAX II (smallt	 rackmount 	 version).   ! 1996 - Alpha PCI 64-275 (OpenVMS)a  E Currently: Two Alpha PC164 systems. One Linux (just set it up: hint ->F use a 7474 TTL chip to do the power switching in a standard ATX PC box with ATX PC power.E supply), and one OpenVMS on which I type this now using Mozilla under C OpenVMS 7.2-1. Still got the 3100. Someone gave me a VAX 4000 pizza 
 box (have not F looked at it yet). There is also a MicroVAX 3400 and TU81 plus someone dumpedF in the middle of my office at work that has been there for a while now? and acts as tablespace. It needs a KLESI QBUS controller as therD current one is broken - they want me to use it to read their rack of
 old magtapes.f  D Pity I've only got the real estate for the current kit - things kind of went=F as they got replaced with (better) things due to space considerations.  E I don't have a WINTEL peecee at home. Used to for cool games before Ip fullytF understood M$ as a company and their morals, however don't now since I& now find WINTEL boxes quite offensive.  F Using new PC PCI widgets on Alpha/OpenVMS systems on the other hand...   ------------------------------    Date: 16 Jul 2001 23:23:49 -0700' From: piet@timmers-it.nl (Piet Timmers)u! Subject: How to shutdown DECeventr= Message-ID: <be44b12d.0107162223.53e05509@posting.google.com>0   Hi,   D When using the diagnose command on one of our clusternodes I get the
 following:  
 $ diagnose  
 DECevent V3.3DD _DIAGNOSE-FAT:  Automatic analysis is still running. Please shutdown and reissuer	  command.K %SYSTEM-F-ABORT, abort  E According to the information found on www.openvms.compaq.com there isaD a  possible problem with the license. To verify this, I have to stop5 DECevent and start it up again. But it does not shut..   $ diagnose shutdown utov13  
 DECevent V3.3a    & Shutting down DECevent on node UTOV13. Shutdown request completed.   ; After this command $ diagnose stil gives the same message. PA Is there an other way to shutdown decevent, whithout shutting the- node?-  
 Greetings,   Piet Timmers   ------------------------------  % Date: Tue, 17 Jul 2001 07:29:49 +0100 5 From: "Steeples, Oliver" <Oliver.Steeples@compaq.com>9% Subject: RE: How to shutdown DECevent-P Message-ID: <F498D199EDB12D468CD2C66680D3080101DDE477@reoexc04.emea.cpqcorp.net>  % Easy one but bit illogical (captain):    @sys$startup:decevent$shutdown   And to startup:c   @sys$startup:decevent$startupN    I If dia is being used on the cluster elsewhere you can get this error (andi- sometimes if the binary errorlog is too big).a   Regards, 	Oliverf   -----Original Message-----4 From: piet@timmers-it.nl [mailto:piet@timmers-it.nl]$ Sent: Tuesday, July 17, 2001 7:24 AM To: Info-VAX@Mvb.Saic.ComI! Subject: How to shutdown DECevent      Hi,o  D When using the diagnose command on one of our clusternodes I get the
 following:  
 $ diagnose  
 DECevent V3.3WD _DIAGNOSE-FAT:  Automatic analysis is still running. Please shutdown and reissuet	  command.i %SYSTEM-F-ABORT, abort  E According to the information found on www.openvms.compaq.com there is D a  possible problem with the license. To verify this, I have to stop5 DECevent and start it up again. But it does not shut.o   $ diagnose shutdown utov13  
 DECevent V3.3r    & Shutting down DECevent on node UTOV13. Shutdown request completed.n  ; After this command $ diagnose stil gives the same message. eA Is there an other way to shutdown decevent, whithout shutting ther node?b  
 Greetings,   Piet Timmers   ------------------------------  % Date: Tue, 17 Jul 2001 10:48:01 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>   Subject: Re: IA64 Rocks My World2 Message-ID: <seY47.836$rc5.60878@news.cpqcorp.net>  G OK.  I'm, not a database expert, but why is this a bad thing?  If TPC-C F truly measures relative performance that could be expected in the realG world - and if partitioning the system gets you this performance - thenfF what's the big deal?  Tru64 clusters do provide a single system image,H making it relatively easy to administer.  On the other hand, if TPC-C isK just another meaningless benchmark number - then lets stop talking about it ) altogether - or define a meaningful test.a      @ Chris Ruemmler wrote in message <9j00of$c44@cello.hpl.hp.com>...+ >In article <3B45E1CD.FBDBC266@us.ibm.com>, * >Greg Pfister  <pfister@us.ibm.com> wrote: >>andrew harrison wrote: >>[snip]: >>> The problem for WildFire is that it was late but  when3 >>> it was announced it wasn't competitive with thet< >>> previous generation of servers from Compaqs competitors. >>>I7 >>> It recently reached the dizzy heights of 230,000 onG4 >>> TPC-C, using the latest 1001 Mhz CPU's, a result7 >>> puts it just ahead of the IBM P680, a machine thatsO) >>> just about to be replaced by Regatta.j >>>R3 >>> When Rob Young started his WildFire boosting inF4 >>> the dark and dim past initial numbers of 200,0004 >>> were being bandied about, I can only assume that4 >>> this number was a design goal that was leaked to1 >>> Rob (unless he made it up). The fact that the 0 >>> initial numbers came out way lower than that1 >>> and then needed OPS in a box tends to suggestt( >>> that they missed their design goals. >>C >>Do you know for a fact that the OPS-in-a-box solution was used inoC >>Wildfire? That's often rather difficult to tell from official TPCk> >>disclosures. If it was used, it of course makes the high TPC6 >>results rather meaningless (in my personal opinion). >>A >Yes, all of the GS320 TPC-C numbers have been a cluster in a cansH >with OPS except the very first one they published and withdrew (becauseB >it was so poor in terms of performance).  If you look at the full
 disclosureI >reports you'll see 8 configuration files for Oracle in the back, one fortF >each instance.  In addition, their "clients" are always configured inA >multiples of 8 which is another hint that partitioning was used.y > E >It is sad that Compaq had to resort to TPC-C tricks like this to trynB >and show performance on the box.  I imagine the "real" single boxC >performance is not terrible, just not leadership.  The NUMA hit on B >that box is really much too large to run TPC-C really well with a >single instance.  >  >--Chris
 >My own views    ------------------------------   Date: 17 Jul 2001 16:28:18 GMT. From: shoens@lenny.sfrn.dnai.com (Kurt Shoens)  Subject: Re: IA64 Rocks My World+ Message-ID: <9j1p32$q1q$1@bob.news.rcn.net>T  2 In article <seY47.836$rc5.60878@news.cpqcorp.net>,4 Fred Kleinsorge <kleinsorge@star.zko.dec.com> wrote:H >OK.  I'm, not a database expert, but why is this a bad thing?  If TPC-CG >truly measures relative performance that could be expected in the real H >world - and if partitioning the system gets you this performance - thenG >what's the big deal?  Tru64 clusters do provide a single system image, I >making it relatively easy to administer.  On the other hand, if TPC-C ishL >just another meaningless benchmark number - then lets stop talking about it* >altogether - or define a meaningful test.  B In the real world (as opposed to benchmark world), customers wouldE prefer not to set up database clusters because they are more complex.2  ? TPC-C is really a system vendor benchmark.  They spend a lot ofR@ time and money getting good TPC-C results for bragging purposes,> but I question how informative TPC-C results are for customers trying to size/select systems.  < Defining a meaningful test is a pretty tall order.  You need? something that predicts performance for the applications that a D lot of people want to do and that isn't easily manipulated to create@ misleading results.  Ideally it wouldn't take six months of work/ to set it up, run it, and validate the results.f  / Of course, all the above just my silly opinion.M   ------------------------------  % Date: Tue, 17 Jul 2001 18:40:47 +0200s* From: Alexis Cousein <al@brussels.sgi.com>  Subject: Re: IA64 Rocks My World/ Message-ID: <3B546A8F.8090608@brussels.sgi.com>r   Fred Kleinsorge wrote:  ? > OK.  I'm, not a database expert, but why is this a bad thing?     C Because in real life, no-one will actually run a database this way,fD because it lacks flexibility and is an administrative nightmare, andE that you cannot predict a workload like you can in a TPC-C benchmark.i  D At least that's what my totally unrepresentative sample of customers0 (and some Sun customers I know) seem to tell me.   -- s? <these messages express my own views, not those of my employer> ) Alexis Cousein				Senior Systems Engineeru. SGI Belgium and Luxemburg		al@brussels.sgi.com   ------------------------------  % Date: Tue, 17 Jul 2001 12:39:48 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>e  Subject: Re: IA64 Rocks My World2 Message-ID: <fTZ47.855$rc5.60849@news.cpqcorp.net>  ? Kurt Shoens wrote in message <9j1p32$q1q$1@bob.news.rcn.net>...  >cC >In the real world (as opposed to benchmark world), customers would F >prefer not to set up database clusters because they are more complex. >3    I If setting up a database cluster increase performance by 2X for example - E then I would expect those highly paid database guru's to get over the2
 "complexity".i  @ >TPC-C is really a system vendor benchmark.  They spend a lot ofA >time and money getting good TPC-C results for bragging purposes,3? >but I question how informative TPC-C results are for customers  >trying to size/select systems.  >     I Then let's just admit that we're just measuring our privates for bragging I rights, and as long as the "rules" of the competition are followed, there  shouldn't be any carping.-  = >Defining a meaningful test is a pretty tall order.  You needi@ >something that predicts performance for the applications that aE >lot of people want to do and that isn't easily manipulated to createiA >misleading results.  Ideally it wouldn't take six months of works0 >to set it up, run it, and validate the results. >t   Sounds good to me.   ------------------------------  % Date: Tue, 17 Jul 2001 12:41:33 -0400O5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>e  Subject: Re: IA64 Rocks My World2 Message-ID: <UUZ47.856$rc5.60905@news.cpqcorp.net>  L Then explain to me why TPC-C numbers have any meaning at all.  Or why OracleC would waste good money developing such a thing - if nobody uses it.r      F Alexis Cousein wrote in message <3B546A8F.8090608@brussels.sgi.com>... >Fred Kleinsorge wrote:s >r@ >> OK.  I'm, not a database expert, but why is this a bad thing? >s >-D >Because in real life, no-one will actually run a database this way,E >because it lacks flexibility and is an administrative nightmare, and F >that you cannot predict a workload like you can in a TPC-C benchmark. >tE >At least that's what my totally unrepresentative sample of customerse1 >(and some Sun customers I know) seem to tell me.  >u >-- @ ><these messages express my own views, not those of my employer>' >Alexis Cousein Senior Systems Engineer . >SGI Belgium and Luxemburg al@brussels.sgi.com >s   ------------------------------  % Date: Tue, 17 Jul 2001 12:56:52 -0400 - From: "Peter Weaver" <peter.weaver@stelco.ca>A  Subject: Re: IA64 Rocks My World4 Message-ID: <o7_47.263375$Z2.3196237@nnrp1.uunet.ca>  ; "Kurt Shoens" <shoens@lenny.sfrn.dnai.com> wrote in message-% news:9j1p32$q1q$1@bob.news.rcn.net...o >...D > In the real world (as opposed to benchmark world), customers wouldG > prefer not to set up database clusters because they are more complex.n >...  L In the real world we put RDB on VMS, then clustering is a non-issue; it just happens.  G One of the sites I do work at has Traditional Oracle installed for somewH unknown reason. Every time we try to talk about clustering the DBA's runI screaming from the room. If they were running RDB I could have added in aoI second (or third, or fourth,... or 100th)VMS box without even telling thelK DBA's that their database was now in a cluster. But then again, my old Beta $ tapes sit unused next to my VHS VTR.   -- Peter WeaverJ Using a WIN NT/WIN 2000 box to manage your VMS systems is like towing your7 mechanic in a 5th wheel motor home behind your Porsche.D   ------------------------------  % Date: Tue, 17 Jul 2001 19:28:04 +0200s* From: Alexis Cousein <al@brussels.sgi.com>  Subject: Re: IA64 Rocks My World- Message-ID: <3B5475A4.80603@brussels.sgi.com>e   Fred Kleinsorge wrote:  A > Kurt Shoens wrote in message <9j1p32$q1q$1@bob.news.rcn.net>...h > D >>In the real world (as opposed to benchmark world), customers wouldG >>prefer not to set up database clusters because they are more complex.o >> >> >  > K > If setting up a database cluster increase performance by 2X for example -eG > then I would expect those highly paid database guru's to get over theb > "complexity".     D Well, they have other things to do than to improve the tpmC/$ metricF of their installations. Actually keeping production databases running,D as an example. As I said -- in real life one cannot always *predict*@ a workload, so one has to make guesstimates and act accordingly.  C Even on single instances you have similar issues: typical DBAs haveS@ better things to do than to split filesystems over separate disk@ sets an iterative process until the load balancing is Just Right> (which may become Really Wrong the day a new user lands). ManyC times, it's more interesting to configure a large pool of disks and G put everything on all the disks -- you do incur more disk head movementd5 as unrelated entities are accessed on the same disks,w9 but at least you know load balancing will be fairly good.l  3 TCP-C has been around for so long that people startiA exploring some of the more exotic stuff impractical for real-life  usage.    K > Then let's just admit that we're just measuring our privates for braggingwK > rights, and as long as the "rules" of the competition are followed, therea > shouldn't be any carping.i    C No argument from me -- it's just that some of your sales colleaguesh> aren't as candid in front of prospects, and seem to imply that6 TPC-C numbers really mean something useful to them ;).   -- h? <these messages express my own views, not those of my employer>l) Alexis Cousein				Senior Systems Engineer . SGI Belgium and Luxemburg		al@brussels.sgi.com   ------------------------------  % Date: Tue, 17 Jul 2001 19:36:22 +0200a* From: Alexis Cousein <al@brussels.sgi.com>  Subject: Re: IA64 Rocks My World/ Message-ID: <3B547796.8080203@brussels.sgi.com>o   Fred Kleinsorge wrote:  ? > Then explain to me why TPC-C numbers have any meaning at all.e    ; Marketing, inertia (TPC-C was designed quite some time ago, A when the numbers achieved weren't as absurdly high), the lack of  C something better if you can't be bothered to design your own tests.t  & The usual stuff with aging benchmarks.   > Or why OracleeE > would waste good money developing such a thing - if nobody uses it.e >     B OPS is used a lot, for other reasons (resiliency, amongst others),F and because scalability of an Oracle instance on some tasks is limited  / *whatever the underlying hardware or software*.p    A OTOH, on most architectures, and depending on the exact workload,2F that cutoff point (after which OPS really becomes necessary to achieveB any kind of scalability) is at CPU counts well into double digits.  ; And, of course, there *are* scenarios where OPS really does > make sense. What I'm saying is that many people for which such9 a setup doesn't make sense seem to look at TPC-C numbers.w   -- a? <these messages express my own views, not those of my employer>e) Alexis Cousein				Senior Systems Engineer1. SGI Belgium and Luxemburg		al@brussels.sgi.com   ------------------------------  % Date: Tue, 17 Jul 2001 13:35:38 -0400u+ From: Russell Crook <Russell.Crook@Sun.COM>h  Subject: Re: IA64 Rocks My World' Message-ID: <3B54776A.88CA9936@Sun.COM>I   Fred Kleinsorge wrote: > N > Then explain to me why TPC-C numbers have any meaning at all.  Or why OracleE > would waste good money developing such a thing - if nobody uses it.t  D I didn't know that the Transaction Performance Council was an Oracle
 front :->   ; Think of TPC benchmarks as "spec benchmarks for databases".m Unfortunately,F the relative tuning/configuration space for a database/computer systemG is very large, which leaves a lot of latitude in massaging the numbers.oA Also, a benchmark is by its very nature is a static target, whileP' customer applications evolve over time.h  C As for meaning... TPC-C started out IIRC as a replacement for TPC-A D and TPC-B, earlier transaction benchmarks. However, the transactions@ in these were very "lightweight", and not very representative of "real world" DB transactions.   H So TPC-C was created as a more representative benchmark.  Unfortunately,B like any benchmark, it has a finite lifetime, as benchmarkers look@ for ways to make their systems look good, sometimes even if thatD makes their benchmark configuration totally unrealistic for customer@ purposes (e.g., non-RAID disk, turn off the redo logs (!), usingC the "cluster cheat", etc.) You really have to dig into the complete E descriptions to find out what the vendor actually did, and then judgea/ if that makes the result a "benchmark special".s  B A similar thing happened to TPC-D, the data warehousing benchmark.> The early numbers probably correlated fairly well to DW system@ performance, but as time went on, "hacks" (as well as legitimateB performance improvements) eventually made the exercise useless forB customers. In particular, some of the "ad hoc" queries (which wereB intended to require significant database sweeps/scans) were turnedF into order(1) table lookups by precomputing all the necessary indexing: information when the database was loaded. Benchmark legal,G but now it's not really an "ad hoc" query anymore (you can't reasonablyn< precompute all possible combinations of indices), making the( benchmark numbers useless for customers.  = TPC-D is now retired, (with at least one company submitting arB "benchmark special" (as above) result just hours before the window3 closed to claim bragging rights - sigh); I'd expectt  TPC-C will follow at some point.     > H > Alexis Cousein wrote in message <3B546A8F.8090608@brussels.sgi.com>... > >Fred Kleinsorge wrote:P > >tB > >> OK.  I'm, not a database expert, but why is this a bad thing? > >  > >iF > >Because in real life, no-one will actually run a database this way,G > >because it lacks flexibility and is an administrative nightmare, andqH > >that you cannot predict a workload like you can in a TPC-C benchmark. > >DG > >At least that's what my totally unrepresentative sample of customersg3 > >(and some Sun customers I know) seem to tell me.  > >o > >--pB > ><these messages express my own views, not those of my employer>) > >Alexis Cousein Senior Systems Engineer(0 > >SGI Belgium and Luxemburg al@brussels.sgi.com > >    -- ,
 Russell Crook-B Not speaking officially for Sun (or anyone else, for that matter).  ; "If you think C++ is not overly complicated, just what is aM@ protected abstract virtual base pure virtual private destructor,G and when was the last time you needed one?" -- Tom Cargill, C++ Journaln   ------------------------------  % Date: Tue, 17 Jul 2001 13:59:18 -0400t5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>   Subject: Re: IA64 Rocks My World2 Message-ID: <U1%47.872$rc5.61019@news.cpqcorp.net>  = Russell Crook wrote in message <3B54776A.88CA9936@Sun.COM>...  >a >Fred Kleinsorge wrote:s >>H >> Then explain to me why TPC-C numbers have any meaning at all.  Or why OracleF >> would waste good money developing such a thing - if nobody uses it. >nE >I didn't know that the Transaction Performance Council was an Oraclee
 >front :-> >     K OK, OK, you knew what I meant...  Why would Oracle spend good money on OPS.s   ------------------------------  % Date: Tue, 17 Jul 2001 12:58:50 -0400 ; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>j# Subject: Re: Is there a WASD forum?-$ Message-ID: <3b546f17$1@news.si.com>  + Doug Mallory (dmallory@interlog.com) wrote:y  * >OSU will not work on VMS 5.5-2 on a vax,   $ Where did you get that silly idea??? -- dA Brian Tillman                   Internet: tillman_brian at si.compA Smiths Aerospace                          tillman at swdev.si.com = 3290 Patterson Ave. SE, MS      Addresses modified to prevent.< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Tue, 17 Jul 2001 08:12:31 -04009( From: Hamlyn Mootoo <univms@bigfoot.com>! Subject: Re: ISV's and VMS future + Message-ID: <3B542BAF.7051BB32@bigfoot.com>>  
 Koloth wrote:r > T > This is true if the only reason for using OpenVMS was for performance.  Now if youS > have the same performance as other systems OpenVMS gives you reliability, ....etcq > T > So the question to the ISV and compaq marketeers is with all else being equal, why/ > would you choose anything other than OpenVMS?u >  > Cass Witkowski > G The same reason your Chrysler salesman doesn't sell Dodge Darts anymoreRE - no market for it.  I've seen Dodge Darts with over 700,000 miles ondG them -same engine, and have a top speed higher than an Intrepid (mainly D becuase the Intrepid accelerator cuts out at about 110 MPH), but tryB selling a new one to someone with a pulse and see how far you get.   HM     > BK4Leg wrote:a > R > > The company I work for has several applications from software vendors, some onR > > VMS and some on UNIX.  I'm talking about applications, not tools or utilities. > >>P > > One rationale for the ISV's who use VMS has been the superior performance ofS > > the Alpha chip and VMS over its UNIX competition - esp Solaris and HP-UX.  With O > > the demise of Alpha, where is the incentive for a vendor to remain on VMS ?i > >aL > > Even more so for the vendors who have both UNIX and VMS flavors of theirQ > > application software, since without the performance advantage of Alpha, therepS > > is little reason to go to VMS.  They can cut costs either by delaying their VMSeN > > versions until well after their UNIX releases, or by eliminating their VMS > > support entirely.r > >nH > > It was nice to hope for a while that Compaq might bring new softwareP > > applications to VMS.  With the demise of the Alpha advantage, I can't expect: > > that many of the current vendors will remain with VMS. > >hS > > If Compaq is sucessful in bringing new applications to VMS, then they should ben > > widely publicizing that. > >w1 > > And BTW, when was the last VMS ad in Forbes ?l > > 
 > > Bernie   ------------------------------   Date: 17 Jul 2001 14:59:51 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)r! Subject: Re: ISV's and VMS futurec+ Message-ID: <9j1jt7$120$2@info.cs.uofs.edu>k  , In article <3B53D0B5.A2BAB07C@telocity.com>,%  Koloth <koloth@telocity.com> writes: U |> This is true if the only reason for using OpenVMS was for performance.  Now if youeT |> have the same performance as other systems OpenVMS gives you reliability, ....etc  K No matter how much peopl ehere try to deny it, other OSes and even Unix aretM more than reliable enough to meet the needs of most businesses.  How reliable I do you need??  Having a machine that that can maintain 18 years of Uptime I does me no good as they shut the power off for one day every year.  WhilekL there are places that can not survive with out 24x365 operation, I think theK majority can live with less and when they factor in the uncertainty that isa? now filling the VMS world, well, we know who is going to loose.p   |> ,U |> So the question to the ISV and compaq marketeers is with all else being equal, why 0 |> would you choose anything other than OpenVMS?  ( Uncertainty about the future of OpenVMS.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   e   ------------------------------  # Date: Tue, 17 Jul 2001 16:53:12 GMT14 From: "Terry C. Shannon" <terryshannon@mediaone.net>! Subject: Re: ISV's and VMS future : Message-ID: <Y3_47.403$N21.410970@typhoon.ne.mediaone.net>  > "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message% news:9j1jt7$120$2@info.cs.uofs.edu...e. > In article <3B53D0B5.A2BAB07C@telocity.com>,' >  Koloth <koloth@telocity.com> writes:i >i > |>L > |> So the question to the ISV and compaq marketeers is with all else being
 equal, why2 > |> would you choose anything other than OpenVMS? >n* > Uncertainty about the future of OpenVMS. >u  G That sums it up pretty nicely. I wonder if Compaq is A) monitoring thisn1 newsgroup and B) will profit from the experience.	   ------------------------------  # Date: Tue, 17 Jul 2001 10:38:23 GMTm' From: Colin Blake <colin@theblakes.com>e/ Subject: Re: looping to delete a directory treel- Message-ID: <3B541552.DFB75E17@theblakes.com>c  / Try DFU. I think it comes on the freeware disk.r   ------------------------------    Date: 17 Jul 2001 11:47:36 -05005 From: carmichael@encompasserve.org (Henry Carmichael)p/ Subject: Re: looping to delete a directory treee3 Message-ID: <hPr+IcmP3pI3@eisner.encompasserve.org>r  c In article <14ce1c21.0107161703.1eb1a528@posting.google.com>, paszty@xoma.com (Nick Paszty) writes:w > hello. > C > i can't seem to delete an entire directory tree if there is stuffTF > (subdirectories or files) in that tree.  here is a tree for example.> > [sasdev.test.testsub.testsubsub]. while at the sasdev level,A > i've tried del/log [.test...]*.*;* but it only deletes the lastnF > directory in the test tree - only testsubsub is deleted.  the systemE > returns an error message saying that testsub was not empty and theneH > goes on to delete testsubsub. i have to recall the command to continueE > deleting up the tree.  i would like to continue to loop the del/logiH > [.test...]*.*;* command until i have wiped out all the directories andD > their contents. is there a system return code i can test to for toF > continue looping and deleting up the tree?  is there a better way to& > perform this simple management task? > 	 > thanks,. >  > nick   Nick:-   please try the following:   J $ del/log [.test...]*.*;*, *.*;*, *.*;*, *.*;*, *.*;*, *.*;*, *.*;*, *.*;*  C The wildcard sequence is repeated eight times to take care of eightnJ possible sublevels.  It should get all the files from all the directories / under test, and all the subdirectories as well.l   Henry T.   ------------------------------  % Date: Tue, 17 Jul 2001 12:32:02 -04008; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com> / Subject: Re: looping to delete a directory treei$ Message-ID: <3b5468d0$1@news.si.com>  B >i can't seem to delete an entire directory tree if there is stuffE >(subdirectories or files) in that tree.  here is a tree for example.n= >[sasdev.test.testsub.testsubsub]. while at the sasdev level,A@ >i've tried del/log [.test...]*.*;* but it only deletes the last; >directory in the test tree - only testsubsub is deleted.  -   This will work:s  # $ set prot=w=d test.dir;[.test...]*e7 $ del [.test...]*.*;*,;*,;*,;*,;*,;*,;*,;*, []test.dir;B -- mA Brian Tillman                   Internet: tillman_brian at si.com-A Smiths Aerospace                          tillman at swdev.si.com"= 3290 Patterson Ave. SE, MS      Addresses modified to prevente< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Mon, 16 Jul 2001 19:11:14 -0700o4 From: Roger Wiechman <Roger_Wiechman@HMC.nospam.Edu>: Subject: Need help with data recovery on failed volume set. Message-ID: <3B539EC2.A01D8D3B@HMC.nospam.Edu>  
 Hello all,  D  I have a volume set of 2 DSP3105S 1GB DEC drives.  Logical volume 1@  has failed.  It spins up sporadically and never loads the heads	 properly.b/  I even swapped drive electronics, to no avail.r  D  Has anyone had any experience with data/drive recovery outfits that
 understand-  the VMS directory structure?  Are there any?   E  Lacking that, what approach could be taken to retrieve data from thet goodG  drive, lacking the master volume?  I have already made a physical copyN of it.  )  Thanks in advance for any and all ideas.o            Roger   ------------------------------  / Date: Tue, 17 Jul 2001 08:39:44 +0200 (MET DST) & From: Rudolf Wingert <win@fom.fgan.de> Subject: No chance for OpenVMS6 Message-ID: <200107170639.IAA19447@sinet1.fom.fgan.de>   Hello,  E today I red, that Andrew Butler of the Gardner Group gives OpenVMS nomC chance to be alive. He wrote, that the "midrange OS VMS will be theaE next, what Compaq will kill. By pressing the user to migrate to a newdF one platform, the installation base will shrink and the user will portG to an other OS. Also the software developer will not see a great market < and in case of this not port there software to OpenVMS IA64.   Regards Rudolf Wingert   ------------------------------  % Date: Tue, 17 Jul 2001 03:21:52 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>t" Subject: Re: No chance for OpenVMS, Message-ID: <3B53E787.52AD90A0@videotron.ca>   Rudolf Wingert wrote:oG > today I red, that Andrew Butler of the Gardner Group gives OpenVMS noa > chance to be alive.s  N Isn't it frustrating when the folks you argued were wrong all these years turn out to be correct after all ?c  H Obviously, Compaq was able to say something that made some VMS loyalistsD totally happy with the situation.  Mind you, if I lived near the VMSN engineers, I might be saying great things about Compaq in the hopes of getting one last VMS job.   N Actions speak louder than words. Compaq may have bribed its key customers, butM the effect will be temporary and I think that those customers will eventuallytL come to the same complusion as Gartner unless Compaq takes some real actions to push VMS.   ------------------------------  % Date: Tue, 17 Jul 2001 10:11:34 -0400g" From: "Hal Kuff" <kuff@tessco.com>" Subject: Re: No chance for OpenVMSO Message-ID: <F5961F121686280E.85D17306A35EE805.6AA2ED1C32FB4FC0@lp.airnews.net>a  " Anyone have a URL on this release?    3 "Rudolf Wingert" <win@fom.fgan.de> wrote in messaget0 news:200107170639.IAA19447@sinet1.fom.fgan.de... > Hello, >hG > today I red, that Andrew Butler of the Gardner Group gives OpenVMS nosE > chance to be alive. He wrote, that the "midrange OS VMS will be thesG > next, what Compaq will kill. By pressing the user to migrate to a new1H > one platform, the installation base will shrink and the user will portI > to an other OS. Also the software developer will not see a great marketd> > and in case of this not port there software to OpenVMS IA64. >n > Regards Rudolf Wingert >t   ------------------------------  % Date: Tue, 17 Jul 2001 10:41:08 -0400h5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>n" Subject: Re: No chance for OpenVMS2 Message-ID: <%7Y47.832$rc5.60862@news.cpqcorp.net>  L You know, when I first started at DEC some 22 years ago, a truly incompetentJ Unit Manager (in the New York District office) left to be a analyst there.F I've always taken everything they say with a grain of salt since then.  K But if everyone starts chanting that VMS is dead, we can make it happen.  I J tend to believe that customers on VMS today, are here because they need to@ be.  Those needs will still be here, and VMS will still be here.       Rudolf Wingert wrote in messageu- <200107170639.IAA19447@sinet1.fom.fgan.de>...  >Hello,e >cF >today I red, that Andrew Butler of the Gardner Group gives OpenVMS noD >chance to be alive. He wrote, that the "midrange OS VMS will be theF >next, what Compaq will kill. By pressing the user to migrate to a newG >one platform, the installation base will shrink and the user will portnH >to an other OS. Also the software developer will not see a great market= >and in case of this not port there software to OpenVMS IA64.l >o >Regards Rudolf Wingert: >6   ------------------------------  % Date: Tue, 17 Jul 2001 10:42:14 -0400g5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>v" Subject: Re: No chance for OpenVMS2 Message-ID: <29Y47.833$rc5.60629@news.cpqcorp.net>  = JF Mezei wrote in message <3B53E787.52AD90A0@videotron.ca>...m  K >Actions speak louder than words. Compaq may have bribed its key customers,h but C >the effect will be temporary and I think that those customers willp
 eventuallyE >come to the same complusion as Gartner unless Compaq takes some realn actionsg
 >to push VMS.     ) Uh, I don't think anyone has been bribed.a   ------------------------------  % Date: Tue, 17 Jul 2001 11:34:42 -0300i) From: fabio_compaq@ep-bc.petrobras.com.brn" Subject: Re: No chance for OpenVMSL Message-ID: <OFA0747654.06903497-ON03256A8C.004FE483@ep-bc.petrobras.com.br>  J It is so obvius that Compaq will not port OpenVMS to Itanium ..... and the	 customersd5 will migrate to Solaris or HPUX ..... not Tru64 ! ! !i  H Well, at least if the decisions will be made in the techincal level, not CEO level !s  G If the decisions will be in the CEO (Capellas) x CEO (customers) level,oD so everybody will port their applications  to WNT 64 bits (Itanium).    , It will be Alpha forever (until 2014 ......)     Regardsr   FC        3 "Hal Kuff" <kuff@tessco.com> em 17/07/2001 11:11:34   . Favor responder a "Hal Kuff" <kuff@tessco.com>             Info-VAX@Mvb.Saic.Coma      " Assunto: Re: No chance for OpenVMS      " Anyone have a URL on this release?    3 "Rudolf Wingert" <win@fom.fgan.de> wrote in messagee0 news:200107170639.IAA19447@sinet1.fom.fgan.de... > Hello, > G > today I red, that Andrew Butler of the Gardner Group gives OpenVMS nomE > chance to be alive. He wrote, that the "midrange OS VMS will be thetG > next, what Compaq will kill. By pressing the user to migrate to a newaH > one platform, the installation base will shrink and the user will portI > to an other OS. Also the software developer will not see a great marketc> > and in case of this not port there software to OpenVMS IA64. >b > Regards Rudolf Wingert >t   ------------------------------  % Date: Tue, 17 Jul 2001 12:13:45 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>s" Subject: Re: No chance for OpenVMS, Message-ID: <3B546436.D2833506@videotron.ca>  * fabio_compaq@ep-bc.petrobras.com.br wrote: > L > It is so obvius that Compaq will not port OpenVMS to Itanium ..... and the > customers.7 > will migrate to Solaris or HPUX ..... not Tru64 ! ! !d  L I have every confidence that VMS will be ported to IA64 and will run on IA64N with only minor differences at the user level. The engineers will overcome the technical problems.a  H The timescale where Compaq would officially pull the plug on VMS is wellI beyond the porting timeline. Compaq seems to have done a very good job ofvF convincing the small group of very profitable customers to have a veryM positive attitude towards this issue so it is a fair bet that they will stickp with VMS for now.o  F However, my opinion is that few of these will bother with Alpha-VMS toE IA64-VMS. They will probably just continue to add processors to theiroM wildfires. (Besides, I doubt that IA64-based wildfires will be available very,N early). New applications will go on other platforms, and this is logical since6 new applications will be available on other platforms.   ------------------------------    Date: 17 Jul 2001 18:10:30 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)" Subject: Re: No chance for OpenVMS* Message-ID: <3b546376$1@news.kapsch.co.at>  j In article <%7Y47.832$rc5.60862@news.cpqcorp.net>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes:L >But if everyone starts chanting that VMS is dead, we can make it happen.  IK >tend to believe that customers on VMS today, are here because they need tosA >be.  Those needs will still be here, and VMS will still be here.a  + I hope, you make this a "want to be there".tE If a "need - but don't want - to be there" is used instead, it means,g1 if the 'need' is gone, the customer is gone, too.- And so the base is shrinking...5   --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888$< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  # Date: Tue, 17 Jul 2001 16:54:33 GMTa4 From: "Terry C. Shannon" <terryshannon@mediaone.net>" Subject: Re: No chance for OpenVMS: Message-ID: <d5_47.405$N21.411520@typhoon.ne.mediaone.net>  3 "Rudolf Wingert" <win@fom.fgan.de> wrote in message"0 news:200107170639.IAA19447@sinet1.fom.fgan.de... > Hello, >-G > today I red, that Andrew Butler of the Gardner Group gives OpenVMS norE > chance to be alive. He wrote, that the "midrange OS VMS will be therG > next, what Compaq will kill. By pressing the user to migrate to a new H > one platform, the installation base will shrink and the user will portI > to an other OS. Also the software developer will not see a great marketl> > and in case of this not port there software to OpenVMS IA64. >k  G Something is missing from this assessment... the obligatory PROBABILITY  FACTOR.V   ------------------------------    Date: 17 Jul 2001 19:18:42 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)" Subject: Re: No chance for OpenVMS* Message-ID: <3b547372$1@news.kapsch.co.at>  q In article <d5_47.405$N21.411520@typhoon.ne.mediaone.net>, "Terry C. Shannon" <terryshannon@mediaone.net> writes:,e >"Rudolf Wingert" <win@fom.fgan.de> wrote in message news:200107170639.IAA19447@sinet1.fom.fgan.de...aH >> today I red, that Andrew Butler of the Gardner Group gives OpenVMS noF >> chance to be alive. He wrote, that the "midrange OS VMS will be theH >> next, what Compaq will kill. By pressing the user to migrate to a newI >> one platform, the installation base will shrink and the user will porttJ >> to an other OS. Also the software developer will not see a great market? >> and in case of this not port there software to OpenVMS IA64.  >vH >Something is missing from this assessment... the obligatory PROBABILITY >FACTOR.  . Yup. I read such an gartner article years ago.; There was an comment on "DEC will drop the Alpha Chip" withy= a probability of 20%. They didn't put it the other way aroundlC ("DEC will continue with the Alpha Chip" with a probability of 80%)pK because that was not the message they wanted the customers to read/believe.r   --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888n< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------    Date: 17 Jul 2001 13:46:31 -0500+ From: young_r@encompasserve.org (Rob Young)-" Subject: Re: No chance for OpenVMS3 Message-ID: <h4HE99O07$P3@eisner.encompasserve.org>n  q In article <d5_47.405$N21.411520@typhoon.ne.mediaone.net>, "Terry C. Shannon" <terryshannon@mediaone.net> writes:@ > 5 > "Rudolf Wingert" <win@fom.fgan.de> wrote in messageT2 > news:200107170639.IAA19447@sinet1.fom.fgan.de...	 >> Hello,m >>H >> today I red, that Andrew Butler of the Gardner Group gives OpenVMS noF >> chance to be alive. He wrote, that the "midrange OS VMS will be theH >> next, what Compaq will kill. By pressing the user to migrate to a newI >> one platform, the installation base will shrink and the user will porthJ >> to an other OS. Also the software developer will not see a great market? >> and in case of this not port there software to OpenVMS IA64.a >> > I > Something is missing from this assessment... the obligatory PROBABILITYn	 > FACTOR.  >   ; 	Which one?  The one they penned in a 1995 timeframe about ,? 	VMS's installed base circa 2000?  The one that Wes bought intoh> 	and the license count would therefore decline to 200K and yet@ 	has held in the 400Ks, hence the starting point for a wonderful> 	Affinity campaign?   Maybe if Gartner states it often enough	A 	and puts it far enough into the future they get it right.  But IaD 	somehow doubt it.  Shoot, they have been pressing people to migrateC 	for quite some time.  Unfortunately (or fortunately as I see it), l> 	VMS does some things very well that others aren't quite doingC 	yet (long distant clusters, for instance).  "Yeah, but just wait."n@ 	Sure... and listening to the conversation over the locked up DB4 	environment . . . "yer gonna get what you deserve."   				Robr   ------------------------------  # Date: Tue, 17 Jul 2001 06:19:04 GMT. From: Dirk Munk <munk@home.nl>K Subject: Re: OpenVMS on IPF (was re: IPF already needs a face-lift for VMS)S' Message-ID: <3B53D8D7.57317363@home.nl>e   Hoff Hoffman wrote:b  r > In article <009FF17F.ED173243@SendSpamHere.ORG>, system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes: > M > :What is to happen, for example, with something as fundamental as queue in-sM > :sertion or removal?  I'm sure something *this* basic has been tossed about-! > :in discussions of this effort.  >mK >   Yes, we have been discussing the implementations that are available fordL >   those interlocked primitives that do not have direct analogs within IPF. >kM >   OpenVMS Alpha, of course, has a PAL-based version of the old VAX hardwareJE >   interlocked instructions, so this is nothing new nor particularlycG >   ground-breaking -- the compilers and the RTL provide access to thisrI >   capability, and I would expect that stuff using this will continue to1 >   operate. >lJ >   For benefit of those that have just joined this IPF and the gazzillionK >   other IPF discussions in progress, Alpha PALcode involves a sequence offK >   Alpha hardware instructions that is loaded into the system by the AlphasI >   console, and that is protected against interrupts.  The IPF analog istH >   arguably the SAL -- IPF has two things called PAL, but these are not; >   as direct an analog to the Alpha PALcode as is the SAL.h >tH >   Whether the particular implementation of these interlocked sequencesI >   ends up in the compilers, in the RTL, in the SAL layer, in some othernG >   component or layer elsewhere, or some combination of these, has noto >   yet been decided.a > A > :Just what things trouble you so that you wish to *break* them?o > K >   The internal implementation of the TDF and of time-keeping would be onerL >   such area -- I'd like to provide a central time-keeping kernel-mode API,G >   switch to UTC for all internal operations, keep a per-process and a.H >   system-wide TDF, while (of course) continuing to present the classicK >   VAX environment for all existing user-mode calls to $bintim and similar I >   calls.  (I have NO idea if I will be able to get this change designedeL >   and implemented into the IPF release, of course.)  There are other areasK >   for enhancements seeing various (and preliminary) discussions, as well.R  m I agree. The subject of good timestamps is one of the most underestimated subjects. How many database and and r program designs will use a UTC based time stamp at the moment ? Just the trouble we have adjusting the system timep when daylight saving time changes occur. Special routines have been designed to "slow down" the time in order tom prevent a change from 03.00 hrs to 02.00 hrs.  Timestamps like they are in use for OSI would be nice I guess.ir And now we are on this subject, Maybe the lexical function F$TIME could be changed a bit ? How about F$TIME("UTC") etc. ?   Regards,   Dirk           >r >aP >  ---------------------------- #include <rtfaq.h> -----------------------------L >       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comP >  --------------------------- pure personal opinion ---------------------------N >    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Tue, 17 Jul 2001 05:42:23 -0400r) From: "Neil Rieck" <n.rieck@sympatico.ca>3K Subject: Re: OpenVMS on IPF (was re: IPF already needs a face-lift for VMS)S; Message-ID: <vMT47.13286$oh4.1234738@news20.bellglobal.com>i  ? "Hoff Hoffman" <hoffman@xdelta.zko.dec.nospam> wrote in messagew, news:i0G47.763$rc5.60120@news.cpqcorp.net...J > In article <009FF17F.ED173243@SendSpamHere.ORG>, system@SendSpamHere.ORG' (Brian Schenkenberger, VAXman-) writes:e >y [snip] >iK >   The internal implementation of the TDF and of time-keeping would be oneeL >   such area -- I'd like to provide a central time-keeping kernel-mode API,G >   switch to UTC for all internal operations, keep a per-process and a-H >   system-wide TDF, while (of course) continuing to present the classicK >   VAX environment for all existing user-mode calls to $bintim and similargI >   calls.  (I have NO idea if I will be able to get this change designedtL >   and implemented into the IPF release, of course.)  There are other areasK >   for enhancements seeing various (and preliminary) discussions, as well..  H When this came up last year I was lead to believe that OpenVMS engineersC were already working on a plan to change all internal clocks to UTCwK (possibly related to DII COE?). Has this change been put on the back burnere or was I mistaken?  
 Neil Rieck Kitchener/Waterloo/Cambridge,- Ontario, Canada.! http://www3.sympatico.ca/n.rieck/5   ------------------------------  # Date: Tue, 17 Jul 2001 12:55:45 GMT = From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)>K Subject: Re: OpenVMS on IPF (was re: IPF already needs a face-lift for VMS)G0 Message-ID: <009FF226.BADE2A56@SendSpamHere.ORG>  j In article <gMH47.782$rc5.60551@news.cpqcorp.net>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes: > / >Brian Schenkenberger, VAXman- wrote in messagem >>J >>Details on SAL and its being an equivalent of Alpha PAL are, at present,J >>sketchy.  Is there another source I can read other that the IPF ARM that& >>will describe SAL is greater detail? >> >  >a) >http://developer.intel.com/design/ia-64/  > C >On the left edge, go to firmware specification, and then to Systems >Abstraction Layer.t >l > G >PAL/SAL provides some of the things done in PAL (like handling machineoM >checks) - but it doesn't provide any of the programmable instruction logic -:A >that is, it isn't a opcode trap and a dispatch to a PAL routine.h  I That's what I was reading and I read that which you are describing above.4I From one of your former posts, I read it as if there was much more to it.g   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM.             J   "And of course, I'm a genius, so people are naturally drawn to my fiery I   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbes    ------------------------------  % Date: Tue, 17 Jul 2001 10:35:21 -0400n5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>sK Subject: Re: OpenVMS on IPF (was re: IPF already needs a face-lift for VMS)e2 Message-ID: <C2Y47.831$rc5.60831@news.cpqcorp.net>   Neil Rieck wrote in message ...d >u >fI >When this came up last year I was lead to believe that OpenVMS engineers:D >were already working on a plan to change all internal clocks to UTCL >(possibly related to DII COE?). Has this change been put on the back burner >or was I mistaken?  >n    K COE doesn't really need this right now.  All it needs is for the time to betK consistant.  But fixing time issues is on the list of projects that we drewvH up that needs to be addressed.  I think the concensus is that VMS should6 keep time in GMT, and convert to local time as needed.   ------------------------------  # Date: Tue, 17 Jul 2001 11:26:18 GMT 2 From: Piyush Avichal<pa@it.singer-friedlander.com>" Subject: Optical Drive for OpenVMS7 Message-ID: <uhV47.22010$Kf3.280158@www.newsranger.com>g   Hello,  I Does anyone know when the 9.1GB software for the USDESIGNS QT9100(C1114R)pN optical drive will be available. I'm not getting much sense from USDesigns. AtO the moment it is only capable of using 4.8GB media. I was told the software forfB 9.1GB media would be out in the summer but I still havent seen it.   Regards,   Piyush.o   ------------------------------  % Date: Tue, 17 Jul 2001 10:31:42 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>i Subject: Re: OT: Dr Who.2 Message-ID: <a%X47.830$rc5.60857@news.cpqcorp.net>  J Actually, the last couple of Doctors were on the lame side.  Even if thereJ were a couple of interesting episodes in there.  Here in the US, Tom Baker* was what made Dr. Who such a cult success.  H My favorite BBC show that I only saw a single run here in the states was7 something like "The Rise and Fall of Reginald Perrine".r  E The BBC show I would like to see continued forever is the Black Adder , series.  Mr. Bean sucked, Black Adder ruled.       Zane H. Healy wrote in message1 <1qO47.8$1w.62549@e3500-chi1.usenetserver.com>...-4 >Robert Deininger <rdeininger@mindspring.com> wrote:D >> That's good.  Stopping Dr. Who was as dumb as ... stopping alpha. > L >Especially when you consider how little they must of spent on most of thoseH >episodes, and how widely it was distributed.  I think the only BBC show thatK >might have been distributed more, might be Eastenders, and I know my local + >PBS station has never shown Eastenders :^(l > F >Personally I think stopping Dr. Who was even dumber than stopping the Alpha.K >On an interesting note, the Dr. Who Magazine that Marvel Comics produce iniL >England has somehow managed to keep going all these years, and the 'furtherH >adventures' books seem to be highly popular (with most of the ones I've read >being quite good).p >e > Zane >f   ------------------------------  # Date: Tue, 17 Jul 2001 16:09:18 GMTR From: dittman@dittman.net  Subject: Re: OT: Dr Who.@ Message-ID: <OqZ47.11695$sx2.583593@e420r-atl3.usenetserver.com>  4 Fred Kleinsorge <kleinsorge@star.zko.dec.com> wrote:L : Actually, the last couple of Doctors were on the lame side.  Even if thereL : were a couple of interesting episodes in there.  Here in the US, Tom Baker, : was what made Dr. Who such a cult success.  J : My favorite BBC show that I only saw a single run here in the states was9 : something like "The Rise and Fall of Reginald Perrine".w  D Our local PBS station shows "The Rise..." every so often.  We get anE infrequent showing of the good series mixed in with never-ending runs % of crap like "Are You Being Served?".   G : The BBC show I would like to see continued forever is the Black Adder.. : series.  Mr. Bean sucked, Black Adder ruled.  D I thought Mr. Bean was okay, but I do prefer the Black Adder series. --   Eric Dittman dittman@dittman.netC   ------------------------------  % Date: Tue, 17 Jul 2001 12:27:32 -0400m2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: OT: Dr Who.L Message-ID: <rdeininger-1707011227320001@user-2iveacr.dialup.mindspring.com>  D In article <a%X47.830$rc5.60857@news.cpqcorp.net>, "Fred Kleinsorge"$ <kleinsorge@star.zko.dec.com> wrote:  L > Actually, the last couple of Doctors were on the lame side.  Even if thereL > were a couple of interesting episodes in there.  Here in the US, Tom Baker, > was what made Dr. Who such a cult success.  F Yes, he was the best one.  I didn't mind Peter Davison.  The last one  never seemed quite right.p  oJ > My favorite BBC show that I only saw a single run here in the states was9 > something like "The Rise and Fall of Reginald Perrine"..  G I've never heard of that one.  How about "A Very Peculiar Practice"?  I)B think you would appreciate the immortal phrase "rinky dinky little
 computer".  G > The BBC show I would like to see continued forever is the Black Adder-. > series.  Mr. Bean sucked, Black Adder ruled.  & There's no substitute for Black Adder!   -- R Robert Deininger rdeininger@mindspring.com7   ------------------------------  % Date: Tue, 17 Jul 2001 16:41:10 +0100,% From: Alan Greig <a.greig@virgin.net>. Subject: Re: OT: Dr Who.8 Message-ID: <fhm8ltccmrcu5p3tslt95gvtb5q5tc1u95@4ax.com>  5 On Tue, 17 Jul 2001 10:31:42 -0400, "Fred Kleinsorge" $ <kleinsorge@star.zko.dec.com> wrote:  K >Actually, the last couple of Doctors were on the lame side.  Even if thereeK >were a couple of interesting episodes in there.  Here in the US, Tom Bakers+ >was what made Dr. Who such a cult success.,  , Tends to be the favourite in the UK as well.   >nI >My favorite BBC show that I only saw a single run here in the states was68 >something like "The Rise and Fall of Reginald Perrine".  E Nearly. "The Fall and Rise of Reginald Perrin". You can see the title1 sequence again at > http://www.bbc.co.uk/cult/ilove/tv/titles/reginaldperrin.shtml    F >The BBC show I would like to see continued forever is the Black Adder- >series.  Mr. Bean sucked, Black Adder ruled.R  D The BBC made a special millennium episode set at Y2K which was shownB in the Millennium Dome then later on tv. It's not very good thoughD compared to the originals. Probably turn up in the US at some point.   >  >, >a >Zane H. Healy wrote in messageo2 ><1qO47.8$1w.62549@e3500-chi1.usenetserver.com>...5 >>Robert Deininger <rdeininger@mindspring.com> wrote:CE >>> That's good.  Stopping Dr. Who was as dumb as ... stopping alpha.o >>M >>Especially when you consider how little they must of spent on most of thoseoI >>episodes, and how widely it was distributed.  I think the only BBC show  >thattL >>might have been distributed more, might be Eastenders, and I know my local, >>PBS station has never shown Eastenders :^( >>G >>Personally I think stopping Dr. Who was even dumber than stopping thep >Alpha.hL >>On an interesting note, the Dr. Who Magazine that Marvel Comics produce inM >>England has somehow managed to keep going all these years, and the 'furtheroI >>adventures' books seem to be highly popular (with most of the ones I'veW >read  >>being quite good). >> >> Zanen >> >1   -- Alan   ------------------------------  % Date: Tue, 17 Jul 2001 13:15:18 -0400d5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>r Subject: Re: OT: Dr Who.2 Message-ID: <yo_47.863$rc5.60601@news.cpqcorp.net>  % Robert Deininger wrote in message ...>E >In article <a%X47.830$rc5.60857@news.cpqcorp.net>, "Fred Kleinsorge"t% ><kleinsorge@star.zko.dec.com> wrote:d >lG >> Actually, the last couple of Doctors were on the lame side.  Even ifo theretG >> were a couple of interesting episodes in there.  Here in the US, Tomb Bakery- >> was what made Dr. Who such a cult success.a >eF >Yes, he was the best one.  I didn't mind Peter Davison.  The last one >never seemed quite right. >%    K Agreed.  A few of the PD episodes (like the escher-like one, can't rememberi# the name offhand) were pretty good.s  K >> My favorite BBC show that I only saw a single run here in the states was : >> something like "The Rise and Fall of Reginald Perrine". >eH >I've never heard of that one.  How about "A Very Peculiar Practice"?  IC >think you would appreciate the immortal phrase "rinky dinky littleA >computer".  >m    L Never heard of that one.  The Reginald Perrin one was about a guy who worked? at an ice cream factory, who decides that he can't take it, and-H "disappears".  Eventually, he ends up creating something called a "grot"G shop - a store that carries only completely useless items, that becomesnI wildly popular.  I was hooked from the moment he see's a hippo every time  his mother-in-law is mentioned.i  H >> The BBC show I would like to see continued forever is the Black Adder/ >> series.  Mr. Bean sucked, Black Adder ruled.d > ' >There's no substitute for Black Adder!l >.    6 Some of my favorite British shows, seen by way of PBS-  : Monty Python (of course, just got the complete 14 DVD set)K A bit of Fry and Lawrie (Stephen Fry and Hugh Lawrie are really funny guys) 4 Dr. Who - excluding the awful first and last Doctors1 The Fall (and subsequent Rise) of Reginald Perrinr Black Addera The Prisoner6 To the Manor Born (well, it was cute the *first* time)   ------------------------------  % Date: Tue, 17 Jul 2001 08:37:48 -0400t3 From: "Brad McCusker" <Brad.McCuskerSP@Mcompaq.com>l7 Subject: Re: Pathworks 6.0D on OpenVMS/vax 7.3 questiont2 Message-ID: <1nW47.814$rc5.60765@news.cpqcorp.net>  H I concur with the licensing issue.   Further, it sounds like this is the only PATHWORKSJ server in the network, so, you probably want to use server based licensing (Don't run the license server).   Regards,   Brad     --( The opinions expressed herein are my own' and do not reflect those of my employerr or anyone else.    BradI "Jean-Francois Marchal" <jean-francois.marchal@x9000.fr> wrote in messageo( news:9ir86o$noc$1@reader1.imaginet.fr...% > This may be a licensing problem ...iC > Check your license on the PC (You can rename the pwliclm.dat file @ > to force PWRK to acquire a new license). You should also checkC > with admin/license ... some licenses might be improperly assignedn# > to users that don't exist anymoreP >w > Jean-Franoisy >eI > "Freddy Meerwaldt" <frederik@freddym.org> a crit dans le message news:H@ > Pine.LNX.4.21.0107132031530.918-100000@firewall.freddym.org... > > Hi!@ > >pI > > I'm not often replying to my own post - but I have to (in this case).n > > I > > > I've recently installed Pathworks 6.0D on a VAXstation 3100 running? > OpenVMS/vax 7.3.F > > > I want to use it as a primary domain contoller in a high school.I > > > I've set up the server properly(???) but every time I try to log onl fromA > a Windows 98 machine into the Domain I get the following error:w > > >P: > > > Network Device Type is incorrect - please try again. > >t. > > I know (at least) _where_ this comes from.J > > I have defined a logon script (add user xyz /script=xyz.bat), which he is > > trying to execute.F > > If I remove the script (/noscript), I can log in without problems.F > > But I want to connect to the Network Share on the VAX I can't (the Networkn/ > > Device Type is incorrect problem persists).cB > > I've tried connecting from a Linux box but had no luck either.0 > > But on the Linux box I got more information:L > > - The error mentioned above means, that I'm trying to mount a file share% > > as a printer share or vice versa.- > > L > > I've tried creating my own share, but I can neither get to my own share, > > nor to the NETLOGON share. > > E > > So it seems the problem is related to sharing - not to the Domain  > > Controlling... > > Many thanks in advance,L
 > > Freddy > >t > > --L > > Geek Code 3.1: GCS s+: a--- C+++ UBOU+++ P-- E--- W++ N w--- V++ PGP- t? > 5? tvc > >a > > J ==========================================================================H > >  Frederik Meerwaldt  ICQ: 83045387  Homepage: http://www.freddym.orgG > >  Bavaria/Germany              OpenVMS and Unix Howtos and much morePH > >  Solaris, HP/UX, AIX, NetBSD, OpenBSD, IRIX, Tru64, OpenVMS, Ultrix, BeOS > >t >w >    ------------------------------  % Date: Tue, 17 Jul 2001 16:34:41 +0100n* From: Andrew Robinson <arobinson@hspg.com> Subject: Pathworks QuestionIM Message-ID: <CDA4BAD1E10ED41181AC00508B6051D3C3E2D0@grumpy.internal.hspg.com>e  6 Please could someone help me with a pathworks problem.  H I Have pathworks V6.0B running on an Alpha 1200 with VMS 7.2-1 patch sys# v9.0 & TCP 5.0A  patch ec0 v5.0-111m  I The problem I have is that NET$BIOS process has tied itself to the second L ethernet adaptor, this mean's that it produces a 350000 block log file every? month. How can I get it to attach to the primary ethernet card?    Netbios = device EWA7u/ all other processes = devices EWB6 up to EWB12)i  8 The EWA adaptor is a DE500 on a KZPCM-DA SCSI controller6 The EWB adaptor is a DE500 on a DE500-BA ethernet card  L I did a search and found a reference to a logical PWRK$NETBIOS$DEVICE but it didn't say how to use it. J To solve the log file problem, I tried to disable NETBIOS in the pathworks, config, but this didn't make any difference.   Thank you in advance   Andrew Robinsonl arobinson@hspg.com  9 "A little Knowledge is dangerous and in my hands lethal!"-   ------------------------------    Date: 17 Jul 2001 18:44:33 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER) Subject: Re: Pathworks Question-* Message-ID: <3b546b71$1@news.kapsch.co.at>  z In article <CDA4BAD1E10ED41181AC00508B6051D3C3E2D0@grumpy.internal.hspg.com>, Andrew Robinson <arobinson@hspg.com> writes:7 >Please could someone help me with a pathworks problem.2  I Consider using the vmsnet.networks.desktop.pathworks group in addition...   I >I Have pathworks V6.0B running on an Alpha 1200 with VMS 7.2-1 patch sysF$ >v9.0 & TCP 5.0A  patch ec0 v5.0-111  M There are a lot more ECOs for OpenVMS V7.2-1 and there is TCPIP V5.1 (ECO 1).  Consider them anyway...f  J >The problem I have is that NET$BIOS process has tied itself to the secondM >ethernet adaptor, this mean's that it produces a 350000 block log file everyo@ >month. How can I get it to attach to the primary ethernet card?  $ With a supported/documented logical.5 And you can check its functionality with SDA>SHOW LANl   >Netbios = device EWA70 >all other processes = devices EWB6 up to EWB12) >t9 >The EWA adaptor is a DE500 on a KZPCM-DA SCSI controllerl7 >The EWB adaptor is a DE500 on a DE500-BA ethernet cardm >sM >I did a search and found a reference to a logical PWRK$NETBIOS$DEVICE but ita >didn't say how to use it.    : 	$ DEFINE/SYSTEM	NETBIOS$DEVICE		EWB0:	!for NB over DECnet? 	$ DEFINE/SYSTEM	PWRK$KNBDAEMON_DEVICE	EWB0:	!for NB over TCPIPa7 	$ DEFINE/SYSTEM PWRK$NETBEUI_DEVICE	EWB0:	!for NetBEUIa  $ see chapter 3.11.1 (pg. 3-33) of theG "AdvancedServerfor OpenVMS/Server Installation and Configuration Guide"-  K >To solve the log file problem, I tried to disable NETBIOS in the pathworksS- >config, but this didn't make any difference.>  H What did you do ? Disable DECnet in ADM/CONF ? Restart Server ? Reboot ?   -- s< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888s< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Tue, 17 Jul 2001 18:31:52 +0100d* From: Andrew Robinson <arobinson@hspg.com> Subject: RE: Pathworks Question M Message-ID: <CDA4BAD1E10ED41181AC00508B6051D3C3E2D2@grumpy.internal.hspg.com>r  7 The three logicals have sorted the problem - Thank you.rK I have lived with this problem a while, because the Digital/Compaq engineeraH who installed the system came in several times to fix it, & then told me/ just to keep deleting the log file every month.dL In answer to how I tried disabling the Netbios - I check the check box whichJ said 'disable Netbios' in the pathworks config, then stopping & restarting pathworks. m   Andrew Robinson    -----Original Message-----0 From: eplan@kapsch.net [mailto:eplan@kapsch.net] Sent: 17 July 2001 17:45 To: Info-VAX@Mvb.Saic.Como Subject: Re: Pathworks Question     
 In articleI <CDA4BAD1E10ED41181AC00508B6051D3C3E2D0@grumpy.internal.hspg.com>, Andrew-% Robinson <arobinson@hspg.com> writes:17 >Please could someone help me with a pathworks problem.n  I Consider using the vmsnet.networks.desktop.pathworks group in addition...e  I >I Have pathworks V6.0B running on an Alpha 1200 with VMS 7.2-1 patch sys $ >v9.0 & TCP 5.0A  patch ec0 v5.0-111  I There are a lot more ECOs for OpenVMS V7.2-1 and there is TCPIP V5.1 (ECOr 1).w Consider them anyway...p  J >The problem I have is that NET$BIOS process has tied itself to the secondG >ethernet adaptor, this mean's that it produces a 350000 block log fileO every'@ >month. How can I get it to attach to the primary ethernet card?  $ With a supported/documented logical.5 And you can check its functionality with SDA>SHOW LANo   >Netbios = device EWA70 >all other processes = devices EWB6 up to EWB12) > 9 >The EWA adaptor is a DE500 on a KZPCM-DA SCSI controller-7 >The EWB adaptor is a DE500 on a DE500-BA ethernet cardu >AJ >I did a search and found a reference to a logical PWRK$NETBIOS$DEVICE but it >didn't say how to use it. @  : 	$ DEFINE/SYSTEM	NETBIOS$DEVICE		EWB0:	!for NB over DECnet? 	$ DEFINE/SYSTEM	PWRK$KNBDAEMON_DEVICE	EWB0:	!for NB over TCPIPl7 	$ DEFINE/SYSTEM PWRK$NETBEUI_DEVICE	EWB0:	!for NetBEUI   $ see chapter 3.11.1 (pg. 3-33) of theG "AdvancedServerfor OpenVMS/Server Installation and Configuration Guide"   K >To solve the log file problem, I tried to disable NETBIOS in the pathworks-- >config, but this didn't make any difference.1  H What did you do ? Disable DECnet in ADM/CONF ? Restart Server ? Reboot ?   -- :< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888 < <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Tue, 17 Jul 2001 12:26:10 +0100 % From: Alan Greig <a.greig@virgin.net>g Subject: Re: PAWZ and ECP -h8 Message-ID: <r488ltk851j0vk4n90t58kuoj4ukha0rvj@4ax.com>  E On Fri, 13 Jul 2001 12:12:13 -0400, rdeininger@mindspring.com (Robertt Deininger) wrote:7     >2I >Very good point.  I really doubt all the alpha --> intel problems peopleiJ >are howling about here are actually critical to anyone in the near term. H >PAWZ is probably similar.  Folks can afford to wait a day (week, month) >before going wild.i  F That's a problem for customers just about to sign a purchase order. DoF they sign now and take things on faith, wait several months to see how, things go or start looking at other options?   -- Alan   ------------------------------  % Date: Tue, 17 Jul 2001 09:57:35 -0400h2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: PAWZ and ECP - L Message-ID: <rdeininger-1707010957350001@user-2iveacr.dialup.mindspring.com>  C In article <r488ltk851j0vk4n90t58kuoj4ukha0rvj@4ax.com>, Alan Greige <a.greig@virgin.net> wrote:   G > On Fri, 13 Jul 2001 12:12:13 -0400, rdeininger@mindspring.com (Robert1 > Deininger) wrote:i >  >  > >.K > >Very good point.  I really doubt all the alpha --> intel problems people3L > >are howling about here are actually critical to anyone in the near term. J > >PAWZ is probably similar.  Folks can afford to wait a day (week, month) > >before going wild.r > H > That's a problem for customers just about to sign a purchase order. DoH > they sign now and take things on faith, wait several months to see how. > things go or start looking at other options?  G A short period of reconsideration is probably wise.  But anyone who wassJ ready to sign a purchase order presumably needs the product soon, not in 3H years.  The recent announcement hasn't really changed plans for the nextE few years.  The EV7 system roadmap has changed only in details.  If a = product was deemed ok before the announcement, it's still ok.   E "Will my product still be supported in 3 or 4 years?" is a reasonablenH question, but the answer was hardly certain before the IPF announcement, and seems about the same now.    IMHO   -- s Robert Deininger rdeininger@mindspring.com:   ------------------------------    Date: 17 Jul 2001 07:13:47 -0700) From: google@mccready.com (Gary McCready)c' Subject: Penance for Compaq's VMS sins?g= Message-ID: <6e64ea70.0107170613.2b16e328@posting.google.com>t  B Most of the converted here agree that Compaq has sinned, and sinceD they cannot go back and undo what has been done, what penance should we ask of them?h  E Translation: since they might feel guilty, what should we ask them toDD do for their VMS crowd that might make life easier, keep our systemsB running better, make us feel that we are not already forgotten and second-class citizens?  F Personally, I would like to see a clear support strategy, ranging fromD what tools Compaq will provide to monitor and tune my system withoutC complications (like having to patch systems before I can use CompaqkC Analyze, etc), to good, robust prior-version support so those of usuE who cannot upgrade will not be forced off the systems earlier than we- would like to.  E In short, I would like them to provide a high-class hospice where VMSe% can die a dignified, pain-free death.m   --Gary McCreadyi2 My opinions have nothing to do with my employer's.   ------------------------------  % Date: Tue, 17 Jul 2001 10:24:41 -0400:5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>c+ Subject: Re: Penance for Compaq's VMS sins?o2 Message-ID: <AUX47.828$rc5.60871@news.cpqcorp.net>  F There is a team of people trying to fix the problems of monitoring and% tuning.  We *do* feel your pain here.s  H We have actually tried real hard, and complicated our lives considerablyG trying to provide new hardware and some features on prior versions.  At L times, we've had to make 4 or 5 versions of a patch, or new hardware supportJ for the various streams of VMS out there, from V7.1-2, to V7.2-1, to V7.3.L We're trying to get that down to a manageable number of streams (and minimalH number of patches needed to apply - like a future 7.2-2 consolidation of$ fixes and performance enhancements).  L V7.1~* lasted a *long* time, as has 7.2~*.  Heck, we still support platformsK with current releases that Tru64 retired a long time ago.  Eventually, whenuK most people have moved forward, it's no longer practical to go back too faro for general support.       Gary McCready wrote in message4 <6e64ea70.0107170613.2b16e328@posting.google.com>...C >Most of the converted here agree that Compaq has sinned, and since>E >they cannot go back and undo what has been done, what penance shouldo >we ask of them? > F >Translation: since they might feel guilty, what should we ask them toE >do for their VMS crowd that might make life easier, keep our systemslC >running better, make us feel that we are not already forgotten andI >second-class citizens?a >vG >Personally, I would like to see a clear support strategy, ranging from E >what tools Compaq will provide to monitor and tune my system withouteD >complications (like having to patch systems before I can use CompaqD >Analyze, etc), to good, robust prior-version support so those of usF >who cannot upgrade will not be forced off the systems earlier than we >would like to.a > F >In short, I would like them to provide a high-class hospice where VMS& >can die a dignified, pain-free death. >  >--Gary McCready3 >My opinions have nothing to do with my employer's.w   ------------------------------  % Date: Tue, 17 Jul 2001 11:10:20 -0400F- From: "Peter Weaver" <peter.weaver@stelco.ca>>+ Subject: Re: Penance for Compaq's VMS sins? 4 Message-ID: <wzY47.263334$Z2.3195529@nnrp1.uunet.ca>  6 "Gary McCready" <google@mccready.com> wrote in message7 news:6e64ea70.0107170613.2b16e328@posting.google.com...sD > Most of the converted here agree that Compaq has sinned, and sinceF > they cannot go back and undo what has been done, what penance should > we ask of them?  >...   - Three hail Mary'sm? - Sacrifice Compaq Analyze on an alter made out of a VAX 11/780C - Give us back DSN 1.2Ci7 - Give us a LA120 for a console instead of a WINTEL boxR; - Remove all references to WIN NT/2000 from all VMS manualstC - Drop the Open prefix so I can write the name the same as I say ita  G And if they still feel guilty after that then come back 'cause I have ae longer list.   -- Peter WeaverJ Using a WIN NT/WIN 2000 box to manage your VMS systems is like towing your7 mechanic in a 5th wheel motor home behind your Porsche.,   ------------------------------  # Date: Tue, 17 Jul 2001 15:02:30 GMTe4 From: "Andy Proctor" <aproctor@removeme.hotmail.com>F Subject: Possibly a dumb question on TCPIP settings for 2 IP Addresses4 Message-ID: <995381733.215609@ananke.eclipse.net.uk>  L First of all thanks to Oswald for solving my last DECNET problem - Rerunning net$configure worked, thanks!e  I Now i have been asked to set up a DS10 with EWA0 and EWB0 (Interfaces WE0wH and WE1) to operate on 2 networks. One is on a LAN, the other a point toC point ethernet connection. I've had EWA0 working for both when usedrK separately, but what is required is to be able to use both networks without K pulling cables around. I thought that i could set up each interface to haveoH a separate IP address. I ran TCPIP$CONFIG and set it up, but it does notJ seem to work. I have the TCPIP services management manual here, am I right4 in now thinking I have to set up a pseudo interface? Do I have to set it up like:    TCPIP> SET NOINTERFACE interface+ TCPIP> SET INTERFACE interface /HOST=host -i1 _TCPIP> /NETWORK_MASK=mask /BROADCAST_MASK=b_maske9 TCPIP> SET CONFIGURATION INTERFACE interface /HOST=host -a1 _TCPIP> /NETWORK_MASK=mask /BROADCAST_MASK=b_maski  K Where "interface" would be the two interfaces I have. Host is the host nameyI (The same regardless) and the masks as I wish. Then I run tcpip$config toJ set the IP addresses, right?  D Or am I just able to set the two interfaces to have the IP addresses required in tcpip$config?   D Will the system be able to know which interface to use for a certain address - done by the masking?  I Sorry for the dumb questions (if they are) but first time I've done this.e   System:i DS10Em VMS7.2-1 TCPIP V5.0ag   Cheers   Andy   ------------------------------   Date: 17 Jul 2001 11:33:25 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)  Subject: Re: RA7x series disks, Message-ID: <9j17q5$1udq$1@info.cs.uofs.edu>  ' In article <3B534DD6.5F5C3CF0@iee.org>,y.  "antonio.carlini" <arcarlini@iee.org> writes: |>   |> e |> Bill Gunshannon wrote:k |>N |> > Maybe you could answer a question or two to help me diagnose the problem.J |> > Are the error codes displayed by the front panel leds the same as theD |> > lights on an RA80/81??  If so, what is a "master/Slave error"?? |> i: |> The disks are (I think) reasonably consistent about howD |> to read the error code (RUN/STOP FAULT READY W-PROT PORT-A PORT-B? |> for the RA7x going from MSB to LSB). But none of the manualsr$ |> I have list "master/slave error"! |> e( |> What hex code do you get? (And out of( |> interest, which manual did you use to |> translate the code?)   B The Fault and Write-Protect indicators light when I push the fault1 button.  I looked it up in the RA81 Users Manual.c   |> 'L |> > The disk seems to spin up OK,  When it reaches full speed it faults andL |> > shuts back down.  based on t his, I assume it is likely electronics andL |> > not mechanics.  There isn't by any chance a physical lock that may haveM |> > been set by the previous owner that is not obviously visible, is there??  |> o/ |> The fault code will lead you to the failing h! |> FRU ... there are only two :-)o  2 Probably would if I had a Maintenance Manual.  :-)   Thanks for the help.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   d   ------------------------------    Date: 17 Jul 2001 00:54:15 -0700 From: pjo@pjo.dk (Peter)= Subject: Returning a value from a C program to a DCL variableS= Message-ID: <4dd0684b.0107162354.67a02f4b@posting.google.com>1  E Has anyone any idea how to return a value from a C program and out tou a DCL symbol/variable such as :g   $ testprogram :== $testprogram
 $ test = 0 $ test = testprogram
 $ sh sym testc   TEST == "10" $e  / where the C program consists of the following :B  
 int main() {p
    return 10;S }2  
 Regards Peter0   ------------------------------  % Date: Tue, 17 Jul 2001 10:29:50 +0200c< From: "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de>A Subject: Re: Returning a value from a C program to a DCL variableI4 Message-ID: <9j0t27$les6e$1@ID-56200.news.dfncis.de>   Peter wrote...F >Has anyone any idea how to return a value from a C program and out to  >a DCL symbol/variable such as : >l >$ testprogram :== $testprogram  >$ test = 0h >$ test = testprograme >$ sh sym test >  TEST == "10"o >$ >W0 >where the C program consists of the following : >  >int main()l >{ >   return 10; >}  8 The return value is stored in the $STATUS variable, i.e.   $ r testprogramc $ sho sym $statusa   $STATUS == "%X0000000A"    cu,o   Martin --J One OS to rule them all       | Martin Vorlaender  |  VMS & WNT programmer7 One OS to find them           | work: mv@pdv-systeme.dem One OS to bring them all      |E( http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------  % Date: Tue, 17 Jul 2001 15:14:32 +0200a2 From: Mike Rechtman <michael.rechtman@digital.com>A Subject: Re: Returning a value from a C program to a DCL variabler+ Message-ID: <3B545658.2A7BE95B@digital.com>   ) Dunno 'bout C, but look up LIB$SET_SYMBOL2# (you're going to need $DESCRIPTORs)    Mike   Peter wrote: > G > Has anyone any idea how to return a value from a C program and out to(! > a DCL symbol/variable such as :  >   > $ testprogram :== $testprogram > $ test = 0 > $ test = testprogram > $ sh sym testo >   TEST == "10" > $p > 1 > where the C program consists of the following :  >  > int main() > {h >    return 10;t > }? >  > Regards Peterd   -- eE ---------------------------------------------------------------------NE Usual disclaimer: All opinions are mine alone, perhaps not even that. ? Mike Rechtman                            *rechtman@tzora.co.il*TF Kibbutz Tzor'a.                          Voice (home): 972-2-9908337  B   "20% of a job takes 80% of the time, the rest takes another 80%"E ---------------------------------------------------------------------! -----BEGIN GEEK CODE BLOCK-----T Version: 3.1: GCM/CS d(-)pu s:+>:- a++ C++ U-- L-- W++ N++ K? w--- V+++$6 PS+ PE-- t 5? X- tv-- b+ DI+ D-- G e++ h--- r+++ y+++@ ------END GEEK CODE BLOCK------e   ------------------------------  # Date: Tue, 17 Jul 2001 16:59:47 GMTh6 From: "Emanuel Marciniak" <emarciniak@patientcare.com> Subject: RSH question..i= Message-ID: <7a_47.4852$A47.3164473@news1.rsm1.occa.home.com>n   Hi,DI  I try to execute the following command from Sun Solaris on Alpha VMS 7.1  and getting error1  E  /bin/su emanuel -c "rsh -n circe repl/bell/user=emanuel 'test test'"<  K %DCL-W-MAXPARM, too many parameters - reenter command with fewer parametersRK %DCL-W-MAXPARM, too many parameters - reenter command with fewer parameters-  \TEST\   \TEST\u  / It works if the message contains only one word.o  ) How can I pass the whole text message ..?s   Thanks for any hint,   Emanuel Marciniak0   ------------------------------  % Date: Tue, 17 Jul 2001 13:16:51 -0400t5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>o Subject: Re: RSH question..i2 Message-ID: <_p_47.864$rc5.60858@news.cpqcorp.net>  F Is there any way to pass a real double quote (") instead of the single quote?    " Emanuel Marciniak wrote in message4 <7a_47.4852$A47.3164473@news1.rsm1.occa.home.com>... >Hi,J > I try to execute the following command from Sun Solaris on Alpha VMS 7.1 >and getting error > F > /bin/su emanuel -c "rsh -n circe repl/bell/user=emanuel 'test test'" >eL >%DCL-W-MAXPARM, too many parameters - reenter command with fewer parametersL >%DCL-W-MAXPARM, too many parameters - reenter command with fewer parameters > \TEST\ > \TEST\ >e0 >It works if the message contains only one word. >o* >How can I pass the whole text message ..? >  >Thanks for any hint,s >. >Emanuel Marciniak >r >g   ------------------------------  % Date: Tue, 17 Jul 2001 14:22:45 -0300t) From: fabio_compaq@ep-bc.petrobras.com.bra Subject: Re: RSH question.. L Message-ID: <OF4B43245F.96458EC4-ON03256A8C.005F6B5D@ep-bc.petrobras.com.br>  7 Try to change the 'test test' statement  by "test test"p   Regardst   FC        G "Emanuel Marciniak" <emarciniak@patientcare.com> em 17/07/2001 13:59:47k  B Favor responder a "Emanuel Marciniak" <emarciniak@patientcare.com>             Info-VAX@Mvb.Saic.Comd       Assunto: RSH question..      Hi, I  I try to execute the following command from Sun Solaris on Alpha VMS 7.1e and getting error   E  /bin/su emanuel -c "rsh -n circe repl/bell/user=emanuel 'test test'"e  K %DCL-W-MAXPARM, too many parameters - reenter command with fewer parameterscK %DCL-W-MAXPARM, too many parameters - reenter command with fewer parameters   \TEST\e  \TEST\   / It works if the message contains only one word.   ) How can I pass the whole text message ..?E   Thanks for any hint,   Emanuel MarciniakP   ------------------------------  # Date: Tue, 17 Jul 2001 17:32:16 GMTf6 From: "Emanuel Marciniak" <emarciniak@patientcare.com> Subject: Re: RSH question.. = Message-ID: <AE_47.4858$A47.3182460@news1.rsm1.occa.home.com>a   Thank you guys..,t  %   "\"test test\""  did the trick.. :)t   Thank you again,   Emanuel     ? Emanuel Marciniak <emarciniak@patientcare.com> wrote in messageE7 news:7a_47.4852$A47.3164473@news1.rsm1.occa.home.com...  > Hi,mK >  I try to execute the following command from Sun Solaris on Alpha VMS 7.1  > and getting error1 >VG >  /bin/su emanuel -c "rsh -n circe repl/bell/user=emanuel 'test test'"- >-B > %DCL-W-MAXPARM, too many parameters - reenter command with fewer
 parametersB > %DCL-W-MAXPARM, too many parameters - reenter command with fewer
 parameters	 >  \TEST\j	 >  \TEST\a >o1 > It works if the message contains only one word.r >n+ > How can I pass the whole text message ..?  >  > Thanks for any hint, >  > Emanuel Marciniak- >n >    ------------------------------  % Date: Tue, 17 Jul 2001 17:00:56 -0700n2 From: Jayasuthan <suthan@eplx01.fairchildsemi.com>( Subject: SNMP monitor decHUB temperature8 Message-ID: <3B54D1B8.19C76F71@eplx01.fairchildsemi.com>   Hi,o  G Did anyone work with command line to get decHUB temperature using SNMP?vF I am working on a Alpha Linux with UCD SNMP, i begin project lately onA LAN monitor. Create a Web base system monitor ,part of it involve1F temperature monitor. I have obtain MIBs from compaq for decHUB, I justB need few guideline, still don't know how to get the parameter fromH DecHUB. If it work I would like to distribute under GPL. I hv done basicD web interface with server alive test using fping and php script on a apache server.  = I actually work to build a OpenVMS monitor on a Linux System.e  
 Thank You,   suthan   ------------------------------  % Date: Tue, 17 Jul 2001 09:07:00 -0400l( From: Hamlyn Mootoo <univms@bigfoot.com>, Subject: Re: SNMP monitor decHUB temperature+ Message-ID: <3B543874.F031372A@bigfoot.com>a   Using a MIB browser, start at:# 1.3.6.1.4.1.36.2.18.11.1.1.1.10.2.1f and navigate further downward.  B This is part of the Chasis MIB version 3 for a DECHUB 900, NOT theG RFC1213 MIB.  It appears to tell whether the temperature sensors are in / a specific state, not quantitative temperature.i   Hope this helps.   HM   Jayasuthan wrote:a >  > Hi,e > I > Did anyone work with command line to get decHUB temperature using SNMP?tH > I am working on a Alpha Linux with UCD SNMP, i begin project lately onC > LAN monitor. Create a Web base system monitor ,part of it involvenH > temperature monitor. I have obtain MIBs from compaq for decHUB, I justD > need few guideline, still don't know how to get the parameter fromJ > DecHUB. If it work I would like to distribute under GPL. I hv done basicF > web interface with server alive test using fping and php script on a > apache server. > ? > I actually work to build a OpenVMS monitor on a Linux System.y >  > Thank You, >  > suthan   ------------------------------  % Date: Tue, 17 Jul 2001 16:53:01 +0200t2 From: "Dr. Otto Titze" <titze@ikp.tu-darmstadt.de> Subject: SS$_GSDFULL3 Message-ID: <3B54514D.24E135AD@ikp.tu-darmstadt.de>e   Hi,h  : sorry about this simple question: What to do when one gets2 ss$_gsdfull - global section descriptor table full   Regards  Otto,  -------------------------------------------, | Dr. Otto Titze, Kernphysik TUD           |, | Schlossgartenstr. 9, D-64289 Darmstadt   |, | titze@ikp.tu-darmstadt.de                |, | Tel: +49(6151)16-2916,FAX:16-4321        |,  -------------------------------------------   ------------------------------  % Date: Tue, 17 Jul 2001 11:16:24 -0400e( From: Hamlyn Mootoo <univms@bigfoot.com> Subject: Re: SS$_GSDFULL+ Message-ID: <3B5456C8.3A497D55@bigfoot.com>   G Increase GBLSECTIONS in sysgen.  You may also need to increase GBLPAGESaG as well, as you may run out of them soon too.  If your application uses G a lot of scratch global sections, you may need to increase GBLPAGFIL as-B well.  To increase GBLSECTIONS, you will need to reboot as it is a$ non-dynamic parameter. You can do a:C WRITE SYS$OUTPUT F$GETSYI("FREE_GBLSECTS") to see if you're out, ore! close to out, of global sections..   HM   "Dr. Otto Titze" wrote:f >  > Hi,i > < > sorry about this simple question: What to do when one gets4 > ss$_gsdfull - global section descriptor table full > 	 > Regardsu > Otto. >  -------------------------------------------. > | Dr. Otto Titze, Kernphysik TUD           |. > | Schlossgartenstr. 9, D-64289 Darmstadt   |. > | titze@ikp.tu-darmstadt.de                |. > | Tel: +49(6151)16-2916,FAX:16-4321        |. >  -------------------------------------------   ------------------------------  % Date: Tue, 17 Jul 2001 11:12:39 -0400f- From: "Richard D. Piccard" <piccard@ohio.edu>a Subject: Re: SS$_GSDFULL( Message-ID: <3B5455E5.6FB2DA19@ohio.edu>   I would start by   $ MCR SYSGEN SYSGEN>  HELP PARAM GBLu  * Then edit MODPARAMS.DAT, run AUTOGEN, etc.                   RDPd     "Dr. Otto Titze" wrote:t   > Hi,o >c< > sorry about this simple question: What to do when one gets4 > ss$_gsdfull - global section descriptor table full >m	 > Regards  > Otto. >  -------------------------------------------. > | Dr. Otto Titze, Kernphysik TUD           |. > | Schlossgartenstr. 9, D-64289 Darmstadt   |. > | titze@ikp.tu-darmstadt.de                |. > | Tel: +49(6151)16-2916,FAX:16-4321        |. >  -------------------------------------------   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  % Date: Tue, 17 Jul 2001 17:29:38 +0200 2 From: "Dr. Otto Titze" <titze@ikp.tu-darmstadt.de> Subject: Re: SS$_GSDFULL3 Message-ID: <3B5459E2.E2BFC022@ikp.tu-darmstadt.de>    Thanks,I   Hamlyn Mootoo schrieb: > I > Increase GBLSECTIONS in sysgen.  You may also need to increase GBLPAGES-I > as well, as you may run out of them soon too.  If your application uses-I > a lot of scratch global sections, you may need to increase GBLPAGFIL as D > well.  To increase GBLSECTIONS, you will need to reboot as it is a& > non-dynamic parameter. You can do a:E > WRITE SYS$OUTPUT F$GETSYI("FREE_GBLSECTS") to see if you're out, or # > close to out, of global sections.7 > / Yes I wondered, on that particular system I hado free_gblsects	467t fre_gblpages	142048," and thought that should be enough.  7 Ok GBLPAGFIL might be a good idea (it has only(?) 1024)    Regards.   Otto   ------------------------------    Date: 17 Jul 2001 11:55:23 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)1 Subject: Re: Sun's take on Compaq's announcement.R* Message-ID: <3b540b8b$1@news.kapsch.co.at>  _ In article <9ivflv$ka3$1@jetsam.uits.indiana.edu>, bdwheele@indiana.edu (Brian Wheeler) writes:t >http://www.sun.com/alpha/ >t@ >All I wonder is:  what does sun do with the traded-in alphas?     Shred them of course. + Not to get them on the (used) market again. F Not to make it possible to create applications for a foreign platform.   -- -< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888s< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Tue, 17 Jul 2001 10:14:43 -0400k5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>w1 Subject: Re: Sun's take on Compaq's announcement.r2 Message-ID: <eLX47.824$rc5.60861@news.cpqcorp.net>  J Maybe they should use them internally to design a Sparc chip that is fast,  or a cache design that works ;-)    D Peter LANGSTOEGER wrote in message <3b540b8b$1@news.kapsch.co.at>...H >In article <9ivflv$ka3$1@jetsam.uits.indiana.edu>, bdwheele@indiana.edu (Brian Wheeler) writes:u >>http://www.sun.com/alpha/o >>? >>All I wonder is:  what does sun do with the traded-in alphas?n >t >Shred them of course., >Not to get them on the (used) market again.G >Not to make it possible to create applications for a foreign platform./ >T >-- = >Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651S< >Network and OpenVMS system manager  Fax.    +43 1 81111-888= ><<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netoI >A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"e   ------------------------------    Date: 17 Jul 2001 18:15:52 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)1 Subject: Re: Sun's take on Compaq's announcement.r* Message-ID: <3b5464b8$1@news.kapsch.co.at>  j In article <eLX47.824$rc5.60861@news.cpqcorp.net>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes:K >Maybe they should use them internally to design a Sparc chip that is fast,1! >or a cache design that works ;-)    With which application ? ;-)    -- 7< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-8881< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Tue, 17 Jul 2001 12:23:45 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>d1 Subject: Re: Sun's take on Compaq's announcement.e2 Message-ID: <cEZ47.854$rc5.60901@news.cpqcorp.net>  G Why something that is OpenSource.  Isn't that obvious?  UNIX is UNIX isa UNIX.w  J Hey, maybe they can use them for those supercomputer sales that they can't) get through the acceptance tests with ;-)o    D Peter LANGSTOEGER wrote in message <3b5464b8$1@news.kapsch.co.at>...E >In article <eLX47.824$rc5.60861@news.cpqcorp.net>, "Fred Kleinsorge">% <kleinsorge@star.zko.dec.com> writes: L >>Maybe they should use them internally to design a Sparc chip that is fast," >>or a cache design that works ;-) >> >With which application ?O >;-) >i >--c= >Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651 < >Network and OpenVMS system manager  Fax.    +43 1 81111-888= ><<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.net>I >A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"    ------------------------------    Date: 17 Jul 2001 12:54:30 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)c1 Subject: Re: Sun's take on Compaq's announcement.-3 Message-ID: <Q0g5NGaUYvT0@eisner.encompasserve.org>   W In article <3b5464b8$1@news.kapsch.co.at>, eplan@kapsch.net (Peter LANGSTOEGER) writes:el > In article <eLX47.824$rc5.60861@news.cpqcorp.net>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes:L >>Maybe they should use them internally to design a Sparc chip that is fast," >>or a cache design that works ;-) >  > With which application ?   The one DEC used for Alpha.6   ------------------------------  % Date: Tue, 17 Jul 2001 12:12:49 +0100a% From: Alan Greig <a.greig@virgin.net> - Subject: Re: Terry Shannon Tech Talk on IA-64 8 Message-ID: <q278ltcrkjp7tamkbqmtmr2bfh2lnk1qtk@4ax.com>  5 On Fri, 13 Jul 2001 13:17:19 -0400, "Fred Kleinsorge"o$ <kleinsorge@star.zko.dec.com> wrote:     >rJ >I'd be more than a little suprised if Intel were to kill the IA64 ISA andL >replace it with the Alpha ISA (not that I wouldn't be happy).  I think thisJ >is just wishful thinking turned into idle speculation by Internet drones.I >Internally, we are planning to port to the IPF ISA, not to some mythical. >hybrid ISA.  > What worries me Fred is that Richard George (Alpha UK BusinessB Manager) confirmed yesterday that the port was targeted at currentA IA64 chips and this would be released to ISVs but commercial shipoD would not happen until IA64 servers incorporating Alpha provided SMTC technology is available commercially. He gave the earliest date forSB this as 2004 and agreed that if Intel (or ISSG who.will design theA machines) caused this to slip to 2005 or 2006 under current plans D there would be no customer available iVMS until that date. He agreed& to feedback customer disquiet at this.   -- Alan   ------------------------------  % Date: Tue, 17 Jul 2001 09:55:44 -0400D5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>-- Subject: Re: Terry Shannon Tech Talk on IA-64u2 Message-ID: <rtX47.820$rc5.60781@news.cpqcorp.net>  I Not that its needed, but just to stick up for Kerry a bit.  I know Kerry,DJ and this guy busts his butt to do his job, and he's pretty good at it.  InH addition, he spends his own time in these forums to try and provide some counter to trolls like Andrew.     _Fred       Main, Kerry wrote in message ... >Bill, >(% >Must you now resort to name calling?  > < >Tsk, Tsk ..why not simply state you disagree and let it go? >rJ >On specific subjects, many of your discussions are pretty good, but may IL >suggest that those you write containing childish, emotional outbursts (suchD >as attached) against those that might disagree with your conspiracy theories= >be taken offline as imho, they do not have a place in c.o.v., >r >Thanks, >  >Kerry MainR >Senior Consultant >Compaq Canada Inc.m >Professional Services >Voice: 613-592-4660 >Fax  :  819-772-7036e >Email: Kerry.Main@Compaq.com  >  >e >-----Original Message-----a- >From: Bill Todd [mailto:billtodd@foo.mv.com]- >Sent: July 16, 2001 3:30 PM >To: Info-VAX@Mvb.Saic.Com. >Subject: Re: Terry Shannon Tech Talk on IA-64 >- >+ >+7 >"Main, Kerry" <Kerry.Main@Compaq.com> wrote in messageDL >news:BE56C50EA024184DAF48F0B9A47F5CF4AD7F0B@kaoexc01.americas.cpqcorp.net.. .- >- >..- >-L >> >>> But instead of even causing you to reexamine what you're saying (evenF >> after being given specific reasons why it's asinine), you just keep saying	 >> it.>>>1 >>I >> Or rather ignoring it (and a good deal of your other "teachings") as IrJ >> personally do not agree with your conspiracy theories. But you are free to" >> express whatever you feel like. >DI >This is not a conspirary theory, but a technical point here (the obvious F >inability to work any major changes into the IA64 architecture in theI >time-frame specified for the VMS port).  You really are a piece of work,oL >accusing others of twisting words when it's virtually *all* you do.  I takeL >back my previous comment:  whoring is a considerably more honest profession >than yours is.m >F >- bills >e >k   ------------------------------  % Date: Tue, 17 Jul 2001 10:05:58 -0400_5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>t- Subject: Re: Terry Shannon Tech Talk on IA-64i2 Message-ID: <1DX47.821$rc5.60599@news.cpqcorp.net>  = JF Mezei wrote in message <3B53998A.F8CC8792@videotron.ca>...  > J >By breaking its commitment to Alpha, Compaq greatly hurt its credibility. AndeE >consdiering that porting VMS to IA64 seems to be an afterthought, itsA >re-enforces the image that Compaq doesn't really care about VMS.d >t    F No more than Tru64, or NSK.  We all found out at the same time.  It isH incredibly POSITIVE that they committed to port VMS, Tru64, and NSK.  IfJ they really wanted to kill VMS, it would have been the perfect opportunityL to say that VMS would continue to be developed on Alpha, but that Marvel/EV7H would be the last Alpha platform and we would continue to support VMS on Alpha for the next 15 years.    I >They may not be actively killing VMS at this point in time, but they arec notoC >doing much to strenghten and expand its market. The fact that theyt
 supposedlyL >visited key customers on the day of Alpha's murder but didn't bother giving thJ >erest of the customers the same information makes one beleive that CompaqE >doesn't care about losing those customers it doesn't feel are "key".   H My understanding is that very few of us knew in advance because of legalG reasons.  The limited time that was provided to tell customers confinedlL things to a small handful of large customers and ISV's - the 80% of revenuesH type.  From what I know, the handful of people who were briefed, and whoH then briefed those customers were flat out trying to get as far down the5 list of customers as they could in the time they had.w  K That doesn't make it right, and I'm sure that the folks here in VMS wish we-L could have had a 1 on 1 meeting with every customer who owns an Alpha system with VMS on it.c   ------------------------------  % Date: Tue, 17 Jul 2001 10:13:38 -040055 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>u- Subject: Re: Terry Shannon Tech Talk on IA-64t2 Message-ID: <eKX47.823$rc5.60738@news.cpqcorp.net>  I I can't possibly guess at the actual target platform, or date (although I7C saw an interesting PP slide show in one of the internet rags).  ThewI engineering focus for the time being has to be on getting it ported to an K existing IPF platform, and getting an SDK release ASAP.  The business folks K can then decide what and when the right production platform would be.  It's L not clear to me what SMT has to do with anything, or why it would be the keyI feature needed for VMS - after all, all it does is to help better utilizeoE chip resources in multithreading - it isn't in itself a single streamrK performance enhancement - or something you can't get by just building a SMP-" or NUMA system with non-SMT chips.  H But I'm not privvy to product planning.  Maybe that is just the date and: chip generation where IPF starts to meet or blow away EV7.   _Fred (NOT speaking for Compaq)i     Alan Greig wrote in message ...t6 >On Fri, 13 Jul 2001 13:17:19 -0400, "Fred Kleinsorge"% ><kleinsorge@star.zko.dec.com> wrote:l >l >n >>K >>I'd be more than a little suprised if Intel were to kill the IA64 ISA and H >>replace it with the Alpha ISA (not that I wouldn't be happy).  I think thisK >>is just wishful thinking turned into idle speculation by Internet drones.kJ >>Internally, we are planning to port to the IPF ISA, not to some mythical
 >>hybrid ISA.g >d? >What worries me Fred is that Richard George (Alpha UK BusinesseC >Manager) confirmed yesterday that the port was targeted at current-B >IA64 chips and this would be released to ISVs but commercial shipE >would not happen until IA64 servers incorporating Alpha provided SMTtD >technology is available commercially. He gave the earliest date forC >this as 2004 and agreed that if Intel (or ISSG who.will design theaB >machines) caused this to slip to 2005 or 2006 under current plansE >there would be no customer available iVMS until that date. He agreedm' >to feedback customer disquiet at this.e >o >--n >Alane   ------------------------------  # Date: Tue, 17 Jul 2001 13:13:19 GMTe8 From: hammond@not@peek.ppb.cpqcorp.net (Charlie Hammond)' Subject: Re: Upgrade PAGEFILE.SYS error 2 Message-ID: <PRW47.816$rc5.60599@news.cpqcorp.net>  1 In article <9ivnno$9mg@dispatch.concentric.net>, p' "Paul Dembry" <pade@trifox.com> writes:.J >I am upgrading an AXP 150 from 7.2-2 to 7.3.  All goes well for the firstC >80% of the execution phase and then I get three messages about thew@ >PAGEFILE.SYS filling up, full, and attempting to continue.  ...  I When you are upgrade an OpenVMS Alpha system, you are booted from the CD. I Since there CD is READ_ONLY, there is no pagefile.sys -- and it certainlyhK can't fill up!  If you are doing things correctly, this probably means that D something has got into an error loop and filled up available memory.  G Suggestions: First, verify that you have completed all the tasks listedeH in the Before You Upgrade chapter of the Upgrade and Installation Guide.) In particular, ANA/DISK your target disk.l  F If this uncovers nothing, restore you backup and boot from the OpenVMSE operating system CD-ROM, as before.  Choose the DCL OPTION and define2 the following logicals:@  0     $ DEFINE /SYS AXPVMS$PCSI_EXECUTE_VERIFY YES,     $ DEFINE /SYS AXPVMS$PCSI_LOG_TRACE BOTH  I This will generate a *LOT* of output, but you should be able to see where,H the error originates.  If it isn't obvious, post some of the output fromG before the hang.  NOTE: If it gets into a loop, you may need to stop itn6 to capture the original error that gets into the loop.   -- LK     Charlie Hammond -- Compaq Computer Corporation -- Pompano Beach  FL USAeH        (hammond@not@peek.ppb.cpqcorp.net -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------  % Date: Tue, 17 Jul 2001 10:49:21 +0200)2 From: "Ren Schelbaum" <rene.schelbaum@datakom.at>. Subject: Re: VMS Delete lots of files quickly?G Message-ID: <3b53fb9d$0$21458$6e365a64@newsreader02.highway.telekom.at>s  6 "Mark Vance" <mvance@iglou.com> schrieb im Newsbeitrag" news:3B5393A7.1050904@iglou.com... >-J > Is there an way, in VMS, to delete files quickly?  Im doing a conversionG > that is going to make millions of small files over its course, and inaI > some small test runs, it takes FOREVER to get a range of files deleted.-H >   Is there any way to prune directories, or delete lots of small files
 > quickly? >, > HEEEEEEELP!!!2 > 	 > Thanks!m >R  K I have had a similar problem. I solved it by simply renaming the directory,3I creating a new one with the original name (for the next run) and deletingLK the files in the renamed directory (in reverse order, that's faster) during  a non-so-busy period.z   regards    Ren Schelbaum Datakom Austria GesmbH   ------------------------------  % Date: Tue, 17 Jul 2001 12:36:37 -0400 ; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>-. Subject: Re: VMS Delete lots of files quickly?$ Message-ID: <3b5469e2$1@news.si.com>  2 >Is there an way, in VMS, to delete files quickly?  H Others have mentioned deleting in reverse alphabetical order.  Two other ways are to use BACKUP and DFU:-  + $ backup [.test...]/delete nl:save.set/save=   $ dfu delete/dir/tree test.dir  ? BACKUP comes with VMS, of course.  DFU is on the freeware disk.g --A Brian Tillman                   Internet: tillman_brian at si.com A Smiths Aerospace                          tillman at swdev.si.com = 3290 Patterson Ave. SE, MS      Addresses modified to prevent9< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Tue, 17 Jul 2001 13:50:32 +0100e% From: Alan Greig <a.greig@virgin.net>e Subject: Re: Want 21164 Alphaw8 Message-ID: <24d8ltctpe4hm8sph144nsgbu99updr93g@4ax.com>  C On 16 Jul 2001 15:52:10 -0700, antonelron@hotmail.com (Anton Elron)e wrote:  D >I am looking for a 21164 Alpha. Are motherboards available or would' >buying a whole used machine be better?  > A > Who would have used 21164 based workstations for sale? (besideso1 >Microway and Aspen, who sell them new, big $$$)?    www.islandco.com   -- Alan   ------------------------------   End of INFO-VAX 2001.394 ************************