1 INFO-VAX	Fri, 06 Jul 2001	Volume 2001 : Issue 371       Contents:1 Re: 3 Reasons why VMS is alive and probably well+ 1 Re: 3 Reasons why VMS is alive and probably well+ 1 Re: 3 Reasons why VMS is alive and probably well+ 1 Re: 3 Reasons why VMS is alive and probably well+ 1 Re: 3 Reasons why VMS is alive and probably well+ 1 Re: 3 Reasons why VMS is alive and probably well+  Re: 4mm tape drive's mechanism Re: A Primrose Path... Re: A Primrose Path... Re: A Primrose Path... RE: A Primrose Path... Re: ASI PDF Viewer Futures+ Re: Compaq ditches Alpha, what does API do? + Re: Compaq ditches Alpha, what does API do?  Re: Compaq switches to IA-64 Re: Compaq switches to IA-64( Re: Compaq switches to IA-64 (VAX float)( Re: Compaq's Alpha design team for sale? RE: DEC Net and TCP/IP Re: DECWORLD 2001 Byte article DECworld 2001? Re: DECworld 2001?3 Re: Disable DELETE, RENAME, and PURGE command verbs 3 Re: Disable DELETE, RENAME, and PURGE command verbs 3 Re: Disable DELETE, RENAME, and PURGE command verbs 3 Re: Disable DELETE, RENAME, and PURGE command verbs 3 Re: Disable DELETE, RENAME, and PURGE command verbs 3 Re: Disable DELETE, RENAME, and PURGE command verbs  DSL for OpenVMS  Re: DSL for OpenVMS  RE: DSL for OpenVMS  Experience with EMC storage  Re: FreeVMS  Re: FUD  Re: FUD  RE: FUD  Re: FUD  Re: FUD  Re: FUD   Re: Full port of VMS to Itanium.  Re: Full port of VMS to Itanium.  Re: Full port of VMS to Itanium.! Re: H-P EOLs PA-RISC Architecture 0 Re: How to (auto-re)submit batch few times a day" Re: I didn't stick it upside down!" Re: I didn't stick it upside down! 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: Need XML Parser for C/C++ / Re: OpenVMS Itanium (So *Microsoft PC* Culture) 7 OpenVMS Training Providers- Are they all created equal? ; Re: OpenVMS Training Providers- Are they all created equal? ; Re: OpenVMS Training Providers- Are they all created equal?  Re: POP server on tcpip5.1 RE: POP server on tcpip5.1 RE: POP server on tcpip5.1( Re: Probably the strangest request yet!!( RE: SSH client, FISH, -- sources anyone? Re: The Alpha/IA64 Hybrid  Re: The Alpha/IA64 Hybrid  Re: The Alpha/IA64 Hybrid  Re: The Alpha/IA64 Hybrid  Re: The Alpha/IA64 Hybrid  Re: The Alpha/IA64 Hybrid 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated 1 Re: The death of VMS has been greatly exaggerated  Re: VAX 4000-106 qbus pinout RE: VAX 4000-106 qbus pinout Re: vax 4000/90 ' VMS-style clustering comes to Linux?!?! + Re: VMS-style clustering comes to Linux?!?!  vms/pine VMSTAR question  Re: VMSTAR question  VT520 F2 key in FMS ; Re: Wailing and moaning.... (was: Compilers go to Intel...) # Re: What about performance issues?? + Re: Why Compaq won't succeed on IA64 either + Re: Why Compaq won't succeed on IA64 either + Re: Why Compaq won't succeed on IA64 either + Re: Why Compaq won't succeed on IA64 either + Re: Why Compaq won't succeed on IA64 either  Re: will the irony never cease  F ----------------------------------------------------------------------  # Date: Thu, 05 Jul 2001 18:10:29 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>: Subject: Re: 3 Reasons why VMS is alive and probably well+7 Message-ID: <p4217.2$e66.13578@typhoon.ne.mediaone.net>   7 "Arne Vajhj" <arne.vajhoej@gtech.com> wrote in message # news:3B449E39.C6682249@gtech.com...  > Bob Marcan wrote:  > > > > C'mon, Arne. Did you read about clustering on Tru64 v5.1a?; > > Did you try v5.1? They have lot inherited from the VMS.  > > > I have noy tried Tru64 clustering, but I have read about it. > 8 > You are correct, that it has inherited a lot from VMS. > C > But what wil prevent Compaq from porting it from Tru64 to Linux ?  >   L Nothing, except perhaps common sense. Why give away the core differentiator?F If you do that, you might as well just contribute the whole OS to Open Source and be done with it!    ------------------------------  $ Date: Thu, 5 Jul 2001 15:50:19 -0400- From: "Peter Weaver" <peter.weaver@stelco.ca> : Subject: Re: 3 Reasons why VMS is alive and probably well+4 Message-ID: <%x317.260109$Z2.3088563@nnrp1.uunet.ca>  ? "Terry C. Shannon" <terryshannon@mediaone.net> wrote in message 1 news:p4217.2$e66.13578@typhoon.ne.mediaone.net...  > 9 > "Arne Vajhj" <arne.vajhoej@gtech.com> wrote in message % > news:3B449E39.C6682249@gtech.com...  >...E > > But what wil prevent Compaq from porting it from Tru64 to Linux ?  > >  > ' > Nothing, except perhaps common sense.  >...  ) Like Arne said, what will prevent Compaq?    -- Peter WeaverJ Using a WIN NT/WIN 2000 box to manage your VMS systems is like towing your6 mechanic in a 5th wheel motor home behind your Porche.   ------------------------------  % Date: Thu, 05 Jul 2001 20:44:33 +0100 % From: Alan Greig <a.greig@virgin.net> : Subject: Re: 3 Reasons why VMS is alive and probably well+* Message-ID: <3B44C3A1.3AF86749@virgin.net>   "Terry C. Shannon" wrote:   9 > "Arne Vajhj" <arne.vajhoej@gtech.com> wrote in message % > news:3B449E39.C6682249@gtech.com...  > > Bob Marcan wrote:  > > @ > > > C'mon, Arne. Did you read about clustering on Tru64 v5.1a?= > > > Did you try v5.1? They have lot inherited from the VMS.  > > @ > > I have noy tried Tru64 clustering, but I have read about it. > > : > > You are correct, that it has inherited a lot from VMS. > > E > > But what wil prevent Compaq from porting it from Tru64 to Linux ?  > >  > N > Nothing, except perhaps common sense. Why give away the core differentiator?H > If you do that, you might as well just contribute the whole OS to Open > Source and be done with it!   L But Compaq have just open sourced the Tru64 clustering code  as far as I canN see. In case you think the announcement just refers to a subset look at the 50N odd powerpoint slides where it becomes obvious everything is going. The SingleL System Image code from Tru64 is going as well. See www.opensource.compaq.com  & If you *know* different please say so.     --
 Alan Greig   ------------------------------  % Date: Thu, 05 Jul 2001 21:27:33 +0200 = From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> : Subject: Re: 3 Reasons why VMS is alive and probably well+) Message-ID: <3B44BFA5.A1DA17BB@gtech.com>    "Terry C. Shannon" wrote: 9 > "Arne Vajhj" <arne.vajhoej@gtech.com> wrote in message % > news:3B449E39.C6682249@gtech.com... E > > But what wil prevent Compaq from porting it from Tru64 to Linux ?  > N > Nothing, except perhaps common sense. Why give away the core differentiator?H > If you do that, you might as well just contribute the whole OS to Open > Source and be done with it!   E It is not my impression that Tru64 are doing that well even with that  piece of VMS technology.  < If Compaq do not develop and push clustering for Linux, then> someone else will. Maybe not quite as good. But how may of the8 decision makers will understand the subtle differences ?   Arne   ------------------------------  # Date: Fri, 06 Jul 2001 01:09:42 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>: Subject: Re: 3 Reasons why VMS is alive and probably well+: Message-ID: <qd817.328$Pf6.310495@typhoon.ne.mediaone.net>  8 "Peter Weaver" <peter.weaver@stelco.ca> wrote in message. news:%x317.260109$Z2.3088563@nnrp1.uunet.ca...A > "Terry C. Shannon" <terryshannon@mediaone.net> wrote in message 3 > news:p4217.2$e66.13578@typhoon.ne.mediaone.net...  > > ; > > "Arne Vajhj" <arne.vajhoej@gtech.com> wrote in message ' > > news:3B449E39.C6682249@gtech.com...  > >...G > > > But what wil prevent Compaq from porting it from Tru64 to Linux ?  > > >  > > ) > > Nothing, except perhaps common sense.  > >... > + > Like Arne said, what will prevent Compaq?  >   ? Touche! Would you like a free side order of NSK with that, Sir?    ------------------------------  % Date: Thu, 05 Jul 2001 21:33:25 -0400 2 From: rdeininger@mindspring.com (Robert Deininger): Subject: Re: 3 Reasons why VMS is alive and probably well+L Message-ID: <rdeininger-0507012133250001@user-2ivea3q.dialup.mindspring.com>  H In article <3B449DBE.AC035245@gtech.com>, Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> wrote:   = > If the cost of a few thousand dollars is significant for an 2 > ISV to decide whether they will support VMS, ...  I A few thousand here, a few thousand there, and pretty soon you're talking  about real money.   H There is always a price level that will cause pain and suffering, and itG varies from one person to the next.  But the cost of license generation H has been causing a lot of folks pain for a long time.  Software PartnersF has been in the VMS software business for a couple of decades, more orJ less.  (I believe they were the example named earlier in this thread.) LMFJ was too expensive for them when they started, and may still be.  Thus theyF and their customers still have to deal with a non-standard mechanism. 8 Silly.  There are better ways for Digipaq to earn money.  F Another Silly Price Scheme can be seen in the Software Product Library CDs.  Why drive customers away?    --   Robert Deininger rdeininger@mindspring.com    ------------------------------  % Date: Thu, 05 Jul 2001 16:07:40 -0500 / From: Chris Scheers <chris@applied-synergy.com> ' Subject: Re: 4mm tape drive's mechanism 3 Message-ID: <3B44D71C.CEB18CC8@applied-synergy.com>    David Lee wrote: > N > I did try to power cycle the drive.  At first, it seemed to work.  The lightM > indicator went off.  But after tape backup is done, I tried dismounting it, 	 > but the F > tape got stuck and the busy light lid up again.  It seemed like some > mechanism L > went bad or something.  Does anybody know what caused this? and how to fix > it?  > Thanks  7 Are you saying that you have a tape stuck in the drive?   F If so, on many DAT drives, if you hold down the eject button while youF power up the drive, the drive will try to eject the tape without doingC anything else.  (On some drives, you seem to need to hold the eject & button down until the tape comes out.)  F If this does not eject the tape, the drive is almost certainly faulty.  
 Good luck!  G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com     Fax: 817-237-3074    ------------------------------  % Date: Thu, 05 Jul 2001 20:16:16 +0100 L From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com> Subject: Re: A Primrose Path... 3 Message-ID: <3B44BD00.43861462@remove_this.sun.com>    Alphaman wrote:  > J > You may not be Devo, and as an comp.sys.OpenVMS visitor, I'd have to sayM > that you are in favor of evo.  But the Q is Devo.  Devolve Digital, Devolve  > AlphaNT, Devolve Alpha...  >  > Yes, what's next?  >    IA-64man ???   Regards  Andrew Harrison  Enterprise IT Architect    ------------------------------  # Date: Thu, 05 Jul 2001 19:30:07 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net> Subject: Re: A Primrose Path... 8 Message-ID: <3f317.73$Pf6.66383@typhoon.ne.mediaone.net>  H "Andrew Harrison SUNUK Consultancy" <ah13473_nospam@remove_this.sun.com>> wrote in message news:3B44BD00.43861462@remove_this.sun.com... >  > Alphaman wrote:  > > L > > You may not be Devo, and as an comp.sys.OpenVMS visitor, I'd have to sayG > > that you are in favor of evo.  But the Q is Devo.  Devolve Digital,  Devolve  > > AlphaNT, Devolve Alpha...  > >  > > Yes, what's next?  > >  >  > IA-64man ???  J Maybe IPFman. With AlphaMan meeting his Omega, there's only two other guys- left in the arena now: POWERman and SPARCman.    ------------------------------  % Date: Thu, 05 Jul 2001 16:40:59 -0300 ) From: fabio_compaq@ep-bc.petrobras.com.br  Subject: Re: A Primrose Path... L Message-ID: <OF828ACEFA.5872850A-ON03256A80.006BED8A@ep-bc.petrobras.com.br>  A In my personal opinion..... Intel should forget this  64 bit chip > and develop the  IA-128. IA-64 is insist in what exist: Alpha.     Regards    FC        I Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com> em  05/07/2001 16:16:16   3 Favor responder a Andrew Harrison SUNUK Consultancy *       <ah13473_nospam@remove_this.sun.com>             Info-VAX@Mvb.Saic.Com        Assunto: Re: A Primrose Path...        Alphaman wrote:  > J > You may not be Devo, and as an comp.sys.OpenVMS visitor, I'd have to sayE > that you are in favor of evo.  But the Q is Devo.  Devolve Digital,  Devolve  > AlphaNT, Devolve Alpha...  >  > Yes, what's next?  >    IA-64man ???   Regards  Andrew Harrison  Enterprise IT Architect    ------------------------------  % Date: Thu, 05 Jul 2001 14:59:32 -0500 + From: Christopher Smith <csmith@amdocs.com>  Subject: RE: A Primrose Path... L Message-ID: <3B55D7F383B0D31197D9009027541CBF0D9D203B@cmiexch1.cmi.itds.com>   > -----Original Message-----; > From: Terry C. Shannon [mailto:terryshannon@mediaone.net]   & > "Andrew Harrison SUNUK Consultancy"    > > IA-64man ???  > > Maybe IPFman. With AlphaMan meeting his Omega, there's only  > two other guys/ > left in the arena now: POWERman and SPARCman.   ! I don't know... how about Z80man?    Regards,   Chris     ! Christopher Smith, Perl Developer  Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");  '       ------------------------------  % Date: Thu, 05 Jul 2001 16:37:32 -0500 / From: Chris Scheers <chris@applied-synergy.com> # Subject: Re: ASI PDF Viewer Futures 3 Message-ID: <3B44DE1C.A1F00F1C@applied-synergy.com>    Robert Deininger wrote:  > B > In article <3B423CEF.94D842A@applied-synergy.com>, Chris Scheers$ > <chris@applied-synergy.com> wrote: > F > > If you are interested in the ASI PDF Viewer or Ghostscript on VMS, > > please see > > @ > >         http://www.applied-synergy.com/pdf/announcement.html > >  >  > Informative but terse. > ) > Any projections on the release of V2.0?   F The impact of creating a shareable image from Ghostscript is currently= being evaulated.  Ghostscript is (mostly) designed for source A compatibility, but isn't really designed for binary compatibility  between versions.   G Once this has been done, I should be able to get back to the group with 
 a projection.   G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com     Fax: 817-237-3074    ------------------------------  % Date: Thu, 05 Jul 2001 17:44:51 -0600  From: yyyc186@mindspring.com4 Subject: Re: Compaq ditches Alpha, what does API do?; Message-ID: <3b44fc3e$1$lllp186$mr2ice@nntp.mindspring.com>   A In <pA0%6.457$9r6.450147@typhoon.ne.mediaone.net>, on 06/29/2001  D    at 02:49 PM, "Terry C. Shannon" <terryshannon@mediaone.net> said:  F I wouldn't bank on VMS getting ported to any other chip.  You have theD nations number one gangster G.Q. Bob sitting on the board...modeling8 himself after Pretty Boy Floyd...just without the banks.   Roland    G >"Bob Kaplow" <kaplow_r@eisner.encompasserve.org.mars> wrote in message . >news:qc0mSGmaDP3W@eisner.encompasserve.org...M >> In article <Pine.SGI.4.21.0106251605080.16346-100000@world.std.com>, Terry * >C Shannon <shannon@world.std.com> writes:@ >> > The EV7 team will put the finishing touches on EV7 and it's >architectural >> > shrink. >> >F >> > Intel now has the EV8 team. EV8 will not be developed; Intel will9 >> > incorporate Alpha technology in post-McKinley chips.  >> > >> >> < >> >> Anyone care to guess what happens to API in this mess? >> >= >> > API NetWorks, ne API, will shuffle off its mortal coils.  >>M >> Another interesting thought: Intel runs their FAB production lines on VMS,eJ >> and is right now converting them from VAX to Alpha. I wonder what Intel >willlF >> do in the future? Build more Alpha chips, just for their own use???  F >Much of the conversion to Alpha is done now. It's not a problem IntelH >will have to worry about for years to come. Assuming that VMS is ported. >to IA-64, it shouldn't be an issue for Intel.       -- i; -----------------------------------------------------------. yyyc186@mindspring.com; -----------------------------------------------------------p   ------------------------------  # Date: Fri, 06 Jul 2001 01:10:48 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>4 Subject: Re: Compaq ditches Alpha, what does API do?: Message-ID: <se817.329$Pf6.311457@typhoon.ne.mediaone.net>  ) <yyyc186@mindspring.com> wrote in messagew5 news:3b44fc3e$1$lllp186$mr2ice@nntp.mindspring.com...AB > In <pA0%6.457$9r6.450147@typhoon.ne.mediaone.net>, on 06/29/2001F >    at 02:49 PM, "Terry C. Shannon" <terryshannon@mediaone.net> said: >iH > I wouldn't bank on VMS getting ported to any other chip.  You have theF > nations number one gangster G.Q. Bob sitting on the board...modeling: > himself after Pretty Boy Floyd...just without the banks.   But his hair is PERFECT!   ------------------------------  % Date: Thu, 05 Jul 2001 14:17:11 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>e% Subject: Re: Compaq switches to IA-64l, Message-ID: <3B44AF26.AF9C6437@videotron.ca>   Vance Haemmerle wrote:I >   You mean from Alpha to IPF?  Unlike the Tru64 side, VMS on Alpha used K > F and G floats by default, and D-floats with a compile switch. I see lotsM> > of VMS users with datafiles having floats in the VAX format.  6 Do G F and IEEE floats take the same number of bytes ?  K If so, then wouldn't it a question of a one off conversion of the data fileMJ with the flot fields convertyed to IEEE binary format and stored "in situ"M with the IA?$ application then just reading a binary value in teh same recorda; location which happens to be in a format IA64 understands ?r  L Would it be a fair assumption that such applications are much more likely toH be home grown with the customers having source code they can recompile ?   ------------------------------  % Date: Thu, 05 Jul 2001 21:32:01 +0200H= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com> % Subject: Re: Compaq switches to IA-64x) Message-ID: <3B44C0B1.9BCEBDF9@gtech.com>r   JF Mezei wrote:  > Vance Haemmerle wrote:K > >   You mean from Alpha to IPF?  Unlike the Tru64 side, VMS on Alpha used M > > F and G floats by default, and D-floats with a compile switch. I see lotse@ > > of VMS users with datafiles having floats in the VAX format. > 8 > Do G F and IEEE floats take the same number of bytes ? > M > If so, then wouldn't it a question of a one off conversion of the data filerL > with the flot fields convertyed to IEEE binary format and stored "in situ"O > with the IA?$ application then just reading a binary value in teh same recordA= > location which happens to be in a format IA64 understands ?c > N > Would it be a fair assumption that such applications are much more likely toJ > be home grown with the customers having source code they can recompile ?  F It is not difficult to convert a file. But it adds to migration costs.  D And there are the issue of FP calculations. A program giving certainB results using G-floats may not necesarrily return the same resultsB using T-floats. In most cases it does not matter. But there may be@ places where heavy testing/certifcation may be required. It also adds to migration costs.   Arne   ------------------------------  % Date: Thu, 05 Jul 2001 14:31:44 -0400m- From: "Richard D. Piccard" <piccard@ohio.edu>>1 Subject: Re: Compaq switches to IA-64 (VAX float)b( Message-ID: <3B44B281.7FE79B3A@ohio.edu>   Nigel Arnot wrote:   > [snip]  J > I also wonder why anyone really wants non-IEEE float, except as a way ofH > accessing old files or writing new ones compatible with an old format,D > or just possibly to talk to a VAX. All these are best addressed byD > converting the float data as you read or write it, and letting the# > rest of the code use native IEEE.c  I An example where bit-for-bit replication might matter:  the FAA-certifiedrN simulation software for ground-control radio antenna radiation patterns, whichN is used to choose locations for the antennas, given local terrain and buildingL locations, to ensure that all places you can taxi a plane will have adequateN signal.  My understanding is that they require re-certification by replicatingM the accepted results for a suite of input data combinations.   I would expectyC other examples in the field of cancer radiation treatment planning.y  N Not that you couldn't do it with IEEE format floating number calculations, butJ rather that having guaranteed bit-for-bit identical results would speed upP re-certification, since you wouldn't have to decide whether the differences were0 significant if there were NO differences at all.  '                                     RDPm   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  % Date: Fri, 06 Jul 2001 09:21:05 +0010s% From: paddy.o'brien@zzz.tg.nsw.gov.aue1 Subject: Re: Compaq's Alpha design team for sale?C5 Message-ID: <01K5LTCCH6UA002DPB@tgmail.tg.nsw.gov.au>c   Alan Greig wrote:g  D >On Thu, 05 Jul 2001 19:46:27 +0010, paddy.o'brien@zzz.tg.nsw.gov.au >wrote:' >, >nJ >>With the Compaq compiler guys going to Inhel, what is being supported?   >Steve oM >>Lionel seems to be one of the first to go.  Will I still have support from hD >>fortran@compaq.com?  Will compaq VMS (and T64) compilers still be  >maintained? > 8 >Steve is going to Intel??  Can you provide a reference?  L Over on comp.lang.fortran, see the thread "Intel to acquire Compaq Fortran"  started by SL on 25-Jun.  M www.mailgate.org/comp/comp.lang.fortran/msg14477.html  for his first message.r  J If you click on Index, you'll get the whole thread with three other posts  from Steve.    Regards, Paddy   ------------------------------   Date: 05 Jul 2001 23:37:48 GMT' From: prosullivan@aol.com (PROSULLIVAN)c Subject: RE: DEC Net and TCP/IP : Message-ID: <20010705193748.18341.00002750@ng-fl1.aol.com>  B I guess you have flushed the session control cache and deleted theN corresponding database file. if you don't delete it, a decnet re-start reloads? the old values. Have you also tried the old set ho scsid trick?o   ------------------------------    Date: 05 Jul 2001 13:11:56 -1000/ From: Jim Thomas <thomas@atlas.cfht.hawaii.edu> ' Subject: Re: DECWORLD 2001 Byte articleh2 Message-ID: <wwhewrm0n7.fsf@atlas.cfht.hawaii.edu>  G >>>>> "Henry" == Henry W Miller <henry.miller@worldnet.att.net> writes:r  :  Henry> "Alan Greig" <a.greig@virgin.net> wrote in message:  Henry> news:uiu5ktg8e6r5a0e1hvotsj71tbmeg4lqjm@4ax.com...C  >> http://www.byte.com/documents/s=716/byt20010622s0005/greer.html   '  Henry>     Ain't technology wonderful?y  1 Yeah, all 15 or so cookies to get to the article.t   Nothead    ------------------------------  $ Date: Thu, 5 Jul 2001 20:50:50 -0400) From: "Neil Rieck" <n.rieck@sympatico.ca>e Subject: DECworld 2001?t; Message-ID: <rT717.26543$g92.3398800@news20.bellglobal.com>    subject: DECworld 2001?-  & You've got to read this to believe it.? http://www.byte.com/documents/s=716/byt20010622s0005/greer.html     
 Neil Rieck Kitchener/Waterloo/Cambridge,b Ontario, Canada.! http://www3.sympatico.ca/n.rieck/.@ http://www3.sympatico.ca/n.rieck/links/compaq_memorial_site.html   ------------------------------  # Date: Fri, 06 Jul 2001 01:13:16 GMTh4 From: "Terry C. Shannon" <terryshannon@mediaone.net> Subject: Re: DECworld 2001?o: Message-ID: <Mg817.332$Pf6.314309@typhoon.ne.mediaone.net>  4 "Neil Rieck" <n.rieck@sympatico.ca> wrote in message5 news:rT717.26543$g92.3398800@news20.bellglobal.com...m > subject: DECworld 2001?u >t( > You've got to read this to believe it.A > http://www.byte.com/documents/s=716/byt20010622s0005/greer.html  >v  - Believe it! Best event I've attended in ages!    ------------------------------   Date: 5 Jul 2001 14:37:31 -0700 - From: afeldman@gfigroup.com (Alan E. Feldman) < Subject: Re: Disable DELETE, RENAME, and PURGE command verbs= Message-ID: <af1e4ce6.0107051337.40464674@posting.google.com>n  a skulker <skulker@mailops.com.yourpants> wrote in message news:<_u%07.2$O_5.1717@nnrp3.sbc.net>...  > Hello all, > G > I was wondering if there is a quick and easy way to do the following:t > W > One of our customers wants to take away the DELETE, RENAME, and PURGE commnads from a _ > subset of users (long story!). He wants to replace the command with a short message that says g > something to the effect "Call me". I have advised this customer against this - but he pays the bills!e  D Assuming your bad users have only normal privileges (I can't imagine; by your post that they have more!) set the protections on $MA RENAME.EXE, DELETE.EXE, PURGE.EXE so that the bad users can't runi those files.  A My test case on V6.2: I created a user with normal privs and gave  RENAME.EXE W=nothing:    $ DIR/DATE/SIZET    Directory SYS$SYSDEVICE:[FELD]  e5 A.B;1                      0   5-JUL-2001 21:27:59.24     Total of 1 file, 0 blocks.% $ DIR/DATE/SIZE SYS$SYSTEM:RENAME.EXEe  n Directory SYS$COMMON:[SYSEXE]t  b> RENAME.EXE;1       insufficient privilege or object protection	 violationr  n Total of 1 file, 0 blocks.
 $ REN A.B B.C-. %DCL-W-ACTIMAGE, error activating image RENAMED -CLI-E-IMGNAME, image file DSA0:[SYS0.SYSCOMMON.][SYSEXE]RENAME.EXE;D -RMS-F-NOPRIV, insufficient privilege or object protection violation $   B You may have to use ACL's or other means to grant privilege to the= RENAME.EXE, etc., to isolate the offending users. Be careful!t  / Just one idea that may or may not work for you.   B As far as replacing it with a message, you'll have to do something  with symbols for those commands.   Disclaimer: JMHO     &-) Alan E. Feldmane afeldman&gfigroup.com    ------------------------------   Date: 5 Jul 2001 15:11:11 -0700d- From: afeldman@gfigroup.com (Alan E. Feldman)d< Subject: Re: Disable DELETE, RENAME, and PURGE command verbs= Message-ID: <af1e4ce6.0107051411.41d991fa@posting.google.com>   a skulker <skulker@mailops.com.yourpants> wrote in message news:<6v%07.3$O_5.1823@nnrp3.sbc.net>...o > Hello all, > G > I was wondering if there is a quick and easy way to do the following:- > W > One of our customers wants to take away the DELETE, RENAME, and PURGE commnads from an_ > subset of users (long story!). He wants to replace the command with a short message that says g > something to the effect "Call me". I have advised this customer against this - but he pays the bills!o  E I can't reply to my previous post right now, but I want to add that IaF certainly don't recommend -- at least not "in general" -- that you setE PROT=W on .EXE's in SYS$SYSTEM. I just did that as an example to show,D that the protection idea works. I'd use ACL's to shut out bad users.   Something like e  ; $ SET ACL /ACL=(IDENTIFIER=BADUSERS,ACCESS=NONE) RENAME.EXE   F will prevent BADUSERS from running RENAME.EXE (assuming they have onlyA normal privileges). You can do the same for DELETE.EXE. Since the0F PURGE command runs DELETE.EXE, that should be it for preventing use ofB those commands. As for the "Call me" message, that's trickier. TheB "separate command tables for those users" idea may work better forD that. I suspect they'd call you even if all that happens is that the> "restricted" commands don't work. Need to know more about your situation to decide that.p  C Another thing you can do is to add something in SYLOGIN.COM that istD run only for the bad users that says something like "DELETE, RENAME,E and PURGE have been disabled for your account. If you need them, callP" me." Seems kind of strange though.   Disclaimer: JMHO     &-) Alan E. Feldmanr afeldman&gfigroup.comP   ------------------------------   Date: 5 Jul 2001 18:40:19 CDT1= From: wayne@tachysoft.xxx.320117.killspam.015d (Wayne Sewell)u< Subject: Re: Disable DELETE, RENAME, and PURGE command verbs. Message-ID: <BukCciEuyv6c@tachxxsoftxxconsult>  \ In article <6v%07.3$O_5.1823@nnrp3.sbc.net>, skulker <skulker@mailops.com.yourpants> writes: > Hello all, > G > I was wondering if there is a quick and easy way to do the following:s > W > One of our customers wants to take away the DELETE, RENAME, and PURGE commnads from an_ > subset of users (long story!). He wants to replace the command with a short message that saysXg > something to the effect "Call me". I have advised this customer against this - but he pays the bills!, > _ > I am planning on these users executing a specific login command procedure that this "point ofe_ > contact" customer will maintain for his subset of users. I know the verbs can be removed fromA > their tables with a  >  > $ SET COMMAND/DELETE=DELETEh > # > and then creating a global symbol  >  > $ DEL*ETE == "Whatever"3 > c > This works OK unless you add a qualifier, such as /LOG. Then you get the "unrecognized qualifier"5 > error. > c > Anyone have a hint on how to handle this? Is there something already out there that will do this?n >     F Well, it depends on the actual value of whatever, but you could do the
 following:  * $ delete = "write sys$output ""call joe""" $ delete call joe
 $ delete /log = %DCL-W-UNDSYM, undefined symbol - check validity and spelling   \LOG\, $ delete = "write sys$output ""call joe"" !"
 $ delete /logp call joe $r  O By adding a comment character to the end of the symbol, you can make it eat the" rest of the line.a   -- .O ===============================================================================1M Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxxA: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-)4O =============================================================================== ? Flounder: "I can't believe I threw up in front of Dean Wormer." >    Otter: "Face it, Flounder.  You threw up *on* Dean Wormer."   ------------------------------  # Date: Fri, 06 Jul 2001 00:32:36 GMTn) From: rob.buxton@wcc.govt.nz (Rob Buxton)h< Subject: Re: Disable DELETE, RENAME, and PURGE command verbs/ Message-ID: <3b450553.2045791@news.wcc.govt.nz>   ? A variation on the theme is to create a separate version of theb8 DCLTables minus the commands specified. say short_tables  A Once done the Users UAF Entry is modified so that they access theiA short_tables. This prevents them easily being able to subvert the- mechanism.    ; Other posters have detailed the ! at the end of the command : substitution which should address the other problem noted.    ) On Thu, 05 Jul 2001 15:14:02 GMT, skulkere& <skulker@mailops.com.yourpants> wrote:   >Hello all,p >gF >I was wondering if there is a quick and easy way to do the following: >4V >One of our customers wants to take away the DELETE, RENAME, and PURGE commnads from a^ >subset of users (long story!). He wants to replace the command with a short message that saysf >something to the effect "Call me". I have advised this customer against this - but he pays the bills! >.^ >I am planning on these users executing a specific login command procedure that this "point of^ >contact" customer will maintain for his subset of users. I know the verbs can be removed from >their tables with a >s >$ SET COMMAND/DELETE=DELETE >." >and then creating a global symbol >l >$ DEL*ETE == "Whatever" >nb >This works OK unless you add a qualifier, such as /LOG. Then you get the "unrecognized qualifier" >error.t >lb >Anyone have a hint on how to handle this? Is there something already out there that will do this? >r6 >Oh yeah, this is a 5.5-2 system (another long story)! >> >Thanks in advance!  >g >skulker  >email replies remove .yourpants >>   ------------------------------  % Date: Thu, 05 Jul 2001 21:50:16 -0400r2 From: rdeininger@mindspring.com (Robert Deininger)< Subject: Re: Disable DELETE, RENAME, and PURGE command verbsL Message-ID: <rdeininger-0507012150160001@user-2ivea3q.dialup.mindspring.com>  M In article <3b450553.2045791@news.wcc.govt.nz>, rob.buxton@wcc.govt.nz wrote:   A > A variation on the theme is to create a separate version of the.: > DCLTables minus the commands specified. say short_tables  I The downside here is the DCL table gets stale.  Any product or OS install-H is likely to update the real version, and then you need to copy, modify,( and re-install it for the special users.  C > Once done the Users UAF Entry is modified so that they access theCC > short_tables. This prevents them easily being able to subvert thel > mechanism.    H If they know how to write a CLD and SET COMMAND, they can make new verbsI that access the orgininal DELETE, etc. images.  So protecting those filesi  against access is also required.  I Personally, I think the best solution for users who can't be trusted withCI their own files is DISUSER.  But some sites do have special requirements.g   -- p Robert Deininger rdeininger@mindspring.comi   ------------------------------  % Date: Thu, 05 Jul 2001 21:09:03 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> < Subject: Re: Disable DELETE, RENAME, and PURGE command verbs' Message-ID: <3B451DBF.55558E7C@fsi.net>r   skulker wrote: >  > Hello all, > G > I was wondering if there is a quick and easy way to do the following:4 > W > One of our customers wants to take away the DELETE, RENAME, and PURGE commnads from a _ > subset of users (long story!). He wants to replace the command with a short message that sayslg > something to the effect "Call me". I have advised this customer against this - but he pays the bills!a > _ > I am planning on these users executing a specific login command procedure that this "point of _ > contact" customer will maintain for his subset of users. I know the verbs can be removed fromt > their tables with a  >  > $ SET COMMAND/DELETE=DELETEo > # > and then creating a global symbol6 >  > $ DEL*ETE == "Whatever"  > c > This works OK unless you add a qualifier, such as /LOG. Then you get the "unrecognized qualifier"  > error. > c > Anyone have a hint on how to handle this? Is there something already out there that will do this?  > 7 > Oh yeah, this is a 5.5-2 system (another long story)!j  G Well, after you explain to the person who pays the bills how badly thismD request can "break" his system, you'll want to go read up on OpenVMSE Security. You should be able to set proper protections, ACEs, etc. onlA files and directories to prevent wayward users from doing serious  damage.u  H If the "subset of users" share a common UIC group, you might try settingF a group level logical for each of DELETE and RENAME (the PURGE commandE uses the DELETE image) such that they point to a program which issuesl@ the message - the program should effectively ignore command lineE qualifiers and parameters by simply not examining them - just displayA the message and exit.    Example:< $ DEFINE/TABLE=LNM$GROUP_000200 DELETE ddcu:<dir>CALL_ME.EXE  H If the "subset of users" do not share a common UIC group, then try using a job level logical instead.   Example:) $ DEFINE/JOB DELETE ddcu:<dir>CALL_ME.EXEa  D Again, I recommend neither approach - but you can try one or both ofG them to show the boos how bad an idea this really is. Better to enforcec1 proper security at the file and directory levels.n   ...IMHO.   --   David J. Dachterar dba DJE Systemsw http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/-  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------   Date: 5 Jul 2001 14:35:56 -0700i1 From: jason_odonnell@erinet.com (Jason O'Donnell)j Subject: DSL for OpenVMS= Message-ID: <5c8ffd05.0107051335.663e7d03@posting.google.com>t  B Does anybody have OpenVMS drivers for any DSL PCI card?  I want to( access the internet from my OpenVMS box.   JMOD   ------------------------------  # Date: Fri, 06 Jul 2001 00:49:33 GMTn6 From: "Andy Bustamante" <a_c_bustamante@earthlink.net> Subject: Re: DSL for OpenVMSD Message-ID: <xW717.4290$G_1.416886@newsread2.prod.itd.earthlink.net>  I Is this an internal PCI DSL modem?  You probably won't find a VMS driver.tC You may be able to find a Linux driver and setup the Intel box as as firewall/router.  J If you have an external modem and need a PPPOE interface I'd recommend youK pick up a Linksys or Netgear DSL router with DSL/Ethernet connections.  YoumF can plug this into your DSL modem and use a native TCPIP stack on yourL system (ucx, tcpip v5x, multinet . . . ).  You can get these routers on sale% for under $100 if you look carefully. H These standalones supposedly have better throughput than the Win32 PPPOE( clients as well, so that's another plus.     --   Andy Bustamante=: Hidden from the wild wild web on a Red Hat router/firewall( remove the ascii-95's to reply by e-mail      > "Jason O'Donnell" <jason_odonnell@erinet.com> wrote in message7 news:5c8ffd05.0107051335.663e7d03@posting.google.com...iD > Does anybody have OpenVMS drivers for any DSL PCI card?  I want to* > access the internet from my OpenVMS box. >e > JMOD   ------------------------------  % Date: Thu, 05 Jul 2001 18:08:22 -0700P! From: Tom Linden <tom@kednos.com>e Subject: RE: DSL for OpenVMS9 Message-ID: <CIEJLCMNHNNDLLOOGNJICENGCOAA.tom@kednos.com>e  F Well, I have an Alcatel 1000 ADSL that I am willing to part with sinceF I can't use it here, had to get a T1.  its an external unit that plugsL in to your hub and to the phone line.  Also have a POTS splitter if you need one. All in original box with docs.% Send me mail privately if interested.    Tom    > -----Original Message-----= > From: Andy Bustamante [mailto:a_c_bustamante@earthlink.net]t' > Sent: Thursday, July 05, 2001 5:50 PMg > To: Info-VAX@Mvb.Saic.Com4 > Subject: Re: DSL for OpenVMS >  >yK > Is this an internal PCI DSL modem?  You probably won't find a VMS driver.SE > You may be able to find a Linux driver and setup the Intel box as aJ > firewall/router. >rL > If you have an external modem and need a PPPOE interface I'd recommend you; > pick up a Linksys or Netgear DSL router with DSL/Ethernetu > connections.  YoutH > can plug this into your DSL modem and use a native TCPIP stack on your> > system (ucx, tcpip v5x, multinet . . . ).  You can get these > routers on sale ' > for under $100 if you look carefully.-J > These standalones supposedly have better throughput than the Win32 PPPOE* > clients as well, so that's another plus. >C >L > -- >t > Andy Bustamante-< > Hidden from the wild wild web on a Red Hat router/firewall* > remove the ascii-95's to reply by e-mail >  >t >r@ > "Jason O'Donnell" <jason_odonnell@erinet.com> wrote in message9 > news:5c8ffd05.0107051335.663e7d03@posting.google.com...>F > > Does anybody have OpenVMS drivers for any DSL PCI card?  I want to, > > access the internet from my OpenVMS box. > >  > > JMOD >i >t   ------------------------------  % Date: Thu, 05 Jul 2001 22:41:45 -0700e. From: "Koloth(Telocity)" <koloth@telocity.com>$ Subject: Experience with EMC storage, Message-ID: <3B454F99.3FC72F60@telocity.com>  E We are investigating EMC as a possibility for storage for our OpenVMStE Alpha system.  We use DSM (Mumps) database.  Does anyone have good or0. bad experiences that they would like to share?   TIAl   Cass Witkowski   ------------------------------  % Date: Thu, 05 Jul 2001 22:40:45 +0100a+ From: "antonio.carlini" <arcarlini@iee.org>- Subject: Re: FreeVMS' Message-ID: <3B44DEDD.7CB0F580@iee.org>    Bill Gunshannon wrote: >  [others wrote:]dK > > : Disassembling VMS and using that information in any way would violatey > >.0 > > That would still be legal in some countries. > E > Yeah, the same countries that allow bootlegging music CDs and DVDs..E > All countries who are signatories of the Berne Convention recognizeeJ > the legal copyrights of other member countries.  That pretty much covers > the civilized world.  4 Reverse engineering is (I think) permitted in Europe0 under some circumstances. I thought that was the case in the US too?A   Just mildly curious.   Antoniob   -- T   ---------------E- Antonio Carlini             arcarlini@iee.org7   ------------------------------  % Date: Thu, 05 Jul 2001 20:21:03 +0100rL From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com> Subject: Re: FUD3 Message-ID: <3B44BE1F.146D320D@remove_this.sun.com>    Christof Brass wrote:s >  > Alphaman wrote:  > >E4 > > Dean Woodward <deanw@rdrop.com> wrote in message' > > news:3B37D8C1.7817B9C4@rdrop.com...cC > > > Um, guys? Anyone notice that Andrew hasn't had a single thinglG > > > to say about the Alpha -> Intel deal?  He doesn't need to, you'reeF > > > all doing his job for him, as usual.  What with Sun and layoffs,7 > > > he may soon fear his own self becoming redundant.o > > J > > He doesn't have to.  Perchance he is a gentleman and is wise enough to  > > respect our mourning period? > > 5 > > Nah, I suspect we'll hear from him shortly... ;^)g > > 	 > > AaronS > > --B > > Aaron Sakovich  http://members.home.net/sakovich/alphaman.htmlB > > Make April 15 just another day:        http://www.fairtax.org/L > > "The supersonic boom should hit you in just a few seconds." (Apollo 440) > > > There is nothing to be happy from SUN people's point of view@ > because SUN will abandon SPARC very soon and Slowaris a little@ > bit later. All what is posted here from SUN people can be used > against the them later..    5 You keep making this claim, you always seem to forgetc3 to explain what leads you to reach this conclusion.h  7 This is the chance you have been waiting for to explaino5 why you reached this conclusion, don't dissapoint as   you have in the past.    Regards  Andrew Harrisonc Enterprise IT Architectu   ------------------------------  % Date: Thu, 05 Jul 2001 20:30:17 +0100ML From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com> Subject: Re: FUD3 Message-ID: <3B44C049.401D8648@remove_this.sun.com>o   "David J. Dachtera" wrote: >  > Dean Woodward wrote: > >PA > > Um, guys? Anyone notice that Andrew hasn't had a single thingnE > > to say about the Alpha -> Intel deal?  He doesn't need to, you'reaD > > all doing his job for him, as usual.  What with Sun and layoffs,5 > > he may soon fear his own self becoming redundant.f > G > Perhaps that is *EXACTLY* what happened. (How's *THAT* for optimism?)  >   4 Nope I am still here. Personnaly I am not supprised 4 that Compaq have ditched Alpha, it didn't make sense if you are a bean counter.  7 The fact that the bean counters ensured that it didn't h; make sense is probably what is really annoying most people p on this group.  8 Incedentally Sun hasn't laid anyone off to my knowledge,< though hoping that someone who you couldn't win an argument < with because you had a sneeking suspicion that he was right 9 all along had been silenced because he had been laid off o5 is pretty sad. I would have expected better from you.C     RegardsN Andrew Harrisonp Enterprise IT Architecti   ------------------------------  % Date: Thu, 05 Jul 2001 15:03:50 -0500s+ From: Christopher Smith <csmith@amdocs.com>  Subject: RE: FUDL Message-ID: <3B55D7F383B0D31197D9009027541CBF0D9D203C@cmiexch1.cmi.itds.com>   > -----Original Message-----) > From: Andrew Harrison SUNUK Consultancya  9 > The fact that the bean counters ensured that it didn't  = > make sense is probably what is really annoying most people   > on this group.  J If I were most people in this group I'd take this as a compliment.  Yes, II think we can safely say that most of our participants are less than happyt$ with bean counters most of the time.   Regards,   Chrisr    ! Christopher Smith, Perl Developer  Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");- '-   ------------------------------  $ Date: Thu, 5 Jul 2001 17:05:52 -0400' From: "Bill Todd" <billtodd@foo.mv.com>e Subject: Re: FUD( Message-ID: <9i2kkb$559$1@pyrite.mv.net>  H "Andrew Harrison SUNUK Consultancy" <ah13473_nospam@remove_this.sun.com>> wrote in message news:3B44C049.401D8648@remove_this.sun.com...   ...e  8 > The fact that the bean counters ensured that it didn't< > make sense is probably what is really annoying most people > on this group.   ...E  I I snipped the balance, which was just typical Andrew.  But he does have a E knack for occasionally hitting the nail squarely on the head (well, InK believe Compaq's bean-counters were in cahoots with its Classic contingent,  but the essence is there).   - bill   ------------------------------  # Date: Fri, 06 Jul 2001 02:47:13 GMTi4 From: "Terry C. Shannon" <terryshannon@mediaone.net> Subject: Re: FUD: Message-ID: <RE917.416$vb6.198965@typhoon.ne.mediaone.net>  H "Andrew Harrison SUNUK Consultancy" <ah13473_nospam@remove_this.sun.com>> wrote in message news:3B44C049.401D8648@remove_this.sun.com... >a > "David J. Dachtera" wrote: > >g > > Dean Woodward wrote: > > > C > > > Um, guys? Anyone notice that Andrew hasn't had a single thingmG > > > to say about the Alpha -> Intel deal?  He doesn't need to, you're F > > > all doing his job for him, as usual.  What with Sun and layoffs,7 > > > he may soon fear his own self becoming redundant.  > >eI > > Perhaps that is *EXACTLY* what happened. (How's *THAT* for optimism?)\ > >  >e5 > Nope I am still here. Personnaly I am not supprisedp6 > that Compaq have ditched Alpha, it didn't make sense > if you are a bean counter. > 8 > The fact that the bean counters ensured that it didn't< > make sense is probably what is really annoying most people > on this group.  J You have a point about the economic futility. Assume CPQ ships 100K AlphasL per year. Assume (being generous) an average config has 5 CPUS. If the AlphaK effort cost CPQ $250M per year, each CPU they shipped would have a $500 R&D= overhead surtax. Ouch.   ------------------------------  $ Date: Thu, 5 Jul 2001 23:54:27 -0400' From: "Bill Todd" <billtodd@foo.mv.com>d Subject: Re: FUD( Message-ID: <9i3cig$mun$1@pyrite.mv.net>  ? "Terry C. Shannon" <terryshannon@mediaone.net> wrote in messagen4 news:RE917.416$vb6.198965@typhoon.ne.mediaone.net... >RJ > "Andrew Harrison SUNUK Consultancy" <ah13473_nospam@remove_this.sun.com>@ > wrote in message news:3B44C049.401D8648@remove_this.sun.com...   ...   : > > The fact that the bean counters ensured that it didn't> > > make sense is probably what is really annoying most people > > on this group. >eL > You have a point about the economic futility. Assume CPQ ships 100K AlphasH > per year. Assume (being generous) an average config has 5 CPUS. If the AlphahI > effort cost CPQ $250M per year, each CPU they shipped would have a $500u R&Dm > overhead surtax. Ouch.  4 Of course, there's the other point of view, as well:  F Let's say that half the $4 billion annual VMS revenue was from serviceK (though that seems kind of high, so feel free to correct it if it is.  ThatoL leaves about $2 billion in annual revenue from VMS system sales alone, whichJ distributed over 50 K systems (since someone recently suggested that Tru64J sold a similar number) - or a bit fewer (since at least a few were sold toA run Linux) - works out to an average of at least $40K per system.r  J If that average system contained 5 CPUs, as you assumed, that 'developmentL tax' imposed by the lowish Alpha run rate was about 6% of the overall systemK price - hardly a definitive difference, and one that could easily have been G halved or quartered with anything like adequate marketing and resulting  increased penetration.  K Or, looking at it from the viewpoint of the lowly DS10, which was something L like a $6K minimal system last I knew, the 'development tax' was about 8% ofF the system cost (and, again could have been far less with even minimal marketing).-  F The only 'futile' aspect was getting Compaq to market Alpha (and VMS):B there was no intrinsic problem with handling the development cost.   - bill   ------------------------------  % Date: Fri, 06 Jul 2001 00:02:11 +0100.L From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com>) Subject: Re: Full port of VMS to Itanium. 3 Message-ID: <3B44F1F3.8DF22FDC@remove_this.sun.com>    Bob Kaplow wrote:  > k > In article <aLMZ6.119$rc5.4462@news.cpqcorp.net>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes:-G > > Nothing in OpenVMS, or Galaxy, is Alpha-specific, except perhaps in,N > > implementation.  With the Alpha design team going to Intel, I would not beM > > suprised to see things from EV7 and EV8 sneak into Itanium in the future.e > N > Will any of them go? And wouldn't Intel have just stolen the technology like( > they did earlier with everything else. > O > > We're pretty confident that by the time we get things ported, it will be in M > > good shape, and EV7 systems will give us a good transition padding with at% > > high-performance Alpha swan-song.R > >t% > >>Andrew Harrisson, is Sun hiring ?s > > ! > > Sun is trying not to lay off.m > L > But Andrew and his buddies will FUD Compaq, VMS, and Tru64 to death before) > the first Itanium / VMS system is sold..     Quiten  = Well I won't, but the sales people who work for Sun will and s< who can refute the facts that they will be able to roll out  to support their "FUD".   ; I have always had a problem with the term "FUD". In my viewi= FUD is not based on hard fact, people who post on this group  ? have in the past been quick to complain that posts which they  !9 find disagreable are FUD despite the fact that the posts t# are undeniably based on hard fact. l  ; Fact, the WildFire systems and Alpha boxes currently being t< sold by Compaq now have a built in obselescence date, which - is no longer determined by the Aplha roadmap.o  5 Fact, IA-64/Tru64/OpenVMS will require new OS's/apps.s  8 Fact this will cost customers money. (Unless Compaq pay)  3 Fact there is a risk associated with the migration h" (Unless Compaq Garantee the risk).  3 Now people may see these points as being FUD, sadlyu they arn't.i   Regards  Andrew Harrisont Enterprise IT Architecti   ------------------------------  % Date: Fri, 06 Jul 2001 00:11:30 +0100mL From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com>) Subject: Re: Full port of VMS to Itanium. 3 Message-ID: <3B44F422.D4D08176@remove_this.sun.com>y   Alan Greig wrote:b > G > On Tue, 26 Jun 2001 17:18:51 -0400, "Bill Todd" <billtodd@foo.mv.com>b > wrote: > 
 > >> iVMS. > >> > >> Let's go trekking ... > >tE > >Is Compaq paying you, or are you simply an idiot with a big mouth?" > >t > Is Sun paying you? >    No	 > >- billg > >y > >>	 > >> John  > >>G > >> Name: John B. Macallister  E-mail: j.macallister1@physics.ox.ac.uk M > >> Post: Nuclear and Astrophysics Laboratory, Keble Road, Oxford OX1 3RH,UKaF > >> Phone: +44-1865-273388 (direct)  273333 (reception)  273418 (Fax) > >  > >x >  > -- > Alan   -- a Andrew Harrisonn Enterprise IT ArchitectE   ------------------------------  # Date: Fri, 06 Jul 2001 01:19:57 GMTt4 From: "Terry C. Shannon" <terryshannon@mediaone.net>) Subject: Re: Full port of VMS to Itanium.l: Message-ID: <1n817.338$Pf6.320308@typhoon.ne.mediaone.net>  H "Andrew Harrison SUNUK Consultancy" <ah13473_nospam@remove_this.sun.com>> wrote in message news:3B44F1F3.8DF22FDC@remove_this.sun.com...   >r< > Fact, the WildFire systems and Alpha boxes currently being= > sold by Compaq now have a built in obselescence date, whichn/ > is no longer determined by the Aplha roadmap.   J You'll need to clarify that point. (The others are perfectly valid.) As itJ was prior to June 25, you'll be able to upgrade a WildFire with EV7 MarvelL drawers. Not that I'd recommend the approach... buy a Marvel and cluster the sucker with your WildFire.  K Whether or not the Big Change occurred on June 25, the post-Marvel platformaK (CPU blades and an InfiniBand mesh architecture) would have been a forklifte upgrade.  J Since the post-Marvel platform was designed to be CPU architecture-neutralG right from the get-go, the IPF fealty imposes no new constraints on the K developers. So all customers face is an OS port and a "simple recompile" of- their apps. ;-}    ------------------------------  % Date: Thu, 05 Jul 2001 20:08:13 +0100tL From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com>* Subject: Re: H-P EOLs PA-RISC Architecture3 Message-ID: <3B44BB1D.1604F8AB@remove_this.sun.com>    PROSULLIVAN wrote: > / > >HP have a server SuperDome which is allreadyx0 > >able to support IA-64 and the IA-64 processor > O > hmm. exactly how many superdomes have HP sold with that whizzy PA-RISC 552MhzL > chip? HP aren't saying.o >   9 It is true that they havn't sold very many, however they i4 do have customers for the product and it does exist.  L > The poor customer who has an N class HP (552Mhz chip) and wants to  expandQ > beyond 8 cpu's has to go to a superdome, running shock horror, the same PA-RISClG > 552Mhz chip. HP banked on Superdome being available with IA64 and arenC > desperately trying to squeeze their PA-RISC until IA64 comes out.i >   B Of course, the delays to IA-64 have lead HP to have to extend the @ life of HP-PA well beyond what they were expecting. It is worth 9 remembering that the HP-PA 8500 was actually designed as S: cost reduced CPU that was to provide EOL support for HP-PA> processor while the IA-64 rolled out. Hence the horribly small; cache 1.5 MB which makes the HP boxes great SPECint enginesc< rather like the WildFire but horribly bad a everything else.    D > That is one reason why HP server sales have taken a nose-dive. The  > price/performance ratio dives. > N > As a result sales in Tru64 and VMS GS machines has increased, despite all of > Compaq's marketing effort.       Regards  Andrew Harrisond Enterprise IT Architecto   ------------------------------  $ Date: Thu, 5 Jul 2001 23:29:09 +01008 From: "Alistair King" <alistair.king@icm-computer.co.uk>9 Subject: Re: How to (auto-re)submit batch few times a day:3 Message-ID: <9i2pnv$376$1@plutonium.btinternet.com>D  H Just watch out for the "Version No" problem when auto submitting command files.  J VMS when you submit a file to a print Queue stores the file by FileID thisH means if you Edit the file after it has been submitted the "old" versionG will be run not the version you have just edited. Also note that if youLD Purge the files the job will fail as it will not be able to find theL submitted file. When this happens which of corse will be a way down the lineL from when you create the procedure this will cause you problems. This is whyK the Small "Stub" procedure that ONLY does the re-submitting and "calls" the F real procedure that is doing the work is the best way of handling this within DCL.   D Below is a script that runs daily, note the code at the start of theJ procedure to make sure that this is the latest version of the file that is running.  
 Alistair Kingn   $m	 $SET NOONm $!D $! Test to check using correct (Latest) version of command procedure $!* $PROCEDURE_NAME=f$environment("PROCEDURE")4 $LAST_VERSION=f$search(f$parse(";0",PROCEDURE_NAME))% $if PROCEDURE_NAME .nes. LAST_VERSION- $then- $ @'LAST_VERSION'  $ exit $endif $! $! $ SAY   :==WRITE SYS$OUTPUTt $ DELETE/SYMB/GLOBAL DOc $ DO  :==CALL DO_COM $!H $ days_to_backup = "/Monday/Tuesday/Wednesday/Thursday/Friday/Saturday/" $ start_hour = "01"a $ stop_backup = "07" $ hourNow = F$CVTime(,,"HOUR")! $ today   = F$CVTime(,,"WEEKDAY") E $ nextDue = F$CVTime("TOMORROW+''start_hour':","ABSOLUTE","DATETIME")r $ Copday  = todayW! $ IF F$Mode() .EQS. "INTERACTIVE"- $ THEN" $  IF HOURNOW .LT. "''START_HOUR'" $  THENmD $   nextDue = F$CVTime("TODAY+''start_hour':","ABSOLUTE","DATETIME") $  ENDIF $ ENDIF0* $ jobName = "Daily Backup (of ''copday')." $! $ SUBMIT/QUEUE=BATCH-iB   /NOPRINT/USER=SYSTEM/LOG_FILE=SYSLOG:'copday/NAME="''jobName'" -:   /RETAIN=ERROR/AFTER="''nextDue'" SYSPRG:DAILY_BACKUP.COM $!7 $ IF F$Mode() .EQS. "INTERACTIVE" THEN GOTO Interactive  $!E $! Only run the backup if it is in the time frame specified, preventsoK $! it running accidentally should the machine go down when the job has beeno	 $! queued 5 $ IF hourNow .GT. "''STOP_BACKUP'" THEN GOTO NOTTHREEr $! Go for it...  $! $!. $  DO SYSPRG:PRE_BACKUP.COM   "Prepare backup" $!6 $ IF F$LOCATE( "/''TODAY'/", "''DAYS_TO_BACKUP'") .ne. f$length("''days_to_backup'")l $ THEN9 $  DO SYSPRG:FULL_BACKUP.COM  "Full backup of every disk"- $ ENDIF  $!4 $  DO SYSPRG:POST_BACKUP.COM  "Tidy up after backup" $! $! $! $ GOTO EndItAlla $! $ NotThree:. $! Wont run as not 3:00 AM $! Send message to warn usersv $ GOSUB WrongTime  $ GOTO EndItAllo $!
 $Interactive:n% $ WRITE SYS$OUTPUT "Backup Submitted"w $ GOTO EndItAll  $ !s $ EndItAll:s $ SET NOVERIFY $ @sys$startup:decw$startupu $ EXIT $!L $!========================================================================== ===  $! Job running at wrong time...e $WrongTime:o" $ OPEN/WRITE MAILMESS MAILMESS.TMP $ WRITE MAILMESSL "=========================================================================== ==="( $ WRITE MAILMESS "Daily Backup Error..." $ WRITE MAILMESS "" : $ WRITE MAILMESS "Daily Backup job commenced at wrong time (",F$TIME(),")...". $ WRITE MAILMESS "Backup has not taken place." $ WRITE MAILMESS ""m $ WRITE MAILMESSL "=========================================================================== ===" $ CLOSE MAILMESS* $ MAIL/SUBJECT="Daily Backup" MAILMESS.TMP9 "@SYSPRIVATE:RESPONSIBLE_STAFF"/pers="Backup @Wrong Time"  $ DELETE MAILMESS.TMP;4 $ REQUEST "Daily Backup Commenced at the Wrong Time" $! $ RETURN $!L $!========================================================================== ===l $! Error running backup...	 $BackErr: " $ OPEN/WRITE MAILMESS MAILMESS.TMP $ WRITE MAILMESSL "=========================================================================== ==="& $ WRITE MAILMESS "Daily Backup Failed" $ WRITE MAILMESS ""o' $ SET MESSAGE SYS$MESSAGE:SYSMGTMSG.EXEH8 $! WRITE MAILMESS "Message is     : ",F$Message(retStat)" $ WRITE MAILMESS "Log file is    :0 SYSLOG:",F$CVTIME("YESTERDAY",,"WEEKDAY"),".LOG"6 $ WRITE MAILMESS "Journal File Is: SYSLOG:THISWEEK.BJL $ WRITE MAILMESS ""3 $ WRITE MAILMESSL "=========================================================================== ===" $ CLOSE MAILMESS* $ MAIL/SUBJECT="Daily Backup" MAILMESS.TMP8 "@SYSPRIVATE:RESPONSIBLE_STAFF"/pers="BACKUP has Failed" $ DELETE MAILMESS.TMP;< $ REQUEST "Daily Backup Failed with - ''F$Message(retStat)'" $! $ RETURN $!   $! $! Description:e $!H $! The purpose of this Command procedure is to automate the running of a dailyoJ $! incremental backup (ie take a copy of those files since the last backup runa $! with /RECORD option). $!I $! When the command procedure is run it immediately submits itself to the  batch * $! queue to run at 3am "Tomorrow" morning. $!H $! If the procedure is being run from an interactive terminal session it then	 $! Exits.o $!- $! The procedure is being run as a batch job:cK $! The Backup is limited to running at 3am to ensure that it only runs whenaK $! the VAX is not being used (in theory at least) and to prevent it runningtF $! should the machine go down during the night and be restarted in the morning. $!D $! If the Hour is not 3 then a Mail message will be sent warning the softwareL $! staff that the backup procedure has run (on the batch queue) at the wrong! $! time. The procedure then exits0 $!@ $! The procedure is a batch job running at three in the morning: $! $!L ============================================================================ =uG $!---------------------------------------------------------------------s9 $! Subroutine To Execute Command Procedure if it is therei! $!   P1 - Command Proceedure Namen $!  P2 - DescriptionG $!---------------------------------------------------------------------a $DO_COM: SUBROUTINEe $  SET NOONa $  IF F$SEARCH(P1) .NES. ""e $  THENP $ SAY ""/ $ SAY F$FAO("!%T",0)," ",P2," (",P1,") STARTED"' $ SAY "" $ @'P1'\ $  ELSEt1 $ SAY F$FAO("!%T",0)," ",P2," (",P1,") Not Found"- $  ENDIF $endsubroutine   ------------------------------  % Date: Thu, 05 Jul 2001 16:21:11 -0500m/ From: Chris Scheers <chris@applied-synergy.com>y+ Subject: Re: I didn't stick it upside down!t3 Message-ID: <3B44DA47.C191B2E6@applied-synergy.com>m   zessin@decus.de wrote:  ? > No problem. I have the correct tool to open the SBB carriers.t   What is the correct tool?   G -----------------------------------------------------------------------.$ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com y   Fax: 817-237-3074a   ------------------------------  % Date: Thu, 05 Jul 2001 21:45:42 -0400w2 From: rdeininger@mindspring.com (Robert Deininger)+ Subject: Re: I didn't stick it upside down!-L Message-ID: <rdeininger-0507012145430001@user-2ivea3q.dialup.mindspring.com>  A In article <3B44DA47.C191B2E6@applied-synergy.com>, Chris Scheers6" <chris@applied-synergy.com> wrote:   > zessin@decus.de wrote: > A > > No problem. I have the correct tool to open the SBB carriers.4 >  > What is the correct tool?a  1 A screwdriver or letter-opener, you silly wabbit!r   :-):   Actually, I have no idea...e   --   Robert Deininger rdeininger@mindspring.com    ------------------------------  % Date: Thu, 05 Jul 2001 20:52:57 +0100lL From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com>  Subject: Re: IA64 Rocks My World3 Message-ID: <3B44C599.DBE01DA1@remove_this.sun.com>p   Eric Smith wrote:f > R > Casper.Dik@Holland.Sun.Com (Casper H.S. Dik - Network Security Engineer) writes:B > > Engineers shouldn't blame everything on marketing; Digital has< > > made serious mistakes when it came to building hardware. > 
 > Such as? > N > Mistakes more serious than hardware engineering at other comparable vendors?     Yes where do you want to start.y  3 Horribly over priced entry level machines compared  2 with the commodity market, but also more tellingly3 when compared with other non Intel systems vendors.e  / A server range that has one competitive product 6 the ES40 surrounded by over priced and underperforming0 or overpriced systems. I refer to the GS160/320 ( in the former and the D's in the latter.  5 And this isn't new, the 8400 after the first 18monthsn5 was a perfomance and price turkey the 8200 even more 1/ so, with the only decent system being the 4100.a  3 Now some of the problems for example pricing could u. be passed off to the marketing folks, but the , GS160/320 architecture can hardly be blamed  on them.   Regards  Andrew Harrisoni Enterprise IT Architecti   ------------------------------  $ Date: Thu, 5 Jul 2001 16:33:41 -0400' From: "Bill Todd" <billtodd@foo.mv.com>i  Subject: Re: IA64 Rocks My World( Message-ID: <9i2io0$425$1@pyrite.mv.net>  H "Andrew Harrison SUNUK Consultancy" <ah13473_nospam@remove_this.sun.com>> wrote in message news:3B44C599.DBE01DA1@remove_this.sun.com... >c > Eric Smith wrote:e > >eL > > Casper.Dik@Holland.Sun.Com (Casper H.S. Dik - Network Security Engineer) writes: D > > > Engineers shouldn't blame everything on marketing; Digital has> > > > made serious mistakes when it came to building hardware. > >r > > Such as? > > G > > Mistakes more serious than hardware engineering at other comparables vendors? >4 >o! > Yes where do you want to start.e >s4 > Horribly over priced entry level machines compared4 > with the commodity market, but also more tellingly5 > when compared with other non Intel systems vendors.a >d1 > A server range that has one competitive productt8 > the ES40 surrounded by over priced and underperforming1 > or overpriced systems. I refer to the GS160/320b* > in the former and the D's in the latter. >e7 > And this isn't new, the 8400 after the first 18months 6 > was a perfomance and price turkey the 8200 even more1 > so, with the only decent system being the 4100.t >k4 > Now some of the problems for example pricing could' > be passed off to the marketing folks,s  ; And that happens to include everything you mentioned above.s    but the- > GS160/320 architecture can hardly be blamedt
 > on them.  J Right - but it can rightly be blamed on Compaq's (and perhaps DEC's beforeJ it) screwed-up management of the development process that stretched it out for additional years.s  L In any event, don't confuse not living up to the 'best-of-breed by far' hypeH of its supporters with being a pig:  Alpha GS systems perform a lot moreL competitively with their competition than the Ultra SPARC processor cores doA with theirs, so if you think GS is a pig, what does that make US?l  L Is Alpha pricing a problem?  In some cases, perhaps.  That has nothing to doH with "serious mistakes when it came to building hardware", which was the topic under discussion.    - bill   > 	 > Regardsy > Andrew Harrisonu > Enterprise IT Architectn   ------------------------------   Date: 5 Jul 2001 21:33:43 GMT N From: Casper.Dik@Holland.Sun.Com (Casper H.S. Dik - Network Security Engineer)  Subject: Re: IA64 Rocks My World0 Message-ID: <9i2mfn$2dp$1@new-usenet.uk.sun.com>  3 [[ PLEASE DON'T SEND ME EMAIL COPIES OF POSTINGS ]]H  ) "Bill Todd" <billtodd@foo.mv.com> writes:t  M >In any event, don't confuse not living up to the 'best-of-breed by far' hypedI >of its supporters with being a pig:  Alpha GS systems perform a lot moretM >competitively with their competition than the Ultra SPARC processor cores doeB >with theirs, so if you think GS is a pig, what does that make US?  > Well, "50%" better performance on TPC-C (still using OPS or isC Oracle 9i do NUMA friendler?) for the latest GS320 vs essentially a = 3-5 year old system, is that really something to be proud of?    Casper --F Expressed in this posting are my opinions.  They are in no way related2 to opinions held by my employer, Sun Microsystems.? Statements on Sun products included here are not gospel and mays be fiction rather than truth.o   ------------------------------  $ Date: Thu, 5 Jul 2001 18:04:51 -0400' From: "Bill Todd" <billtodd@foo.mv.com>c  Subject: Re: IA64 Rocks My World( Message-ID: <9i2o2u$7g3$1@pyrite.mv.net>  J "Casper H.S. Dik - Network Security Engineer" <Casper.Dik@Holland.Sun.Com>; wrote in message news:9i2mfn$2dp$1@new-usenet.uk.sun.com... 5 > [[ PLEASE DON'T SEND ME EMAIL COPIES OF POSTINGS ]]- >-+ > "Bill Todd" <billtodd@foo.mv.com> writes:: >3J > >In any event, don't confuse not living up to the 'best-of-breed by far' hypeK > >of its supporters with being a pig:  Alpha GS systems perform a lot more1L > >competitively with their competition than the Ultra SPARC processor cores doD > >with theirs, so if you think GS is a pig, what does that make US? >>@ > Well, "50%" better performance on TPC-C (still using OPS or isE > Oracle 9i do NUMA friendler?) for the latest GS320 vs essentially a ? > 3-5 year old system, is that really something to be proud of?   K Are you genetically incapable of sticking to the subject?  The question wasdG not pride, not blow-away vs. simply-competitive performance, but market-K viability.  And Alpha provides perfectly adequate performance, including in L its GS configurations, for market viability - in contrast to your "EngineersJ shouldn't blame everything on marketing; Digital has made serious mistakesJ when it came to building hardware" silliness that was the root of Andrew's equally-off-the-mark response.   - bill   >m > Casper > --H > Expressed in this posting are my opinions.  They are in no way related4 > to opinions held by my employer, Sun Microsystems.A > Statements on Sun products included here are not gospel and mayt > be fiction rather than truth.v   ------------------------------  # Date: Thu, 05 Jul 2001 23:31:02 GMTA; From: Mark Garrett <Mark.Garrett@wedontwantyourspam.com.au>e  Subject: Re: IA64 Rocks My WorldC Message-ID: <B76B35D1.19A5A%Mark.Garrett@wedontwantyourspam.com.au>v  H in article 9i2mfn$2dp$1@new-usenet.uk.sun.com, Casper H.S. Dik - NetworkJ Security Engineer at Casper.Dik@Holland.Sun.Com wrote on 06/07/2001 07:33:  5 > [[ PLEASE DON'T SEND ME EMAIL COPIES OF POSTINGS ]]I > + > "Bill Todd" <billtodd@foo.mv.com> writes:r > O >> In any event, don't confuse not living up to the 'best-of-breed by far' hypeeK >> of its supporters with being a pig:  Alpha GS systems perform a lot more O >> competitively with their competition than the Ultra SPARC processor cores do-D >> with theirs, so if you think GS is a pig, what does that make US? > @ > Well, "50%" better performance on TPC-C (still using OPS or isE > Oracle 9i do NUMA friendler?) for the latest GS320 vs essentially ac? > 3-5 year old system, is that really something to be proud of?S > H Oracle 8.1.7 and onward are meant to be NUMA friendly, I as yet have not& stats on how effective they have been.    
     Cheers         Mark :)e   ------------------------------  # Date: Fri, 06 Jul 2001 02:56:32 GMTa4 From: "Terry C. Shannon" <terryshannon@mediaone.net>  Subject: Re: IA64 Rocks My World: Message-ID: <AN917.419$vb6.203975@typhoon.ne.mediaone.net>  J "Casper H.S. Dik - Network Security Engineer" <Casper.Dik@Holland.Sun.Com>; wrote in message news:9i2mfn$2dp$1@new-usenet.uk.sun.com...S5 > [[ PLEASE DON'T SEND ME EMAIL COPIES OF POSTINGS ]]l >i+ > "Bill Todd" <billtodd@foo.mv.com> writes:e > J > >In any event, don't confuse not living up to the 'best-of-breed by far' hypeK > >of its supporters with being a pig:  Alpha GS systems perform a lot moreiL > >competitively with their competition than the Ultra SPARC processor cores doD > >with theirs, so if you think GS is a pig, what does that make US? >h@ > Well, "50%" better performance on TPC-C (still using OPS or isE > Oracle 9i do NUMA friendler?) for the latest GS320 vs essentially ah? > 3-5 year old system, is that really something to be proud of?e  I No, but it *is* ammunition in the benchmarketing wars. CPQ may be able to K keep its powder dry for a few months, but IBM's S85 is gonna whip the pantsf off the WildFire.i   ------------------------------  # Date: Thu, 05 Jul 2001 20:41:40 GMT + From: Ryan Moore <rmoore@rmoore.dyndns.org>5& Subject: Re: Need XML Parser for C/C++< Message-ID: <Pine.LNX.4.31.0107051335140.1902-100000@jaipur>   > Subodh Damle wrote:eC > > Does anyone know any XML Parser for C/C++ available under VMS ? P > > Alternatively, has anyone tried porting any parser library for UNIX to VMS ?  A I got expat 1.95.1 to compile under VMS.  The test program in thel distribution seems to work.h  F First, get the distribution at http://sourceforge.net/projects/expat/.( The libraries are in the lib/ directory.  > I used the following config.h file in the top level directory:  7 /* config.h.  Generated automatically by configure.  */cB /*================================================================ ** Copyright 2000, Clark Cooperi ** All rights reserved." **J ** This is free software. You are permitted to copy, distribute, or modifyI ** it under the terms of the MIT/X license (contained in the COPYING file  ** with this distribution.)n ** ** */  4 /* Define to empty if the keyword does not work.  */ /* #undef const */  7 /* Define if you have a working `mmap' system call.  */T /*#define HAVE_MMAP 1*/e  8 /* Define to `long' if <sys/types.h> doesn't define.  */ /* #undef off_t */  < /* Define to `unsigned' if <sys/types.h> doesn't define.  */ /* #undef size_t */C  2 /* Define if you have the ANSI C header files.  */ #define STDC_HEADERS 1  B /* Define if your processor stores words with the most significantB    byte first (like Motorola and SPARC, unlike Intel and VAX).  */ /* #undef WORDS_BIGENDIAN */  - /* Define if you have the bcopy function.  */  /*#define HAVE_BCOPY 1*/  3 /* Define if you have the getpagesize function.  */O /*#define HAVE_GETPAGESIZE 1*/  / /* Define if you have the memmove function.  */S #define HAVE_MEMMOVE 1  4 /* Define if you have the <fcntl.h> header file.  */ #define HAVE_FCNTL_H 1  5 /* Define if you have the <unistd.h> header file.  */  /*#define HAVE_UNISTD_H 1*/    #define XML_NS #define XML_DTDT   #ifdef WORDS_BIGENDIAN #define XML_BYTE_ORDER 21N #elsej #define XML_BYTE_ORDER 12/ #endif   #define XML_CONTEXT_BYTES 1024   #ifndef HAVE_MEMMOVE #ifdef HAVE_BCOPYN) #define memmove(d,s,l) bcopy((s),(d),(l))k #elset+ #define memmove(d,s,l) ;punting on memmove;! #endif   #endif   =========================d  C And the file descrip.mms in the lib directory to compile the files:P   CFLAGS =N /define=package="""expat"""/define=version="""expat_1.95.1_vms"""/include=[] -,                 /include=[-]/names=shortened  > .obj.c : $(CC) $(CFLAGS) /obj=$(MMS$TARGET_NAME) $(MMS$SOURCE)   all : stuffK         libr expat.olb *.obj"         @ write sys$output "Done!"  + stuff : xmlparse.obj xmlrole.obj xmltok.obj    xmlparse.obj : xmlparse.co   xmlrole.obj : xmlrole.ct   xmltok.obj : xmltok.c    xmltok_impl.obj : xmltok_impl.c    xmltok_ns.o : xmltok_ns.cc     ======================  D To compile, go into the "lib" directory and make the file expat.olb:   $ libr/crea expat.olbr   Then compile with MMS:   $ mms=  H Or use CC directly with the qualifier in the CFLAGS section above.  Some; routine names are greater than 31 characters, so I used theA0 '/names=shortened' qualfier to compile the code.   -Ryan=   ------------------------------  % Date: Thu, 05 Jul 2001 23:32:09 +0100lL From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com>8 Subject: Re: OpenVMS Itanium (So *Microsoft PC* Culture)3 Message-ID: <3B44EAE9.5AE67DB8@remove_this.sun.com>m   "John D. Peedle" wrote:= > ? > David Mathog <mathog@seqaxp.bio.caltech.edu> wrote in message ( > news:9hanos$rg3@gap.cco.caltech.edu... > <snip>L > > 1.  It wasn't enough to get Compaq to commit to using IA64.  They had toK > >     go the extra mile and coopt (destroy) the Alpha development team so!N > >     that nobody else could buy it, and or all of Compaq, and use the Alpha > >     to beat on IA64. > M > Whilst Intel are hardly the basket to which I prefer to transfer my eggs, IEK > think you are perhaps leaving the Q out of the blame cycle too much here.=< > Capellas still hasn't learned some very important lessons. > J > 1.   He can't compete with Dell. He has no concept of how to even start.J > 2.   The author/inspiration of Genesis Ch 1 was not William H Gates III.I > 3.   Trying to turn Compaq into 'The leading IT Solutions company' willSK > require someone with the skill and balls of Lou Gerstner. I don't see any=> > evidence that Compaq has anybody like that on their payroll.L > 4.   Most people who buy Highly Available systems do not expect to see the$ > Intel warning label on the outsideK > 5.   ComputaCentre (Arguably the real leading IT solutions company in theg/ > UK) is not for sale at a price he can afford.i8 > 6.   Sun and IBM are laughing all the way to the bank. > : > Andrew, where are you? You should be having a field day!    2 To be honest it is of no immediate interest to me.  5 The customer I am responsible for delivering all the s/ systems infrastructure to isn't a Compaq Alpha  1 customer now, wasn't in the past and won't be on !) the desktop (IA-64) for at least 5 years.r  5 However the fact that Compaq has dropped Alpha hasn'tp/ gone un-noticed by a number of our sales peoplec0 an I doubt if they will let Compaq off the hook.  2 In a recession the best news you can get is either1 an upturn in the economy or the news that one of  - your closer competitors has voluntarely self t destructed.    Regrds ! Andrew Harrisoni Enterprise IT Architectg   ------------------------------   Date: 5 Jul 2001 11:43:28 -0700u% From: s.joseph@lph.org (Shawn Joseph)w@ Subject: OpenVMS Training Providers- Are they all created equal?= Message-ID: <c7ea9784.0107051043.25bc92c0@posting.google.com>=  A I am a newbie in the OpenVMS arena and my employer has graciouslyu? agreed to send me to some instructor led training.  I have been-F reviewing the following providers and wondered if any of you out thereF could give me some feedback on them: Parsec Group, MindIQ and Bruden. E I would like to hear any and all feedback, positive and negative, andO> whether one is better than the other and why.  If I understandF correctly Compaq doesn't actually do the training, it is farmed out to7 one of the above mentioned companies.  Is that correct?t  
 Respectfully,F Shawn Joseph   ------------------------------   Date: 5 Jul 2001 15:12:11 -050019 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) D Subject: Re: OpenVMS Training Providers- Are they all created equal?3 Message-ID: <s$DactBOI06z@eisner.encompasserve.org>:  e In article <c7ea9784.0107051043.25bc92c0@posting.google.com>, s.joseph@lph.org (Shawn Joseph) writes:-C > I am a newbie in the OpenVMS arena and my employer has graciously A > agreed to send me to some instructor led training.  I have been H > reviewing the following providers and wondered if any of you out thereH > could give me some feedback on them: Parsec Group, MindIQ and Bruden. G > I would like to hear any and all feedback, positive and negative, andt@ > whether one is better than the other and why.  If I understandH > correctly Compaq doesn't actually do the training, it is farmed out to9 > one of the above mentioned companies.  Is that correct?   @ I have seen people from Parsec and Bruden presenting sessions at> DECUS symposia, which gives them added stature in my view.  Of> course the instructor you get may not be the one who presented= at DECUS, but I think they are both small-enough outfits thati& peer-pressure has some part in hiring.   ------------------------------   Date: 5 Jul 2001 14:34:07 -0700w1 From: jason_odonnell@erinet.com (Jason O'Donnell)tD Subject: Re: OpenVMS Training Providers- Are they all created equal?= Message-ID: <5c8ffd05.0107051334.1b07f9ef@posting.google.com>p  j s.joseph@lph.org (Shawn Joseph) wrote in message news:<c7ea9784.0107051043.25bc92c0@posting.google.com>...C > I am a newbie in the OpenVMS arena and my employer has graciously A > agreed to send me to some instructor led training.  I have beeniH > reviewing the following providers and wondered if any of you out thereH > could give me some feedback on them: Parsec Group, MindIQ and Bruden. G > I would like to hear any and all feedback, positive and negative, andi@ > whether one is better than the other and why.  If I understandH > correctly Compaq doesn't actually do the training, it is farmed out to9 > one of the above mentioned companies.  Is that correct?t >  > Respectfully,  > Shawn Joseph  F We use PARSEC.  They do a wonderful job.  I investigated the few firmsD that give OpenVMS training and liked what I saw from PARSEC better. B Bruce Ellis(?) of Bruden actually wrote the official Compaq courseF materials.  I did not feel they were as good as they should/could have been.c   JMOD   ------------------------------  $ Date: Thu, 5 Jul 2001 23:41:15 +01008 From: "Alistair King" <alistair.king@icm-computer.co.uk># Subject: Re: POP server on tcpip5.1d3 Message-ID: <9i2qep$qhj$1@neptunium.btinternet.com>W  L You can set Pop up with a "Unique" account or you can use there standard VMSH account. Do note that if a mail has been red by VMSMAIL you can not thenK read it with POP mail this is a real problem if a user reads a mail with anD@ attachment under VMS as they can not then get at the attachment.  K Also note that you do not have to use the VMS user name as the Mail address L you can set up aliases in any format using forward command in mail such as:-    . MAIL SET FORWARD ALISTAIR/USER="ALISTAIR.KING"   Hope this helpsr  
 Alistair Kingd    . "Tom Linden" <tom@kednos.com> wrote in message3 news:CIEJLCMNHNNDLLOOGNJICEMJCOAA.tom@kednos.com...dI > So the user does not have a separate POP account, pre se, but has a VMSh
 > account? >l > > -----Original Message-----= > > From: Robert Deininger [mailto:rdeininger@mindspring.com]e) > > Sent: Thursday, July 05, 2001 9:33 AMo > > To: Info-VAX@Mvb.Saic.Comh' > > Subject: Re: POP server on tcpip5.1  > >e > >eH > > In article <CIEJLCMNHNNDLLOOGNJIAEMGCOAA.tom@kednos.com>, Tom Linden > > <tom@kednos.com> wrote:h > >0E > > > Just installed tcpip5.1 on 7.3 AXP.  How do I configure SMTP too > > support POPb
 > > > clientse? > > > logging (to SMTP) from their PCs? Specifically, suppose ah > > client is assigned: > > > and IP and node name, NODE, such that his address is > > John@NODE.kednos.com > > > ? > > > 1.  Do I have to run TCPIP$config to make the association H > > > xx.xxx.xxx.xxx  NODE.kednos.com ?    Is there no HOSTS source file here?t > >eI > > See the SET HOST command in TCPIP.  This is like the unix HOSTS file,u buttJ > > it's not a plain text file you edit.  You use the command interface toB > > change it.  You don't need to do this if you have a nameserver
 configured > > on the network.e > > L > > > 2.  How do I configure SMTP to allow the client to login from his mail
 > > > programl > > > to retrieve is mail? > >tL > > You'll have to enable and start the POP service, if you haven't already. > >h > > $ @sys$manager:tcpip$config  > >   then option 3 on the menun6 > >      then option 13 is POP (in the version I have) > >-( > > Note that POP is separate from SMTP. > >  > >mD > > Also, the POP user will have to have a VMS account, with a valid > > password, etc. > >o > >yB > > > I find the documentation much too general.  This is a simple > > routine task.I > > > Is there anyB > > > How To, documentation anywhere?  I looked in the VMS FAQ but > > nothing there. > > G > > Well, POP is in the TCPIP management manual.  I don't know if it isiJ > > sufficiently step-by-step.  But I figured it out, so the docs must not beH > > too bad.  My intuition on things tcpip is not very good -- I have to rely9 > > on the manuals.  Maybe I've just gotten used to them.n > >  > > -- > > Robert Deininger > > rdeininger@mindspring.como > >  >e   ------------------------------  % Date: Thu, 05 Jul 2001 15:57:30 -0700o! From: Tom Linden <tom@kednos.com>s# Subject: RE: POP server on tcpip5.1 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIIENCCOAA.tom@kednos.com>,  I Well, ideally, I would like to set up POP accounts for users that have non VMSs account.   > -----Original Message-----? > From: Alistair King [mailto:alistair.king@icm-computer.co.uk] ' > Sent: Thursday, July 05, 2001 3:41 PM  > To: Info-VAX@Mvb.Saic.Comr% > Subject: Re: POP server on tcpip5.1. >> > A > You can set Pop up with a "Unique" account or you can use thereu > standard VMSJ > account. Do note that if a mail has been red by VMSMAIL you can not then@ > read it with POP mail this is a real problem if a user reads a > mail with anB > attachment under VMS as they can not then get at the attachment. >d@ > Also note that you do not have to use the VMS user name as the > Mail address? > you can set up aliases in any format using forward command ina > mail such as:- >s >o0 > MAIL SET FORWARD ALISTAIR/USER="ALISTAIR.KING" >o > Hope this helps> >  > Alistair Kingr >r >m0 > "Tom Linden" <tom@kednos.com> wrote in message5 > news:CIEJLCMNHNNDLLOOGNJICEMJCOAA.tom@kednos.com... K > > So the user does not have a separate POP account, pre se, but has a VMSo > > account? > >a  > > > -----Original Message-----? > > > From: Robert Deininger [mailto:rdeininger@mindspring.com] + > > > Sent: Thursday, July 05, 2001 9:33 AMh > > > To: Info-VAX@Mvb.Saic.Comf) > > > Subject: Re: POP server on tcpip5.1n > > >m > > >iJ > > > In article <CIEJLCMNHNNDLLOOGNJIAEMGCOAA.tom@kednos.com>, Tom Linden > > > <tom@kednos.com> wrote:  > > >sG > > > > Just installed tcpip5.1 on 7.3 AXP.  How do I configure SMTP tow > > > support POP- > > > > clients A > > > > logging (to SMTP) from their PCs? Specifically, suppose ai > > > client is assigned< > > > > and IP and node name, NODE, such that his address is > > > John@NODE.kednos.com > > > >eA > > > > 1.  Do I have to run TCPIP$config to make the association J > > > > xx.xxx.xxx.xxx  NODE.kednos.com ?    Is there no HOSTS source file > here?r > > >aK > > > See the SET HOST command in TCPIP.  This is like the unix HOSTS file,s > but L > > > it's not a plain text file you edit.  You use the command interface toD > > > change it.  You don't need to do this if you have a nameserver > configured > > > on the network.t > > >/@ > > > > 2.  How do I configure SMTP to allow the client to login > from his maill > > > > program@ > > > > to retrieve is mail? > > >v= > > > You'll have to enable and start the POP service, if youh > haven't already. > > >w! > > > $ @sys$manager:tcpip$configp! > > >   then option 3 on the menuy8 > > >      then option 13 is POP (in the version I have) > > > * > > > Note that POP is separate from SMTP. > > >o > > > F > > > Also, the POP user will have to have a VMS account, with a valid > > > password, etc. > > >r > > >?D > > > > I find the documentation much too general.  This is a simple > > > routine task.o > > > > Is there anyD > > > > How To, documentation anywhere?  I looked in the VMS FAQ but > > > nothing there. > > > I > > > Well, POP is in the TCPIP management manual.  I don't know if it is-L > > > sufficiently step-by-step.  But I figured it out, so the docs must not > beJ > > > too bad.  My intuition on things tcpip is not very good -- I have to > rely; > > > on the manuals.  Maybe I've just gotten used to them.  > > >r > > > -- > > > Robert Deininger > > > rdeininger@mindspring.comn > > >e > >s >c >.   ------------------------------   Date: 5 Jul 2001 19:27 CDT' From: carl@gerg.tamu.edu (Carl Perkins)s# Subject: RE: POP server on tcpip5.1e, Message-ID: <5JUL200119273706@gerg.tamu.edu>  % Tom Linden <tom@kednos.com> writes...iJ }Well, ideally, I would like to set up POP accounts for users that have no }VMS	 }account.t  E As far as most things on VMS are concerned, an account is an account.,@ (There are exceptions - various bits of software that let you do? things only via that software that autorize users independantly @ of the VMS user authorization - but this is fairly pointless andB it is usually included in software ported from some other platform/ just to be complete and completely compatable.)p  B Just create a regular account for each POP user. You don't have to@ allow them interactive access if you don't want them logging in.E It's the easiest way. It is also the cleanest way - all your accountso? are managed identically, instead of using different methods forN different types of account.0  C Besides, if they are getting mail it has to be stored on the systemeF somewhere. Unless you buy some add-on mail system, that means the mailF will be stored in the usual VMS Mail files. These files have to belongJ to someone. That someone is the unique account (it has to be unique unlessG you want multiple people competing to see who gets to download the mailhE out of a shared account) which has the e-mail address pointing at it. B Thus, you *need* a regular VMS account so that it can own the mailG files. But you don't have to give the account interactive access, which=@ prevents the user from logging in and doing stuff interactively.  G If you have money, and the time and inclination, to get additional mailrG software then you may be able to dispense with the regular VMS account.nI But I don't know for certain that any such product allows you to do this.oF You might check out PMDF (at www.process.com), which can be configuredE to use it's own mail message store instead of the usual VMS Mail file*H method. Since it handles the SMTP mail and has its own POP server it mayD allow you do create POP-only "accounts". But I wouldn't count on it.   --- Carl   ------------------------------  % Date: Thu, 05 Jul 2001 23:11:26 +0100 + From: "antonio.carlini" <arcarlini@iee.org> 1 Subject: Re: Probably the strangest request yet!!s& Message-ID: <3B44E60E.628D939@iee.org>   Bill Gunshannon wrote:; > I have a desire to have available, but not full-time, then@ > following VAX OSes: MicroVMS4.5, OpenVMS-6.2 and Ultrix32-3.0.  6 The *easy* way is three disks hanging off one machine, but you knew that.  E >   While I know that I can run Ultrix32 diskless from another UltrixhC > server, is it possible to run all of these OSes diskless with then0 > server being another VAX running OpenVMS-7.1??  6 If Ultrix loads via MOP and picks up its filesystem(s)6 via NFS, then I expect that OpenVMS could load Ultrix.% I bet there are some wrinkles though!   $ > So the question(s) work out to be:L >    Can OpenVMS serve diskless machines running OSes and versions different >    than itself??  3 It can server MOP Loads to whatever nodes ask. This . is how terminal servers and routers of various kinds load their OS.  * The hard part comes once you go beyond the* initial MOP-loading stage. (Ultrix may not need to do that, I don't know).d  ) OpenVMS uses MOP for the initial load buto% picks up the majority of itself using=( the LAVC stuff (NISCS?). This only works( if you are loading into the same cluster" which means you need to be running# compatible versions (ortherwise you. crash one or both nodes).a  & Instead of booting into a cluster you & can (once beyond the initial MOP load)( carry on via LAD to pick up an exec from' an infoserver (network storage device).i  ) In fact, an Infoserver can serve multiplef# OpenVMS versions to various clientsl+ (over the last few years the OpenVMS binaryu& CDs include a bootable root to exploit this capability).   / If infoservers can also MOP load stuff (I don'te+ think they can but I may be wrong) then youe' could pick up Ultrix too (but you would-+ still probably have to NFS serve stuff for   Ultrix from elsewhere).m  , Currently OpenVMS cannot be a LAD server but' there was a plan to add this capabilitye' to OpenVMS Alpha. If that ever happens, ' you should be able to get close to whats you want to do.h  , (MicroVMS V4.5 may be a shade to old to cope- anyway. My recollection is that LAVC crept ine at round about V4.5B or V4.5C)..   Antonio>   -- .   ---------------t- Antonio Carlini             arcarlini@iee.org    ------------------------------   Date: 5 Jul 2001 23:58:47 GMTm5 From: ccburgess@uqstu.jdstory.uq.edu.au (Ian Burgess)d1 Subject: RE: SSH client, FISH, -- sources anyone?a- Message-ID: <9i2uvn$lt$1@bunyip.cc.uq.edu.au>a   I (Ian Burgess) wrote:? >I've just scanned the web for the source of a good SSH client.f >A recurring message is...N >"The OpenVMS SSH1 client is done by Christer Weinigel and Richard Levitte and/ >is available at  http://www.free.lp.se/fish/."o< >It seems unanimous that FiSH is the way to go, but there isL >a sad story -- a disk head crash has put LP (www.free.lp.se) out of action.@ >If someone has the distribution from there I would be grateful.& >I think Richard Levitte would be too.
 >Ian Burgess a  < Thanks to those who replied privately and to the news group.  & I now have Fish installed and running.   comp.os.vms is great!n   Ian Burgess  University of Queensland I.Burgess[at]its.uq.edu.au www.its.uq.edu.aut   ------------------------------  % Date: Thu, 05 Jul 2001 13:06:13 -0500.' From: Greg Pfister <pfister@us.ibm.com>o" Subject: Re: The Alpha/IA64 Hybrid* Message-ID: <3B44AC95.115D1D6B@us.ibm.com>   Larry Kilgallen wrote: > s > In article <fyt07.406522$eK2.82328712@news4.rdc1.on.home.com>, "Yousuf Khan" <ykhan@nospam.home.com.spam> writes:pJ > > "Larry Kilgallen" <Kilgallen@eisner.decus.org.nospam> wrote in message1 > > news:xu1946w16jvw@eisner.encompasserve.org...t9 > >> > Sun, HP, and Veritas would beg to differ with you.n > >>@ > >> Sun and HP are late to the game and trying to make excuses. > > O > > I'm afraid not, it truly is a form of clustering, whether it fits into your  > > narrow definition or not.o > >oB > >> Another way to configure failover.  Certainly not clustering,8 > >> no matter what Veritas uses for their product name. > >-4 > > Certainly is, your narrow view needs broadening. > > > So far as I know, the term was first used by VMS about 1985,A > for what they offered then (not materially different from today@4 > with regard to the capabilities under discussion). > 6 > Was there a prior use of "cluster" in this context ? > > > If not, then lesser copycat efforts are not really clusters.  < I definitely credit DEC with coining the term "cluster." The8 first public use of the term I've found was in: Nancy P.A Kronenberg, Henry M. Levy, and William D. Strecker. "Vaxclusters:O: A Closely-Coupled Distributed System." ACM Transactions on) Computer Systems, 4(3):130146, May 1986.t  @ That doesn't mean they invented the *concept* of clustering. I'd= put that way back sometime in the early 60s or even 50s, when-A some unknown customer first said "We really need a spare computer = if we're going to be sure of doing X on time. Probably shouldR: make some use of it in the meantime." There were groups of: mainframes being used as units with JES (Job Entry System); fanning out the work appropriately to them back in the 70s,o probably pre-dating VMS itself.s  6 Of course, if your personal definition of "cluster" is? "VMSCluster," and nothing else need apply, you're hardly likelynA to agree with the above. You're of course free to use the term asi> you want, but I think holding onto that definition will hinder. your communication with the rest of the world.   Greg Pfister   ------------------------------  % Date: Thu, 05 Jul 2001 22:30:28 +0200e& From: Michael Joosten <joost@c-lab.de>" Subject: Re: The Alpha/IA64 Hybrid$ Message-ID: <3B44CE64.751F@c-lab.de>   David J. Dachtera wrote: >  > Jack Patteeuw wrote: > >e > > Bill Todd wrote: > > .B  $ > Typical head-in-the-sand attitude. > U > > Sigh, and most Un*x and MS admins still have never heard of a ISAM file structure T > > and are happily paying millions to Oracle, MS, etc., etc. because no Un*x vendorR > > would **DARE** to draw the ire of one of their major vendors by including ISAMT > > support "native".  Other programmers are happy with their jobs security of "roll > > your own" db's !!!!  > G > Once upon many moons ago, one or more UN*X distro "came with" C-ISAM.f0 > Been too long - don't remember which or why... >   D This might be the times when UNIX just made its entree in commercialC computing, the BISON group, /usr/group, then finally OSF. The earlyaG X/Open standardisation efforts also resulted in a book covering C-ISAM.e  ; From my limited experience, most manufacturers of so-calledyG minicomputers then engaging in UNIX usually had ISAM available at least D as layered product, in order to provide their existing COBOL or justF typical commercial applications on the new OS. Relational databases atC that time just began to enter the scene, because they were probablysE munching away far too many cycles from the main applications to stillt: provide a decent response time to the dozens of terminals.  E Even in '94 I was told that some small vendor just used ISAM, because ) the SQL stuff was just too big and slow. e  G And nowadays... What do you expect? Students must know about relational F DBs, and their professors like to tell them about nifty extensions forD multimedia or even OODBs, so there is of course no/few time to teach+ them about such obsolete stuff like ISAM !?i   -- i* Michael Joosten, SBS C-LAB, joost@c-lab.de* Fuerstenallee 11, 33094 Paderborn, Germany, Phone: +49 5251 606127, Fax: +49 5251 6060658 C-LAB is a cooperation of University Paderborn & SIEMENS   ------------------------------  # Date: Thu, 05 Jul 2001 21:04:14 GMTh From: LBohan@dbc.spam_less..com " Subject: Re: The Alpha/IA64 Hybrid8 Message-ID: <lfk9ktkrcjek0v0p2kptenhcuj69u5r7pc@4ax.com>  D On Thu, 05 Jul 2001 22:30:28 +0200, Michael Joosten <joost@c-lab.de> wrote:   >David J. Dachtera wrote:e >> a >> Jack Patteeuw wrote:g >> > >> > Bill Todd wrote:  >> > . >A% >> Typical head-in-the-sand attitude.K >>  V >> > Sigh, and most Un*x and MS admins still have never heard of a ISAM file structureU >> > and are happily paying millions to Oracle, MS, etc., etc. because no Un*x vendoreS >> > would **DARE** to draw the ire of one of their major vendors by including ISAM-U >> > support "native".  Other programmers are happy with their jobs security of "rolly >> > your own" db's !!!! >> oH >> Once upon many moons ago, one or more UN*X distro "came with" C-ISAM.1 >> Been too long - don't remember which or why...t >> d >.E >This might be the times when UNIX just made its entree in commerciallD >computing, the BISON group, /usr/group, then finally OSF. The earlyH >X/Open standardisation efforts also resulted in a book covering C-ISAM. >s< >From my limited experience, most manufacturers of so-calledH >minicomputers then engaging in UNIX usually had ISAM available at leastE >as layered product, in order to provide their existing COBOL or justxG >typical commercial applications on the new OS. Relational databases at D >that time just began to enter the scene, because they were probablyF >munching away far too many cycles from the main applications to still; >provide a decent response time to the dozens of terminals.h > F >Even in '94 I was told that some small vendor just used ISAM, because* >the SQL stuff was just too big and slow.  >oH >And nowadays... What do you expect? Students must know about relationalG >DBs, and their professors like to tell them about nifty extensions for E >multimedia or even OODBs, so there is of course no/few time to teach , >them about such obsolete stuff like ISAM !?    , for those of you with greater experience w/ , more varied versions of UNIX  than myself,    3 how do the DBM-type api's fit into this picture?   .  + are they typically part of the C run-time ?c  * are the DBM api's portable bwtween Unices?  ! Could one move a DBM mapped file  + (backing store file, *.dir?) btween unices?V   ------------------------------  % Date: Fri, 06 Jul 2001 00:28:51 +0200e& From: Michael Joosten <joost@c-lab.de>" Subject: Re: The Alpha/IA64 Hybrid$ Message-ID: <3B44EA23.3F78@c-lab.de>    LBohan@dbc.spam_less..com wrote: >   - > for those of you with greater experience w/t, > more varied versions of UNIX  than myself, > B Not that I'm an expert, but I just found the old XPG3 book on Data Management.     2 > how do the DBM-type api's fit into this picture? >   F Compared with C-ISAM they are very poor. C-ISAM has about 20 functionsH in its API, locking, erasing of records, multi-part keys, even secondary$ index files for a single ISAM file.   H So, one could think of libdbm as a 'poor university's implementation' of ISAM...-  - > are they typically part of the C run-time ?-  F libdbm.a is a BSDism, so I'd guess that early SystemV R2 and 3 did notD come with it - unless they were 'value-added'... Even then, the hashF file is very 'holey' (large empty gaps of NULs), so that large indicesG consume a lot of space if the filesystem does not support sparse files.o@ E.g., backing up the index files of old SunOS/Solaris AnswerBook installations was no joy.t   > , > are the DBM api's portable bwtween Unices?   Fairly the same. o > " > Could one move a DBM mapped file- > (backing store file, *.dir?) btween unices?e  H Mmm. Officially no. If the word sizes and endianess is the same, perhaps> yes. This also depends on the alignment rules of the compiler.     -- p* Michael Joosten, SBS C-LAB, joost@c-lab.de* Fuerstenallee 11, 33094 Paderborn, Germany, Phone: +49 5251 606127, Fax: +49 5251 6060658 C-LAB is a cooperation of University Paderborn & SIEMENS   ------------------------------  # Date: Fri, 06 Jul 2001 02:30:19 GMT-" From: amolitor-at@visi-dot-com.com" Subject: Re: The Alpha/IA64 Hybrid1 Message-ID: <%o917.9797$B7.1708420@ruti.visi.com>   O In article <9i0k2s$i8u$1@pyrite.mv.net>, Bill Todd <billtodd@foo.mv.com> wrote:  >  >FAB/RAB/NAM/XAB     	AIIIIIGGGGGGGGG!G 	AIIIIIGGGGGGGGG!> 	AIIIIIGGGGGGGGG!  	AIIIIIGGGGGGGGG!o 	AIIIIIGGGGGGGGG!  	AIIIIIGGGGGGGGG!p 	AIIIIIGGGGGGGGG!H 	AIIIIIGGGGGGGGG!M  ? 	Don't say stuff like that without giving us a warning, spoilerM space, something ;)p   ------------------------------  % Date: Fri, 06 Jul 2001 02:55:59 +0000a( From: Dennis Ritchie <dmr@bell-labs.com>" Subject: Re: The Alpha/IA64 Hybrid- Message-ID: <3B4528BF.7128BDA7@bell-labs.com>h  " Michael Joosten wrote [of libdbm]:  ...H > Compared with C-ISAM they are very poor. C-ISAM has about 20 functionsJ > in its API, locking, erasing of records, multi-part keys, even secondary% > index files for a single ISAM file.e > J > So, one could think of libdbm as a 'poor university's implementation' of	 > ISAM...h  + libdbm is certainly no substitute for ISAM.t  / > > are they typically part of the C run-time ?  > H > libdbm.a is a BSDism, so I'd guess that early SystemV R2 and 3 did notF > come with it - unless they were 'value-added'... Even then, the hashH > file is very 'holey' (large empty gaps of NULs), so that large indicesI > consume a lot of space if the filesystem does not support sparse files.PB > E.g., backing up the index files of old SunOS/Solaris AnswerBook > installations was no joy.   C For the record, libdbm is a Ken-Thompsonism, was in Seventh Edition G and 32V, and picked up by Berkeley from there.  It doesn't seem to haveo> been adopted for the early System V releases, but it reappearsA in System Vr4 in a place suggesting that it repatriated from UCB.V   	Dennis    ------------------------------  % Date: Thu, 05 Jul 2001 20:37:23 +0100nL From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com>: Subject: Re: The death of VMS has been greatly exaggerated3 Message-ID: <3B44C1F3.72185B4D@remove_this.sun.com>    "Main, Kerry" wrote: >  > Aaron, > M > >>> It's how many customers will wait 18 months or 2 years so that they cani > port their apps to > yet another platform.>>> > $ > re: porting to another platform .. > L > I am not saying that the entire picture is rosy (it still has much left to? > be filled in), but is not every vendor in the same position ?  >  > - Win32 --> Win64r  3 Doesn't require new binaries because you can still c run win32 apps.h     > - IBM --> Power4  6 Doesn't require new binaries its the same architecture   > - IBM --> IA64    0 > - IBM MVS --> new 64bit mainframe architecture   Preseves the same binaries   > - Sparc II --> Sparc III  6 Doesn't require new binaries its the same architecture   > - HP PA --> IA64  3 Doesn't require new binaries its is able to decode m and run HP-PA binaries on IA-64lK > - Alpha --> IA64-2 (or whatever the enhanced IA64 + Alpha architecture isP	 > called)d >   , Requires a completely new os and ported apps  I > Given that, in most cases, various levels of emulation capabilities are M > stopgaps until the code can be recompiled into a native image, do Customersd; > have much of a choice with whatever platform they choose?g >   : You are whistling in the wind. Anything that requires new 2 hardware, a new os and all the applications to be 1 ported when compared with the list you presented  2 from the other vendors is at a huge dissadvantage.     Regardsu Andrew Harrisonu Enterprise IT Architectl   ------------------------------  % Date: Thu, 05 Jul 2001 20:41:46 +0100eL From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com>: Subject: Re: The death of VMS has been greatly exaggerated3 Message-ID: <3B44C2FA.35C8C0A3@remove_this.sun.com>    John Vottero wrote:e > @ > "Malcolm Dunnett" <nothome@spammers.are.scum> wrote in message( > news:MFVbm5gV2I4b@malvm5.mala.bc.ca...3 > > In article <tji3n47ge8e60b@news.supernews.com>,s/ > >     "John Vottero" <John@mvpsi.com> writes:w > > >rE > > > 32 bit processors are a dead end for general purpose computing.  > >(E > >     Which problem in particular do you see 32bit processors beinga@ > > inadequate for? Will those problems need to be solved by the$ > > average desktop system customer? > >e > N > It's not that all (or even many) problems need more than 32 bits.  It's thatL > most computers are used to do more than one thing.  A 32 bit chip can onlyI > address 4GB of RAM.  Every day another pointy haired boss says "Tell mehL > again why I can't put more than 4GB of RAM in the server.", "Can't we just$ > by machine with more DIMM slots?". >     > This still does not address the point which is that 99.99% of : desktop apps and users of desktops apps do not need 64 bit: addressing. They might need more than 4 GB of RAM but you # don't need a 64bit system for that.s  8 And while this ratio improves if you beleive that 64bit 6 is the be-all and end-all when you look at servers it 7 probably only improves to 80% of server apps don't need  64 bit addressing.   Regards  Andrew Harrisonx Enterprise IT Architecto   ------------------------------  % Date: Thu, 05 Jul 2001 21:49:57 +0200w= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>g: Subject: Re: The death of VMS has been greatly exaggerated) Message-ID: <3B44C4E5.E2D11E39@gtech.com>   ( Andrew Harrison SUNUK Consultancy wrote: > > - HP PA --> IA64 > 4 > Doesn't require new binaries its is able to decode! > and run HP-PA binaries on IA-64t  M > > - Alpha --> IA64-2 (or whatever the enhanced IA64 + Alpha architecture is  > > called)m > . > Requires a completely new os and ported apps  C I guess one has to be a SUN sales guy to understand why HP-UX/HP-PAoA binaries can run on HP-UX/IA-64 but VMS/Alpha binaries cannot runn on VMS/IA-64 !   Arne   ------------------------------  $ Date: Thu, 5 Jul 2001 17:11:24 -0400' From: "Bill Todd" <billtodd@foo.mv.com>r: Subject: Re: The death of VMS has been greatly exaggerated( Message-ID: <9i2kum$5k4$1@pyrite.mv.net>  7 "Arne Vajhj" <arne.vajhoej@gtech.com> wrote in messagee# news:3B44C4E5.E2D11E39@gtech.com...r* > Andrew Harrison SUNUK Consultancy wrote: > > > - HP PA --> IA64 > >u6 > > Doesn't require new binaries its is able to decode# > > and run HP-PA binaries on IA-64  >lL > > > - Alpha --> IA64-2 (or whatever the enhanced IA64 + Alpha architecture is
 > > > called)a > >b0 > > Requires a completely new os and ported apps >bE > I guess one has to be a SUN sales guy to understand why HP-UX/HP-PA C > binaries can run on HP-UX/IA-64 but VMS/Alpha binaries cannot run  > on VMS/IA-64 !  J You can, of course, run anything on anything, given sufficient motivation.L But IA64 was specifically designed to support running PA-RISC binaries (justJ how much emulation will be required I don't know, but one should assume itE will be as minimal as was possible) whereas IA64 was almost certainly:A designed with no thought whatsoever about running Alpha binaries.n   There is a difference.   - bill   >t > Arne   ------------------------------  % Date: Thu, 05 Jul 2001 23:13:45 +0200m= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>n: Subject: Re: The death of VMS has been greatly exaggerated) Message-ID: <3B44D889.1D7CAF60@gtech.com>t   John Vottero wrote: H >                                                 A 32 bit chip can only > address 4GB of RAM.o   ????  ? 32 bit and 64 bit are traditionally always referring to virtuald address space.  = There are no direct correlation between virtual address spacen and physical memory.   The best example is THE VAX.  : The latest VAX'es supported 34 bit physical address space,= which give max. og 16 GB memory (I think the VAX 10000 series ; actually could be had with up to 14 GB memory). The virtual:@ address space was ofcourse 4 GB. But that is 2 GB shared S0 & S1> address space and 2 GB process-specific P0 & P1 address space.; So you only need 6 processes to map 14 GB physical memory !2   Arne   ------------------------------  % Date: Fri, 06 Jul 2001 00:33:58 +0100rL From: Andrew Harrison SUNUK Consultancy <ah13473_nospam@remove_this.sun.com>: Subject: Re: The death of VMS has been greatly exaggerated3 Message-ID: <3B44F966.2215937A@remove_this.sun.com>u   Arne Vajhj wrote: > * > Andrew Harrison SUNUK Consultancy wrote: > > > - HP PA --> IA64 > >t6 > > Doesn't require new binaries its is able to decode# > > and run HP-PA binaries on IA-64o > O > > > - Alpha --> IA64-2 (or whatever the enhanced IA64 + Alpha architecture isa
 > > > called)a > >s0 > > Requires a completely new os and ported apps > E > I guess one has to be a SUN sales guy to understand why HP-UX/HP-PArC > binaries can run on HP-UX/IA-64 but VMS/Alpha binaries cannot runl > on VMS/IA-64 !  5 I guess you don't know that IA-64 has the ability to n decode an run HP-PA binaries.   8 I also guess that you don't know that no IA-64 processor1 has the ability to decode and run Alpha binaries.e  1 This lack of knowledge on your part puts you at ai4 dissadvantage that you cannot recover from, even HP 1 sales people are better informed than you in thisr1 case so be more polite about sales people in the t future.T     Regardsn Andrew Harrisons Enterprise IT Architectt   ------------------------------  $ Date: Thu, 5 Jul 2001 19:42:30 -0400% From: "John Vottero" <John@mvpsi.com> : Subject: Re: The death of VMS has been greatly exaggerated/ Message-ID: <tk9uqekep98507@news.supernews.com>s  F When I said "32 bit chip", I meant a chip that uses 32 bits to addressL physical memory.  I don't know how you really decide if a chip is 32, 34, 365 or 64 bit.  Maybe pick it up and look underneath?  :)c  7 "Arne Vajhj" <arne.vajhoej@gtech.com> wrote in messages# news:3B44D889.1D7CAF60@gtech.com...u > John Vottero wrote: J > >                                                 A 32 bit chip can only > > address 4GB of RAM.  >  > ???? >iA > 32 bit and 64 bit are traditionally always referring to virtualo > address space. > ? > There are no direct correlation between virtual address space- > and physical memory. >  > The best example is THE VAX. > < > The latest VAX'es supported 34 bit physical address space,? > which give max. og 16 GB memory (I think the VAX 10000 series = > actually could be had with up to 14 GB memory). The virtual B > address space was ofcourse 4 GB. But that is 2 GB shared S0 & S1@ > address space and 2 GB process-specific P0 & P1 address space.= > So you only need 6 processes to map 14 GB physical memory !a >  > Arne   ------------------------------  # Date: Thu, 05 Jul 2001 23:45:45 GMTb4 From: "Terry C. Shannon" <terryshannon@mediaone.net>: Subject: Re: The death of VMS has been greatly exaggerated: Message-ID: <J_617.277$Pf6.237394@typhoon.ne.mediaone.net>  H "Andrew Harrison SUNUK Consultancy" <ah13473_nospam@remove_this.sun.com> wrote in message   (re: porting and migration)u >m; > You are whistling in the wind. Anything that requires newr3 > hardware, a new os and all the applications to bei2 > ported when compared with the list you presented4 > from the other vendors is at a huge dissadvantage.  L Absolutely. Whilst I think IBM and HP are in the best position to capitaliseJ on this newfound opportunity, the VMS installed base harvest is big enoughH for all and sundry. I have no doubt that IBM, HP, and SUN are proceedingL accordingly, whether Compaq is cognizant of the magnitude of the exposure it$ now faces is a matter of conjecture.   ------------------------------  % Date: Thu, 05 Jul 2001 21:21:09 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>s: Subject: Re: The death of VMS has been greatly exaggerated, Message-ID: <3B451283.9682264D@videotron.ca>   "Terry C. Shannon" wrote:nN > Absolutely. Whilst I think IBM and HP are in the best position to capitaliseL > on this newfound opportunity, the VMS installed base harvest is big enoughJ > for all and sundry. I have no doubt that IBM, HP, and SUN are proceedingN > accordingly, whether Compaq is cognizant of the magnitude of the exposure it& > now faces is a matter of conjecture.    I Consider the fud that customers have generated on their own. Imagine what-. Compaq's competitors will be able to generate.  L I have a feeling that Compaq has identified the VMS sites it intends to keepF and will gladly let the smaller ones go away. Tandem survived with 3-4N customers per medium size city, and I guess Compaq will model VMS after TandemC : a tiny number of very important customers who will get individual J relationship with Compaq. I would not be susprised if those customers haveG been NDA'ed to the max about what is coming and are OK with it. And the>M smaller unimportant customers will leave due to the FUD that Compaq generatedo8 on purpose because of the way it announced the bad news.  M Think about it: no need for a big sales force, no need for marketing. Just anpK account manager for each city, and the hardware/software support folks that K visit customers. (and hardware support will probably be bundled (bungled ?) # into the wintel junk to save money.t   ------------------------------  $ Date: Thu, 5 Jul 2001 21:36:02 -0400' From: "Bill Todd" <billtodd@foo.mv.com>s: Subject: Re: The death of VMS has been greatly exaggerated( Message-ID: <9i34et$hc2$1@pyrite.mv.net>  0 "John Vottero" <John@mvpsi.com> wrote in message) news:tk9uqekep98507@news.supernews.com... H > When I said "32 bit chip", I meant a chip that uses 32 bits to addressK > physical memory.  I don't know how you really decide if a chip is 32, 34,c 367 > or 64 bit.  Maybe pick it up and look underneath?  :)s  E Most of the world decides based on the size of the unit on which mostrI machine instructions can operate (the machine's 'word size'), which oftenpJ but not always corresponds to the virtual address range made available (orE the physical address range available without remapping the underlying F physical memory).  Thus the 8086 was a 16-bit machine despite having aE 20-bit address bus (some at the time wanted to call the 8088 an 8-bitfI machine because of its 8-bit I/O architecture, but reason prevailed), the K PDP-11 a 16-bit machine despite having up to a 22-bit physical address bus,fK the VAX a 32-bit machine despite having up to a 34- (35-?) bit address bus, H the Xeon a 32-bit machine despite having up to a 36-bit address bus, theL PDP-10 a 36-bit machine despite (at least in its earlier incarnations, IIRC)5 supporting only an 18 bit process address space, etc.l  K The way you decide the bit-width class a processor falls into is to use theu& definition the rest of the world uses.   - bill   ------------------------------  % Date: Thu, 05 Jul 2001 21:44:14 -040082 From: rdeininger@mindspring.com (Robert Deininger): Subject: Re: The death of VMS has been greatly exaggeratedL Message-ID: <rdeininger-0507012144140001@user-2ivea3q.dialup.mindspring.com>  5 In article <3B451283.9682264D@videotron.ca>, JF Mezeit% <jfmezei.spamnot@videotron.ca> wrote:?    K > Consider the fud that customers have generated on their own. Imagine whatn0 > Compaq's competitors will be able to generate.  F Actually, I doubt the highly-paid professionals can do better than you+ have.  They'll be happy to tie your record.l   -- i Robert Deininger rdeininger@mindspring.com    ------------------------------  $ Date: Thu, 5 Jul 2001 22:00:03 -0400' From: "Bill Todd" <billtodd@foo.mv.com>s: Subject: Re: The death of VMS has been greatly exaggerated( Message-ID: <9i35rv$i9s$1@pyrite.mv.net>  ? "Robert Deininger" <rdeininger@mindspring.com> wrote in messageoF news:rdeininger-0507012144140001@user-2ivea3q.dialup.mindspring.com...7 > In article <3B451283.9682264D@videotron.ca>, JF Mezeir' > <jfmezei.spamnot@videotron.ca> wrote:  >a >yH > > Consider the fud that customers have generated on their own. Imagine what2 > > Compaq's competitors will be able to generate. > H > Actually, I doubt the highly-paid professionals can do better than you- > have.  They'll be happy to tie your record.e  G Au contraire:  the professionals won't be at all fettered by the truth,rK whereas we've been careful to stick to it (or as close to it as is possible J to guess, given that Compaq has never been all that forthcoming in details2 save for the ones that later turn out to be lies).   - bill   >u > -- > Robert Deininger > rdeininger@mindspring.com9   ------------------------------  # Date: Thu, 05 Jul 2001 23:00:39 GMTB. From: "mulp" <michaelpettengill@earthlink.net>% Subject: Re: VAX 4000-106 qbus pinoutAD Message-ID: <rk617.5883$oa1.561257@newsread1.prod.itd.earthlink.net>  7 "Dave McDonald" <browser@acenet.co.za> wrote in messageG' news:9hrjj3$85c$1@ctb-nnrp1.saix.net...o? > This may not be the ideal group, but there's no hardware.vax.- >-H > I'm looking for the pinout of the Qbus expansion connectors on the VAX 4000 > 105 and 106.  A I noticed VAX hardware microfiche listed on ebay - search for vax8   ------------------------------  % Date: Fri, 06 Jul 2001 00:58:40 +0100d5 From: "Steeples, Oliver" <Oliver.Steeples@compaq.com>n% Subject: RE: VAX 4000-106 qbus pinoutoN Message-ID: <F498D199EDB12D468CD2C66680D308018B1568@reoexc04.emea.cpqcorp.net>  " this site has loads of Qbus stuff:  ' http://www.vaxarchive.org/hw/index.htmlf       -----Original Message-----3 From: mulp [mailto:michaelpettengill@earthlink.net]p$ Sent: Friday, July 06, 2001 12:01 AM To: Info-VAX@Mvb.Saic.Coms% Subject: Re: VAX 4000-106 qbus pinout     7 "Dave McDonald" <browser@acenet.co.za> wrote in messageh' news:9hrjj3$85c$1@ctb-nnrp1.saix.net...t? > This may not be the ideal group, but there's no hardware.vax.i >hH > I'm looking for the pinout of the Qbus expansion connectors on the VAX 4000 > 105 and 106.  A I noticed VAX hardware microfiche listed on ebay - search for vaxd   ------------------------------  # Date: Fri, 06 Jul 2001 01:51:42 GMTb$ From: Scott Vieth <svieth@wi.rr.com> Subject: Re: vax 4000/90) Message-ID: <3B451A44.4B057E97@wi.rr.com>p  N I know.  I upgraded a couple 4000/90s with Nemonix motherboards which took theM systems up to something like 50 VUPs (can't remember the exact number off the O top of my head).  Those systems could hold their own compute-wise against a VAXj 6630.:  
 -scott :^)   mark wrote:e  ? > Thats not nice, were a real masterpiece in their time...lololm > 3 > "Scott Vieth" <svieth@wi.rr.com> wrote in messageh% > news:3B413979.614FB00A@wi.rr.com...rN > > You're chasing a VAXstation 4000 model 90?  It shouldn't take you too long > to* > > catch it.  They can't run very fast... > >s > > -Scott ;^) > >  > > mark wrote:i > >l > > > Hi there,o > > >fK > > > I am chasing a VAX 4000/90 with tape drive preferably, I thought that  > this > > > was a good place to ask. > > >yK > > > If you think that you can help and you are in Australia please let mee > know.s > > >e > > > Thanks > >    ------------------------------  # Date: Fri, 06 Jul 2001 01:46:36 GMT $ From: Scott Vieth <svieth@wi.rr.com>0 Subject: VMS-style clustering comes to Linux?!?!) Message-ID: <3B451913.A69CA58C@wi.rr.com>-  / This little goodie showed-up in my inbox today:t& ====================================== In this issue:  ! * VAX clustering makes a comebackI * Links related to Linux * Featured reader resource  ? _______________________________________________________________e TECHNOLOGY & VENDOR NEWS ALERTSp7 Sign-up to receive e-mail news alerts on LANs, Storage,e? Network/Systems Management, The Edge, Cisco and Microsoft. Keeph; abreast of the most significant developments of the week int< these specific technologies and for these vendors! Subscribe& today at http://nww1.com/go/ad082.html? _______________________________________________________________ $ Today's focus: A blast from the past   By Phil Hochmuth  9 Devotees of Digital Equipment were probably saddened when = Compaq announced last week that it would phase out its 64-bitU? Alpha chip by 2004. So goes another old DEC technology into the: history books.  = However, one group of programmers is bringing another old DECm9 technology - namely, VAX clustering - back into use, thisD time with Linux.  : Under a project called Distributed Lock Manager (DLM), the7 creators tout the software as "an implementation of then; classic VAX cluster locking semantics for a Linux cluster."   = VAX is DEC's old mini-computer which housed the VMS operatinge: system and for years was the standard in high-availability5 computing in large enterprises, rivaled only by IBM'sl
 mainframe.  = The machine was known for its superb clustering capabilities,?= which allowed between a dozen and several hundred machines tor< be grouped together to act as one large system. Nodes in the< cluster could be brought up and down without notice from end; users, making the cluster very reliable and scalable. Whileg> clustering technology exists today for Unix, Windows and Linux< platforms, many long-time technology observers and users say$ that only the VAX ever got it right.  > Users with knowledge of VAX cluster administration should find9 DLM software familiar. DLM consists of software, a kernel ? module and daemon, plus shared API. The shared API runs on eachh; client that controls the locking and unlocking of resourcesn> hosted by the cluster, such as applications and files. The DLM7 software also manages recovery of failed nodes, and thee% addition of new nodes to the cluster.   ; The DLM code was previously used to support IBM's AIX-basedu9 high-availability system. However, with support for eight== nodes, DLM does not match the clustering scale of the VAX, ore7 even the commercial clustering systems available today.e> However, for users looking to put together a smallish, highly-? available Linux cluster with Linux, DLM might be worth checkingo out.  > While documentation for the software is thin, DLM creators say? some documentation from IBM on its high-availability clustering : is complimentary to the DLM software, and is available at:9 http://www.rs6000.ibm.com/software/downloads/ha44lock.pdfo  ? _______________________________________________________________n To contact Phil Hochmuth:   6 Phil Hochmuth is a staff writer for Network World, and1 a former systems integrator. You can reach him atm mailto:phochmut@nww.com.? _______________________________________________________________y RELATED EDITORIAL LINKSh   Check out the DLM web page8 http://oss.software.ibm.com/developerworks/projects/dlm/   Download the softwaresL http://oss.software.ibm.com/developerworks/project/showfiles.php?group_id=27   &release_id=65   ------------------------------  $ Date: Thu, 5 Jul 2001 22:14:55 -0400' From: "Bill Todd" <billtodd@foo.mv.com>h4 Subject: Re: VMS-style clustering comes to Linux?!?!( Message-ID: <9i36nr$iof$1@pyrite.mv.net>  G Another sterling example of Compaq's VMS marketing:  the author clearlysK thinks that VMS clustering (and VMS itself) faded with the VAX ('blast from  the past' indeed).  K Without having followed up the references, it looks as if this is IBM's AIXaK version of the DLM, created around 1994 to support OPS on SP systems (whichlL qualify as 'clusters', albeit somewhat different ones).  If so, this will beI at least the second VMS DLM imitator on Linux, since IIRC Stephen TweedieaG has been creating something similar.  And Compaq's "Single System ImagelG Clusters for Linux" open-source project is also contributing a DLM (and1L associated cluster file system), though I don't happen to know if it's a VMS
 DLM clone.   - bill  1 "Scott Vieth" <svieth@wi.rr.com> wrote in message # news:3B451913.A69CA58C@wi.rr.com...n1 > This little goodie showed-up in my inbox today: ( > ====================================== > In this issue: > # > * VAX clustering makes a comebackt > * Links related to Linux > * Featured reader resource > A > _______________________________________________________________c! > TECHNOLOGY & VENDOR NEWS ALERTSH9 > Sign-up to receive e-mail news alerts on LANs, Storage,sA > Network/Systems Management, The Edge, Cisco and Microsoft. Keepr= > abreast of the most significant developments of the week ina> > these specific technologies and for these vendors! Subscribe( > today at http://nww1.com/go/ad082.htmlA > _______________________________________________________________G& > Today's focus: A blast from the past >- > By Phil Hochmuth   ...    ------------------------------   Date: 5 Jul 2001 13:13:39 -0700a% From: jliddil@liddil.com (Jim Liddil)> Subject: vms/pined< Message-ID: <6740c7a9.0107051213.4ca7c3b@posting.google.com>  A I ahve an account on a vms system. Previously PINE worked fine (vH? 4.21L).  Now the system has been upgraded to VMS 7.3 and when I 8 compose or reply  I get a reg dump like this.  Any help?   Jim   ; %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtuale address=000000000000& 0081, PC=FFFFFFFF8098A960, PS=0000001B North Haven, CTr2   Improperly handled condition, image exit forced.1     Signal arguments:   Number = 0000000000000005P1                         Name   = 000000000000000CM1                                  0000000000000000 1                                  0000000000000081n1                                  FFFFFFFF8098A960a1                                  000000000000001Bb  C     Register dump:nd      ^R Read File ^Y Prev Pg   ^K Cut Text  ^Oo Postpone9     R0  = FFFFFFFF809A7760  R1  = 000000007AEB87C8  R2  =v 000000007BF77768ll9     R3  = 0000000000000005  R4  = 000000000135F7D0  R5  =i 00000000000000009     R6  = 0000000000000000  R7  = 000000000000014D  R8  =e 000000007AEB8B989     R9  = 0000000000000000  R10 = 0000000000F36448  R11 =0 00000000013E48A89     R12 = 0000000000000000  R13 = 0000000000000000  R14 =a 000000000121D9E89     R15 = 000000000001C130  R16 = 0000000000000081  R17 =r 00000000000010009     R18 = 0000000000000005  R19 = 0000000000000001  R20 =a 00000000000000009     R21 = 000000007AEB87F8  R22 = 0000000000000001  R23 =  00000000000000009     R24 = FFFFFFFFFFFFFFFF  R25 = 0000000000000001  R26 =j FFFFFFFF809A77B89     R27 = 000000007BF77768  R28 = 0000000000000000  R29 =m 000000007AEB87609 $ daSP  = 000000007AEB8750  PC  = FFFFFFFF8098A960  PS  =  100000000000001B   ------------------------------  $ Date: Thu, 5 Jul 2001 15:21:52 -0400 From: "John McDen" <jj@jj.moc> Subject: VMSTAR question+ Message-ID: <9i2el1$8fa$1@bob.news.rcn.net>e  H How do I take a back of all the files in the main dir and all the subdirI with the VMSTAR.exe utility. When I try to backup it only takes the filescD from the current dir and just copies the empty dir without any files   Please help me ....    ------------------------------  % Date: Thu, 05 Jul 2001 21:17:12 -0500B1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: VMSTAR question' Message-ID: <3B451FA8.78A64BD8@fsi.net>e   John McDen wrote:s > J > How do I take a back of all the files in the main dir and all the subdirK > with the VMSTAR.exe utility. When I try to backup it only takes the filesuF > from the current dir and just copies the empty dir without any files >  > Please help me .....   This works for me:  . $ VMSTAR -CF tar_archive_name [.dirname...]*.*   -- t David J. Dachtera  dba DJE Systems: http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/l  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------   Date: 5 Jul 2001 13:19:05 -0700I+ From: stephane_paquin@hotmail.com (SPaquin)w Subject: VT520 F2 key in FMS< Message-ID: <fdd7874.0107051219.11e2a3d5@posting.google.com>   Hi !  E On our old VT420, the F2 key was defined as F2( F2=F2 ) In FMS forms,0< it was returning 44 as field terminator and it was processed accordingly.  E On the new VT520, I am having trouble defining the same behavior. HowrE can I define the VT520 F2 key so that it sends the same key code thana the VT420 F2=F2 definition ?   Any help will be appreciated.d Stphane Paquin    ------------------------------  % Date: Thu, 05 Jul 2001 15:36:45 -0400o- From: "Richard D. Piccard" <piccard@ohio.edu>oD Subject: Re: Wailing and moaning.... (was: Compilers go to Intel...)( Message-ID: <3B44C1C1.24C7F5A1@ohio.edu>   Larry Kilgallen wrote:   > [snip]  ? > I would strongly suggest, then, that you _read_ what has beenl6 > posted here in the interim, before writing about it.   [snip]    m I don't know about others, but my newsfeed for some months now has exhibited a delay between posting date andmm availability for viewing that has a standard deviation on the order of ten days, with out-of-sequence arrivalto routine.  It is quite possible (in theory) then, to be completely caught up with what is available on your newsN@ server and still not have seen all the replies that others have.  +                                         RDPt   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  $ Date: Thu, 5 Jul 2001 16:57:48 -0400' From: "Bill Todd" <billtodd@foo.mv.com>-, Subject: Re: What about performance issues??( Message-ID: <9i2k58$544$1@pyrite.mv.net>  5 "Keith Parris" <kparris@my-deja.com> wrote in messaget7 news:cb85fed2.0107050734.4eca2e2c@posting.google.com...t4 > "Bill Todd" <billtodd@foo.mv.com> wrote in message$ news:<9ht7va$qi9$1@pyrite.mv.net>...9 > > "Keith Parris" <kparris@my-deja.com> wrote in messaged; > > news:cb85fed2.0107030817.7abe3d38@posting.google.com...iJ > > > And this level of I/O performance was being achieved with the older,H > > > slower HSJ50 and CI hardware, not the hot new Fibre Channel stuff;J > > > with no XFC; with plain old RMS indexed files, without using the newG > > > Fast_IO interface that is significantly streamlined compared withl > > > $QIO.  > >c< > > But, I suspect, by taking advantage of mirrored, stable, controller-levelI > > write-back caching.  Suitable hardware can always make up for a greato dealK > > in the way of software inadequacy - but most of the discussion involvedaJ > > write performance on lesser hardware where enabling write-back caching would K > > potentially compromise integrity.  (Yes, to some extent read-caching aspD > > well, and XFS has helped in that area, but it did take a while.) >iE > HSJ50s don't have mirrored cache.  We covered for that by shadowing @ > across different pairs of controllers, so that the one time weE > actually did suffer an HSJ50 cache module failure, Volume ShadowingsH > did its job, kicked out all the units on the failed controller, and we/ > continued to run on the other shadow members.   K That's certainly reasonable, given *independent and reliable* power to eachrJ controller.  But it still qualifies as a moderately high-end configurationL in my book.  Unix's approach has the advantage of working well right down to" the stand-alone-workstation level.   >aH > In my measurements, write-back cache didn't seem to make a significantE > difference in performance on solid-state disk units.  Based on thatsH > data, when we built our 2nd cluster, we disabled caching on SSD units,5 > leaving all the cache space for the magnetic disks.b  I Using controller-level write-back cache for SSDs shouldn't be expected to.E make a noticeable difference:  you've already paid the price of going3L through RMS, the file system, the driver, and the controller hardware to getI to it, and the additional overhead of going the rest of the way should be H minor.  Main-memory file-level cache (as Unix uses), however, might wellH make a noticeable difference (you could tell by setting up RMS to do the: caching and comparing uncached SSD performance with that).   >t% > This workload had about 30% writes.s >rH > My point was that to make a blanket statement that I/O on VMS is slow,@ > inefficient, etc. based on anecdotal evidence (as has happenedE > frequently in this ng recently) doesn't present a true and complete'
 > picture.  I I don't think that specific head-to-head comparisons on the same hardwareeJ using the same code merit the adjective 'anecdotal', and that's what DavidC Mathog has been presenting.  Nor do I think that explanation of thetE disparities based on analysis of what each system is doing internally ! constitutes 'anecdotal evidence'.e  I There may have been general allegations to the effect that "I/O on VMS is E slow", but certainly not from me (though David has suggested that RMS@G contributes significantly to the code-path length he's observing to the7E point that it makes a noticeable difference in operations that do notbI involve any disk access, and I do find that credible).  My point has beentL the difference in file-access performance (to normal disks, not SSDs) due toB the difference in Unix caching/buffering policies vs. those used - especially by default - by VMS.i  >   If VMS I/O was really slow and inefficient, we wouldn't have? > been able to drive the SSDs to the level of I/O rates we saw.0  J The whole discussion has been comparative, not absolute.  You can't make aL statement that's relevant to it without running comparative tests on VMS andJ Unix and including configuration data with the results so that the reasons can be analyzed.   - bill   >  > KeithO   ------------------------------  % Date: Fri, 06 Jul 2001 00:59:53 +0100 : From: andrew harrison <ah13473_nospam@remove_this.sun.com>4 Subject: Re: Why Compaq won't succeed on IA64 either3 Message-ID: <3B44FF79.900B5F71@remove_this.sun.com>    Larry Kilgallen wrote: > l > In article <xyq_6.2220$T_2.374135@news.uswest.net>, "Michael D. Ober" <mdo.@.wakeassoc.com.nospam> writes: > >hC > > "Robert Deininger" <rdeininger@mindspring.com> wrote in messageiJ > > news:rdeininger-2706011519320001@user-2ivebp7.dialup.mindspring.com...O > >> In article <9hd6i0$1v7@gap.cco.caltech.edu>, mathog@seqaxp.bio.caltech.edui > >> wrote:> > >> > >>C > >> > Think of it is insurance - you don't really want your family L > >> > to collect on your life insurance, but you'd be a fool not to pay it. > > SoL > >> > when 2003 hits, if Microsoft announces that they've just made PowerPC > > thesO > >> > primary platform for Windows and are stopping all further development on L > >> > IA64, Compaq, and more importantly, VMS, stands a chance of surviving > > it.  > >>M > >> I assume you're making the same insurance suggestion to HP and the othercF > >> unix folks who will certainly (according to you) beat Compaq intoN > >> IA64-land?  Compaq's risk at this point isn't much different than severalN > >> other companies'.  Compaq might have fired more rounds into their feet to? > >> get here, but they find themselves in a fairly large boat.h > >> > >aN > > The difference here is that IBM, HP, et.al won't have killed their primaryP > > processor/OS lines.  Compaq will be the only one without the life preserver. > < > Are you ignoring the HP EOL notice for PA-RISC processors,E > or are you claiming that the HP3000 is HP's primary processor line?L    9 Yes and no, people invest in an ABI/API, HP are claiming t4 that due to their port of HP-UX to IA-64 and IA-64's5 ability to decode and run HP-PA binaries that peopless0 investment is protected. They are claiming that / it does not matter which processor you choose. _  / This may or may not be BS, it is however betterH, BS than the OpenVMS/Alpha->OpenVMS/IA-64 BS.     Regards  Andrew HarrisonS Enterprise IT Architect    ------------------------------  $ Date: Thu, 5 Jul 2001 21:56:42 -0400) From: "Neil Rieck" <n.rieck@sympatico.ca> 4 Subject: Re: Why Compaq won't succeed on IA64 either; Message-ID: <bR817.26942$g92.3465289@news20.bellglobal.com>_  L I have no problem with Compaq playing both sides of the field. It only makesD sense that they leverage their compiler (and OS) "know how" on every platform in the marketplace.  L I do have a problem with them cancelling EV8 etc. If it was costing too muchI to develop new Alphas, then they could develop new evolutions at a slower J pace. With EV8 as the center piece, they could sell off the Alpha divisionL to some one who might want to really run with it (Samsung, IBM, AMD, Apple?)J for big bucks rather than letting the technology wither on the vine. Also,K if new Alpha development continued then Compaq would have something to fall 3 back onto if IA-64 is delayed or just doesn't sell.t  K It is my belief that glueless SMP with EV7, and SMT with EV8, will beat thelG pants off EPIC and whatever "Intel macroprocessing" is supposed to meannF anyway so slower major evolutions probably wouldn't make that big of a% difference in the vendor race anyway.e   ***d  F ps. Has anyone seen the prices of Itanium chips yet. If not, check out Intel's price list here:4 http://www.xbitlabs.com/news/story.html?id=994263536  G Does anybody in this news group believe that the $4200 models will everaL sell? Money is tight everywhere and may continue to be so for a long time toL come. If I was running Compaq I'd keep all my bases covered so I'd have moreI options (what's that quote about the "principle of the excluded middle"?)e   ***u  
 Neil Rieck Kitchener/Waterloo/Cambridge,I Ontario, Canada.! http://www3.sympatico.ca/n.rieck/u@ http://www3.sympatico.ca/n.rieck/links/compaq_memorial_site.html   ------------------------------  # Date: Fri, 06 Jul 2001 02:53:38 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>4 Subject: Re: Why Compaq won't succeed on IA64 either: Message-ID: <SK917.417$vb6.203126@typhoon.ne.mediaone.net>  4 "Neil Rieck" <n.rieck@sympatico.ca> wrote in message5 news:bR817.26942$g92.3465289@news20.bellglobal.com...bH > I have no problem with Compaq playing both sides of the field. It only makesaF > sense that they leverage their compiler (and OS) "know how" on every > platform in the marketplace. > I > I do have a problem with them cancelling EV8 etc. If it was costing tooy muchK > to develop new Alphas, then they could develop new evolutions at a slowerwL > pace. With EV8 as the center piece, they could sell off the Alpha divisionF > to some one who might want to really run with it (Samsung, IBM, AMD, Apple?)tL > for big bucks rather than letting the technology wither on the vine. Also,H > if new Alpha development continued then Compaq would have something to fall5 > back onto if IA-64 is delayed or just doesn't sell.   J The lack of a fallback or "Plan B" strategy is worrisome. Compaq is now atI the mercy of Intel and its ability to deliver competitive processors in arK timely and predictable manner. It's not nice guys who finish last, its guys-$ who practice unilateral disarmament.   >5I > It is my belief that glueless SMP with EV7, and SMT with EV8, will beatu thenI > pants off EPIC and whatever "Intel macroprocessing" is supposed to meantH > anyway so slower major evolutions probably wouldn't make that big of a' > difference in the vendor race anyway.i  C That was CPQ's party line through June 24. One wonders if CPQ knows . something about McKinley that we don't know...   ------------------------------  $ Date: Fri, 6 Jul 2001 00:05:46 -0400' From: "Bill Todd" <billtodd@foo.mv.com>o4 Subject: Re: Why Compaq won't succeed on IA64 either( Message-ID: <9i3d7n$n8v$1@pyrite.mv.net>  ? "Terry C. Shannon" <terryshannon@mediaone.net> wrote in message 4 news:SK917.417$vb6.203126@typhoon.ne.mediaone.net... >s6 > "Neil Rieck" <n.rieck@sympatico.ca> wrote in message7 > news:bR817.26942$g92.3465289@news20.bellglobal.com...    ...V  K > > It is my belief that glueless SMP with EV7, and SMT with EV8, will beat  > the4K > > pants off EPIC and whatever "Intel macroprocessing" is supposed to meanyJ > > anyway so slower major evolutions probably wouldn't make that big of a) > > difference in the vendor race anyway.a > E > That was CPQ's party line through June 24. One wonders if CPQ knows_0 > something about McKinley that we don't know...  D Since when has it appeared that Compaq knew anything (except its own' internal idiocies) that we didn't know?t  B McKinley is expected to be about twice as fast as Itanic, which myE not-very-informed impression is should make it about equal in integer L performance to EV69(?) fully-adapted to .18 micron technology (faster on FP,H but most of the non-HPTC world can't make very much use of that).  GivenK that the performance-enhancement path *beyond* McKinley is to say the least-I murky (unlike that for Alpha beyond EV7), McKinley would have to blow theiK doors off the above expectations just to stay even with EV7, let alone formi+ a base from which IA64 could challenge EV8.s   - bill   ------------------------------  # Date: Fri, 06 Jul 2001 04:50:44 GMTo. From: "mulp" <michaelpettengill@earthlink.net>4 Subject: Re: Why Compaq won't succeed on IA64 eitherD Message-ID: <Esb17.4835$G_1.467946@newsread2.prod.itd.earthlink.net>  4 "Neil Rieck" <n.rieck@sympatico.ca> wrote in message5 news:bR817.26942$g92.3465289@news20.bellglobal.com... H > I have no problem with Compaq playing both sides of the field. It only makes F > sense that they leverage their compiler (and OS) "know how" on every > platform in the marketplace. >0I > I do have a problem with them cancelling EV8 etc. If it was costing too  muchK > to develop new Alphas, then they could develop new evolutions at a sloweriL > pace. With EV8 as the center piece, they could sell off the Alpha division  H For almost 4 years, Alpha development has been slowed down.  The lawsuitK against Intel, which might have been justified, didn't help, but as soon as7I the announcement was made to sell part of the chip biz to Intel, followed K soon after by the proposed sale to Compaq, things really started suffering.   K This started a series of big slips in EV6, Wildfire, EV7, and by extension,1E EV8.  People bailed, designs were switched from one process vendor to0! another, people demoralized, etc.   8 Compaq had at least three years to correct this problem.   And they did on June 25.  K One thing that DEC was accused of was making and remaking the same decisionAL over and over, but this happened because there was considerable debate about+ the merits of doing or not doing something.F  I Compaq doesn't debate decisions - they sort of make a decision and let iteI stand for a few years until its clear that it was a bad decision and thenVA they make another decision and let it stand for a few more years.   K Whether there is a viable migration plan for moving VMS and Tru64 operatingVF systems, applications, and customers while maintaining some reasonableH amount of business in the process and using the new platform to grow the business is an open question.-  K We have been debating the issues here for two weeks with no clear concensuse on the eventual outcome.&   - some think that this will be worse'   - some think that this will be bettern5   - many think that it won't make a bit of differencei1   - there are probably quite a few who don't knowV  L My sense of the discussion is that the prospects of the outcome being better is a small minority.  F Certainly a major negative factor in forming an opinion is the lack of information.  H The history of the past three years is not encouraging - what changed inI just the last 3 months.  If the decision to kill Alpha is based on CompaqoI deciding that the PC business is not the future, and that the dotcom meltnL down means that Alpha won't get market share, then why did it take a year toG figure this out.  Why wasn't there more discussion of the alternatives.o  I Years of poor decisions and/or execution by DEC, by Digital, by Compaq...y   Why will this be better?   ------------------------------  $ Date: Thu, 5 Jul 2001 21:47:53 -0500% From: "Rich Jordan" <rjordan@mcs.net>V' Subject: Re: will the irony never cease05 Message-ID: <hC917.14271$j02.214743@news.goodnet.com>r  % Terry C. Shannon wrote in message ...  >F. >"Nic Clews" <nclews@csc.com> wrote in message8 >news:a720d610.0107050332.21d3b705@posting.google.com...@ >> mathog@seqaxp.bio.caltech.edu (David Mathog) wrote in message) >news:<9hlbbi$keb@gap.cco.caltech.edu>...s   ...mF >> > At least there was an emergency contact number supplied below the boilerK >> > plate, so I guess they have a few people working this week.  Still, it1G >> > seems like a good 7 days not to have to try to employ your service:
 >contract. >>G >> I had to raise a call at 0700 hours Mountain Time on July 4th, and I5I >> had a specialist on the problem in less than 30 minutes, and complete,g3 >> correct problem diagnosis by 0745 Mountain Time.h >> >tJ >A reseller friend of mine had a similar experience on the Fourth of July.F >The marketeers might be hibernating, but the services folks are stillB >manning the frontiers of availability and the bastions of uptime. >oJ Thats better than I was getting the week before.  30 minutes on hold and IL hadn't got a call screener yet.  Second problem call too 15 minutes on hold.L In all the years we've had Digital service and the descendant (in many ways)F Compaq service, I  had never had to wait more than 3-4 minutes before.J Maybe all the management types and beancounters should take more vacations/ like this one... it seems to improve service :)c   Rich Jordan  rjordan@mcs.net    ------------------------------   End of INFO-VAX 2001.371 ************************ersity   ------------------------------  $ Date: Thu, 5 Jul 2001 16:57:48 -0400' From: "Bill Todd" <billtodd@foo.mv.com>-, Subject: Re: What about performance issues??( Message-ID: <9i2k58$544$1@pyrite.mv.net>  5 "Keith Parris" <kparris@my-deja.com> wrote in messaget7 news:cb85fed2.0107050734.4eca2e2c@posting.google.com...t4 > "Bill Todd" <billtodd@foo.mv.com> wrote in message$ news:<9ht7va$qi9$1@pyrite.mv.net>...9 > > "Keith Parris" <kp