1 INFO-VAX	Thu, 12 Jul 2001	Volume 2001 : Issue 383       Contents: .NET on VMS 5 Re: Alpha fiasco looses $30M deal to IBM in Singapore 5 Re: Alpha fiasco looses $30M deal to IBM in Singapore ! Re: Attunity Connect and Oracle 8 ! Re: Attunity Connect and Oracle 8  Re: CETS2001 InquiryD Re: Compilers go to Intel, hobbyist, CSLG, etc. goes out the window? Re: Consultant needed "C-GOD"  Re: Copying "foreign" tapes... CSWS V1.1 and cert sellers! Re: DEC Notes available, someone? ! Re: DEC Notes available, someone? ! Re: DEC Notes available, someone? ! Re: DEC Notes available, someone? ! RE: DEC Notes available, someone? ! Re: DEC Notes available, someone?  Re: Experience with EMC storage  Re: Experience with EMC storage  Re: Experience with EMC storage  Re: Experience with EMC storage  Fortran... IRAD50  Re: Fortran... IRAD50  Re: Fortran... IRAD50  Re: Fortran... IRAD50  Re: FUD  Re: FUD  Re: FUD  Re: FUD  Re: FUD  Re: Future support of VAX-VMS  Re: Future support of VAX-VMS  Re: Future support of VAX-VMS 
 Re: Hobbyists 
 Re: Hobbyists 
 Re: Hobbyists 
 Re: Hobbyists  RE: IA64 Rocks My World  RE: IA64 Rocks My World  Re: IA64 Rocks My World  Re: Intel/Alpha announcment  RE: Intel/Alpha announcment  Re: Intel/Alpha announcment  Re: Memory Channel II . Re: Memory Channel II (and an Itanium comment). Re: Memory Channel II (and an Itanium comment). Re: Memory Channel II (and an Itanium comment) PAWZ dicontinued Re: PAWZ dicontinued Re: PAWZ dicontinued1 Re: Some thoughts on the recent turn of events... 1 RE: Some thoughts on the recent turn of events... 1 Re: Some thoughts on the recent turn of events... 1 Re: Some thoughts on the recent turn of events... $ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-64$ Re: Terry Shannon Tech Talk on IA-641 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: VMS V7.3 SPD Error Re: VT520 F2 key in FMS  Re: VXT 2000 with VMS?!  Re: VXT 2000 with VMS?!  Re: What will we see at 2004  F ----------------------------------------------------------------------  % Date: Wed, 11 Jul 2001 14:49:27 -0500 : From: "Scandora, Anthony \(35048\)" <scandora@cmt.anl.gov> Subject: .NET on VMS+ Message-ID: <9iiak8$ilm$1@milo.mcs.anl.gov>   L Ximian and the Free Software Foundation just announced the Mono project, "an? effort to create a free software (sometimes called open source) 7 implementation of the .NET Development Framework."  See K http://www.ximian.com/mono/ for details.  This is after Microsoft and Corel ' announced a .NET porting effort to BSD.   C If a highly visible anti-Microsoft group is working to attract .NET J application developers to GNU/Linux, and even Microsoft is at least givingF the appearance of working to attract .NET developers to BSD, one mightK believe that developers will be writing .NET applications.  If so, wouldn't J VMS be a good platform for .NET?  Compaq certainly is well occupied at theL moment with the Itanium announcement, but I sure would like to see a .NET on$ VMS announcement in the near future.  1 Tony Scandora, Argonne National Lab, 630-252-7541  scandora@cmt.anl.gov   ------------------------------  # Date: Wed, 11 Jul 2001 18:43:10 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>> Subject: Re: Alpha fiasco looses $30M deal to IBM in Singapore7 Message-ID: <27137.5$XT4.14431@typhoon.ne.mediaone.net>   2 "Alan Greig" <a.greig@virgin.net> wrote in message2 news:l4boktgqjugb4epqj3vd1hcv8tcipv0ev8@4ax.com...   <yadda yadda yadda SNIP> >  > > G > >Letting Alpha die and all of the O/S platforms which boot upon it is  clear L > >and decisive proof.  If they keep on this course, your newsletter will beK > >undergoing another name change to "Shannon Knows Nothing".  Not meant to  beG > >a slur against you -- simply that there will be nothing left for you  report.  >  > Shannon Knew Stuff?   J Oh, there would be a seamless transition through the intermediate "ShannonJ Knew Compaq" branding. Still, there are more secure career paths, and lord- knows there are things that pay a lot better.     A > Beavis and Butthead do Compaq - the story of Michael 'Butthead' C > Capellas and Mike 'Beavis' Winkler tour of destruction of Compaq?  > That SUCKS!!!!!!!   L The irony is, a guy I once partnered with (before his Illuminata got big andH famous) and I thought of doing a Beavis and Butt-head skit about Digital marketing and strategy...   K Oh, by the way, The Inquirer got it a bit wrong. Never was a $30M deal with J the Singapore govt. A couple of smallish HPTC bids in play ( a few millionJ bucks), but IBM was favored to win those right from the get-go. June 25 is not a factor in either bid.    ------------------------------  # Date: Wed, 11 Jul 2001 18:44:38 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>> Subject: Re: Alpha fiasco looses $30M deal to IBM in Singapore7 Message-ID: <q8137.6$XT4.15200@typhoon.ne.mediaone.net>   2 "Alan Greig" <a.greig@virgin.net> wrote in message2 news:9ucokt8vh104nekvu0j1i8j9vpb031cq2b@4ax.com...F > On 10 Jul 2001 17:09:37 -0700, morton_john@hotmail.com (John Morton) > wrote: >  > >Compaq Death Spiral > >  > L >http://www.washingtonpost.com/wp-srv/aponline/20010710/aponline171821_000.h tm > >  > >The Associated Press ( > >Tuesday, July 10, 2001; 5:18 p.m. EDT > > H > >HOUSTON &#8211;&#8211; In its second wave of staffing cuts this year,F > >Compaq Computer Corp. said Tuesday it was trimming 4,000 more jobs,D > >bringing the total number of workers it plans to cut this year to	 > >8,500.  > H > I wonder if this number includes those transferring to Intel. Probably > not.  K I suspect you are right. $185M of restructuring is dedicated to writing off ) EV8 and future Alpha development, though.    ------------------------------  % Date: Wed, 11 Jul 2001 14:29:21 -0400 2 From: "Sue Skonetski" <susan.skonetski@compaq.com>* Subject: Re: Attunity Connect and Oracle 82 Message-ID: <_V037.581$rc5.39908@news.cpqcorp.net>   Ed,   3 I am on the phone with Ron from Attunity right now.   F The OPP (On Platform Package) which is a special version of Attunity'sG Connect product that was previously defined by Compaq and bundeled into K every version of OpenVMS V7.1 and beyond) this version provides data access  to Oracel 8i datasources  " Ron's phone number is 781-359-3584* email address is 'Ron.Denaro@attunity.com'   Hope this helps,   Sue       1 "Ed wilts" <ewilts@mediaone.net> wrote in message 7 news:e12df3dd.0107110918.3ced16de@posting.google.com... D > The Attunity Connect product from Compaq currently has support forE > Oracle 7, but nothing in the docs has anything about Oracle 8.  Has H > anybody made it work with Oracle 8 or know of Compaq's plans to update > it?  > 	 > Thanks,  >    .../Ed  >  > mailto:ewilts@ewilts.org   ------------------------------  % Date: Wed, 11 Jul 2001 22:37:46 -0400 * From: Eric Ebinger <eebinger@telocity.com>* Subject: Re: Attunity Connect and Oracle 88 Message-ID: <002e01c10a7b$a40f5d10$0200a8c0@teamrdb.com>  ; And would someone explain to me WHY the On Platform Package 0 for OpenVMS doesn't include an interface to Rdb?   Eric Ebinger   ----- Original Message -----0 From: Sue Skonetski <susan.skonetski@compaq.com> To: <Info-VAX@Mvb.Saic.Com> & Sent: Wednesday, July 11, 2001 2:29 PM* Subject: Re: Attunity Connect and Oracle 8     > Ed,  > 5 > I am on the phone with Ron from Attunity right now.  > H > The OPP (On Platform Package) which is a special version of Attunity'sI > Connect product that was previously defined by Compaq and bundeled into F > every version of OpenVMS V7.1 and beyond) this version provides data access > to Oracel 8i datasources > $ > Ron's phone number is 781-359-3584, > email address is 'Ron.Denaro@attunity.com' >  > Hope this helps, >  > Sue  >  >  > 3 > "Ed wilts" <ewilts@mediaone.net> wrote in message 9 > news:e12df3dd.0107110918.3ced16de@posting.google.com... F > > The Attunity Connect product from Compaq currently has support forG > > Oracle 7, but nothing in the docs has anything about Oracle 8.  Has J > > anybody made it work with Oracle 8 or know of Compaq's plans to update > > it?  > >  > > Thanks, 
 > >    .../Ed  > >  > > mailto:ewilts@ewilts.org >  >  >    ------------------------------    Date: 11 Jul 2001 18:17:30 -0700( From: kparris@my-deja.com (Keith Parris) Subject: Re: CETS2001 Inquiry = Message-ID: <cb85fed2.0107111717.3f7b0e5a@posting.google.com>   e medleyb@ev1.net (Bert Medley) wrote in message news:<Xns90DB696E0427Dmedleybev1net@207.218.245.68>... K > It's been a while since I've been to a "Decus" symposium.  Do they still  " > have the "war stories" sessions?  E You're probably thinking of the "Magic" sessions, or perhaps 'Chiller 	 Theatre'. D I never cared much for the "bribe the judges with alcohol" tradition@ of affecting a contestant's score, but the war stories were both educational and entertaining.   C At CETS2000, the program team was reportedly reluctant to devote an F evening time slot to a Magic session.  I don't think they realized theC value that 'Magic' has had to attendees as a part of the culture of D the user group.  If you feel it would be worthwhile to have one thisC time around, as I do, it would be a good idea to submit an official E suggestion for this type of session at www.cets2001.org -- follow the . "Call for papers -- Submit session idea" link.C ------------------------------------------------------------------- C Keith Parris | parris at encompasserve dot org | VMS consulting on: C Clusters, Disaster Tolerance, Internals, Performance, Storage & I/O    ------------------------------  % Date: Wed, 11 Jul 2001 21:50:39 +0100 + From: "antonio.carlini" <arcarlini@iee.org> M Subject: Re: Compilers go to Intel, hobbyist, CSLG, etc. goes out the window? ' Message-ID: <3B4CBC1F.D18CA814@iee.org>    Alan Greig wrote: G > If contracts have been signed which specify things in black and white F > then it is not useful in a practical sense. In a let off steam senseF > it might be but that will die down if and only if all of this works.G > I'll take your hint and try and leave you folks to it until some more  > news good or bad appears.   . In my (one) experience of a similar situation,3 things *do* get missed. I would be surprised if the 5 written contract tied down every last detail. I would 1 hope that it is sufficiently flexible that little 2 niggles can be sorted out in a reasonable fashion.   Antonio    --     --------------- - Antonio Carlini             arcarlini@iee.org    ------------------------------  % Date: Wed, 11 Jul 2001 21:07:17 -0400 " From: "Hal Kuff" <Kuff@Tessco.Com>& Subject: Re: Consultant needed "C-GOD"O Message-ID: <E5CEF0388FBC579D.A49454EA2B8D31C2.9D9BA1B0DA43C5D9@lp.airnews.net>   L     Can you e-mail me directly? The below address minus span doesn't work...? "Hoff Hoffman" <hoffman@xdelta.zko.dec.nospam> wrote in message , news:kk_27.557$rc5.39691@news.cpqcorp.net... > In articleI <DBC18F524F0B2718.BED63845A9F43E05.0E74C2202F38BFB0@lp.airnews.net>, "Hal  Kuff" <kuff@tessco.com> writes:  > J > :    We need the services of a "C-GOD" with experience in Oracle/Rdb andI > :SQLMOD.... basically someone to review our use of  Compaq C and Oracle  RDB I > :coding standards... This person should also be extremely proficient in  using D > :the VMS Debugger with C as we have one or two tough bugs to slay. > : F > :    Engagement period is at least one week, possibly two.  Reply by e-mail > :and not via this group. >  > B >   Not sure I count as a C or SQL$PRE diety...  I do have passing
 experienceG >   with C, OpenVMS, Rdb, embedded SQL, and the OpenVMS debugger, among  other H >   topics.  I've some experience with hunting for bugs in C and SC code using K >   the debugger and using other techniques.  (But I also have some hot Rdb L >   and some OpenVMS-on-IPF work running, thus my availability would be, um,K >   subject to Compaq management approval, um, limited and, um, expensive).  :-)  > E >   And we now return to our regularly scheduled dose of reality: The  OpenVMS L >   Ask The Wizard area's topic (1661) has a list of some of the more commonI >   errors found lurking on C.  qv: http://www.openvms.compaq.com/wizard/  > ( >  ---------------------------- #include' <rtfaq.h> ----------------------------- L >       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com, >  --------------------------- pure personal# opinion --------------------------- 1 >    Hoff (Stephen) Hoffman   OpenVMS Engineering  hoffman#xdelta.zko.dec.com >    ------------------------------  % Date: Wed, 11 Jul 2001 17:20:15 -0500 * From: cjt & trefoil <cheljuba@prodigy.net>' Subject: Re: Copying "foreign" tapes... + Message-ID: <3B4CD11F.10879CC9@prodigy.net>   D Are they even 9-track tapes?  As I recall, not all Cyber tapes were.   "James L. Wiley" wrote:  > M > I am trying to figure out how to copy some tapes from a non-VMS system (CDC M > Cyber) to files on disk.  I don't want any kind of conversion or changes to M > the data.  I even want all the label information and any end of file marks, N > etc to go along.  My plan is to download these "container" files to a PC andH > burn CD's.  The tape drive on the VMS system is a SCSI attached TSZ07. > M > I have tried mounting a tape with the /over=all parameter, but I get errors , > if I just try to use the VMS copy command. > 0 > Could someone point me in the right direction? >  > Thanks >  > James L. Wiley > Wiley@Tarleton.edu   ------------------------------  # Date: Wed, 11 Jul 2001 22:14:25 GMT + From: kuhrt@encompasserve.org (Marty Kuhrt) # Subject: CSWS V1.1 and cert sellers 3 Message-ID: <TANTx3WWo9O1@eisner.encompasserve.org>   D I'm working on the latest and greatest Apache server (CSWS V1.1) and> was thinking of serving up some pages with HTTPS.  So far, forC testing, I have been using self-signed certificates.  When I deploy F this so that people from the outside world can get to it, I'm assuming> that I have to get a certificate from a trusted source.  I'm a@ complete newbie to this stuff, so it could be that I'm not quite getting it.   B Anyone have any pros or cons for the various places that do this? C I've seen VeriSign and Thawte offer this for different prices, $499 < and $199, IIRC.  Why the cost difference?  Any other places?   Thanks in advance, Marty    ------------------------------  # Date: Wed, 11 Jul 2001 18:08:50 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)* Subject: Re: DEC Notes available, someone?2 Message-ID: <SC037.579$rc5.39773@news.cpqcorp.net>  g In article <h83pktohgm23b6dqecghpq7pnv6k50h6h3@4ax.com>, Steve Lionel <Steve.Lionel@compaq.com> writes:   I :... Most of us who use Windows boxes use an internally written NT Notes  L : client which is very nice.  I don't know if it can be released externally  : - I'll ask the author.    I   The external version of the referenced tool is on the current Freeware.   <   http://www.openvms.compaq.com/freeware/freeware50/ntnotes/  I :Someone also recently developed a Notes SERVER that ran on Windows, but  ; :he left the company and I don't think it was finished off.   H   I tried to get that server for the Freeware (and I know where the codeG   is), but (IIRC) that code implementation involved some licensed code.     N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  , Date: Wed, 11 Jul 2001 18:45:46 +0200 (CEST)- From: Freddy Meerwaldt <frederik@freddym.org> * Subject: Re: DEC Notes available, someone?K Message-ID: <Pine.LNX.4.21.0107111844180.11814-100000@firewall.freddym.org>    Hi!   G > >There was a NetNotes product from OS Technologies that was patterned F > >after DEC Notes. A VMS system running DEC Notes could be a NetNotesH > >client. DEC was using NetNotes on Wintel boxes for some internal DEC  > >Notes conferences.  > G > Yes, but I understand that it is no longer supported.  Most of us whoiF > use Windows boxes use an internally written NT Notes client which isF > very nice.  I don't know if it can be released externally - I'll askF > the author.  Someone also recently developed a Notes SERVER that ranG > on Windows, but he left the company and I don't think it was finished  > off.   And I suppose that's good!B I really don't want to run a DEC Notes server on a Windows *shake* Machine.7 I certainly don't want to run anything under Windows...   
 Best Regards,M 	Freddy    -- xN Geek Code 3.1: GCS s+: a--- C+++ UBOU+++ P-- E--- W++ N w--- V++ PGP- t? 5? tv  J ==========================================================================D  Frederik Meerwaldt  ICQ: 83045387  Homepage: http://www.freddym.orgC  Bavaria/Germany              OpenVMS and Unix Howtos and much more I  Solaris, HP/UX, AIX, NetBSD, OpenBSD, IRIX, Tru64, OpenVMS, Ultrix, BeOSR   ------------------------------  % Date: Wed, 11 Jul 2001 15:56:06 -0400e, From: Steve Lionel <Steve.Lionel@compaq.com>* Subject: Re: DEC Notes available, someone?8 Message-ID: <rmbpktscrpf87le6e9781o4g5jl1ct9tom@4ax.com>  ; On Wed, 11 Jul 2001 18:45:46 +0200 (CEST), Freddy Meerwaldty <frederik@freddym.org> wrote:m     >And I suppose that's good!dC >I really don't want to run a DEC Notes server on a Windows *shake*n	 >Machine..8 >I certainly don't want to run anything under Windows...  B Oh, c'mon.  It isn't that bad!  Windows 2000 is really quite nice.F I'm still very fond of VMS, but I use a Windows 2000 system as my mainD "computer" and it works very well.  Windows and OpenVMS coexist very nicely.e- Steve Lionel (mailto:Steve.Lionel@compaq.com)o Fortran Engineering * High-Performance Technical Computing Group& Compaq Computer Corporation, Nashua NH  6 Compaq Fortran web site: http://www.compaq.com/fortran   ------------------------------  % Date: Wed, 11 Jul 2001 16:13:00 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> * Subject: Re: DEC Notes available, someone?' Message-ID: <3B4CC15C.241BA6CC@fsi.net>    Steve Lionel wrote:? > = > On Wed, 11 Jul 2001 18:45:46 +0200 (CEST), Freddy Meerwaldt- > <frederik@freddym.org> wrote:- >  > >And I suppose that's good!aE > >I really don't want to run a DEC Notes server on a Windows *shake*) > >Machine.m: > >I certainly don't want to run anything under Windows... > D > Oh, c'mon.  It isn't that bad!  Windows 2000 is really quite nice.H > I'm still very fond of VMS, but I use a Windows 2000 system as my mainF > "computer" and it works very well.  Windows and OpenVMS coexist very	 > nicely.w  9 How do I control a W2K program from an OpenVMS batch job?    --   David J. Dachterar dba DJE SystemsD http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/e  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.o   ------------------------------  % Date: Wed, 11 Jul 2001 16:19:41 -0500 + From: Christopher Smith <csmith@amdocs.com>S* Subject: RE: DEC Notes available, someone?L Message-ID: <3B55D7F383B0D31197D9009027541CBF0D9D205D@cmiexch1.cmi.itds.com>   > -----Original Message-----5 > From: Steve Lionel [mailto:Steve.Lionel@compaq.com]A  = > On Wed, 11 Jul 2001 18:45:46 +0200 (CEST), Freddy Meerwaldt1 > <frederik@freddym.org> wrote:e  : > >I certainly don't want to run anything under Windows...  D > Oh, c'mon.  It isn't that bad!  Windows 2000 is really quite nice.H > I'm still very fond of VMS, but I use a Windows 2000 system as my mainF > "computer" and it works very well.  Windows and OpenVMS coexist very	 > nicely.   I What planet do you live on, and how advanced is your computing technologynL that you don't have to put up with waiting 15 seconds or so for a list of --# oh, say, 5 files -- in a directory?h   Regards,   Chris     ! Christopher Smith, Perl Developern Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");  'w  h   ------------------------------  % Date: Wed, 11 Jul 2001 23:00:08 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)* Subject: Re: DEC Notes available, someone?L Message-ID: <rdeininger-1107012300080001@user-2iveapg.dialup.mindspring.com>  E In article <rmbpktscrpf87le6e9781o4g5jl1ct9tom@4ax.com>, Steve Lionel   <Steve.Lionel@compaq.com> wrote:  D > Oh, c'mon.  It isn't that bad!  Windows 2000 is really quite nice.H > I'm still very fond of VMS, but I use a Windows 2000 system as my mainF > "computer" and it works very well.  Windows and OpenVMS coexist very	 > nicely.t  + Oh no.  Steve is already being assimilated.i   -- d Robert Deininger rdeininger@mindspring.com    ------------------------------  # Date: Wed, 11 Jul 2001 22:51:35 GMTd0 From: "Richard B. Gilbert" <dragon5311@home.com>( Subject: Re: Experience with EMC storage( Message-ID: <3B4CD8CB.EF2A5529@home.com>  H Now why would anyone call Compaq for EMC support?!?!?!  EMC supports its6 stuff.  They even have someone knowledgable about VMS.  L And frankly, applications that choke on a device name are BROKEN!  To do so,F an application would have to be so badly written that the market would& probably be limited to Outer Mongolia!  M I have an EMC 3630 array that has been running for two years now with no down;F time attributable to it.  Every once in a while EMC wants to replace a5 component which they do without needing to shut down.   J No, I don't like not being able to manage it myself.  The upside is that IM don't have to.  If the company wants to pay that much premium to make my lifey easier, I'm all for it.    Scott Vieth wrote:  C > 1) Are you an IDX customer?  We had a small problem when we firsteI > fired-up our new StorageWorks storage with our IDX system.  We logged a:J > support call with IDX (we thought that our version of ISM was choking onC > the $1$DGAnn: device names) and their response was "we don't havewJ > hardware like that at our location to replicate your configuration."  IfJ > they have trouble helping out with Compaq fc storage, imagine how far upK > sh*t creek you'll be when you attach EMC to your bread-and-butter system.v > H > 2) There was post just a week ago about someone who was having troubleG > getting EMC storage to work with VMS.  When that person called CompaqsI > support, the support folks told him to take a hike (after they heard it"H > was non-Compaq storage).  Is this the kind of support you want on yourK > mission critical systems?  I'd rather get some sleep at night and be ableyD > to go out of town on vacation once a year without worrying that myK > storage is going to go belly-up and Compaq is going to turn their back onp > me.  >a > Buy StorageWorks: sleep well..% > Buy EMC: learn to live on the edge.  >- > -Scott >- > "Koloth(Telocity)" wrote:1 >uI > > We are investigating EMC as a possibility for storage for our OpenVMS:I > > Alpha system.  We use DSM (Mumps) database.  Does anyone have good ora2 > > bad experiences that they would like to share? > >h > > TIAf > >a > > Cass Witkowski   ------------------------------  # Date: Thu, 12 Jul 2001 01:53:10 GMTs" From: Ed Wilts <ewilts@ewilts.org>( Subject: Re: Experience with EMC storage; Message-ID: <aq737.6753$GI4.346611@typhoon.mn.mediaone.net>:   Rob Young wrote:  C > Actually, Keith Parris put out an excellent post (here I believe)iA > that pointed out the not so obvious.  His suggestion.... DO NOTiG > set preferred on units!!!  Let them go where they may.  Set preferredy@ > is a "trick" that tells VMS the unit/device is *only* avaiableB > on the path you select.  The problem there of course is failover? > (and I forget the rest of the details but essentially, it mayo > not go over there at all).    L I don't often disagree with Keith since he usually proves me wrong, but I'm L going to disagree here.  We have all our HSJ-50 controllers use a preferred I path and have verified that they really do failover properly - we've had cK enough controllers crash over the years that I know that this works.  SHOW  - DEVICE even shows the correct alternate path.p           .../Ed   --   Ed Wilts, Mounds View, MN, USA mailto:ewilts@ewilts.org   ------------------------------    Date: 11 Jul 2001 22:07:52 -0700( From: kparris@my-deja.com (Keith Parris)( Subject: Re: Experience with EMC storage= Message-ID: <cb85fed2.0107112107.4ebd22cf@posting.google.com>n  e Ed Wilts <ewilts@ewilts.org> wrote in message news:<aq737.6753$GI4.346611@typhoon.mn.mediaone.net>...2N > I don't often disagree with Keith since he usually proves me wrong, but I'm N > going to disagree here.  We have all our HSJ-50 controllers use a preferred K > path and have verified that they really do failover properly - we've had rM > enough controllers crash over the years that I know that this works.  SHOW r/ > DEVICE even shows the correct alternate path.   > If you leave HSx preferred paths set and it works fine in yourF environment, that's great.  That's the way it's supposed to work.  AndF in the case of an HSx crash (or any sort of total controller failure),D the other controller can tell pretty well that the failed controller> is dead, and my experience agrees with yours in those cases -- failover occurs as it should.s  B It's just that over a period of 4-5 years (during which I used HSxF preferred paths extensively), I finally had experienced too many casesF where failover didn't occur, and over time the circumstancial evidenceB grew that failures were more likely when an HSx preferred path wasF involved.  My growing impression of this link was clinched by a coupleD of cases where I had units on the same controller pair which did NOTF have a preferred path set and failed over just fine, while other unitsC on the same controller pair which DID have a preferred path set got @ stuck and didn't fail over properly.  In another good example, aD controller didn't crash, but two of its back-end SCSI bus interfaces< failed, while the other four SCSI busses were just fine; theD controller concluded that it was OK, and that all the disks on those@ two busses had obviously failed, when in fact the disks were allF reachable through the other controller in the pair, but the controller> with the failure saw no need to give up the units, which had a? preferred path set to it, and thus access to devices on those 2f shelves was lost.   E Since the term "preferred path" can apply in different ways, and even C the commands at the controller and DCL level are very similar, I'll  clarify what I recommend:-  D What I advise against is leaving a preferred path set (all the time)E down at the HSx controller level.  I see no problem with setting (and8D leaving in place) a preferred path up at the host (VMS) level, using; SYS$EXAMPLES:PREFER.MAR or the more-recent DCL command $SETw@ PREFERRED_PATH, since VMS feels free to disregard the host-levelF preferred path setting if necessary in searching for a working path to
 fail over to.E  @ I also see no problem with using the HSx SET unit PREFERRED_PATH8 command on HSx controllers to achieve an even balance ofE units/workload across the two controllers in each pair; I just adviseeF doing an HSx SET unit NOPREFERRED_PATH command soon after the failoverF to the desired unit has occurred, so VMS has the freedom to use eitherF controller as a path to the unit if it needs to after a failure later.C ------------------------------------------------------------------- C Keith Parris | parris at encompasserve dot org | VMS consulting on: C Clusters, Disaster Tolerance, Internals, Performance, Storage & I/On   ------------------------------    Date: 12 Jul 2001 01:09:20 -0500+ From: young_r@encompasserve.org (Rob Young)r( Subject: Re: Experience with EMC storage3 Message-ID: <Zm3dFbtg+mgI@eisner.encompasserve.org>s  ` In article <aq737.6753$GI4.346611@typhoon.mn.mediaone.net>, Ed Wilts <ewilts@ewilts.org> writes: > Rob Young wrote: > D >> Actually, Keith Parris put out an excellent post (here I believe)B >> that pointed out the not so obvious.  His suggestion.... DO NOTH >> set preferred on units!!!  Let them go where they may.  Set preferredA >> is a "trick" that tells VMS the unit/device is *only* avaiableoC >> on the path you select.  The problem there of course is failover-@ >> (and I forget the rest of the details but essentially, it may >> not go over there at all).  p > N > I don't often disagree with Keith since he usually proves me wrong, but I'm N > going to disagree here.  We have all our HSJ-50 controllers use a preferred K > path and have verified that they really do failover properly - we've had 1M > enough controllers crash over the years that I know that this works.  SHOW e/ > DEVICE even shows the correct alternate path.o >   < 	For clarification, here is what Keith said a few weeks ago.= 	Also, IIRC Keith was working with a ton of HSJ50 controllerso 	at a client site.    ( From: Keith Parris (kparris@my-deja.com) Subject: Re: Disk Farm Problem   Newsgroups: comp.os.vmso Date: 2001-06-22 11:50:00 PST   e  5 Since you have HB RAID SW, you might also look in thesC SYS$MANAGER:RAID$DIAGNOSTIC*.LOG files on each node for more clues.s  C Because of all the problems I saw in practice, I now recommend thataC folks do NOT leave a HSx controller preferred-path setting in place ? (in other words, go ahead and use it to fail drives over to the > controller you wish, for load-balancing purposes, but then set? NOPREFFERRED_PATH to let it come under VMS' control if needed).-  C I've seen too many failures where the HSx controllers couldn't tell D which one of the two of them would work, but VMS could figure it out: and recover if you allowed it the freedom to switch paths.  A The way HSx preferred paths work is that the two controllers in aaD dual-redundant pair conspire to lie to VMS and one of them tells VMS? it does not have access to the unit when in fact it could.  If, E between the two controllers, they decide they need to fail over, thengF the controllers switch roles.  Problem is, there are too many possibleD failure scenarios where the controller thinks it is just fine, thankF you very much, but can't reach the drives, yet the other controller inC the pair, which can reach them just fine, declines to take over the D units because the first controller (which has the preferred path set? to it) seems to be functioning just fine as far as it can tell.    Keithu   ------------------------------  % Date: Wed, 11 Jul 2001 21:59:56 -0400u- From: Michael Austin <miaustin@bellsouth.net>g Subject: Fortran... IRAD50- Message-ID: <3B4D049B.884FD9A8@bellsouth.net>q  C What in the world is IRAD50?  And how do I convert the following tot something that F90 understands?i   CALL IRAD50(3,'MCR',DSPNAM(1)) CALL IRAD50(3,'...',DSPNAM(2))   Michael Austin   ------------------------------   Date: 11 Jul 2001 21:38 CDTo' From: carl@gerg.tamu.edu (Carl Perkins)e Subject: Re: Fortran... IRAD50- Message-ID: <11JUL200121380689@gerg.tamu.edu>e  _ In article <3B4D049B.884FD9A8@bellsouth.net>, Michael Austin <miaustin@bellsouth.net> writes...tD }What in the world is IRAD50?  And how do I convert the following to  }something that F90 understands? }  }CALL IRAD50(3,'MCR',DSPNAM(1))  }CALL IRAD50(3,'...',DSPNAM(2))m }  }Michael Austine    H This is aparently for producing "Radix 50" data. (Not that I expect that" knowing this will be much help...)  / The Compaq Fortran web site has some info. See:.  ' http://www.compaq.com/fortran/examples/-   specifically,   5 http://www.compaq.com/fortran/examples/rad50_subs.fora   There's also a little info on:  ( www.compaq.com/fortran/migrating-va.html  - (which also points to the above source code).    --- Carl   ------------------------------   Date: 12 Jul 2001 02:51:00 GMT@ From: groth@pupgg.princeton.edu (Edward J. Groth   609-258-4361) Subject: Re: Fortran... IRAD50, Message-ID: <9ij3ak$55r$1@cnn.Princeton.EDU>  ] In article <3B4D049B.884FD9A8@bellsouth.net>, Michael Austin <miaustin@bellsouth.net> writes: D >What in the world is IRAD50?  And how do I convert the following to  >something that F90 understands? >  >CALL IRAD50(3,'MCR',DSPNAM(1))t >CALL IRAD50(3,'...',DSPNAM(2))  >u >Michael Austind >  >e >s  @ Well, "radix 50" was a way of storing 3 characters in 2 bytes. I* assume DSPNAM above are two byte integers.  > I think you could actually represent 40 characters: upper case= alpha, numerals, and 4 punctuation characters. (blank, period'
 whatever).  < 40 is octal 50 and I think that's where radix 50 cones from.  ? Anyway, it all depends what DSPNAM is to be used for - it looks % like it's receiving the string MCR...d  	 					- Ed    --  J /------------------------------------------------------------------------\J | Edward J. Groth, Assoc. Chr. | Phone: 609-258-4361   Fax: 609-258-6853 |J | Physics Dept., Jadwin Hall   | URL: http://pupgg.princeton.edu/~groth/ |J | Princeton University         | SPAN/HEPNET:  PUPGG::GROTH=44117::GROTH |J | Princeton, NJ 08544          | Internet:     groth@pupgg.princeton.edu |   ------------------------------  % Date: Wed, 11 Jul 2001 23:10:23 -0400u2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: Fortran... IRAD50L Message-ID: <rdeininger-1107012310230001@user-2iveapg.dialup.mindspring.com>  < In article <3B4D049B.884FD9A8@bellsouth.net>, Michael Austin <miaustin@bellsouth.net> wrote:w  E > What in the world is IRAD50?  And how do I convert the following toE! > something that F90 understands?n >   > CALL IRAD50(3,'MCR',DSPNAM(1))  > CALL IRAD50(3,'...',DSPNAM(2))  P "Radix 50" packs 3 characters (from a reduced character set) into 16 bits, IIRC.; I think it was developed by DEC, but don't hold me to that.m  G This looks like a function to convert normal ascii strings to radix 50.   G Your post reminds me that radix 50 support has gone away in some of thePC new Fortran compilers.  I don't recall if Alpha did away with RAD50s; completely, or if it's still supported in the F77 compiler.r  G I think your best bet is to find out what happens to DSPNAM, and recode0F that part to use normal character strings.  DSPNAM looks like an array@ holding something to be DiSPlayed; look for a write statement orF something.  I suspect you'll find 'MCR...' is output somewhere, unless DSPNAM is modified later.f  H You're lucky the radix 50 strings are expressed as ASCII to start with. G Otherwise you'd have to figure out which characters were represented bym the bits...n   -- t Robert Deininger rdeininger@mindspring.com    ------------------------------  % Date: Wed, 11 Jul 2001 15:55:45 -0500t1 From: "David J. Dachtera" <djesys.nospam@fsi.net>e Subject: Re: FUD' Message-ID: <3B4CBD51.B4B10A3C@fsi.net>    Bob Kaplow wrote:e > m > In article <OF42A73B4F.2E7BE924-ON80256A78.00440329@qedi.quintiles.com>, steven.reece@quintiles.com writes:yN > > Calm down Aaron.  At some point _everybody_ would be surprised, except theH > > people that made the decision and the person that thought of it as a# > > possibility in the first place.  >  > OLE! > H > It could have been handled the same way as the VAX to ALpha migration: > N > 1) Announce now that VMS, Tru64, NSK would be ported to IA64. Continue Alpha& > as competing alternate archetecture. > L > 2) 5 years after IA64/VMS et all are ESTABLISHED TECHNOLOGY, *IF* Alpha isJ > dwindling in sales and performance, announce end of life for Alpha a few > years down the road. > M > Where do you think Alpha, VMS, Tru64, and DEC/Cpq would be today if in 1990 L > digital had announced a new 64 bit processor was being developed, and thatM > VAX development was ending and you HAD to port to the new chip immediately.r: > And then that new chip was several years late to market. > N > Announcing a port to IA64 could have been great news for the installed base.M > Coupling it with killing the current processor before its replacement is an M > established product is ultimate stupidity. The Q have killed the goose thatl > laid the golden egg.  ) ...and the bad news from Intel continues:t  @ from THE INFOWORLD SCOOP A.M. EDITION, Wednesday, July 11, 2001:  $ INTEL STOPS SHIPMENTS OF 900MHZ XEON  ( Posted at July 10, 2001 03:42 PM Pacific  1 A GLITCH DISCOVERED in one of Intel's Xeon servere. processors caused the chip maker on Tuesday to. announce it has stopped shipments of the chip.  0 "We are temporarily withholding shipments of the2 Pentium III Xeon 900MHz [processor] with 2MB of L2) cache," said Intel spokesman Bill Kircos.e     For the full story:nR http://www.infoworld.com/articles/hn/xml/01/07/10/010710hnintelglitch.xml?0711weam   (Sorry if the URL wrapped...)S  F ...but, sure, we'll be on Itanic - oops, I mean Unobtainium - by 2004!  H ...and BTW, I have this great piece of land in Arid-zona that you simply
 GOTTA see!   --   David J. Dachteraa dba DJE Systemsa 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: Wed, 11 Jul 2001 20:06:20 GMTs4 From: "Terry C. Shannon" <terryshannon@mediaone.net> Subject: Re: FUD8 Message-ID: <0l237.15$XT4.49599@typhoon.ne.mediaone.net>  F "Bob Kaplow" <kaplow_r@eisner.encompasserve.org.mars> wrote in message- news:6UcEQILloPsD@eisner.encompasserve.org... ; > In article <3B396846.7F2786B3@prodigy.net>, cjt & trefoiln <cheljuba@prodigy.net> writes:. > > Maybe it's "Shannon Knows Commoditization" >aH > Or Shannon Knows Capellas. His commentary the past few weeks certainlyK > sounds more like "party line" than it does "not authorized by, affiliatedtK > with, or endorsed by". It almost reminds me of the kind of commentary you'6 > read in "Stereo Review"; straight vendor propiganda.  I No, it's Shannon Knows Spelling. "Propaganda" does not contain the lettera "i,"  K I haven't a clue what Compaq's "party line" is right now. I doubt they evenh
 have one. ;-}i   ------------------------------  # Date: Wed, 11 Jul 2001 22:28:24 GMTe4 From: "Terry C. Shannon" <terryshannon@mediaone.net> Subject: Re: FUD9 Message-ID: <cq437.76$XT4.111458@typhoon.ne.mediaone.net>i  < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3B4CBD51.B4B10A3C@fsi.net...a > Bob Kaplow wrote:  > >)L > > In article <OF42A73B4F.2E7BE924-ON80256A78.00440329@qedi.quintiles.com>," steven.reece@quintiles.com writes:L > > > Calm down Aaron.  At some point _everybody_ would be surprised, except the(J > > > people that made the decision and the person that thought of it as a% > > > possibility in the first place.s > >s > > OLE! > >eJ > > It could have been handled the same way as the VAX to ALpha migration: > >iJ > > 1) Announce now that VMS, Tru64, NSK would be ported to IA64. Continue Alphar( > > as competing alternate archetecture. > >sK > > 2) 5 years after IA64/VMS et all are ESTABLISHED TECHNOLOGY, *IF* AlphaK isL > > dwindling in sales and performance, announce end of life for Alpha a few > > years down the road.  I That's the way I would have preferred to do it, or, more accurately, have L seen Compaq do it. Two downsides: people would immediately begin to questionJ CPQ's commitment to Alpha, CPQ would have been forced to sustain its heftyK expenditures in the Alpha effort rather than starting to reduce them almostC immediately.  D One of a number of questions from the latest issue of my newsletter:  J No Exit: Compaq made a conscious decision to embrace IPF across the board.B Under the still-murky terms of its agreement with Intel, Compaq isH immediately abandoning the development of future-generation Alpha chips.D What's more, Compaq is transferring to Intel the critical resources,A technology, and intellectual property necessary for ongoing AlphaaL development. By pursuing a "unilateral disarrmament" CPU strategy, Compaq isC rendering itself completely dependent on Intel's ability to deliverpL industry-competitive IPF products in a timely and predictable manner. There'K s little doubt that Intel can deliver competitive 64-bit CPUs, but Compaq's J termination of the Alpha program limits the firm's choices: in the absenceL of Alpha, Compaq no longer has an "Option B" architectural contingency plan.9 A devastating strategic blunder? What sayeth you, Compaq?o   ------------------------------  % Date: Wed, 11 Jul 2001 19:57:02 -04002' From: "Bill Todd" <billtodd@foo.mv.com>K Subject: Re: FUD( Message-ID: <9iiosv$860$1@pyrite.mv.net>  ? "Terry C. Shannon" <terryshannon@mediaone.net> wrote in messagel3 news:cq437.76$XT4.111458@typhoon.ne.mediaone.net...7 >G> > "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message# > news:3B4CBD51.B4B10A3C@fsi.net...e > > Bob Kaplow wrote:n   ...s  L > > > It could have been handled the same way as the VAX to ALpha migration: > > >iL > > > 1) Announce now that VMS, Tru64, NSK would be ported to IA64. Continue > Alpha * > > > as competing alternate archetecture. > > >EG > > > 2) 5 years after IA64/VMS et all are ESTABLISHED TECHNOLOGY, *IF*d Alphas > isJ > > > dwindling in sales and performance, announce end of life for Alpha a fewt > > > years down the road. > K > That's the way I would have preferred to do it, or, more accurately, havenE > seen Compaq do it. Two downsides: people would immediately begin toe questionL > CPQ's commitment to Alpha, CPQ would have been forced to sustain its heftyF > expenditures in the Alpha effort rather than starting to reduce them almost > immediately.  K In other words, it would have been somewhat like the transition from VAX touJ Alpha:  have the new platform in place, see how people like it, and *then*G (*if* most people like it) announce phase-out of the old one.  Which is H exactly what Bob said, and which nothing you responded with negated as aF reasonable approach (hell, that approach would have *strengthened* theH perception of Compaq's commitment to Alpha, since the funding would haveB clearly been over-and-above that required for minimal sustenance).   >oF > One of a number of questions from the latest issue of my newsletter: >lL > No Exit: Compaq made a conscious decision to embrace IPF across the board.D > Under the still-murky terms of its agreement with Intel, Compaq isJ > immediately abandoning the development of future-generation Alpha chips.F > What's more, Compaq is transferring to Intel the critical resources,C > technology, and intellectual property necessary for ongoing AlphaoK > development. By pursuing a "unilateral disarrmament" CPU strategy, Compaqa isE > rendering itself completely dependent on Intel's ability to deliverLG > industry-competitive IPF products in a timely and predictable manner.s There'D > s little doubt that Intel can deliver competitive 64-bit CPUs, but Compaq'sL > termination of the Alpha program limits the firm's choices: in the absenceH > of Alpha, Compaq no longer has an "Option B" architectural contingency plan.c; > A devastating strategic blunder? What sayeth you, Compaq?r  J I think at least a lot of people don't see the above as the major problem. Instead, they aske  K 1.  Why did Compaq choose to abandon a piece of proprietary technology withmL as much potential for profit as Alpha had?  At a minimum, it stood to hold aK smallish but very profitable chunk of the 64-bit market for the foreseeableRD future - more than adequate to fund its development with even modestI marketing effort by Compaq.  At best, it had the potential to box in IA64tK between IA32 (and AMD-64) on the low side and Alpha (and POWER) on the high-J side, keep Intel from grabbing any really dominant market share with IA64,L and thus leave a large slice of the pie for Compaq, IBM, and Sun (because itL always does, regardless of how uncompetitive US may be technically) to divvy up.5  I 2.  What does this say about Compaq's attitude toward its other lucrativeh? but currently-niche proprietary technology (VMS in particular)?s  1 3.  What does this say about Compaq's competence?a  G 4.  What does the way the decision was handled (unilaterally) say aboutAJ Compaq's attitude toward its 'commitments' to customers (and its integrity in general)?   - bill   ------------------------------  % Date: Wed, 11 Jul 2001 20:01:36 -0400 ' From: "Bill Todd" <billtodd@foo.mv.com>t Subject: Re: FUD( Message-ID: <9iip5f$8dl$1@pyrite.mv.net>  < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3B4CBD51.B4B10A3C@fsi.net...e   ...s  J > ...and BTW, I have this great piece of land in Arid-zona that you simply > GOTTA see!  I Now, if Compaq were handling it, it would turn out to be at the bottom of  Lake Powell.   - bill   >o > -- > David J. Dachteraa > dba DJE Systemsd > http://www.djesys.com/ > < > Unofficial Affordable OpenVMS Home Page and Message Board:! > http://www.djesys.com/vms/soho/H >eH > This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings > is to be expected. >iB > Feel free to exercise your rights of free speech and expression. >oH > However, attacks against individual posters, or groups of posters, are > strongly discouraged.m   ------------------------------  % Date: Wed, 11 Jul 2001 15:44:50 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>h& Subject: Re: Future support of VAX-VMS' Message-ID: <3B4CBAC2.B883E8D1@fsi.net>s   Bob Kaplow wrote:  > h > In article <3B381520.B55B16E@toyvax.Tucson.AZ.US>, Vance Haemmerle <vance@toyvax.Tucson.AZ.US> writes: > > Fred Kleinsorge wrote: > >sI > >> Did anywhere in my note I indicate otherwise?  I'll repeat, "Our VAXuO > >> customers are important to us".  I would expect us to provide "support VMSeE > >> on the VAX "through the year 2010" with "preventive, performancet, > >> optimization, and remedial services."". > >VE > >   Through some third-party organization like Computer Associates?i > N > Isn't posting that company name to this newsgroup the equivalent of invoking > Godwin's law?  > > > THey've screwed us 3 (almost 4) times already. No thank you!  F Dunno if you've seen the news re: CA: one of the major shareholders isA launching a proxy bid to oust the current BOD and emplace its ownh candidates.l  < from THE INFOWORLD SCOOP P.M. EDITION, Monday, July 9, 2001:  5 CA SHAREHOLDER WYLY FILES PRELIMINARY PROXY STATEMENTb WITH SEC  ( Posted at July 09, 2001 07:27 AM Pacific  7 COMPUTER ASSOCIATES SHAREHOLDER Sam Wyly on Monday tooke2 the next step in his battle to gain control of the6 software giant by filing a preliminary proxy statement2 that could be used to unseat the company's current board of directors.-  7 Wyly's investment group, called Ranger Governance, senti. its proxy statement to the U.S. Securities and5 Exchange Commission (SEC) for review and approval. Ifs4 cleared by the SEC, the proxy will then be mailed to6 CA shareholders before the company's annual meeting on/ Aug. 29. The proxy statement contains a list ofA5 proposed board of directors -- including Wyly -- that-6 Ranger Governance suggests should replace CA's current
 directors.     For the full story: N http://www.infoworld.com/articles/hn/xml/01/07/09/010709hnwylysec.xml?0709mnpm   (Sorry if the URL wrapped...)e  A Wyly's pissed 'cuz the company is shafting the shareholders whileaH padding executive salaries and bonuses. Guess he knows how the customers feel...9   -- s David J. DachteraR dba DJE SystemsA http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/o  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.o   ------------------------------    Date: 11 Jul 2001 17:14:45 -0400/ From: jordan@lisa.gemair.com (Jordan Henderson)-& Subject: Re: Future support of VAX-VMS* Message-ID: <9iifk5$ptj$1@lisa.gemair.com>  ' In article <3B4CBAC2.B883E8D1@fsi.net>,j0 David J. Dachtera <djesys.nospam@fsi.net> wrote: >Bob Kaplow wrote: >>  i >> In article <3B381520.B55B16E@toyvax.Tucson.AZ.US>, Vance Haemmerle <vance@toyvax.Tucson.AZ.US> writes:e >> > Fred Kleinsorge wrote:  >> >J >> >> Did anywhere in my note I indicate otherwise?  I'll repeat, "Our VAXP >> >> customers are important to us".  I would expect us to provide "support VMSF >> >> on the VAX "through the year 2010" with "preventive, performance- >> >> optimization, and remedial services."".  >> >F >> >   Through some third-party organization like Computer Associates? >> 4O >> Isn't posting that company name to this newsgroup the equivalent of invokingk >> Godwin's law? >> c? >> THey've screwed us 3 (almost 4) times already. No thank you!r >iG >Dunno if you've seen the news re: CA: one of the major shareholders is-B >launching a proxy bid to oust the current BOD and emplace its own >candidates. > = >from THE INFOWORLD SCOOP P.M. EDITION, Monday, July 9, 2001:  >h6 >CA SHAREHOLDER WYLY FILES PRELIMINARY PROXY STATEMENT	 >WITH SEC/ > ) >Posted at July 09, 2001 07:27 AM PacificT >P8 >COMPUTER ASSOCIATES SHAREHOLDER Sam Wyly on Monday took3 >the next step in his battle to gain control of the-7 >software giant by filing a preliminary proxy statement23 >that could be used to unseat the company's currentn >board of directors. >r8 >Wyly's investment group, called Ranger Governance, sent/ >its proxy statement to the U.S. Securities andB6 >Exchange Commission (SEC) for review and approval. If5 >cleared by the SEC, the proxy will then be mailed to57 >CA shareholders before the company's annual meeting onh0 >Aug. 29. The proxy statement contains a list of6 >proposed board of directors -- including Wyly -- that7 >Ranger Governance suggests should replace CA's current6 >directors.  >h >  >For the full story:O >http://www.infoworld.com/articles/hn/xml/01/07/09/010709hnwylysec.xml?0709mnpmc >s >(Sorry if the URL wrapped...) >eB >Wyly's pissed 'cuz the company is shafting the shareholders whileI >padding executive salaries and bonuses. Guess he knows how the customers  >feel... >i  A Actually, the primary reason he _says_ he's doing this is becaused@ of the bad reputation that CA has with customers.  Now, this mayD be more pretext than pity, but that's what I read about the takeover bid.  C Now, he's also claiming that shareholder value is being negatively  B affected by bad customer reputation.  What a concept, satisfy your@ customers and shareholder value goes up...  What a maverick this Sam Wyly is...   >--  >David J. Dachtera >dba DJE Systems >http://www.djesys.com/i >.; >Unofficial Affordable OpenVMS Home Page and Message Board:-  >http://www.djesys.com/vms/soho/ > G >This *IS* an OpenVMS-related newsgroup. So, a certain bias in postingsm >is to be expected.R >FA >Feel free to exercise your rights of free speech and expression.  > G >However, attacks against individual posters, or groups of posters, are, >strongly discouraged.   -Jordan Hendersonm jordan@greenapple.como   ------------------------------   Date: 11 Jul 2001 18:14 CDTa' From: carl@gerg.tamu.edu (Carl Perkins)h& Subject: Re: Future support of VAX-VMS- Message-ID: <11JUL200118145163@gerg.tamu.edu>n  5 "David J. Dachtera" <djesys.nospam@fsi.net> writes... B }Wyly's pissed 'cuz the company is shafting the shareholders whileI }padding executive salaries and bonuses. Guess he knows how the customersI }feel... }  }--  }David J. Dachtera  G He also says that studies indicate that CA is their own customers' mosteC hated vendor - it's the vendor they would most like to dump if theyp> were not locked into using their software for whatever reason.   --- Carl   ------------------------------  % Date: Wed, 11 Jul 2001 21:59:09 +0100 + From: "antonio.carlini" <arcarlini@iee.org>p Subject: Re: Hobbyists' Message-ID: <3B4CBE1D.58D4F8CC@iee.org>o   Jason O'Donnell wrote:= > I cut my teeth on an 11/750.  How does one of those comparesD > performance wise to say an AlphaServer 1000?  I have seen a lot of > 1000s on eBay for sale.   & The VAX-11/750 compares quite well, at( least in the nostalgia, noise-generation and looking big(gish) stakes.m  & In the computing performance stakes it% lags behind at something like 0.1% to2# 0.5% of the AlphaServer 1000. Don'tH even ask to compare it against   a fast modern Alpha :-)T   AntonioE   -- e   ---------------r- Antonio Carlini             arcarlini@iee.orga   ------------------------------  % Date: Wed, 11 Jul 2001 22:01:59 +0100 + From: "antonio.carlini" <arcarlini@iee.org>n Subject: Re: Hobbyists' Message-ID: <3B4CBEC7.1BE3D226@iee.org>B   Keith Parris wrote:hH > Smallest cluster I've heard of was two Tadpole Alpha laptops connectedG > by a cross-over cable on an airline flight.  (This cluster also holdse > the altitude record.)p  + For some reason, this reminds me of the old . joke about always carrying an explosive device) when flying, because the chances of thereD( being *two* explosive devices on any one flight is vanishingly small :-)    Antonioe   -- n   ---------------t- Antonio Carlini             arcarlini@iee.orgO   ------------------------------  % Date: Wed, 11 Jul 2001 22:47:16 -0400.2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: HobbyistsL Message-ID: <rdeininger-1107012247160001@user-2iveapg.dialup.mindspring.com>  < In article <cb85fed2.0107110852.ebe5d70@posting.google.com>,) kparris@my-deja.com (Keith Parris) wrote:n  "H > Smallest cluster I've heard of was two Tadpole Alpha laptops connectedG > by a cross-over cable on an airline flight.  (This cluster also holdse > the altitude record.)s  G Are you sure about the altitude record?  Seems like some military folks6H might have flown a VMS cluster at high altitude.  But they wouldn't tell us, would they?n   -- w Robert Deininger rdeininger@mindspring.comp   ------------------------------  % Date: Wed, 11 Jul 2001 22:24:49 -0600t From: yyyc186@mindspring.com Subject: Re: Hobbyists; Message-ID: <3b4d26d6$2$lllp186$mr2ice@nntp.mindspring.com>c  ; In <9iei3c$i9lpb$1@ID-56200.news.dfncis.de>, on 07/10/2001 iF    at 11:33 AM, "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de> said:t  C I have an AlphaPC164 and an Alpha Station 3000.  The lather will be E auctioned off on ebay in the coming months once I get it all put back ? together with the bigger disk drives.  It will also include thee proprietary monitor with cable.:   Roland   >Jason O'Donnell wrote...4C >> It seems that there are several OpenVMS Hobbyists in this group.iE >> I was just curious what systems people have in their homes runningw >> OpenVMS.  Any GS series?e  B >Sure would be nice, but no. I have a VAXstation 4000-60 and a DEC
 >3000-300.   >cu,	 >  Martin    -- m; -----------------------------------------------------------x yyyc186@mindspring.com; -----------------------------------------------------------a   ------------------------------  % Date: Wed, 11 Jul 2001 16:14:19 -0400d+ From: "Main, Kerry" <Kerry.Main@compaq.com>   Subject: RE: IA64 Rocks My WorldR Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4AD7EF7@kaoexc01.americas.cpqcorp.net>  : >>> No I am not, it is allowed by the rules for TPC-C. <<<  
 Thank you.  K >>> I am objecting to the idea which you seem to sunscribe to which is thatnJ Cluster TPC-C and single system non clustered TPC-C results are in any wayK comparable they are not for all the reasons I have covered in detail in them	 past. <<<T  B One approach does performance and availability. Your approach does performance alone.  K I know, I know .. A single server should not be allowed to use the featurestL that were designed for it (partitions, shared memory, using shared memory asL a high speed communication interconnect etc) in order to maximize its singleI server number. A real mans TPC benchmark only scales up in a straight SMPm config .. right?  H If a Customer can achieve his goals of single server performance and yet; have additional availability thrown in, is this a negative?I  L Anyway, this is a never ending argument that could go for a long time and is not likely to be settled here.     Regards,  
 Kerry Main Senior Consultantt Compaq Canada Inc. Professional Services  Voice: 613-592-4660a Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----7 From: andrew harrison [mailto:andrew.nospam@uk.sun.com]t Sent: July 11, 2001 7:50 AM' To: Info-VAX@Mvb.Saic.Coma  Subject: Re: IA64 Rocks My World     "Main, Kerry" wrote: >  > Greg,o > I > re: OPS in a box .. I think Andrew is trying to state somehow that a HW L > partitioned system using separate OS's within the same system (and OPS) isJ > somehow not a valid benchmark. His "Customer complexity" arguments refer toJ > setting up Oracle Parallel Server (which any good Oracle DBA can do) and inK > fine tuning the app to run in a NUMA environment (which any Customer withn ae > NUMA server needs to do).J >   B No I am not, it is allowed by the rules for TPC-C. I am objecting 9 to the idea which you seem to sunscribe to which is that t; Cluster TPC-C and single system non clustered TPC-C resultsr; are in any way comparable they are not for all the reasons u% I have covered in detail in the past.   I > As most here are aware, you can scale up (big SMP) or you can scale outIL > (cluster approach). Both have merit in different application environments. > K > Given that one of the reasons for doing HW partitioned systems is to haveoB > separate OS environments, but one single view of the data, imho,L > illustrating the overall performance one can get out of a single system is > fine.y > E > This is exactly what Customers doing server consolidation today aree lookinga > for. > F > [hopefully, we can avoid another "what is a real mans TPC benchmark" > discussion.. ] > 
 > Regards, >  > Kerry Main > Senior Consultanti > Compaq Canada Inc. > Professional Servicese > Voice: 613-592-4660  > Fax  :  819-772-7036 > Email: Kerry.Main@Compaq.com >  > -----Original Message-----0 > From: Greg Pfister [mailto:pfister@us.ibm.com] > Sent: July 6, 2001 12:06 PMe > To: Info-VAX@Mvb.Saic.Comc" > Subject: Re: IA64 Rocks My World >  > andrew harrison wrote: > [snip]? > > Designing a memory subsystem that had poorer memory latencyo? > > than a 4 year old Origin 2000 was hardly something that you   > > can blame on the marketeers. > >tC > > > BTW, from what I also hear, SUN is having similar engineerings > difficultiesJ > > > making large scale USIII system work. Remeber those old quotes about theCH > > > systems scaling to 1000 processors etc? Where are the systems? Oh, > right,? > > > of course you can cluster lots of small ones together ;-)o > > >  > >h9 > > All the major systems vendors have suffered delays inh< > > introducing their latest high end systems. The SuperDome; > > was late, the SunFires were late and WildFire was late.n > @ > Careful with that "all." I don't think IBM's been late yet. At> > least not recently, since TPTB finally decided that the Un*x) > market was something worth going after.U > : > > The problem for WildFire is that it was late but  when3 > > it was announced it wasn't competitive with the < > > previous generation of servers from Compaqs competitors. > >a7 > > It recently reached the dizzy heights of 230,000 on 4 > > TPC-C, using the latest 1001 Mhz CPU's, a result7 > > puts it just ahead of the IBM P680, a machine thatsp) > > just about to be replaced by Regatta.  > >o3 > > When Rob Young started his WildFire boosting in?4 > > the dark and dim past initial numbers of 200,0004 > > were being bandied about, I can only assume that4 > > this number was a design goal that was leaked to1 > > Rob (unless he made it up). The fact that theA0 > > initial numbers came out way lower than that1 > > and then needed OPS in a box tends to suggestd( > > that they missed their design goals. > C > Do you know for a fact that the OPS-in-a-box solution was used inlC > Wildfire? That's often rather difficult to tell from official TPCe> > disclosures. If it was used, it of course makes the high TPC6 > results rather meaningless (in my personal opinion). >  > Greg Pfister   -- v Andrew Harrison- Enterprise IT Architecto   ------------------------------  % Date: Wed, 11 Jul 2001 16:02:38 -0400e+ From: "Main, Kerry" <Kerry.Main@compaq.com>h  Subject: RE: IA64 Rocks My WorldR Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4AD7EF6@kaoexc01.americas.cpqcorp.net>  I >>> Oh come on Kerry try to do better. Or ar you saying that Wintel boxes:' that make up the vast majority of your @% revenues are in fact high margin.  <<-  L Come on Andrew .. surely you are not going to say that the total NSK, Tru64,K and OpenVMS revenues are not as significant to Compaq as the Wintel stuff ?   H I know it suits your purpose to try and position these revenues as beingH insignificant, but surely BILLIONS of dollars is not  something that any& company would view as insignificant ..  L ok, maybe I am missing something - tell me how you think BILLIONS of dollars are insignificant.   :-)e  
 Kerry Main Senior Consultant/ Compaq Canada Inc. Professional Services  Voice: 613-592-4660w Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----7 From: andrew harrison [mailto:andrew.nospam@uk.sun.com]o Sent: July 11, 2001 7:47 AMd To: Info-VAX@Mvb.Saic.Como  Subject: Re: IA64 Rocks My World     "Main, Kerry" wrote: >  > Andrew, Andrew ..1 > < > You stated the total market - please re-read your posting. > : > >>>  In the first quarter this year everyone has taken a2 > > beating but Compaq more than everyone else the2 > > total market declined by 2% but Compaq led the > > way with a 6% decline.>>>o > 8 > I simply pointed out the inaccuarcy of this statement. > J > Regardless, everyone can spin analyst numbers to meet the goal they want to
 > achieve. >   1 Since this newsgroup concentrates on OpenVMS and M2 Alpha don't you think that using the numbers that 1 actually refer to OpenVMS/Alpha revenues and the d/ market they compete in might have been a betterp5 choice than using numbers that consist predominantly ' of Wintel boxes.    3 I chose the former, you chose the latter what does  
 that say ????   4 > >>> Hum, care to comment on a business model where2 > revenues are increasing in your high volume very7 > low margin business while they are decreasing in youro$ > lower volume high margin units.<<< > G > Sure, not a good one to be in. Compaq is certainly not following that2 model,G > so who is that you are referring to that is following this strategy ?d >   9 Oh come on Kerry try to do better. Or ar you saying that L4 Wintel boxes that make up the vast majority of your " revenues are in fact high margin.    Regardsi Andrew Harrisony Enterprise IT Architectu   ------------------------------  % Date: Wed, 11 Jul 2001 19:40:46 -0400i' From: "Bill Todd" <billtodd@foo.mv.com>,  Subject: Re: IA64 Rocks My World( Message-ID: <9iinue$7mf$1@pyrite.mv.net>  6 "Main, Kerry" <Kerry.Main@compaq.com> wrote in messageL news:BE56C50EA024184DAF48F0B9A47F5CF4AD7EF6@kaoexc01.americas.cpqcorp.net... >nK > >>> Oh come on Kerry try to do better. Or ar you saying that Wintel boxese( > that make up the vast majority of your' > revenues are in fact high margin.  <<r >tG > Come on Andrew .. surely you are not going to say that the total NSK,/ Tru64,K > and OpenVMS revenues are not as significant to Compaq as the Wintel stuffy ?p  F Andrew doesn't have to say it:  Compaq makes it crystal-clear by their actions.   >dJ > I know it suits your purpose to try and position these revenues as beingJ > insignificant, but surely BILLIONS of dollars is not  something that any( > company would view as insignificant ..  $ Not any competently-managed company.   >VF > ok, maybe I am missing something - tell me how you think BILLIONS of dollarsf > are insignificant.  # What you're missing is objectivity.    - bill   >s > :-)t >d > Kerry Main > Senior Consultantr > Compaq Canada Inc. > Professional Services- > Voice: 613-592-4660e > Fax  :  819-772-7036 > Email: Kerry.Main@Compaq.com   ------------------------------  % Date: Wed, 11 Jul 2001 22:26:26 +0100 + From: "antonio.carlini" <arcarlini@iee.org><$ Subject: Re: Intel/Alpha announcment' Message-ID: <3B4CC482.77C0DEBB@iee.org>    Elliott Roper wrote: > I > C'mon. Teco's good, but not *that* good. I've looked at snippets of thetI > source for both. Surely Rick Murphy never did it in one glorious macro?     Given that there is a TECO macro! which will compute pi to as many   decimal places as you care, I % do not find it at all hard to believeV that it could do something ash mechanical as MACRO-11 tov MACRO-32 :-)   Antonioi     --     ---------------e- Antonio Carlini             arcarlini@iee.orgr   ------------------------------  # Date: Wed, 11 Jul 2001 23:03:46 GMT 9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)i$ Subject: RE: Intel/Alpha announcment3 Message-ID: <BuNgSt58QPZD@eisner.encompasserve.org>o  z In article <3B55D7F383B0D31197D9009027541CBF0D9D2057@cmiexch1.cmi.itds.com>, Christopher Smith <csmith@amdocs.com> writes: >> -----Original Message-----o/ >> From: kaplow_r@eisner.encompasserve.org.mars9 > H >> IIRC TECO-32 came about from a TECO macro that translated MACRO-11 to> >> MACRO-32, which was written specifically to port TECO from  >> Compatibility? >> mode to native mode on the MicroVAX. An old friend of mine, a >> Rick Murphy, dids >> the port... > N > So if I read this thread correctly, we have TECO originally being written inN > MACRO-11, later translated to MACRO-32 using _itself_ as the translator, andM > then, the MACRO-32 being lost (or hopeless to compile on an Alpha?), porteds/ > to Alpha with VEST and some special trickery?b > B > Maybe it should be re-written in something of a higher level. :)  @ Before attempting to rewrite anything, it is necessary to have aC specification of the proper behavior.  Otherwise one would get into D the problem that occurred when Mail was rewritten in C from Bliss --7 lots of "features" upon which users depended went away./   ------------------------------  # Date: Thu, 12 Jul 2001 01:18:37 GMTe- From: goathunter@goatley.com (Hunter Goatley)o$ Subject: Re: Intel/Alpha announcment1 Message-ID: <3b4cfa7c.201932323@news.process.com>h  J On Wed, 11 Jul 2001 23:03:46 GMT, Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) wrote:s  A >Before attempting to rewrite anything, it is necessary to have a D >specification of the proper behavior.  Otherwise one would get intoE >the problem that occurred when Mail was rewritten in C from Bliss --n8 >lots of "features" upon which users depended went away.  @ "Features" that they refuse to put back in, because "it works asC it's documented to work."  And why did they rewrite it in C?  I wasdA told that it was because the new people coming in only knew C andeA didn't know BLISS.  Well, duh.  Apparently the new people weren'ts* capable of learning something new, either.   Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/t9 goathunter@goatley.com     http://www.goatley.com/hunter/    ------------------------------  % Date: Wed, 11 Jul 2001 20:02:25 -0400s  From: Kuff@Tessco.Com (Hal Kuff) Subject: Re: Memory Channel IIO Message-ID: <285EFFAB55A0E358.282D0B1D1EFA5BD6.FC00C70B9985913E@lp.airnews.net>r  ,    is the consenus that it works with a hub?        J In article <MOEAJKGGEIMGCCPEPJBHAENGFAAA.arturo.saavedra@wcom.com>, arturo* saavedra <arturo.saavedra@wcom.com> wrote:  G > We run a pair of GS80s with memory channel.. and have problems duringeM > reboots in which we bugcheck continuously until we reset both ports via the K > mc_diag command from console.  Engineering is in the process of rewritingiM > the driver for 7.2-1h1.  Part of our problem is that we are running withoutd2 > a MC Hub which was recommended but not required. >  >  > -----Original Message-----) > From: Hal Kuff [mailto:Kuff@Tessco.Com]e& > Sent: Tuesday, July 10, 2001 4:57 PM > To: Info-VAX@Mvb.Saic.Come > Subject: Memory Channel II >  > A > In a powerpoint presentation on relative performance of ClusteruM > Interconnects for OpenVMS 7.3 the engineer mentions that the Memory Channel L > II driver for 7.2x had issues.....  Does anyone know how stable the driver > for 7.2-1H1 is?a   ------------------------------  # Date: Thu, 12 Jul 2001 00:17:05 GMT / From: "Tom Simpson" <simpsont@xxx.mediaone.net>e7 Subject: Re: Memory Channel II (and an Itanium comment)oG Message-ID: <50637.11847$cK1.3340718@typhoon.jacksonville.mediaone.net>m  J I can verify that the "hubless" memory channel configuration has "issues".  H We were lucky the get 30 days of continuous uptime before Compaq finallyI admitted there were problems if there was no hub in the configuration anduI provided us with a hub.  It's going on 4 months now since our last crash.rL That's VMS 7.2-1 (and VMS 7.1-2) in a 2-node ES40 cluster with a shared SCSID storageworks disks and HSZ controllers and of course, memory channel interfaces.    Regards, Tomr  - "Nic Clews" <nclews@csc.com> wrote in messagel7 news:a720d610.0107110200.2f7a12f9@posting.google.com...f= > arturo saavedra <arturo.saavedra@wcom.com> wrote in messagei? news:<MOEAJKGGEIMGCCPEPJBHAENGFAAA.arturo.saavedra@wcom.com>...-I > > We run a pair of GS80s with memory channel.. and have problems during-K > > reboots in which we bugcheck continuously until we reset both ports via  theiC > > mc_diag command from console.  Engineering is in the process ofg	 rewritingnG > > the driver for 7.2-1h1.  Part of our problem is that we are running< without04 > > a MC Hub which was recommended but not required. >oH > In a cluster involving ES40's, again without a hub (may be the issue?)? > at VMS 7.1-2 we see the MC drop out from time to time, but itnG > generally doesn't cause any grief because there is fast ethernet (two  > of) plus CI in the cluster.  >eE > Tru64 exclusively uses MC for clustering, so it's most likely not at; > hardware issue, but to feedback info from a Tru64 ItaniumeE > presentation, they will have MC and Infiniband as supported cluster- > interconnects. >uH > I would assume that VMS support for MC would be present in the ItaniumD > architecture machines, hopefully with debugged MC driver software! >iF > I'm also speculating that Infiniband will be a supported VMS clusterE > interconnect at or shortly after the Itanium VMS releases, and thiseG > was hinted at back at a UK DECUS clustering event back in February. I H > can't imagine it being 'backported' to Alpha systems. I'm increasinglyG > of the opinion that all the clews (sic) were and have been around for F > a while, and that VMS will quite definitely move forwards on Itanium1 > [in ways it couldn't under Alpha]. Vive la VMS!  >I* > Regards, Nic Clews CSC Computer Sciences > nclews at csc dot com>   ------------------------------    Date: 12 Jul 2001 01:13:09 -0500+ From: young_r@encompasserve.org (Rob Young),7 Subject: Re: Memory Channel II (and an Itanium comment)-3 Message-ID: <VQi1bR111zWC@eisner.encompasserve.org>-  y In article <50637.11847$cK1.3340718@typhoon.jacksonville.mediaone.net>, "Tom Simpson" <simpsont@xxx.mediaone.net> writes: L > I can verify that the "hubless" memory channel configuration has "issues". > J > We were lucky the get 30 days of continuous uptime before Compaq finallyK > admitted there were problems if there was no hub in the configuration and K > provided us with a hub.  It's going on 4 months now since our last crash.hN > That's VMS 7.2-1 (and VMS 7.1-2) in a 2-node ES40 cluster with a shared SCSIF > storageworks disks and HSZ controllers and of course, memory channel
 > interfaces.  > 
 > Regards, > Toma >   ? 	Was this your only interconnect or did you also have NI turnedl@ 	on to use the Network for cluster communication?  Where can you= 	reference any online or DSNlink details about such "issues?"e  
 			Thanks,   				Robm   ------------------------------    Date: 12 Jul 2001 01:19:48 -0500+ From: young_r@encompasserve.org (Rob Young)C7 Subject: Re: Memory Channel II (and an Itanium comment)i3 Message-ID: <SxCUcAgMZWnP@eisner.encompasserve.org>   y In article <50637.11847$cK1.3340718@typhoon.jacksonville.mediaone.net>, "Tom Simpson" <simpsont@xxx.mediaone.net> writes:gL > I can verify that the "hubless" memory channel configuration has "issues". > J > We were lucky the get 30 days of continuous uptime before Compaq finallyK > admitted there were problems if there was no hub in the configuration andqK > provided us with a hub.  It's going on 4 months now since our last crash.sN > That's VMS 7.2-1 (and VMS 7.1-2) in a 2-node ES40 cluster with a shared SCSIF > storageworks disks and HSZ controllers and of course, memory channel
 > interfaces.a >   = 	I missed something from earlier that Nic wrote, found below:t   >>I >> In a cluster involving ES40's, again without a hub (may be the issue?)o@ >> at VMS 7.1-2 we see the MC drop out from time to time, but itH >> generally doesn't cause any grief because there is fast ethernet (two >> of) plus CI in the cluster. >>  : 	So to rephrase that question... If what Nic says is true,A 	are you using the Network to cluster also?  And if not, why not? @ 	But if you are and it was still crashing, can you confirm that?   				Robi   ------------------------------    Date: 11 Jul 2001 16:14:13 -0700+ From: dwaybright@dollar.com (Don Waybright)a Subject: PAWZ dicontinuedv= Message-ID: <9ef96daa.0107111514.18395d43@posting.google.com>a  E Just received word that Compaq has discontinued PAWZ.  They layed offv= the entire PAWZ team, and you won't find the link to the PAWZt= documention on the OVMS System Management Tools page anymore.o  / What is the future of the ECP Data collector?  f  & I wonder if they will sell PAWZ to CA?   ------------------------------  # Date: Thu, 12 Jul 2001 01:38:37 GMTs" From: Ed Wilts <ewilts@ewilts.org> Subject: Re: PAWZ dicontinuedo; Message-ID: <xc737.6750$GI4.346611@typhoon.mn.mediaone.net>i   Don Waybright wrote:    ( > I wonder if they will sell PAWZ to CA?  L I heard that after CA acquired DecPS, the issue hit the top-10 list for DEC C executives when CA started seriously screwing their newly acquried @J customers.  I seriously doubt that DEC/Compaq will make the same mistake, E nor will the customers be stupid enough to fall for the screwing and S gouging again.  -         .../Ed (obviously an ex-CA customer) o   -- : Ed Wilts, Mounds View, MN, USA mailto:ewilts@ewilts.org   ------------------------------    Date: 11 Jul 2001 22:13:17 -0700( From: kparris@my-deja.com (Keith Parris) Subject: Re: PAWZ dicontinuedh= Message-ID: <cb85fed2.0107112113.2caaae2c@posting.google.com>i  p dwaybright@dollar.com (Don Waybright) wrote in message news:<9ef96daa.0107111514.18395d43@posting.google.com>...G > Just received word that Compaq has discontinued PAWZ.  They layed off ? > the entire PAWZ team, and you won't find the link to the PAWZl? > documention on the OVMS System Management Tools page anymore.c > 1 > What is the future of the ECP Data collector?  a > ( > I wonder if they will sell PAWZ to CA?  D I sent some suggestions in for PAWZ improvements last month, and theF response from the group manager was this: "Prem Sinha is leading a new8 company that will be taking care of PAWZ in the future."  E That seems to rule out another CA fiasco.  But it also puts PAWZ into F the 3rd-party software category, as opposed to a vendor-supplied tool.  E I'm told that a copy of the source code for CP/Collect and CP/Analyze0> was obtained by the VMS organization before the layoffs.  ThatD probably means the ECP data collector will continue to be available.C -------------------------------------------------------------------WC Keith Parris | parris at encompasserve dot org | VMS consulting on:0C Clusters, Disaster Tolerance, Internals, Performance, Storage & I/On   ------------------------------  # Date: Wed, 11 Jul 2001 18:48:01 GMT) From: LBohan@dbc.spam_less..com : Subject: Re: Some thoughts on the recent turn of events...8 Message-ID: <ug7pkt47okkdca2036mcl1i32rd96rtnng@4ax.com>  3 On Wed, 11 Jul 2001 14:25:42 +0100, andrew harrison-! <andrew.nospam@uk.sun.com> wrote:i    C >The idea that Compaq made this announcemnt without first assessingo8 >the intentions of their major ISV's is also ludicrous.  >u >Regards >Andrew Harrison >Enterprise IT Architect  1 yep. and  the impressions I garned from various  I( 3rd-party/channel/Vars news-magazines,  6 that at least some of them, are seriously ticked off, 9 enough so to consider dumping CPQ from their offerings ..k   ------------------------------  % Date: Wed, 11 Jul 2001 16:30:19 -0400e+ From: "Main, Kerry" <Kerry.Main@compaq.com>y: Subject: RE: Some thoughts on the recent turn of events...R Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4AD7EF8@kaoexc01.americas.cpqcorp.net>   Andrew,u  H >>> I agreed with you up till this. Of course you should expect concrete. timetables plus migration details. You cannot E expect Compaqs customer base to survive on what has been announced sov far.<<<b  K As was stated a number of times in various threads (Fred, Hoff, Steve L) in D this ng, the Engineering and Mgmt folks are working on these now andB additional information will be made available as soon as possible.  ( What part of those threads did you miss?   Regards,  
 Kerry Main Senior Consultanto Compaq Canada Inc. Professional Servicesm Voice: 613-592-4660q Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----7 From: andrew harrison [mailto:andrew.nospam@uk.sun.com]b Sent: July 11, 2001 9:26 AM  To: Info-VAX@Mvb.Saic.ComK: Subject: Re: Some thoughts on the recent turn of events...     Robert Deininger wrote:d > 9 > In article <3B474299.AFCFEDB2@ui.urban.org>, Jim Beckero > <jbecker@ui.urban.org> wrote:o > H > > Not exactly a reply to Ken's message -- I'm just swiping his subject > > line...a > >hI > > Our organization was, before the recent turn of events, just about tomD > > buy another Alpha running OpenVMS. We're not a big shop, so thisJ > > wasn't a trivial decision. We like VMS and we like the Alpha, but whenA > > a vendor announces plans to discontinue one platform and portcF > > "everything" to another platform to be made by some other company,C > > it's just good business sense for us to re-assess our decision.e >  > <snip> > H > You raised many good points.  Some of them are directly under compaq'sL > control.  I suggest you put the questions for Compaq in a nice letter, andE > send it to Rich Marcello.  Make it plain that your next alphaserver B > purchase is ON HOLD until you see concrete movement in the right
 > directions.  > J > Keep in mind that it is less than 2 weeks since the public announcement.5 > You shouldn't expect concrete timetables yet, IMHO.t >   < I agreed with you up till this. Of course you should expect 7 concrete timetables plus migration details. You cannot  9 expect Compaqs customer base to survive on what has been r announced so far.I  7 The idea that Compaq has made this announcment without ,5 haveing answers to these questions is almost worth a d class action suit.    J > Similar letters should go to the vendors of your important applications.K > They may need even more time, since they have to read Compaq's tea leavesB$ > before they worry about their own. >   B The idea that Compaq made this announcemnt without first assessing7 the intentions of their major ISV's is also ludicrous. o   Regardsu Andrew Harrisong Enterprise IT Architecte   ------------------------------  % Date: Wed, 11 Jul 2001 19:29:39 -0400 ' From: "Bill Todd" <billtodd@foo.mv.com>c: Subject: Re: Some thoughts on the recent turn of events...( Message-ID: <9iin9i$7e3$1@pyrite.mv.net>  6 "Main, Kerry" <Kerry.Main@compaq.com> wrote in messageL news:BE56C50EA024184DAF48F0B9A47F5CF4AD7EF8@kaoexc01.americas.cpqcorp.net...	 > Andrew,l > J > >>> I agreed with you up till this. Of course you should expect concrete/ > timetables plus migration details. You cannotbG > expect Compaqs customer base to survive on what has been announced so 	 > far.<<<f >nJ > As was stated a number of times in various threads (Fred, Hoff, Steve L) inF > this ng, the Engineering and Mgmt folks are working on these now andD > additional information will be made available as soon as possible. >o* > What part of those threads did you miss?  J The "We had this planned out well enough to give people at least the ghost: of a chance of believing we knew what we were doing" part?   - bill   >E
 > Regards, >  > Kerry Main > Senior ConsultantT > Compaq Canada Inc. > Professional Services  > Voice: 613-592-4660r > Fax  :  819-772-7036 > Email: Kerry.Main@Compaq.com   ------------------------------  # Date: Thu, 12 Jul 2001 05:11:19 GMTo. From: "Duane Sand" <duane.sand@mindspring.com>: Subject: Re: Some thoughts on the recent turn of events...A Message-ID: <Xja37.176293$%i7.115053738@news1.rdc1.sfba.home.com>-   > "Main, Kerry" wroten, > > What part of those threads did you miss?   Bill Todd wrote:L > The "We had this planned out well enough to give people at least the ghost< > of a chance of believing we knew what we were doing" part?  @ Yup.  We're all looking for those lost threads (inside and out)!  
   -- Duane   ------------------------------  # Date: Wed, 11 Jul 2001 20:00:43 GMTr4 From: "Terry C. Shannon" <terryshannon@mediaone.net>- Subject: Re: Terry Shannon Tech Talk on IA-64s8 Message-ID: <Lf237.13$XT4.47471@typhoon.ne.mediaone.net>  4 "D.Webb" <david20@alpha1.mdx.ac.uk> wrote in message% news:9ihu0s$ikh$1@aquila.mdx.ac.uk...oD > In article <3b4c27b4$1@news.kapsch.co.at>, eplan@kapsch.net (Peter LANGSTOEGER) writes:H > >In article <uUM27.550$l%.214419@typhoon2.gnilink.net>, "Jeff Killeen" <Jeff@IDM-IO.com> writes:e > >>Revised... > >R% > >Time to jump in (for a second) ;-)g > >e: > >>This is the first in a series of "Tech Talks" by Terry< > >>Shannon. This Tech Talk will be on IA-64: Good Questions > >>You Will Need to Answere > >>! > >>We live in interesting times.  > >m< > >Nuclear (or Genome) technology/times is interesting, too.J > >But I so far haven't seen anything really positive from this invention,) > >only a really big problem for mankind.h > >c8 > >'Interesting' doesn't neccessarily mean 'positive' !!9 > >And in the case of Alpha/IA64 I still only have fears.- > >- > >- >-9 > "May you live in interesting times" is a chinese curse.u >rJ > Historically "interesting times" are those times when disaster strikes - > plagues, wars etc2  ( Which ties back to the curse, of course. >3 >: > >TG > >Why does everyone seem to assume that IA64 would be the only chip of I > >the future ? I think the opposite is true. I don't think that IA64 canoG > >totally replace IA32 within this decade !! So, INTEL would have thiscH > >price/performance problems for their 64bit chip, too, for many years.D > >And is a decade later on the market (=> has no applications whichF > >requires IA64). And is slower than Alpha (for yet another couple of years).r > >tK > >For me, Q has killed the now much brighter future for the Alpha (because J > >INTEL would have not satisfied all the perceptions of all the customers, > >in the new CPU of the chip market leader)  F Given the current Alpha installed base and shipment rate, and absent aJ DRAMATIC incease in adoption, Compaq really couldn't justify the continuedG investment. Decisions that were made in the early 1990s (let's build ansG architecture and not ensure that it becomes a volume platform) mid-90's5K (let's get suckered into a one-sided Alliance for Enterprise Computing) andnF near the end of the decade (let's buy Digital and then marginalize its5 flagship architecture) have finally taken their toll.u   ------------------------------  % Date: Wed, 11 Jul 2001 16:25:18 -0500l1 From: "David J. Dachtera" <djesys.nospam@fsi.net> - Subject: Re: Terry Shannon Tech Talk on IA-64 ' Message-ID: <3B4CC43E.C26C5777@fsi.net>    "Terry C. Shannon" wrote:s > 6 > "D.Webb" <david20@alpha1.mdx.ac.uk> wrote in message' > news:9ihu0s$ikh$1@aquila.mdx.ac.uk...CF > > In article <3b4c27b4$1@news.kapsch.co.at>, eplan@kapsch.net (Peter > LANGSTOEGER) writes:J > > >In article <uUM27.550$l%.214419@typhoon2.gnilink.net>, "Jeff Killeen" > <Jeff@IDM-IO.com> writes:u > > >>Revised... > > > ' > > >Time to jump in (for a second) ;-)  > > >e< > > >>This is the first in a series of "Tech Talks" by Terry> > > >>Shannon. This Tech Talk will be on IA-64: Good Questions > > >>You Will Need to Answern > > >># > > >>We live in interesting times.  > > >-> > > >Nuclear (or Genome) technology/times is interesting, too.L > > >But I so far haven't seen anything really positive from this invention,+ > > >only a really big problem for mankind.5 > > >8: > > >'Interesting' doesn't neccessarily mean 'positive' !!; > > >And in the case of Alpha/IA64 I still only have fears.s > > >  > > >r > >k; > > "May you live in interesting times" is a chinese curse.  > >tL > > Historically "interesting times" are those times when disaster strikes - > > plagues, wars etco  H Economically, these are "interesting" times, if you equate "interesting" with "challenging":s   o Endless streams of lay-offsr o Webvan goes belly-up) o Sears gives up on their Homelife storesc> o Compaq cuts another 10%+, EOLs Alpha and partners with Intel o et cetera...  * > Which ties back to the curse, of course. > [snip]H > Given the current Alpha installed base and shipment rate, and absent aL > DRAMATIC incease in adoption, Compaq really couldn't justify the continuedI > investment. Decisions that were made in the early 1990s (let's build anuI > architecture and not ensure that it becomes a volume platform) mid-90'snM > (let's get suckered into a one-sided Alliance for Enterprise Computing) andfH > near the end of the decade (let's buy Digital and then marginalize its7 > flagship architecture) have finally taken their toll.a  6 Maybe we could give 'em a collective, "We told ya so!"   -- ) David J. Dachtera  dba DJE Systemsu http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/n  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.s   ------------------------------  # Date: Wed, 11 Jul 2001 22:38:08 GMTu4 From: "Terry C. Shannon" <terryshannon@mediaone.net>- Subject: Re: Terry Shannon Tech Talk on IA-64w9 Message-ID: <kz437.77$XT4.116642@typhoon.ne.mediaone.net>y  < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3B4CC43E.C26C5777@fsi.net...k > "Terry C. Shannon" wrote:g > >t8 > > "D.Webb" <david20@alpha1.mdx.ac.uk> wrote in message) > > news:9ihu0s$ikh$1@aquila.mdx.ac.uk...eH > > > In article <3b4c27b4$1@news.kapsch.co.at>, eplan@kapsch.net (Peter > > LANGSTOEGER) writes:L > > > >In article <uUM27.550$l%.214419@typhoon2.gnilink.net>, "Jeff Killeen" > > <Jeff@IDM-IO.com> writes:a > > > >>Revised... > > > >-) > > > >Time to jump in (for a second) ;-)1 > > > >0> > > > >>This is the first in a series of "Tech Talks" by Terry@ > > > >>Shannon. This Tech Talk will be on IA-64: Good Questions > > > >>You Will Need to Answern > > > >>% > > > >>We live in interesting times.v > > > >a@ > > > >Nuclear (or Genome) technology/times is interesting, too.C > > > >But I so far haven't seen anything really positive from thiss
 invention,- > > > >only a really big problem for mankind.e > > > >o< > > > >'Interesting' doesn't neccessarily mean 'positive' !!= > > > >And in the case of Alpha/IA64 I still only have fears.o > > > >m > > > >  > > >u= > > > "May you live in interesting times" is a chinese curse.E > > >sD > > > Historically "interesting times" are those times when disaster	 strikes -e > > > plagues, wars etcs > J > Economically, these are "interesting" times, if you equate "interesting" > with "challenging":  >u > o Endless streams of lay-offsi > o Webvan goes belly-up+ > o Sears gives up on their Homelife storess@ > o Compaq cuts another 10%+, EOLs Alpha and partners with Intel > o et cetera... >J, > > Which ties back to the curse, of course.
 > > [snip]J > > Given the current Alpha installed base and shipment rate, and absent aD > > DRAMATIC incease in adoption, Compaq really couldn't justify the	 continued0K > > investment. Decisions that were made in the early 1990s (let's build annK > > architecture and not ensure that it becomes a volume platform) mid-90'shK > > (let's get suckered into a one-sided Alliance for Enterprise Computing)  andqJ > > near the end of the decade (let's buy Digital and then marginalize its9 > > flagship architecture) have finally taken their toll.h >o8 > Maybe we could give 'em a collective, "We told ya so!" >i  G Yeah, I'm sure we could. Compaq could have revived the Alpha effort andnI proliferated the architecture's acceptance, but the firm's actions duringeL the first few post-Compaqtion months sealed Alpha's fate. Recall that CompaqD was, among other things, porting its Nonstop SQL database to WNT andK fortifying WNT with a cluster file system and real clustering capabilities.aF Had the Sculptor project continued, Alpha today would be the referenceI platform for Win64. A certain ex-DEC ex-Tandem ex-Compaq senior executiveeI who shall remain nameless pulled the plug on the whole scheme, so here we  are today...   ------------------------------  % Date: Wed, 11 Jul 2001 19:35:17 -0400-' From: "Bill Todd" <billtodd@foo.mv.com>7- Subject: Re: Terry Shannon Tech Talk on IA-64a( Message-ID: <9iink6$7el$1@pyrite.mv.net>  ? "Terry C. Shannon" <terryshannon@mediaone.net> wrote in message 2 news:Lf237.13$XT4.47471@typhoon.ne.mediaone.net...   ...h  H > Given the current Alpha installed base and shipment rate, and absent aL > DRAMATIC incease in adoption, Compaq really couldn't justify the continued
 > investment.t  K The first premise is dubious (since it certainly appears that overall Alpha-F was reasonably profitable, which is more than you can say for a lot ofJ Compaq's endeavors of late).  The second is what really pisses people off:E if Compaq had made more than token efforts as recently as last month,yJ there's every reason to believe that Alpha adoption *would* have increased7 dramatically, and this was 100% under Compaq's control.r  <  Decisions that were made in the early 1990s (let's build anI > architecture and not ensure that it becomes a volume platform) mid-90'sbI > (let's get suckered into a one-sided Alliance for Enterprise Computing). and(H > near the end of the decade (let's buy Digital and then marginalize its7 > flagship architecture) have finally taken their toll.   K Yeah, right.  Let's see:  we inherited this full Paul Revere silver serviceuI from our grandmother, but she hadn't polished it in years - throw it out.t   - bill   ------------------------------  % Date: Wed, 11 Jul 2001 19:39:10 -04006' From: "Bill Todd" <billtodd@foo.mv.com>m- Subject: Re: Terry Shannon Tech Talk on IA-64o( Message-ID: <9iinre$7ma$1@pyrite.mv.net>  ? "Terry C. Shannon" <terryshannon@mediaone.net> wrote in messageo3 news:kz437.77$XT4.116642@typhoon.ne.mediaone.net...    ...h  I > Yeah, I'm sure we could. Compaq could have revived the Alpha effort andeK > proliferated the architecture's acceptance, but the firm's actions during,; > the first few post-Compaqtion months sealed Alpha's fate.h  G Bullshit.  What sealed Alpha's fate was the announcement two weeks ago."I Alpha managed to ride through all the other mismanagement with its market F share, technical credibility, and profits pretty much intact, just not growing.   - bill   ------------------------------  # Date: Thu, 12 Jul 2001 03:34:05 GMT . From: "mulp" <michaelpettengill@earthlink.net>- Subject: Re: Terry Shannon Tech Talk on IA-64 ? Message-ID: <NU837.20$pv.9753@newsread1.prod.itd.earthlink.net>-  1 "Jeff Killeen" <Jeff@IDM-IO.com> wrote in messagea3 news:JZE27.8229$Tl4.2871005@typhoon2.gnilink.net...tL > InfiniBand - a new low latency high bandwidth I/O and interconnect fabric.C > It incorporates many of the concepts of Compaq's Tandem-developeda	 ServerNet D > architecture. InfiniBand will show up first as an I/O subsystem in ProLiantI > systems, and then will be adopted as a cluster interconnect. It is good0 > stuff.  L The way to understand Servernet and Infiniband is to think about CI-2, CI-3, CI-4.n   > L > QuickBlade, BladeRunner, etc. Code names for systems that will rely on CPUD > modules (called Blades) that sit on an InfiniBand "backplane." RLX  G Infiniband is a switched I/O architecture with functions that one would-  expect in a greatly enhanced CI.  I As such, a "blade" solution glued together with Infiniband will look moree* like a cluster than like a multiprocessor.  J EV7, on the other hand, integrates a switched memory interface with a eachK node having an I/O hose which accesses memory thru the same switched memory 
 interface.   ------------------------------  # Date: Thu, 12 Jul 2001 03:50:42 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>- Subject: Re: Terry Shannon Tech Talk on IA-64 : Message-ID: <m8937.123$XT4.278959@typhoon.ne.mediaone.net>  9 "mulp" <michaelpettengill@earthlink.net> wrote in messages9 news:NU837.20$pv.9753@newsread1.prod.itd.earthlink.net...t3 > "Jeff Killeen" <Jeff@IDM-IO.com> wrote in messager5 > news:JZE27.8229$Tl4.2871005@typhoon2.gnilink.net... F > > InfiniBand - a new low latency high bandwidth I/O and interconnect fabric.?E > > It incorporates many of the concepts of Compaq's Tandem-developedf > ServerNetiF > > architecture. InfiniBand will show up first as an I/O subsystem in
 > ProLiantK > > systems, and then will be adopted as a cluster interconnect. It is goodn
 > > stuff. >uH > The way to understand Servernet and Infiniband is to think about CI-2, CI-3,h > CI-4.t >  > >eJ > > QuickBlade, BladeRunner, etc. Code names for systems that will rely on CPUnF > > modules (called Blades) that sit on an InfiniBand "backplane." RLX > I > Infiniband is a switched I/O architecture with functions that one would " > expect in a greatly enhanced CI.  A And IIRC InfiniBand will serve as the next-generation VMS cluster 
 interconnect.n   ------------------------------  # Date: Thu, 12 Jul 2001 03:57:19 GMT . From: "mulp" <michaelpettengill@earthlink.net>- Subject: Re: Terry Shannon Tech Talk on IA-64 ? Message-ID: <ze937.37$IJ.8622@newsread1.prod.itd.earthlink.net>n  ? "Terry C. Shannon" <terryshannon@mediaone.net> wrote in messagei2 news:Lf237.13$XT4.47471@typhoon.ne.mediaone.net...H > Given the current Alpha installed base and shipment rate, and absent aL > DRAMATIC incease in adoption, Compaq really couldn't justify the continuedI > investment. Decisions that were made in the early 1990s (let's build an I > architecture and not ensure that it becomes a volume platform) mid-90's I > (let's get suckered into a one-sided Alliance for Enterprise Computing)e andnH > near the end of the decade (let's buy Digital and then marginalize its7 > flagship architecture) have finally taken their toll.n  F It wasn't so much a matter of failing to make Alpha volume, as much asL making sure that VMS revenue declined in the hopes that it would go away and% somehow make Digital Unix successful.e  1 Compaq has diluted the Tandem uniqueness as well.   J The problem is that customers and ISVs would not cooperate and today thereI is a massive amount of revenue and profit/profit potential in the VMS and  Tandem businesses.  K A few changes in the slides replaces Alpha with IA64 and magically a lot ofaL money is saved in the engineering costs.  But on the other hand, the revenueE coming from the chip and system sales goes away as well.   Maybe $250-L million was spent per year, but the net cost was a lot less.  So with a costJ of switching to a new architecture not having been calculated, but my swagH puts it at at least half a billion.  So the switch reaches break even atH best 10 years down the road, assuming that Compaq stays in business that long.i  K And if you say, "but Intel is going pay for the porting", I say, well, theyuC should pay for the half billion in porting because this is to theirYL advantage, but they should also pay a billion or two for the IP that they atE getting in the deal.  My guess is that Intel is only going pay a halfrH billion, but somehow that is going to end up being in reduced Intel chipG prices which will mean that the Intel side of the business will be moreaJ competitive with Dell while the VMS, Tandem, and Tru64 businesses struggle! with the lost customers and ISVs.n  J In the end, Dell will go to Intel and say, give us the same price you give Compaq or we switch to AMD.l   ------------------------------  # Date: Thu, 12 Jul 2001 05:21:12 GMTd. From: "mulp" <michaelpettengill@earthlink.net>- Subject: Re: Terry Shannon Tech Talk on IA-64yA Message-ID: <cta37.199$IJ.27624@newsread1.prod.itd.earthlink.net>   ? "Terry C. Shannon" <terryshannon@mediaone.net> wrote in messageo4 news:m8937.123$XT4.278959@typhoon.ne.mediaone.net...C > And IIRC InfiniBand will serve as the next-generation VMS clustere > interconnect.,  J I wouldn't hold my breath.  Almost all the people who did the work on SCA,L SCS, NISCS, CI, etc. have left.  Support of Fibre Channel has consumed a lotI of resources as has the new NISCS enhancements.  It should be interestingeD finding some people who know a lot about Infiniband architecture andH implementation and VMS internals and cluster architecture to develop theK mapping between the two systems.  The people involved in that probably haveaF a year's worth of up front study and planning before starting to code.   ------------------------------  % Date: Wed, 11 Jul 2001 18:33:03 -0400h+ From: "Main, Kerry" <Kerry.Main@compaq.com><: Subject: RE: The death of VMS has been greatly exaggeratedR Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4AD7EFA@kaoexc01.americas.cpqcorp.net>   Andrew, Andrew ...  K >>> So at the moment you have no idea what the transition costs will be fore' a customer to move to IA-64 from Alpha.e  K Since this is the case isn't a bit early for Compaq to be talking about thesJ benefits of moving to IA-64. Without being able to judge the costs you areH not providing your customers with enough information to make an informed
 choice.<<<  L As stated a number of times in this ng, the appropriate folks are working onJ the plans, and these will be made available as soon as possible. Does this3 need to be repeated a number of times more for you?   L The fact that over time, AIX, HPUX, Linux64 (IBM and Compaq and other vendorL 64bit Linux's), Win64, OpenVMS, Tru64 UNIX, and Tandem NSK all have plans toE support the IA64 platform, has got to say something about goodness inu+ protecting future Customer HW investments.    G In addition, with the exception of MS, each OS provider has a backup or I alternate 64bit HW plan (Power4, Alpha EV7, PA, MIPS) that will allow thehK Customer to migrate when they feel the timing is right - whether it is in 3o years or 5+ years.  L However, as stated before, more detailed information on all of this is still" cooking and will be released asap.   Regards,  
 Kerry Main Senior Consultantf Compaq Canada Inc. Professional Servicesh Voice: 613-592-4660u Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----7 From: andrew harrison [mailto:andrew.nospam@uk.sun.com]t Sent: July 11, 2001 7:34 AM  To: Info-VAX@Mvb.Saic.Comi: Subject: Re: The death of VMS has been greatly exaggerated     "Main, Kerry" wrote: > 	 > Andrew,  > K > >>> You are whistling in the wind. Anything that requires new hardware, acJ > new os and all the applications to be ported when compared with the list youeA > presented from the other vendors is at a huge dissadvantage.<<<t > L > Given this announcement was only made a week ago, I would suggest that youJ > do not make comparitive statements about what is available today vs whatE > will be available in 2-3 years when the VMS/IA64 offerings start toy appear.f >   8 This cuts both ways Kerry as you should appreciate only 	 too well.C  ? Perhaps that person who has most to learn from the last months "9 events is you, interesting that you don't appear to have.$  H > Will there be emulation built in for OpenVMS Customers or a translator (aka1 > vest) or an FX!32 type utility ?  I don't know.r >   < So at the moment you have no idea what the transition costs 3 will be for a customer to move to IA-64 from Alpha.   9 Since this is the case isn't a bit early for Compaq to bei7 talking about the benefits of moving to IA-64. Without h8 being able to judge the costs you are not providing your6 customers with enough information to make an informed  choice.g    H > In addition, while HW/OS vendors may state app's will run in emulation mode, ? > will the applicable SW vendor fully support this environment?e > E > Some might, but if the FX!32 experience is any guage, they may not.t > H > My point is that for full vendor support and the best performance, allK > platforms require their important applications to be recompiled in nativerI > mode on the new target platforms- surely you do not disagree with this?e >   E You are right, but it also depends on the ratio of natively compiled cH performance to emulated and how performance critical the particular app E is on the overall software stack. If performance is not to awfull and  the vendor  < supports an emulated mode then it would be Ok for some apps.    F > Imho, its way to early to say what the strategy for this will be forG > OpenVMS. The Eng folks are working out their strategy and will updates  > Customers as soon as possible. > J > However, they have stated support for mixed IA64-Alpha clusters (similar toI > VAX-Alpha clusters) - can the same be said for mixed Sparc II and SPARCu IIIdJ > servers in the latest Sun cluster offerings? Will a SPARC II applicationG > running in emulation mode be supported in a SPARC III Sun cluster? [Ie reallyF > dont know these answers and would appreciate an update or pointer on this..]v >   F This is an entirely different propostion, the both the UltraII and the UltraIIII support the SPARC V9 architecture and are binary compatible. This is not fD what you are providing with mixed IA-64/Alpha clusters. You might as well /C ask if Comapq support mixing EV5 and EV6 nodes in the same cluster,i which  I assume you do.  G > The OpenVMS Eng folks have a great deal of experience in dealing with.L > porting issues and mixed cluster support, so I am confident that with themJ > working closely with the Alpha EV8 team that is going to work with Intel onL > improving the Itanium architecture, any technical issues will be resolved. >   D This misses the point, it does not really matter how good your port B of OpenVMS is or how performant or how timely, I would assume the E best case scenario if I was you. What matters is what your key ISV's n? do and this is not something that OpenVMS engineering or Intel   can influence directly.0   Regardsi Andrew Harrison  Enterprise IT Architect4   ------------------------------  # Date: Thu, 12 Jul 2001 00:01:32 GMT   From: jlsue <jlsuexxxz@home.com>: Subject: Re: The death of VMS has been greatly exaggerated8 Message-ID: <manpkt8p2glpttmlj1r48inkkothfuudqi@4ax.com>  / On Fri, 06 Jul 2001 16:08:56 +0200, Arne Vajhjh <arne.vajhoej@gtech.com> wrote:e   >Arne Vajhj wrote: = >> > I also guess that you don't know that no IA-64 processor 6 >> > has the ability to decode and run Alpha binaries. >> i< >> I do not know whether the IA-64's in 2003 will have Alpha! >> emulation capabilities or not.  >> B? >> But ofcourse Intel and Compaq top-executives are sharing alle5 >> the details of their talks with you, so you know !  > = >And just to make it completely clear: my point is that it is = >pretty disgusting to see a SUN sales guy tell VMS users whats; >exactly Compaq and Intel will do without any foundation ins >the publisized material.r >EB >I am very sure that the Compaq way wil be recompilation and maybe5 >software emulation, but not hardware emulation. This @ >was the decision for VAX->Alpha and I do not think the decision >wil be different this time. >u  F Hey Arne, Andrew's been doing that for years.  He's repeatedly told usD what Compaq Management's plans were, when there was no communication& affect anywhere to back up his claims.  C This is why his comments are most often described as FUD (as in, no E basis in fact, but trying to scare people away from the competition).a   ------------------------------  # Date: Thu, 12 Jul 2001 00:01:33 GMTo  From: jlsue <jlsuexxxz@home.com>: Subject: Re: The death of VMS has been greatly exaggerated8 Message-ID: <4cnpktc4odgqkve5pgddkqr1t3m5lnlqaq@4ax.com>  C Sure, they miss chances to improve things, but they do get the porte/ done quicker... which is a big issue right now.   E I remember one of the quotes at a (then) DECUS session discussing thetE VAX-to-Alpha port was "Just port it" (a take-off on the Nike "Just doo it" ad).    E On Fri, 06 Jul 2001 21:53:10 -0400, rdeininger@mindspring.com (Robertt Deininger) wrote:n  P >In article <009FE9D4.D7BB65F8@SendSpamHere.ORG>, system@SendSpamHere.ORG wrote: > I >> In article <3b463834.9185828@news.process.com>, goathunter@goatley.como >(Hunter Goatley) writes:s8 >> >On Fri, 06 Jul 2001 21:39:41 GMT, "Terry C. Shannon" ><terryshannon@mediaone.net>
 >> >wrote: >> >>sN >> >>Gotta agree with you on that one. I understand that when the VAX to AlphaM >> >>port was done, all the VMS device drivers and much of the OS itself werel >> >>recoded in C, >> >N >> >That's not true.  A few drivers, maybe, and some pieces of the OS, but the# >>                         ^-- new :& >> >majority is still BLISS and MACRO. >> .L >> Recently, I was looking through the source listings to see if/where some K >> mod_STD$routine was used.  Much to my chagrin, DEC created them but theyiM >> seldom use them.  Why?  Because so much of OpenVMS was ported with little eK >> more than a compile.  In the Macro-32 case, the addition of .directives CJ >> to assist the compiler with knowledge it might no be able to gleen from% >> scanning and parsing the source.  ( >cC >I guess that's good and bad news.  If they could port once by just,J >compiling, they can almost certainly port the module the same way again. / >That's good, since it leads to a quicker port.C >a9 >The bad part is that they missed a chance to upgrade thehK >quality/maintainability of these modules, and will probably miss many suchd >chances again this time.s >0I >On the other hand, if it ain't broke, don't fix it still applies.  Every(< >time someone tweeks a module, there's a chance to break it.   ------------------------------  % Date: Wed, 11 Jul 2001 22:12:11 +0100c+ From: "antonio.carlini" <arcarlini@iee.org>e Subject: Re: VMS V7.3 SPD Errori' Message-ID: <3B4CC12B.C60F619C@iee.org>a   Robert Deininger wrote:aI > How about a locally-attached booted disk, containing an image backup ofeE > the VMS CD (created on another VMS system)?  This is the method youi9 > suggested for CD-less alpha systems, and it works fine.-   How about hanging said image 1 (or even the actual CD)B off an Infoserver.  % Then you can be unsupported twice :-)T   Antonio    --     ---------------$- Antonio Carlini             arcarlini@iee.orgr   ------------------------------    Date: 11 Jul 2001 12:38:21 -0700+ From: stephane_paquin@hotmail.com (SPaquin)'  Subject: Re: VT520 F2 key in FMS< Message-ID: <fdd7874.0107111138.2038506b@posting.google.com>  o stephane_paquin@hotmail.com (SPaquin) wrote in message news:<fdd7874.0107051219.11e2a3d5@posting.google.com>...l > Hi ! > G > On our old VT420, the F2 key was defined as F2( F2=F2 ) In FMS forms,o> > it was returning 44 as field terminator and it was processed > accordingly. > G > On the new VT520, I am having trouble defining the same behavior. How-G > can I define the VT520 F2 key so that it sends the same key code than  > the VT420 F2=F2 definition ? >  > Any help will be appreciated.T > Stphane Paquin   A For the record, I have finally fond the way to define VT420 F2=F2J behavior on the VT520.  @ The sequence to enter in the UDK for unshifted F2 key is ESC[12~  6 Select Host only and save. I have found a reference at7 www.cs.utk.edu/~shuford/terminal/dec_keyboards_news.txt      Stphane Paquinq   ------------------------------  % Date: Wed, 11 Jul 2001 14:02:02 -0400i+ From: John Eisenschmidt <jeisensc@aaas.org>m  Subject: Re: VXT 2000 with VMS?!# Message-ID: <sb4c5c77.082@aaas.org>t   Hoff dreams...  F Performing Cluster Wide Shadow Merge on node ICEBERG running OpenVMS =
 Itanic 7.5  E OpenVMS (TM) Itanic Operating System, Version V7.5   -- System Dump =  Analysis Crashdump Summary Information:  + Crash Time:         11-JUL-2001 11:18:29.49w/ Bugcheck Type:     INSF_OCN, Insufficient Oceano, Node:              ICEBERG  (Cluster Member)7 CPU Type:          Intel Itanic IA-64 Reference 800 MHzl VMS Version:       V7.5t  Current Process:   SHADOW_SERVER/ Current Image:     SYS$SYSTEM:SHADOW_SERVER.EXEo$ Failing PC:        FFFFFFFF.802D544C$ Failing PS:        18000000.00000803L Module:            SYS$SHDRIVER    (Link Date/Time: 16-OCT-2000 08:53:35.67= )  Offset:            0007144C   + Boot Time:          10-JUL-2001 08:39:58.00n* System Uptime:               1 02:38:31.49  J >>> Hoff Hoffman <hoffman@xdelta.zko.dec.nospam> 07/11/2001 1:40:44 PM >>>L In article <Pine.LNX.4.21.0107111631140.11106-100000@firewall.freddym.org>,=0  Freddy Meerwaldt <frederik@freddym.org> writes:  J :unreliable sources were telling me that it is possible to boot a VMS on a :VXT 2000+.t :Is that true?     Not that I am aware of.=20  :   AFAIK, the VXT2000 is unlike any VAXstation or MicroVAX.  G   This may well be some confusion with the predecessor VT1300 series=20cH   X terminals, a model which was a hardware variant of the VAXstation=20    3100 model 38 SPX workstation.  5 :It would be cool to have a VXT2000+ in a VMS-Cluster   G   For some definitions of "cool", of course.  Mine involves having a=20IF   new [censored] system booted into the local cluster.  Whoops, sorry,   can't talk about that. :-)  F   As for some of the other VXT questions, the VXT software and related#   stuff is on the OpenVMS Freeware.O    L  ---------------------------- #include <rtfaq.h> --------------------------= --- L       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com =   =20-L  --------------------------- pure personal opinion ------------------------= ---iL    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.co= me   ------------------------------  % Date: Wed, 11 Jul 2001 22:44:12 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)  Subject: Re: VXT 2000 with VMS?!L Message-ID: <rdeininger-1107012244140001@user-2iveapg.dialup.mindspring.com>  P In article <sb4c5c77.082@aaas.org>, John Eisenschmidt <jeisensc@aaas.org> wrote:   > Hoff dreams... > H > Performing Cluster Wide Shadow Merge on node ICEBERG running OpenVMS = > Itanic 7.5 > G > OpenVMS (TM) Itanic Operating System, Version V7.5   -- System Dump =o
 > Analysis  > Crashdump Summary Information: > - > Crash Time:         11-JUL-2001 11:18:29.49i1 > Bugcheck Type:     INSF_OCN, Insufficient Oceant. > Node:              ICEBERG  (Cluster Member)9 > CPU Type:          Intel Itanic IA-64 Reference 800 MHz  > VMS Version:       V7.5v" > Current Process:   SHADOW_SERVER1 > Current Image:     SYS$SYSTEM:SHADOW_SERVER.EXEf& > Failing PC:        FFFFFFFF.802D544C& > Failing PS:        18000000.00000803N > Module:            SYS$SHDRIVER    (Link Date/Time: 16-OCT-2000 08:53:35.67= > )n  J So, you claim they've been secretely working on the itanic port since last October?   :-)e   -- c Robert Deininger rdeininger@mindspring.como   ------------------------------  % Date: Wed, 11 Jul 2001 22:24:20 -0600  From: yyyc186@mindspring.com% Subject: Re: What will we see at 2004-; Message-ID: <3b4d2680$1$lllp186$mr2ice@nntp.mindspring.com>.  B In <iLz27.18119$gb6.1943932@news20.bellglobal.com>, on 07/10/2001 9    at 05:22 AM, "Neil Rieck" <n.rieck@sympatico.ca> said:   I >Consider this: when VAX/VMS was rolled out in 77-78 did DEC announce thehJ >death of PDP (even if they were planning to eventually kill it off and/orH >sell the 16-bit  technology to someone else)? When Alpha was rolled outG >in 92-93 did DEC announce the future death of VAX? So why can't Compaq 1 >play their cards a little closer to their chest?n  ' Yes, they did...it was called the 9000.n   Roland -- h; -----------------------------------------------------------e yyyc186@mindspring.com; -----------------------------------------------------------m   ------------------------------   End of INFO-VAX 2001.383 ************************ lost customers and ISVs.n  J In the end, Dell will go to Intel and say, give us the same price you give Compaq or we switch to AMD.l   ------------------------------  # Date: Thu, 12 Jul 2001 05:21:12 GMTd. From