1 INFO-VAX	Thu, 07 Mar 2002	Volume 2002 : Issue 130       Contents:# Re: Alpha CPU Architecture question ) Re: Any virus scan software for VMS smtp? ) Re: Any virus scan software for VMS smtp? 2 Re: best/most flexible preserve VMS files onto CD?2 Re: best/most flexible preserve VMS files onto CD?* Re: checking decc compiler version in mms?3 Re: Clustering beginnings - Clair Grant might know. 3 Re: Clustering beginnings - Clair Grant might know. 3 Re: Clustering beginnings - Clair Grant might know. 3 Re: Clustering beginnings - Clair Grant might know. 3 Re: Clustering beginnings - Clair Grant might know. 3 Re: Clustering beginnings - Clair Grant might know. 3 Re: Clustering beginnings - Clair Grant might know. 3 Re: Clustering beginnings - Clair Grant might know. 3 Re: Clustering beginnings - Clair Grant might know. 3 Re: Clustering beginnings - Clair Grant might know. 3 Re: Clustering beginnings - Clair Grant might know. . Re: DCL challenge of the day: dates comparison. Re: DCL challenge of the day: dates comparison. Re: DCL challenge of the day: dates comparison5 RE: DCL Minute of the GMT+1 nite: nr of lines (again)  Re: emacs on VMS, Re: EV8 and McKinley analysis by Paul DeMone Re: expensive procedure calls & Re: GEMBASE in GS Servers *** HELP ***# HP Unix servers plunge below $1,000 ; Import & Purchase Department - for DEPB  DFRC  SIL & DUTY F  Incremental BACKUP question   Re: interrupt state CPU activity Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles Re: Itanium troubles' Re: LIB$CVT_FROM_INTERNAL_TIME - year?? + Re: Microsoft Curries Favor With Undergrads + Re: Microsoft Curries Favor With Undergrads + Re: Microsoft Curries Favor With Undergrads + Re: Microsoft Curries Favor With Undergrads 8 Re: Microsoft Curries Favor With Undergrads (correction)" MicroVAX 3100/30 to connect to ISP& Re: MicroVAX 3100/30 to connect to ISP& Re: MicroVAX 3100/30 to connect to ISP& Re: MicroVAX 3100/30 to connect to ISP& Re: MicroVAX 3100/30 to connect to ISP& Re: MicroVAX 3100/30 to connect to ISP& Re: MicroVAX 3100/30 to connect to ISP& Re: MicroVAX 3100/30 to connect to ISP MicroVAX Operating Temperature" Re: MicroVAX Operating Temperature" Re: MicroVAX Operating Temperature" RE: MicroVAX Operating Temperature" Re: MicroVAX Operating Temperature- Re: Need help on Alpha:  Calling C from MACRO 3 Re: Netscape annoyance with OpenVMS CGI executables 3 Re: Netscape annoyance with OpenVMS CGI executables 3 Re: Netscape annoyance with OpenVMS CGI executables  OpenVMS jobs available - FYI& Re: Pathworks with Windows 2000 client= Re: provide remote administration facility on a C/C++ process P Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of  Compaq AcP Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of Compaq AcqP Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of Compaq AcqP Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of Compaq AcqP Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of Compaq AcqP Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of Compaq Acq% Re: Real-time scheduling with sockets 	 Re: RWAST ! Re: Standalone backup question... ' Re: Steps for installation OVMS 6.2 1H3  SWCC Re: SWCC" Re: TCPIP$FTP hangs in MUTEX state Re: UNMESSAGE for Alpha ?  Re: VAX Pascal Re: vax to tcp/ip  VMS DEBUG V7.2 bugs me! 2 RE: What VMS version runs on a Digital Server 3000P Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startuP Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startuP Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startuP Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startuP Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startuP Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startu  F ----------------------------------------------------------------------  # Date: Wed, 06 Mar 2002 23:25:38 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman), Subject: Re: Alpha CPU Architecture question2 Message-ID: <Szxh8.665$fL6.10015@news.cpqcorp.net>  ] In article <a5o828$u38$1@newsreader.mailgate.org>, David Harrold <DHarrold@wi.rr.com> writes: M :We had a crash yesterday on one of our GS-160s and the reason came back from O :the support center as a BCache failure on one of the CPUs.  SO what exactly is  :the BCache?  C   This is the Backup Cache -- this is the off-processor cache.  The F   on-processor cache is refered to by various names, including pcache.C   VAX and Alpha systems have had several layers of caching for many E   years now -- caches close to the processor are typically limited by C   the processor chip area available and thus fairly small, and are  I   typically far faster than going off-chip.  Main memory is the furthest  E   away from the microprocessor, and thus the slowest.  Bcache exists  G   between these two extremes, and typically uses more expensive memory  D   technologies than does main memory -- faster than main memory, and   larger than pcache...   B   Main memory failures can sometimes mimic bcache failures, too.     But I digress...  O :And more generally, is there any publicly available documentation on the Alpha * :CPUs that describe these types of things?  G   Do the pointers in the OpenVMS Frequently Asked Questions (FAQ) help? F   (There are a variety of documents available -- this question is moreD   specific to the Alpha microprocessor design than to the particularD   system implementation -- it is the microprocessor that has supportI   for the pcache and (typically) the ability to be connected to a backup  I   cache.  Though the sizes of the bcache can and do vary by the platform,    of course.    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: Thu, 7 Mar 2002 10:55:05 +0000 (UTC) From: david20@alpha1.mdx.ac.uk2 Subject: Re: Any virus scan software for VMS smtp?+ Message-ID: <a67gu9$6ik$1@aquila.mdx.ac.uk>   h In article <d7791aa1.0203060727.681eeb96@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:R >david20@alpha1.mdx.ac.uk wrote in message news:<a64tu7$al1$1@aquila.mdx.ac.uk>...k >> In article <d7791aa1.0203051612.1843853c@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:   J I'm cross posting this to vmsnet.mail.mx to see whether anyone there knowsG whether MX can be setup to virus scan incoming mail in a similar way to  PMDF  
 David Webb VMS and Unix team leader CCSS Middlesex University  @ >> >Called Compaq, Sophos and others and can't find a virus scan@ >> >product that runs on vms for vms smtp based mail servers ...= >> >get the typical "why not put another windoze or linux box ; >> >in front of vms and run 2 servers?" to which I respond  A >> >why should I be forced to run 2 smtp servers, especially when B >> >the windoze linux front would be exposing me to the very thingB >> >I avoid running a single vms server, shutdowns ... does anyone >> >know of anything for vms?  >>  M >> Sophos should have told you that their product runs on VMS. We use it with + >> with the PMDF mailserver product on VMS. D >> See http://www.sophos.com/products/software/antivirus/savvms.html >>   >>  N >> Your only problem if you don't use PMDF might be in integrating the productP >> to scan your incoming Mime encoded mail. PMDF has special inbuilt facilities P >> - the conversion channels - which split out each attachment , Mime decodes itJ >> and then allows a program such as a Virus scanner to be run against it.Q >> The Attachment can then be replaced, deleted etc according to what the scanner 7 >> has found and Mime encoding put back on as required.  >>  Q >> I don't know what facilities the various TCPIP stacks (UCX, MULTINET, TCPWARE) 2 >> and software such as MX provide in this regard. >>   >>  
 >> David Webb  >> VMS and Unix team leader  >> CCSS  >> Middlesex University  > I >that's a problem as I just spoke w/Sophos Britain w/their vms expert and F >he said vsweep is currently locked to pmdf, but they were consideringD >doing an smtp version for tcpware/ucx/multinet on vms ... if vms isI >going to drive ecommerce, Compaq better get some mail solutions for smtp F >instead of everyone having to buy a second mail gateway like pmdf ...   ------------------------------  " Date: Thu,  7 Mar 02 17:02:36 +100 From: rok@nuk.uni-lj.si 2 Subject: Re: Any virus scan software for VMS smtp?& Message-ID: <3c879d35.0@NUK.Uni-Lj.Si>  K >I'm cross posting this to vmsnet.mail.mx to see whether anyone there knows H >whether MX can be setup to virus scan incoming mail in a similar way to >PMDF   D  Yes, it's possible through SITE interface. As Sophos does not parseC multipart MIME files you have to break message into separate parts, . decode and scan them (just as PMDF has to do).   Regards,  D Rok Vidmar                       Internet:  rok.vidmar@nuk.uni-lj.si; National and University Library  Phone:     +386 1 421 5461 ; Turjaska 1, SI-1000 Ljubljana    Fax:       +386 1 421 5464  Slovenia   ------------------------------   Date: 7 Mar 2002 06:37:45 -0600 - From: Kilgallen@SpamCop.net (Larry Kilgallen) ; Subject: Re: best/most flexible preserve VMS files onto CD? 3 Message-ID: <rKWfDEcxPDF8@eisner.encompasserve.org>   [ In article <3C86EC4D.3DB7C663@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:   G > The guys at TECsys turned me onto a method of co-mapping a CD so that H > ISO side points to the extents of the "VMS side". Trouble there is, of- > course, the record format problem remains.    A The ISO 9660 standard provides for storage of RMS record formats, B so there should be no problem with any standards-compliant reader.   ------------------------------  # Date: Thu, 07 Mar 2002 00:47:51 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman); Subject: Re: best/most flexible preserve VMS files onto CD? 2 Message-ID: <XMyh8.670$fL6.10245@news.cpqcorp.net>  b In article <d0e744c9.0203051416.75258c@posting.google.com>, leeroth@my-deja.com (Lee Roth) writes:B :I have some old TK50 & TK70 tapes that were made with VMS BACKUP.> :I would like to 'restore' these files onto a Microvax II hardD :drive (temporarily), and then ship them to a Windoze PC for burning	 :onto CD.      Please don't.     F   Please use a direct-connected CD-R on a SCSI-based OpenVMS system.    B   Yes, you can get a Windows-based burner to work -- you will haveC   to learn rather more about the target file structure for the CD-R C   and the record structures of the files involved, in particular.   E   You will particularly want to avoid an ISO-9660 format with some of H   the extensions -- OpenVMS cannot read various of these ISO-9660 disks.  D   Using the direct access to the CD-R, you can burn ODS-2 or ODS-5, B   the native OpenVMS structures.  (There are also tools to create A   ISO-9660 CD-R media around, too.)  (ODS-2 and ODS-5 can coexist A   with ISO-9660, if you really want to get fancy.)  As for access B   to ODS-2, there are PC tools to read it on the OpenVMS Freeware.  B   There are more than two or three discussions of CD-R on OpenVMS,A   transfering files for CD-R on Windows-based burners, and all of @   the "fun" that can be present.  (As soon as you move outside a>   BACKUP saveset or a correctly-created zip archive on another@   platform, you now start to have problems with the file record @   structure.  Yes, even the record structure of sequential text $   files can get corrupted this way.)  5   Details on the OpenVMS CD-R options are in the FAQ.     N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Wed, 06 Mar 2002 23:43:12 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)3 Subject: Re: checking decc compiler version in mms? 2 Message-ID: <kQxh8.667$fL6.10313@news.cpqcorp.net>  u In article <8369d643.0203011343.1395a528@posting.google.com>, forrest.cahoon@merrillcorp.com (Forrest Cahoon) writes: M :I want to implement a check in mms such that if the decc compiler version is 2 :>= 6.5, /optimize is used, otherwise /nooptimize. : - :Does anyone have a good idea how to do this?   E   There's no particularly modular nor portable approach.  Well, short A   of the simple expedient of installing Compaq C V6.5, of course. B   The ugly approach involves parsing the output -- this works now,D   but is not supported and could break -- or rummaging around in theG   image or image headers.  I've already had to fix code from one bunch  F   that got "creative" with a compiler-related check, and got "nailed" >   during a compiler upgrade.  I don't recommend this approach.  A   Within C code, things get far easier -- there is a preprocessor     define (__DECC_VER) available.  F   I will assume there is some unstated reason not to use /OPT on olderG   releases -- when I've seen software with special requirements for the J   optimizer, it has tracked back to bugs in the application code far more I   regularly than to bugs in the optimizer.  Put another way, what are you H   up to, and why do you think you need this?  (If there's a good reason,+   well, it might be worth implementing it.)   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: Thu, 07 Mar 02 08:58:58 GMT  From: jmfbahciv@aol.com < Subject: Re: Clustering beginnings - Clair Grant might know.+ Message-ID: <a67i66$mg8$1@bob.news.rcn.net>   / In article <3C86B841.FB66242@bellatlantic.net>, +    bad bob <sfmc68@bellatlantic.net> wrote:  >  >  >"Edward C. Bailey" wrote: >>  : >> >>>>> "Alan" == Alan Greig <a.greig@virgin.net> writes: >>  J >> Alan> On Tue, 5 Mar 2002 18:33:10 +0100, "Dr. Dweeb" <Dweeb@NoSpam.com> >> Alan> wrote:  >>  3 >> >> Hmm, clustering in 1980 ?  Is that accurate ?  >>  J >> Alan> By 1982 I had seen a presentation on TOPS-20 clustering (CI, starH >> Alan> couplers, hsc-50 controllers). I believe the TOPS-20 clusteringF >> Alan> shipped before that for VMS. but they could have been done in >> Alan> parallel. >>  K >> Well, in '81 I was working on Galaxy support in MR1, and I recall seeing F >> both "NI" and "CI" cables/hardware in the LCG lab, so it certainly  existed  >> at that time. >>  H >> Before and after this time I worked with VMS and I can tell you that  "real"K >> clustering under VMS (ie, shared read/write file acess between different F >> cluster members) was quite late -- the functionality that VMS firstH >> supported was simply device access over CI.  No shared read/write wasH >> possible because VMS hadn't moved away from ACP processes towards theD >> distributed lock manager approach they eventually implemented to 	 arbitrate  >> shared device access. >>  G >> My recollection was that TOPS-20 clustering did ship before VMS had   "real"J >> clustering; whether it shipped before VMS claimed CI device support, I  doK >> not know.  My cynical side wonders if VMS' CI device support was rushed   out I >> to claim "first support", but I have no data to back that up.  It may   haveJ >> been as simple as the fact that the hardware was ready to be sold, and  the H >> VAX product line had no way to generate sales on this stuff until VMS> >> provided at least some level of support for the hardware... >> oD >That is my recollection too.  The HSC stuff was one of the keys forA >CI connected storage, and that set included various models that c? >transitioned to COlorado in the late 70s iirc, I recall RichieiC >Lary coming in to help with 11/60 microcode, and DMC/KMC microcode @ >around 76 from ColoSprings.  I admit my dates may be off a bit, >maybe a year, but...w  C The clustering project was separate from the CI-HSC implementation.gB TW did the CI-HSC on both TOPS-10 and TOPS-20.  The -10 was first,> then he did the -20.  Look at the dates on the monitor sources? and you'll get an idea of when he did his work.  I didn't thinkw> that the clustering project was dependent on the CI being done= (but I could be very wrong...it was hard enough keeping trackb of the -10 bit flows).   /BAH  ' Subtract a hundred and four for e-mail.n   ------------------------------  # Date: Thu, 07 Mar 2002 12:43:09 GMT ' From: bad bob <sfmc68@bellatlantic.net>n< Subject: Re: Clustering beginnings - Clair Grant might know.0 Message-ID: <3C8760BE.85FA077C@bellatlantic.net>   jmfbahciv@aol.com wrote: > 1 > In article <3C86B841.FB66242@bellatlantic.net>,a- >    bad bob <sfmc68@bellatlantic.net> wrote:a ><<snip to save my eyes>>vF > >That is my recollection too.  The HSC stuff was one of the keys forB > >CI connected storage, and that set included various models thatA > >transitioned to COlorado in the late 70s iirc, I recall Richie E > >Lary coming in to help with 11/60 microcode, and DMC/KMC microcodeuB > >around 76 from ColoSprings.  I admit my dates may be off a bit, > >maybe a year, but...u > E > The clustering project was separate from the CI-HSC implementation. D > TW did the CI-HSC on both TOPS-10 and TOPS-20.  The -10 was first,@ > then he did the -20.  Look at the dates on the monitor sourcesA > and you'll get an idea of when he did his work.  I didn't thinki@ > that the clustering project was dependent on the CI being done? > (but I could be very wrong...it was hard enough keeping track0 > of the -10 bit flows). >  > /BAH Barb,t> You jogged my memory too, I believe you are correct, there was= all the SMP stuff, clustering (Loosely Coupled Systems is thec9 term that comes to mind again), and all that going on in V< what my memory recalls as overlapping activities.  It really was a busy time! bobi    ) > Subtract a hundred and four for e-mail.u   ------------------------------  % Date: Thu, 07 Mar 2002 14:15:45 +0000 % From: Alan Greig <a.greig@virgin.net> < Subject: Re: Clustering beginnings - Clair Grant might know.8 Message-ID: <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com>  C On Thu, 07 Mar 2002 12:43:09 GMT, bad bob <sfmc68@bellatlantic.net>  wrote:     >Barb,? >You jogged my memory too, I believe you are correct, there wase> >all the SMP stuff, clustering (Loosely Coupled Systems is the: >term that comes to mind again), and all that going on in = >what my memory recalls as overlapping activities.  It reallyP >was a busy time!v  C And TOPS-10/TOPS-20 customers were getting lots of presentations onVE all the upcoming goodies and how this would all scream along with theP new "Jupiter" hardware.    Then it ended.  9 So am I right in summing up, consensus is that the CompaqoB (Digital/DEC) claim to have invented 'modern' clustering on VMS inE 1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS  around that time period?     -- Alan   ------------------------------  # Date: Wed, 06 Mar 2002 22:38:46 GMTT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)< Subject: Re: Clustering beginnings - Clair Grant might know.2 Message-ID: <WTwh8.659$fL6.10288@news.cpqcorp.net>   ..  D   The following is specific to OpenVMS VAX -- to VAX/VMS, as it was    then known -- clustering...n  H   CI780 support was first in VAX/VMS V3.1 (1983), HSC50 in V3.3 (1983), F   full (write-shared) clustering was in V4.0 (1984).  Between V3.3 andG   prior to V4.0, the VAX/VMS releases permitted storage configurations  D   with one-write-accessor-multiple-readers.   (At least one regular I   reader has been using this "ancient" storage configuration for reasons sG   not germaine to this discussion, and the reader encountered problems -I   with the behaviour of the Fibre Channel storage and mount verification.z   But I digress.)   B   Please follow the "history" links in the FAQ for related details   and historical information..  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: Thu, 07 Mar 2002 09:46:27 -0500u* From: Clair Grant <grant@evms.zko.dec.com>< Subject: Re: Clustering beginnings - Clair Grant might know.0 Message-ID: <3C877D43.52090610@evms.zko.dec.com>  P I believe that TOPS-20 did demo at DECUS multiple host operating systems sharingN files on the CI prior to the first public VMS demo. I think VMS was ready with straight HSC storage first.n  P Although it never became a product we had VMS and TOPS-20 systems on the same CIO in the lab. They knew how to keep out of each other's way when it came to filesaR and we could do DECnet over the CI between the operating systems. I always thoughtP this had great potential. Some of the things we used to experiment with in thoseM days on the CI have become reality in today's fibre channel SAN environments.e   Clairg       Hoff Hoffman wrote:t   > .. >FE >   The following is specific to OpenVMS VAX -- to VAX/VMS, as it wase >   then known -- clustering...  >eI >   CI780 support was first in VAX/VMS V3.1 (1983), HSC50 in V3.3 (1983),yH >   full (write-shared) clustering was in V4.0 (1984).  Between V3.3 andH >   prior to V4.0, the VAX/VMS releases permitted storage configurationsE >   with one-write-accessor-multiple-readers.   (At least one regular J >   reader has been using this "ancient" storage configuration for reasonsH >   not germaine to this discussion, and the reader encountered problemsK >   with the behaviour of the Fibre Channel storage and mount verification.- >   But I digress.)- >-D >   Please follow the "history" links in the FAQ for related details >   and historical information.- >-P >  ---------------------------- #include <rtfaq.h> -----------------------------L >       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comP >  --------------------------- pure personal opinion ---------------------------N >    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Thu, 07 Mar 2002 15:03:56 +0000.( From: Nic Clews <sendspamhere@127.0.0.1>< Subject: Re: Clustering beginnings - Clair Grant might know.) Message-ID: <3C87815C.49A97A73@127.0.0.1>o   Alan Greig wrote:e > ; > So am I right in summing up, consensus is that the Compaq D > (Digital/DEC) claim to have invented 'modern' clustering on VMS inG > 1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS  > around that time period?  F Just for interest, in the preface of "VAXCLUSTER PRINCIPLES" that I amB totally indebted to Steve Reece for the copy, Roy G. Davis writes:  H "In the early 1980s, Digital Equipment Corporation developed an entirelyF new technology for multiprocessor computing called clustering. DigitalH released the first implementation of this technology in late 1983. Known% as the VAXcluster configuration, ..."    March 25th 1993.   -- d( Regards, Nic Clews CSC Computer Sciences nclews at csc dot com    ------------------------------  ! Date: Thu, 07 Mar 02 12:01:58 GMT/ From: jmfbahciv@aol.com.< Subject: Re: Clustering beginnings - Clair Grant might know.+ Message-ID: <a67st8$k89$1@bob.news.rcn.net>n  0 In article <3C8760BE.85FA077C@bellatlantic.net>,+    bad bob <sfmc68@bellatlantic.net> wrote:e >- >i >jmfbahciv@aol.com wrote:t >> m2 >> In article <3C86B841.FB66242@bellatlantic.net>,. >>    bad bob <sfmc68@bellatlantic.net> wrote: >><<snip to save my eyes>>G >> >That is my recollection too.  The HSC stuff was one of the keys foriC >> >CI connected storage, and that set included various models thathB >> >transitioned to COlorado in the late 70s iirc, I recall RichieF >> >Lary coming in to help with 11/60 microcode, and DMC/KMC microcodeC >> >around 76 from ColoSprings.  I admit my dates may be off a bit,  >> >maybe a year, but... >> sF >> The clustering project was separate from the CI-HSC implementation.E >> TW did the CI-HSC on both TOPS-10 and TOPS-20.  The -10 was first,xA >> then he did the -20.  Look at the dates on the monitor sourceseB >> and you'll get an idea of when he did his work.  I didn't thinkA >> that the clustering project was dependent on the CI being done-@ >> (but I could be very wrong...it was hard enough keeping track >> of the -10 bit flows).r >> a >> /BAHe >Barb,? >You jogged my memory too, I believe you are correct, there wasi> >all the SMP stuff, clustering (Loosely Coupled Systems is the: >term that comes to mind again), and all that going on in 2 >what my memory recalls as overlapping activities.  ? Clustering was the -20 group's answer to our SMP.  You wouldn'tv> believe the "my thingie's better than your thingie" arguments.   > ..  It really  >was a busy time!   ? Yup.  It was one of the good side effects of canceling Jupiter.i; We got got funding for projects that had nothing to do with 	 hardware.1   /BAH  ' Subtract a hundred and four for e-mail.    ------------------------------   Date: 7 Mar 2002 10:20:05 -06004+ From: young_r@encompasserve.org (Rob Young)K< Subject: Re: Clustering beginnings - Clair Grant might know.3 Message-ID: <UT3MR+tKUPa1@eisner.encompasserve.org>s  T In article <3C87815C.49A97A73@127.0.0.1>, Nic Clews <sendspamhere@127.0.0.1> writes: > Alan Greig wrote:  >>  < >> So am I right in summing up, consensus is that the CompaqE >> (Digital/DEC) claim to have invented 'modern' clustering on VMS in H >> 1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS >> around that time period?  > H > Just for interest, in the preface of "VAXCLUSTER PRINCIPLES" that I amD > totally indebted to Steve Reece for the copy, Roy G. Davis writes: > J > "In the early 1980s, Digital Equipment Corporation developed an entirelyH > new technology for multiprocessor computing called clustering. DigitalJ > released the first implementation of this technology in late 1983. Known' > as the VAXcluster configuration, ..."t >   > 	This is very strange... back to that great work: "VAX OpenVMS. 	at 20", edited by Andy Goldstein, on page 60:    	VMS to OpenVMS:  Major Releases  ? 	VMS V3 April 1982  - 10,000 license (no mention of clustering)D  ( 	VMS V4 September 1984 - 40,000 licenses  9 	o  Support for new processor - VAX 8600, MV I/II, (v4.1)e 		VAXstation I/II (v4.1) 	o  VAXclustersr 	o  Connection manager 	o  Distributed lock manager% 	o  Distributed file system (F11BXQP)n 	o  Security enhancementst+ 	o  Command line editing and command recalli 	o  LAT server 	o  ACL implentedO! 	o  Cluster wide operator controlx 	o  Variable prompt stringsa   ---u  G 	So maybe it was internal then, but not out in the wild until SeptembernB 	1984.  Assuming Andy got the ship date right and none of this was# 	picked up on prior to publication.-   					Rob   ------------------------------  # Date: Thu, 07 Mar 2002 16:30:09 GMTw' From: bad bob <sfmc68@bellatlantic.net>O< Subject: Re: Clustering beginnings - Clair Grant might know.0 Message-ID: <3C8795F4.174ED987@bellatlantic.net>   Alan Greig wrote:  > E > On Thu, 07 Mar 2002 12:43:09 GMT, bad bob <sfmc68@bellatlantic.net>n > wrote: >  > >Barb,A > >You jogged my memory too, I believe you are correct, there wasa@ > >all the SMP stuff, clustering (Loosely Coupled Systems is the; > >term that comes to mind again), and all that going on ind? > >what my memory recalls as overlapping activities.  It reallyy > >was a busy time!e > E > And TOPS-10/TOPS-20 customers were getting lots of presentations oniG > all the upcoming goodies and how this would all scream along with thea > new "Jupiter" hardware.w >  > Then it ended. > ; > So am I right in summing up, consensus is that the Compaq-D > (Digital/DEC) claim to have invented 'modern' clustering on VMS inG > 1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS0 > around that time period?E I believe your statement of summary is factual, accurate and correct.  bobc  i > -- > Alan   ------------------------------  ! Date: Thu, 07 Mar 02 12:50:52 GMTe From: jmfbahciv@aol.com5< Subject: Re: Clustering beginnings - Clair Grant might know.+ Message-ID: <a67vou$p8j$5@bob.news.rcn.net>c  8 In article <73te8u0qj8skv08tmmiif9agura0jl30bc@4ax.com>,)    Alan Greig <a.greig@virgin.net> wrote:zD >On Thu, 07 Mar 2002 12:43:09 GMT, bad bob <sfmc68@bellatlantic.net> >wrote:a >  >u >>Barb,o@ >>You jogged my memory too, I believe you are correct, there was? >>all the SMP stuff, clustering (Loosely Coupled Systems is ther; >>term that comes to mind again), and all that going on in d> >>what my memory recalls as overlapping activities.  It really >>was a busy time! > D >And TOPS-10/TOPS-20 customers were getting lots of presentations onF >all the upcoming goodies and how this would all scream along with the >new "Jupiter" hardware.  B Nope.  That's marketing.  It didn't affect software development...< much.  We're talking about actual projects we were doing and selling.   >o >Then it ended.  >o: >So am I right in summing up, consensus is that the CompaqC >(Digital/DEC) claim to have invented 'modern' clustering on VMS inpF >1980 should be modified to on TOPS-10/TOPS-20 closely followed by VMS >around that time period?   < Not TOPS-10.  We did SMP instead.  TOPS-20 did clustering to2 achieve their brand of multi-CPU configurations.     /BAH         >- >-  ' Subtract a hundred and four for e-mail.-   ------------------------------  % Date: Thu, 07 Mar 2002 17:45:03 +0000.( From: Nic Clews <sendspamhere@127.0.0.1>< Subject: Re: Clustering beginnings - Clair Grant might know.( Message-ID: <3C87A71F.C5A7EDA@127.0.0.1>   Rob Young wrote: > G >         This is very strange... back to that great work: "VAX OpenVMS87 >         at 20", edited by Andy Goldstein, on page 60:n  ; Tremendous book. Has the PDF vanished from the website now?$  ) >         VMS to OpenVMS:  Major ReleasesF1 >         VMS V4 September 1984 - 40,000 licensesa ...i >         o  VAXclusters >         o  Connection manager>% >         o  Distributed lock managerr. >         o  Distributed file system (F11BXQP) ...8* >         o  Cluster wide operator control > P >         So maybe it was internal then, but not out in the wild until SeptemberK >         1984.  Assuming Andy got the ship date right and none of this was>, >         picked up on prior to publication.  H Putting Clair's response together with some other elements in VAXcluster> Principles, device communication (HSC) is vitally important inG clustering to Davis. The book goes into fascinating detail about device" communication / DSA.  E On the other hand, it could be a typo! September is late (ish) in the>H year by my estimation, but there's just a fingers width between 3 and 4.  3 I'm speculating wildly so I'd probably better quit!l -- s( Regards, Nic Clews CSC Computer Sciences nclews at csc dot comu   ------------------------------  % Date: Thu, 07 Mar 2002 10:02:43 +0100 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>v7 Subject: Re: DCL challenge of the day: dates comparisonc' Message-ID: <3C872CB3.19A2B84E@aaa.com>e  @ I'd rather see a picture of Peter opening the a box of chocolate posted  :-)    Didier Morandi wrote:iQ > Peter Weaver has sent me a very nice solution, do you all allow him to post it?  > He is shy... > :-)    ------------------------------  $ Date: Thu, 7 Mar 2002 11:31:14 -0500- From: "Peter Weaver" <peter.weaver@stelco.ca>v7 Subject: Re: DCL challenge of the day: dates comparisonm2 Message-ID: <wBMh8.27106$X2.282723@nnrp1.uunet.ca>  3 "Jan-Erik Sderholm" <aaa@aaa.com> wrote in messagea! news:3C872CB3.19A2B84E@aaa.com...>B > I'd rather see a picture of Peter opening the a box of chocolate
 > posted  :-)o   >...  E That would be my wife opening it, I'll be lucky if I get to smell thet" chocolates let alone taste them :)   --L Opinions are my own, and do not reflect the opinions of my employer, whoever that may be.H BTW: If anyone knows who that will after 1-APR-2002, please let me know.   ------------------------------   Date: 7 Mar 2002 10:21:11 -0800e+ From: pierre.bru@spotimage.fr (Bru, Pierre)l7 Subject: Re: DCL challenge of the day: dates comparison2< Message-ID: <1c0e37b1.0203071021.7edd0d0@posting.google.com>  Z Didier Morandi <Didier.Morandi@gmx.ch> wrote in message news:<3C865F5A.69B3F033@gmx.ch>... > yep, me again. > * > The smallest number of lines to do this:  & is this an Obfuscated DCL Contest ? :)  e& > enter starting date (JJMMAA): 160302& > enter ending date   (JJMMAA): 080402 >a > 1. turn dates to VMS format-  - does 160302 mean 16-MAR-2002 or 16-MAR-1902 ?r   % > 2. check that dates are valid datesj > 3. check that d2 >= d13 > 4. compute nr of days between d1 and d2, included.E > 5. set matrix of days per month to 1 where digits are within periode > 8 >                        12345678901234567890123456789019 > Example march_array = "0000000000000001111111111111111"a8 >         april_array = "111111110000000000000000000000" > - > Valid for all dates in the year, of course.dK > A Swiss chocolate box at Easter for the best solution, shipment included.e= > (Hein Van Den Heuvel is *not* admitted to the challenge :-)n >  > D.   Pierre.    ------------------------------   Date: 7 Mar 2002 06:53:23 -0600  From: briggs@encompasserve.org> Subject: RE: DCL Minute of the GMT+1 nite: nr of lines (again)3 Message-ID: <ofSYjpM3lhnC@eisner.encompasserve.org>   V In article <6MAR200222382156@gerg.tamu.edu>, carl@gerg.tamu.edu (Carl Perkins) writes:E > It should also be somewhat more efficient than the (not so bad) DCLtH > presented by Didier which copies each record to the null device, whichH > is not quite as fast as it sounds like it should be, and then parses aG > text file in DCL. SEARCH is probably faster that the Didier's DCL for,D > small files, but at some point the cost of doing the comparison toG > try to find the string is probably be higher than the copying to null  > and parsing a text file.  B If you search for "", the comparison overhead in SEARCH is minimalB since each record matches on the zero'th byte.  Use /MATCH=NAND if@ you want to make sure that you go down the "does not match" code path.   F I got a speed up from 2.6 CPU seconds down to .9 CPU seconds searchingD a 32 megabyte file for "" /MATCH=NAND versus "gorkasdfasdf" with the default /MATCH selection.6  G Wall clock speed up corresponded closely, going from 17.7 down to 16.1.e  F Then I boosted my RMS default multiblock count from the system defaultG of 16 to a process default of 127.  That took the direct I/O count down  from 4097 to 517.m  F The wall clock speed went from 16.1 seconds down to 11.7.  At a guess,B at that point we're pushing the disk transfer rate (RZ74 via HSJ40D through CI on a busy production system) and are unlikely to get much further improvement by tuning.  E Just for grins, I boosted the RMS default multibuffer count to 3, butaD SEARCH was apparently either already double-buffering or not readingC ahead because there was no performance gain.   I assume that SEARCHM+ does double buffered read-ahead by default.    	John Briggs   ------------------------------   Date: 7 Mar 2002 14:52:54 GMTa1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon): Subject: Re: emacs on VMSa, Message-ID: <a67us6$2hcd$1@info.cs.uofs.edu>  , In article <6MAR200222111603@gerg.tamu.edu>,*  carl@gerg.tamu.edu (Carl Perkins) writes:% |> david20@alpha1.mdx.ac.uk writes...gO |> }Looks like Staroffice will never happen. It appears Sun is no longer making 1 |> }it freely available except for Solaris see :-. |> } e5 |> }http://www.theregister.co.uk/content/4/24192.htmlt |> } r |> }Title :- |> }  6 |> }"Sun to charge for StarOffice (Linux and Windows)" |> } u |> }David Webb |> eH |> OpenOffice is still free. Practically the same thing, at this point - |> but they may now diverge.  B Like Netscape 6 and Mozilla??  And which one will improve the most and the quickest??   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   O   ------------------------------   Date: 7 Mar 2002 09:21:41 -0800-( From: bob@instantwhip.com (Bob Ceculski)5 Subject: Re: EV8 and McKinley analysis by Paul DeMonen= Message-ID: <d7791aa1.0203070921.5a7c7f68@posting.google.com>-  n "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote in message news:<VTue8.142$fL6.2592@news.cpqcorp.net>...G > You are missing one ingredient.  It also must be able to run Windows.k >   G WRONG!!!  That is the kind of thinking that destroyed DEC and now Q ...sC           All you need w/VMS is to supply us an easy way to develop B           for a windows client (graphics) and link that to VMS ...C           Decwindows was not the answer, and new products like com IE           and ericoms coming vt sessions over html browsers is a stepeE           step in the right direction, but windows is and always wills@           be a client ... you have to create an easy developmentH           environment in VMS to interface to windows easily!  That's it!   ------------------------------   Date: 7 Mar 2002 08:43:58 -0800f( From: fox@crisp.demon.co.uk (Paul D Fox)& Subject: Re: expensive procedure calls= Message-ID: <772053ad.0203070843.6343da89@posting.google.com>   h Bruce Hoult <bruce@hoult.org> wrote in message news:<bruce-4E3854.15405504032002@copper.ipg.tsnz.net>...I > As far as I know, the only dynamically scoped Lisp still in common use mI > is Emacs Lisp.  And pretty much everyone (including RMS) would like to   > change that. > J > Common Lisp does still have the ability to declare particular variables 9 > to be dynamically scoped (as does Perl, using 'local').n  A I have my own language, called crunch, which is part of the CRiSPgC editor suite, which supports dynamic scoping...and I wish it didnt!o  @ crunch evolved from a Lisp-like language (lots of (..) ) into anA ANSI/ISO-C like interpreted language. At the time, adding dynamict< symbol evaluation was about 3 lines of code - nice and easy.  E But its a maintainance nightmare - it produces 'surprising' behaviour-B (in 100,000 lines of code its maybe used 10 times, so one tends toA miss this fact). It also makes it very difficult to optimise. ThecA virtual machine architecture is not brilliant, but this mechanism:; means that it is difficult to produce a JVM style byte codec interpreter.  D There are ways around it. It can be a very poweful feature, but fromE my own experience, it wasnt 'thought out enough' to see the pitfalls,cC and I think the evaluation and addressing modes of modern languages'A (designed by people much better than myself), show its not a good1 thing.  F The primary reason for the languages existance was that Lisp-code is aB pain to read and debug, especially when mixing languages (CRiSP is7 written in C). Ones mind gets caught by those brackets.   > Interestingly, a language like C or C++ - probably has no moreD brackets than Lisp, its just that they are mixed ( '()' and '{}' andF '[]') so the brain doesnt see a huge forest of brackets, but can groupF the elements together more easily, making it easier to comprehend at a
 higher level.,  B At the end of the day, all the languages are equivalent - the main> difference is emphasis on different layout styles and personal8 expertise in writing and deciphering other peoples code.   ------------------------------  # Date: Thu, 07 Mar 2002 10:00:15 GMT 1 From: "Kari Keronen" <kari.keronen@radiolinja.fi> / Subject: Re: GEMBASE in GS Servers *** HELP ***i5 Message-ID: <PSGh8.11485$zZ5.235775@news.kpnqwest.fi>B  J We have two GS160. Both have two Galaxy partitions and then each partitionL in one GS160 is in cluster with its counterpart in other GS160. We generatedL SEND codes for each 4 nodes when the CPU configuration was what we generallyG thought it would usually be and send them to Ross (actually through ourcI application provider, because GEMBASE is part of that application). AftercL receiving working licenses there is no problem switching CPUs between Galaxy( instances, GEMBASE doesn't seem to mind.  J I also don't know what actual hardware information the licensing is based,J but it looks like there is not much hardware checking after the license is	 accepted.t   -Kari-  ; "Fabio Cardoso" <fabiopenvms@yahoo.com.br> wrote in messaged: news:20020306172711.10414.qmail@web20208.mail.yahoo.com...2 > Anyone here is using GEMBASE in GS-80 servers ??. > We are trying to know if this software works > fine in GS systems.l >)2 > We will upgrade our AS4100 to a GS-80. Following2 > Compaqs license model, it is one machine and we1 > will transfer all the licenses from one machinee! > to the two instances of the GS.a >a4 > But Ross dont have a model for licensing (upgrade)- > to the GS machines - following my reseller.o >s/ > They said : if you will upgrade to a ES-40 nom! > problem, just pay the transfer.e >r- > But to transfer the license to a GS machine * > (two instances) they dont know how to do. > the licensing (ie. in terms of $$$ cost) and/ > they dont know how the GEMBASE will genereate15 > the licenses codes because it generates it based in  > hardware configuration.-- > They suppose if I generate the license codeo1 > in one intance with 4 CPUs and 6 GB RAM, and IfC5 > i change this configuration using the Galaxy soft.,-# > the GEMBASE can stop to run (???)e >g >  > Any ideas please ????A >7 >2 >.	 > Regardsl >  > FC >h > =====. > ========================== > Fbio dos Santos Cardoso > OpenVMS System Manager > Rio de Janeiro - Brazili > fabiopenvms@yahoo.com.br > ========================== > 4 > __________________________________________________ > Do You Yahoo!?9 > Try FREE Yahoo! Mail - the world's greatest free email!i > http://mail.yahoo.com/   ------------------------------  % Date: Thu, 07 Mar 2002 10:41:36 -0800o' From: David Mathog <mathog@caltech.edu>9, Subject: HP Unix servers plunge below $1,000* Message-ID: <3C87B460.C26B56A@caltech.edu>  @ HP has followed Sun into the "affordable" end of the Unix market with sub $1000 models.  See:  ,   http://zdnet.com.com/2100-1103-853856.html   for more specs see:i  N   http://www.hp.com/products1/servers/rackoptimized/rp2400/specifications.html  B Sure, neither HP nor Sun's offerings are bleeding edge technology,L but buying one is cheap and easy and since (at least in Sun's case) they useL standard PC components they are also cheap to upgrade.  Compaq is nowhere inB this market - I tossed out a flyer from them on Monday targeted atB universities which had lots of Windows stuff and one page offeringL to sell DS10Ls at more than $6000 entry price.  And while it's true that theM DS10L is faster in the CPU department than the Sun (maybe not the HP though),  itsa: 33 Mhz IDE controller is only half the speed of the ATA/66F that Sun offers, and the new HPs appear to have integrated Ultra2 SCSI controllers!N For disk limited operations the HP will be vastly better, and the Sun somewhat better,-# than the much more expensive DS10L.0  ? The best bang for the buck is still Linux boxes on AMD or Intel C but the cheapie HP and Sun are close enough so that these companies-? can at least fend off some of the low end Linux competition and D offer a growth path into more lucrative hardware for those who triedF their low end systems and liked them. Compaq offers nothing comparableC and I guess figures (incorrectly) that if we don't buy Alphas we'llp be buying Q PCs instead.  C Well, maybe if HP does succeed in buying the Q and doesn't wipe outrH VMS after all we can maybe look  forward to a low cost VMS entry system?A I'm not holding my breath though - way too many maybe's and if's.s   Regards,   David Mathog mathog@caltech.edu   ------------------------------  % Date: Thu, 07 Mar 2002 15:56:59 +0529n3 From: "padmavati_1@vsnl.net" <padmavati_1@vsnl.net> D Subject: Import & Purchase Department - for DEPB  DFRC  SIL & DUTY F7 Message-ID: <20020307102333.B3D96D3A7@mmb4.vsnl.net.in>/  G <META HTTP-EQUIV="Content-Type" CONTENT="text/html;charset=iso-8859-1">r> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">: <HTML xmlns:v = "urn:schemas-microsoft-com:vml" xmlns:o = 4 "urn:schemas-microsoft-com:office:office" xmlns:w = Q "urn:schemas-microsoft-com:office:word"><HEAD><TITLE>Padmavati Enterprise</TITLE>$L <META content="text/html; charset=iso-8859-1" http-equiv=Content-Type><BASE 2 href="file://C:\WINDOWS.000\Desktop\Girish Bhai\">( <META content=Word.Document name=ProgId>0 <META content="Microsoft Word 9" name=Generator>7 <META content="Microsoft Word 9" name=Originator><LINK cG href="./Padmavati%20Enterprise_files/filelist.xml" rel=File-List><LINK  T href="./Padmavati%20Enterprise_files/editdata.mso" rel=Edit-Time-Data><!--[if !mso]>
 <STYLE>v\:* {a 	BEHAVIOR: url(#default#VML) }a o\:* { 	BEHAVIOR: url(#default#VML) }c w\:* { 	BEHAVIOR: url(#default#VML) }n shape {l 	BEHAVIOR: url(#default#VML) }s </STYLE>$ <![endif]--><!--[if gte mso 9]><xml>  <o:DocumentProperties>'   <o:Author>ICS-786</o:Author>!   <o:Template>Normal</o:Template>e&   <o:LastAuthor>ICS-786</o:LastAuthor>   <o:Revision>2</o:Revision>   <o:TotalTime>74</o:TotalTime>i-   <o:Created>2001-12-15T11:38:00Z</o:Created>i1   <o:LastSaved>2001-12-15T11:38:00Z</o:LastSaved>n   <o:Pages>1</o:Pages>   <o:Words>162</o:Words>"   <o:Characters>926</o:Characters>   <o:Company>WAPDC</o:Company>   <o:Lines>7</o:Lines>    <o:Paragraphs>1</o:Paragraphs>7   <o:CharactersWithSpaces>1137</o:CharactersWithSpaces>B   <o:Version>9.2720</o:Version>h  </o:DocumentProperties>* </xml><![endif]--><!--[if gte mso 9]><xml>  <w:WordDocument>1   <w:EnvelopeVis/>  </w:WordDocument> </xml><![endif]--> <STYLE>: <!--  /* Style Definitions */( p.MsoNormal, li.MsoNormal, div.MsoNormal 	{mso-style-parent:""; 	margin:0in; 	margin-bottom:.0001pt;y 	mso-pagination:widow-orphan;w 	font-size:12.0pt; 	font-family:"Times New Roman";G, 	mso-fareast-font-family:"Times New Roman";} h1 	{mso-style-next:Normal; 	margin:0in; 	margin-bottom:.0001pt;  	mso-pagination:widow-orphan;  	page-break-after:avoid; 	mso-outline-level:1;  	font-size:27.0pt; 	mso-bidi-font-size:12.0pt;e 	font-family:"Times New Roman";  	mso-font-kerning:0pt; 	font-weight:normal;}( h2 	{mso-style-next:Normal; 	margin:0in; 	margin-bottom:.0001pt;  	mso-pagination:widow-orphan;e 	page-break-after:avoid; 	mso-outline-level:2;  	font-size:12.0pt; 	font-family:"Times New Roman";n 	text-decoration:underline;i 	text-underline:single;} h3 	{mso-style-next:Normal; 	margin-top:12.0pt;m 	margin-right:0in; 	margin-bottom:3.0pt;  	margin-left:0in;n 	mso-pagination:widow-orphan;  	page-break-after:avoid; 	mso-outline-level:3;n 	font-size:13.0pt; 	font-family:Arial;} h4 	{mso-style-next:Normal; 	margin:0in; 	margin-bottom:.0001pt;i 	mso-pagination:widow-orphan;b 	page-break-after:avoid; 	mso-outline-level:4;  	font-size:12.0pt;  	font-family:"Times New Roman";} h5 	{mso-style-next:Normal; 	margin-top:0in; 	margin-right:0in; 	margin-bottom:0in;- 	margin-left:4.0in;a 	margin-bottom:.0001pt;2 	text-align:justify; 	mso-pagination:widow-orphan;? 	page-break-after:avoid; 	mso-outline-level:5;h 	mso-layout-grid-align:none; 	font-size:12.0pt;  	font-family:"Times New Roman";} h6 	{mso-style-next:Normal; 	margin:0in; 	margin-bottom:.0001pt;i 	text-align:justify; 	mso-pagination:widow-orphan;r 	page-break-after:avoid; 	mso-outline-level:6;  	mso-layout-grid-align:none; 	font-size:12.0pt;  	font-family:"Times New Roman";}. p.MsoBodyText, li.MsoBodyText, div.MsoBodyText
 	{margin:0in;  	margin-bottom:.0001pt;e 	text-align:center;m 	mso-pagination:widow-orphan;1 	font-size:16.0pt; 	mso-bidi-font-size:12.0pt;l 	font-family:"Times New Roman";a, 	mso-fareast-font-family:"Times New Roman";}1 p.MsoBodyText2, li.MsoBodyText2, div.MsoBodyText2o
 	{margin:0in;d 	margin-bottom:.0001pt;u 	text-align:center;  	mso-pagination:widow-orphan;m 	font-size:12.0pt; 	font-family:"Times New Roman";p, 	mso-fareast-font-family:"Times New Roman";} a:link, span.MsoHyperlinkr
 	{color:blue;  	text-decoration:underline;  	text-underline:single;}$ a:visited, span.MsoHyperlinkFollowed 	{color:purple;a 	text-decoration:underline;e 	text-underline:single;} @page Section1 	{size:8.5in 11.0in;! 	margin:1.0in .75in 1.0in 1.25in;i 	mso-header-margin:.5in; 	mso-footer-margin:.5in; 	mso-paper-source:0;}r div.Section1 	{page:Section1;}w -->i </STYLE> <!--[if gte mso 9]><xml>/  <o:shapedefaults v:ext="edit" spidmax="2050"/>m* </xml><![endif]--><!--[if gte mso 9]><xml>  <o:shapelayout v:ext="edit">n"   <o:idmap v:ext="edit" data="1"/>#  </o:shapelayout></xml><![endif]-->2< <META content="MSHTML 5.00.2614.3500" name=GENERATOR></HEAD>F <BODY bgColor=#ffffff lang=EN-US link=blue style="tab-interval: .5in" 
 vLink=purple>2 <DIV>&nbsp;</DIV>eA <H1 align=center class=Section1 style="TEXT-ALIGN: center"><SPAN ez style="BACKGROUND: black; COLOR: white; FONT-SIZE: 24pt; mso-bidi-font-size: 12.0pt; mso-highlight: black">Padmavati<SPAN ? style="mso-spacerun: yes">&nbsp; </SPAN>Enterprise</SPAN><SPAN  Y style="COLOR: white; FONT-SIZE: 24pt; mso-bidi-font-size: 12.0pt"><o:p></o:p></SPAN></H1>  <P class=MsoBodyText><B><SPAN P style="FONT-SIZE: 13pt; mso-bidi-font-size: 12.0pt">Duty Free Import Licences / 4 D.E.P.B / Drawback/DFRC / <o:p></o:p></SPAN></B></P> <P class=MsoBodyText><B><SPAN B style="FONT-SIZE: 13pt; mso-bidi-font-size: 12.0pt">Clearing<SPAN > style="mso-spacerun: yes">&nbsp; </SPAN>&amp; Forwarding<SPAN @ style="mso-spacerun: yes">&nbsp; </SPAN>Agents.</SPAN></B><SPAN J style="FONT-SIZE: 13pt; mso-bidi-font-size: 12.0pt"><o:p></o:p></SPAN></P>Y <P class=MsoNormal><!--[if gte vml 1]><v:line id="_x0000_s1026" style='position:absolute;hO  z-index:1' from="-18pt,5.5pt" to="531pt,5.5pt"/><![endif]--><![if !vml]><SPAN uh style="HEIGHT: 9px; LEFT: -25px; POSITION: relative; TOP: 6px; WIDTH: 735px; mso-ignore: vglayout"><IMG H height=3 src="cid:003901c18d39$43986e40$0a64a8c0@mtnl.net.in" width=735 b v:shapes="_x0000_s1026"></SPAN><![endif]><![if !supportEmptyParas]>&nbsp;<![endif]><o:p></o:p></P>6 <DIV class=MsoBodyText style="TEXT-ALIGN: left"><SPAN n style="BACKGROUND: black; COLOR: white; FONT-SIZE: 11pt; mso-bidi-font-size: 12.0pt; mso-highlight: black">1, P Humera Arcade, Ground floor, 2/4, Israel Mohalla, Samuel Street, Masjid Bunder, " Mumbai &#8211; 400009</SPAN><SPAN Z style="COLOR: white; FONT-SIZE: 11pt; mso-bidi-font-size: 12.0pt"><o:p></o:p></SPAN></DIV>6 <H6 class=MsoBodyText2 style="TEXT-ALIGN: left"><SPAN : style="FONT-SIZE: 10pt; mso-bidi-font-size: 12.0pt"><SPAN 3 style="mso-spacerun: yes">&nbsp;</SPAN></SPAN></H6> 6 <H6 class=MsoBodyText2 style="TEXT-ALIGN: left"><SPAN : style="FONT-SIZE: 10pt; mso-bidi-font-size: 12.0pt"><SPAN 2 style="mso-spacerun: yes"></SPAN>Tel.</SPAN><SPAN M style="FONT-SIZE: 10pt; mso-bidi-font-size: 12.0pt"> Office: 91-22-3710825 / u9 3781915 / 3702102. TeleFax. : 91-22-3710825/2080404<SPAN  I style="mso-spacerun: yes">&nbsp; </SPAN>Res.: 91-22-4142643/4144112<SPAN fE style="mso-spacerun: yes">&nbsp;&nbsp; </SPAN><o:p></o:p></SPAN></H6>rA <H6 align=left class=MsoBodyText2 style="TEXT-ALIGN: left"><SPAN I: style="FONT-SIZE: 10pt; mso-bidi-font-size: 12.0pt"><SPAN M style="mso-spacerun: yes">&nbsp;</SPAN>Mobile : 9820292949 / 9820049216<SPAN :5 style="mso-spacerun: yes">&nbsp; </SPAN></SPAN><SPAN a: style="FONT-SIZE: 11pt; mso-bidi-font-size: 12.0pt"><SPAN ? style="mso-spacerun: yes">&nbsp;&nbsp;</SPAN>Email</SPAN><SPAN u9 style="FONT-SIZE: 11pt; mso-bidi-font-size: 12.0pt">: <A aK href="mailto:girishblodaya@hotmail.com">girishblodaya@hotmail.com</A><SPAN p- style="mso-spacerun: yes">&nbsp; </SPAN>/ <A vR href="mailto:padmavati_1@vsnl.net">padmavati_1@vsnl.net</A><o:p></o:p></SPAN></H6> <P class=MsoNormal><B><SPAN style="FONT-SIZE: 11pt; mso-bidi-font-size: 12.0pt"><![if !supportEmptyParas]>&nbsp;<![endif]></SPAN></B><![if !supportEmptyParas]><![endif]></P>e <P class=MsoNormal><SPAN -M style="FONT-SIZE: 14pt; mso-bidi-font-size: 12.0pt">To,<o:p></o:p></SPAN></P>ws <H4 class=Section1>The Director (Import / Purchase Dept.<![if !supportEmptyParas]><![endif]>)&nbsp;<o:p></o:p></H4>.L <P class=MsoNormal><![if !supportEmptyParas]><![endif]>&nbsp;<o:p></o:p></P>& <P class=MsoNormal>Dear Sir/Madam,</P>p <P class=MsoNormal><![if !supportEmptyParas]><![endif]>&nbsp;<o:p></o:p></P><![if !supportEmptyParas]><![endif]>Q <P class=MsoNormal>&nbsp;Sub.&nbsp;<STRONG><U>Duty free pass book (DEPB) Special M Import Licence Duty Free Advance Licences</U><U><![if !supportEmptyParas]>&nbsp;<![endif]><o:p></o:p></U></STRONG></P><![if !supportEmptyParas]><![endif]> <P aZ class=MsoNormal><![if !supportEmptyParas]><![endif]><STRONG>&nbsp;</STRONG><o:p></o:p></P>- <P class=MsoNormal>Please contact us for:</P> L <P class=MsoNormal><![if !supportEmptyParas]><![endif]>&nbsp;<o:p></o:p></P>T <P class=MsoNormal><B><U><o:p></o:p></U></B></P><![if !supportEmptyParas]><![endif]>8 <P class=MsoNormal><STRONG><U>D E P B :</U></STRONG></P>L <P class=MsoNormal>&nbsp;<o:p></o:p></P><![if !supportEmptyParas]><![endif]>N <P class=MsoNormal>We offer DEPB licences at the best rates for DEBIT against Q your various imports. As per the current customs duty structure the <STRONG>Nett  O Savings is 4 to 5 % </STRONG>on the Nett payable duty if&nbsp; debited against h! duty free pass book (D.E.P.B)</P>- <P class=MsoNormal>&nbsp;</P>3L <P class=MsoNormal>&nbsp;<o:p></o:p></P><![if !supportEmptyParas]><![endif]>J <H2 class=Section1>Special Import Licence / Duty Free Advance Licence</H2> <H2 class=Section1>&nbsp;</H2>Q <DIV class=Section1>We also offer S.I.L at most reasonable premiums and also can sM offer specific licences for duty free imports (Against Advance Licence)</DIV>a  <DIV class=Section1>&nbsp;</DIV>@ <DIV class=Section1><STRONG><U>Other Services</U></STRONG></DIV>  <DIV class=Section1>&nbsp;</DIV>K <DIV class=Section1>We also provide <STRONG>Clearing Services </STRONG>for  e Imports Cargo at Nava Sheva, Mumbai &amp; Mumbai Air Cargo.</DIV><![if !supportEmptyParas]><![endif]>u <P class=MsoNormal>&nbsp;</P>w^ <P class=MsoNormal><STRONG><U>Liason Work</U></STRONG></P><![if !supportEmptyParas]><![endif]>A <P class=MsoNormal style="mso-layout-grid-align: none">&nbsp;</P>uK <P class=MsoNormal style="mso-layout-grid-align: none">Liason work for all g( D.G.F.T and Customs Related Matters.</P>A <P class=MsoNormal style="mso-layout-grid-align: none">&nbsp;</P> = <H2 class=Section1 style="mso-layout-grid-align: none"><SPAN eG style="TEXT-DECORATION: none; text-underline: none">Thanking you,<SPAN :vstyle="mso-tab-count: 5">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;  </SPAN></SPAN></H2>lL <H2 class=Section1 style="mso-layout-grid-align: none">Yours Sincerely,</H2>D <DIV class=Section1 style="mso-layout-grid-align: none">&nbsp;</DIV>+ <H6 class=Section1>Girish Lodaya&nbsp;</H6>u <H6 class=Section1><SPAN y.style="mso-tab-count: 5">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;  </SPAN></H6> <H6 class=Section1><SPAN r9 style="mso-tab-count: 5"></SPAN>&nbsp;</H6></BODY></HTML>c   ------------------------------  % Date: Thu, 07 Mar 2002 09:15:39 -0800,7 From: David Spencer <spencer@pageweavers.spaamfree.com>r$ Subject: Incremental BACKUP questionB Message-ID: <070320020915393529%spencer@pageweavers.spaamfree.com>  X I'm getting frustrated with an unintended behavior of BACKUP. My objective is to performX incremental backups on a small directory tree where one of the subdirectories gets a fewL files added each day. It's something like this (generalized for discussion):  
 disk:[top] disk:[top.html]d disk:[top.images]a disk:[top.pdf]  V Mostly the files in the heirarchy don't change. However, there are usually a few filesV added daily to the images directory. The incremental backup command I'm using is this:  D $ backup disk:[top...]*.*;/modified/since=yesterday saveset.bck/save  Q However, since files have been added to the images directory, images.dir has been S modified. Thus, I'm getting *all* of the files within disk:[top.images] each time Io run this incremental backup.  P Is there a simple way of modifying my backup commands in a generic way that willW only truly pick up the new and/or modified files rather than everything in the modified 
 subdirectory?d   Many thanks for any assistance,      Dave Spencer, PageWeavers  OpenVMS Alpha 7.2-1    ------------------------------   Date: 7 Mar 2002 01:40:33 -0800c- From: denis.fayaud@netspace.mc (Denis Fayaud)l) Subject: Re: interrupt state CPU activityn= Message-ID: <93820504.0203070140.16f2922a@posting.google.com>r   rdeininger@mindspring.com (Robert Deininger) wrote in message news:<rdeininger-0802022226400001@1cust160.tnt2.nashua.nh.da.uu.net>...nC > In article <pYX88.68$Jh4.575@news.cpqcorp.net>, "Fred Kleinsorge"d& > <kleinsorge@star.zko.dec.com> wrote: > M > >V7.3-1.  Note that release code names are used, since we can always change-L > >the release "version" at the last minute.  V7.3R is the "Remedial" stream > >for V7.3. > K > ... And the fixes in the remedial stream will wind up in ECO kits at someaI > unspecified future time.  Software maintenance customers who need a fixdF > before an ECO kit is ready can likely get early access through their$ > regular software service contacts.   Any news about this patch ?    ------------------------------    Date: 07 Mar 2002 08:40:27 +0100& From: "Ketil Malde" <ketil+@ii.uib.no> Subject: Re: Itanium troublesE. Message-ID: <egpu2g7rmc.fsf@sefirot.ii.uib.no>  0 peterm@CSUA.Berkeley.EDU (Peter Mardahl) writes:  O >> Is it correct to state that such a metric gives some insight on what happense" >> when you raise the clock rate ?  F > Congratulations, you've demonstrated very clearly the uselessness ofB > comparing different CPU families using MHz, or "work per clock".  ? I thought the point was to show how the various families scale,   i.e. view the results by family:   >> hp733		0.78  (itanium)  >> hp800		0.76  (itanium)i  0 and see how efficiency drops as clock increases.   -kzm --  H If I haven't seen further, it is by standing in the footprints of giants   ------------------------------    Date: 07 Mar 2002 08:52:31 +0100& From: "Ketil Malde" <ketil+@ii.uib.no> Subject: Re: Itanium troublesh. Message-ID: <egk7so7r28.fsf@sefirot.ii.uib.no>  1 mccalpin@gmp246.austin.ibm.com (McCalpin) writes:   , > There was even a suggestion that Intel was> > claiming (implicitly) a SPECint2000 number in the 750 range.  F Which is about what you get from a 2GHz P4 today, right?  Will that beC sufficient to capture any market share, I wonder, given that by the F time McKinley hits the shelves, P4s and AMDs will probably at least be+ around 3GHz (or with a 3000 'plus'-rating).o   -kzm -- rH If I haven't seen further, it is by standing in the footprints of giants      ------------------------------  % Date: Thu, 07 Mar 2002 09:17:35 +0100 E From: Jan C. =?iso-8859-1?Q?Vorbr=FCggen?= <jvorbrueggen@mediasec.de>u Subject: Re: Itanium troublesw+ Message-ID: <3C87221F.5C8C318A@mediasec.de>e   >> If the IA-64oN >> render farm needs more CPUs than an equivalent Alpha render farm, so be it. >              ^^^^^^^^^B >> They'll get the job done and be a good advertisement for IA-64.7 >                                    ^^^^^^^^^^^^^^^^^^ H > Obviously, my reading abilities are deteriorating quickly, these days.  H Not at all. The advertisement is in a well-known Hollywood company usingK IA64. The fact that they need more processors than the competition for the aL same task is unimportant to the customers this is addressing (CTOs and otherB decision makers), because for these that is not a figure of merit.   	Jan   ------------------------------  # Date: Thu, 07 Mar 2002 10:47:13 GMT * From: "Bill Todd" <billtodd@metrocast.net> Subject: Re: Itanium troublesPA Message-ID: <RyHh8.96672$nz4.8574936@bin4.nnrp.aus1.giganews.com>:  0 "Paul DeMone" <pdemone@igs.net> wrote in message! news:3C86FC3C.D996D8DD@igs.net...e >m > Bill Todd wrote: > >c > [...]m7 > > Intel originally hoped to ship McKinley at 1.4 GHz.  >T > Where do you get this from?s  J While I can't remember exactly where it came from, there's a good chance IJ got it from your realworldtech article of 3/14/01 - which now has acquiredB an addendum indicating that no such figure was actually specified.   - bill   ------------------------------  % Date: Thu, 07 Mar 2002 11:19:17 +0000t& From: Ken Green <Ken.Green@kgcc.co.uk> Subject: Re: Itanium troublesi* Message-ID: <3C874CB5.6E35963E@kgcc.co.uk>   Stuecheli wrote:   > Peter Boyle wrote: >:' > > On Tue, 5 Mar 2002, JF Mezei wrote:: > >#( > > > "Scandora, Anthony (35048)" wrote: > >a > >ML > > Ermm... remember that the Itanic debuted at the top of the SPEC-FP list. > >  >A- > From spec org,            base peak hw aval/- > hp 733 Itanium              494   577  6/01t- > hp 800 Itanium              526   610  6/01i0 > intel 1700 p4                  650   659  4/010 > intel 2200 p4                  766   777  1/02) > alpha 833 21264B        643   784  6/01>( > alpha 1000 21264C      776   960 10/01$ > ibm 1.3 GHz Power4 1098 1169 12/01/ > sun 900 ultra III            369    410  8/01g* > sun 1050 ultra III Cu   701    827  2/02 >eR > Itanic was not at the top of the list actually.  Beat out by intels own p4.  AndR > then total whooped by other in the next few months.  Mckinley better show higherP > results, but it has quite a way to go just to catch up with the other allready > released products. >. > Jeff  1 Hmmmmm are we looking at the same set of results?;  O HP rx4610 800MHz 1 CPU   SPECfp2000_base 701  SPECfp2000 701 test date May 2001nK Intel P4 1.7Ghz 1 CPU SPECfp_base2000 650 SPECfp2000 659 test date Aug 20011  Q When the Itanic scores were first posted I'm pretty sure they were the top of thei chart,Q but it didn't last long, they never do, HP's PA8700 was top of the SPECint tablesd for soM short a period of time they didn't even manage to get a press release out :-)e  O For McKinley's performance we will just have to wait until those people running # McKinley boxes are allowed to talk.g   Cheers   Keni   ------------------------------  % Date: Thu, 07 Mar 2002 13:30:50 +0100a* From: Alexis Cousein <al@brussels.sgi.com> Subject: Re: Itanium troublesl/ Message-ID: <3C875D7A.9070001@brussels.sgi.com>h   JF Mezei wrote:t > Peter Mardahl wrote: > G >>The moral of the story is that work/MHz, and MHz itself, is a USELESSeA >>measure of performance except within the SAME processor family,l >> > N > Since IA64 is , for all practical purposes, still vapourware, I am trying toL > figure out what the "beta" version running at 833mhz will come out as when+ > they do the process shrink with McKinley.t > P > If they build IA64 with the same process technology as the 8086-P4s, will they3 > be able to match the clock speed with the 8086s,    E I see you have little idea on how the design of a pipeline influencesvD the clock rate. It's not only a measure of "chip complexity", but ofD actual design decisions on how many stages to build in the pipeline.  A In other words, "MHz" tells you how many latches there are in thehE pipeline and what process technology you're using -- but little else.d  E Super-deep pipelines may give you an edge in the MHz pissing contest,B? but the actual latency for an operation will stay the same, and-@ you'll have more and more trouble actually filling the pipeline.  G  > If they build IA64 with the same process technology as the 8086-P4s, =  > will they be able to match the clock speed with the 8086s,s  C No. But then, that has little to do with actual performance, so why-B bother? They're two architectures built on different architecturalF decisions, that's all; one of the symptoms will be the MHz difference.   -- i? <these messages express my own views, not those of my employer> ) Alexis Cousein				Senior Systems Engineerl. SGI Belgium and Luxemburg		al@brussels.sgi.com   ------------------------------  * Date: Thu, 7 Mar 2002 13:40:47 +0000 (UTC)! From: Erik Corry <erik@arbat.com>  Subject: Re: Itanium troublesn, Message-ID: <a67qkv$ceu$1@news.net.uni-c.dk>  @ In comp.sys.intel JF Mezei <jfmezei.spamnot@videotron.ca> wrote:  ! > Not sure if this is valid, BUT:m   It isn't  P > If one compares the ratio of spec / clock, one might get an idea of efficiency
 > of chip.   Not really.i   --   Erik Corry erik@arbat.com";   Interviewer:  "Real programmers use cat as their editor."HN   Bill Joy:     "That's right! There you go! It is too much trouble to say ed,K                  because cat's smaller and only needs two pages of memory."l   ------------------------------  * Date: Thu, 7 Mar 2002 12:41:03 +0000 (UTC)/ From: mccalpin@gmp246.austin.ibm.com (McCalpin)o Subject: Re: Itanium troubles 1 Message-ID: <a67n4v$jg8$1@ausnews.austin.ibm.com>n  : In article <Pine.SOL.4.43.0203062107160.5823-100000@erik>,$ Erik Magnuson  <erik@cts.com> wrote:$ >On Thu, 7 Mar 2002, McCalpin wrote: >o. >>In article <3C868FC7.5C55EC3F@videotron.ca>,1 >>JF Mezei  <jfmezei.spamnot@videotron.ca> wrote:  >>Q >>>What puzzles me is that, although this sample is too small to draw any serious N >>>conclusions, both intel chips see a lower spec/mhz rating as the clock rate >>>icreases. >>E >>This is the "normal" direction -- memory is not getting faster veryAE >>fast, so the portion of the time waiting on memory increases as the 9 >>CPU speed increases.  This leads to lower "efficiency".g >> >1F >Ralph Grisham's book on CDC 6000 series assembly language mentions anJ >access time of 500 ns for the main core memory (in 1965). The best DRAM's, >are maybe 100 times faster after 37 years?   @ The fastest current uniprocessor systems have actual load-to-useA memory latency of slightly over 100 ns.    Multiprocessor systemsx? range from the 250 ns range (typical for 4-way systems) all then/ way up to 300-400 ns for 8-way and larger SMPs.e  2 From http://www.psych.usyd.edu.au/pdp-11/core.html? 	Core memory is also fairly slow, with the early systems havingg> 	a cycle time of up to 6s. By the early 1970's, the cycle was: 	down to 1.2s, and within a few years it halved to 600ns.  @ Other references agree with the 6 microsecond cycle time for theB late 1950's, (e.g., the AN/FSQ-7, designed by MIT and built by IBM4 for the SAGE project -- initially deployed in 1958).  D So we got a factor of 10 in the 20 years from the late 1950's to theD late 1970's, and then a factor a six or so in the 25 years since theF late 1970's.   Of course the next point is revelant to the discussion:    >It should be noted that the maxD >memory configuration is less than the on-chip cache of a few recent >microprocessors.1  C Latency for a fixed size memory or a fixed cost memory has improveduB much more rapidly.  For example the IBM p680 has 16 MB of L2 cacheF per cpu at an access time of about 20 ns, and the IBM p690 has ~1.5 MBB of L2 cache (on chip) at an access time of under 10 ns.  The core 6 memories from the 1950's-1970's were never this large.  E The original (1976) Cray-1 had up to 32 MB of RAM at a latency of 150 F ns using bipolar memory.  So we have about a 8x improvement in latencyB in 25 years for memories of the "10's of MB" size.  Still not very impressive.;  D The Cray T90, using an all-SRAM main memory, achieved 101 ns latencyB for memories up to 8 GB.  This 1996-era achievement is still aboutB the top of the list for a large SMP -- of course it was not cheap.   -- r9 John D. McCalpin, Ph.D.           mccalpin@austin.ibm.com:F Senior Technical Staff Member     IBM POWER Microprocessor Development-     "I am willing to make mistakes as long ast1      someone else is willing to learn from them."p   ------------------------------  % Date: Thu, 07 Mar 2002 09:59:35 -0600t# From: Stuecheli <jeffas@us.ibm.com>> Subject: Re: Itanium troublese* Message-ID: <3C878E67.A909167D@us.ibm.com>   Douglas Siebert wrote:  ' > Stuecheli <jeffas@us.ibm.com> writes:s >o >m ><J > You are looking at "HW available" dates, not test dates.  The P4 has hadJ > new compilers that bumped up its numbers, the 1.7GHz P4 at the time thatJ > results for Itanium were submitted was only 593.  Also, your numbers forG > Itanium are wrong, it had 701 peak for HP rx4610, test date May 2001,tG > HW availability June 2001.  The Alpha 833 was test date June 2001, HW;. > availability June 2001 (SW not until August) >tH > You should have been checking the Q2 2001 (and previous) SPEC results,G > not looking at all results and trying to work back.  The fact Itanium H > topped the list in SPECfp was discussed here in comp.arch when it cameI > out, though it was quite obvious that it wouldn't last there long (just!< > awaiting the Alpha shrink, faster P4s, POWER4, etc., etc.) >> > --J > Douglas Siebert                          dsiebert@excisethis.khamsin.net >tL > A good friend will help you move, a true friend will help you move a body.  P I guess I should have looked at the back level results.  For my memory I did notS think the Itanium was higher than anything.  Never really thought of the Itanium asd leadership SpecFp performace.s  S It is interesting that the 701 peak result for the hp rx4610 has been revised to be:T quite a bit less.  I assume this is due to not shipping that system?  So they postedT the result, yet could not really delever the result, thus had to remove it.  That isF a bit worse than just having higher results posted from other vendors.   Jeff   ------------------------------   Date: 7 Mar 2002 18:15:15 GMTV( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: Itanium troublese0 Message-ID: <a68anj$f4h$1@pegasus.csx.cam.ac.uk>  N In article <86elixkipd.fsf@mihalis.net>, Chris Morgan  <cm@mihalis.net> wrote:+ >nmm1@cus.cam.ac.uk (Nick Maclaren) writes:u >iE >> Do you have any evidence for the assumption that Intel are copyingmF >> AMD's design?  A lot of people seem to assume it, but I can find noC >> evidence for it - there is no obvious reason that it couldn't be-- >> an incompatible 64-bit extension to IA-32.M >oF >There was a substantial discussion on just this topic only a month orD >so ago. In fact there is some evidence and it's not hard to find if >you know where to look :P > - >http://www.google.com/search?q=intel+yamhill:  B I have done that more than once, but I never found anything that ID would call evidence - only speculation.  Note that I am SPECIFICALLY? referring to the issue of whether Intel is using AMD's ISA or a&" 64-bit IA-32 extension of its own.  8 If you have found some, I should appreciate a reference.     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679"   ------------------------------   Date: 7 Mar 2002 18:19:36 GMTi( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: Itanium troublesG0 Message-ID: <a68avo$fap$1@pegasus.csx.cam.ac.uk>  H In article <Pine.GSO.4.33.0203061626480.19910-100000@holyrood.ed.ac.uk>,. Peter Boyle  <pboyle@holyrood.ed.ac.uk> wrote: >bI >> >The remarks about the Montecito a week or so earlier were interesting&E >> >but bizarre - I have great difficulty in believing that the Alpha&D >> >people and technology can be got on board fast enough to deliverE >> >much of consequence (let alone EPIC+SMT) in 2004.  But it is well-H >> >known that executives confuse easily, even when they aren't spinning >> >a line.  >>E >> I thought that they said that Alpha technology would exist in some G >> chip by 2004.  That doesn't have to mean that work from Alpha has to F >> moved over by then, just that a chip which is due in 2004 will haveG >> technology which could currently be described as `Alpha Technology'.tH >> Hasn't there been a patent sharing agreement between Compaq and IntelI >> in place for sometime?  Since DEC sold their fab to Intel before being.I >> taken over by Compaq?  Am I remembering that sequence of events right?  >eB >I would guess if there is "alpha  technology" Intel is after (forJ >purpose other than ritual sacrifice), then on that time scale it may well2 >be the cache coherency protocol and interconnect,% >rather than a whole new core design.t >h9 >This is much closer to being a bolt-on addition, and has'A >helps IA-64 move into the market Intel has been proposing for it 7 >(and allows an even bigger premium to be charged :( ).m  E You could well be right.  We know that the main technology that Intelp: wanted was the wetware, but there is a bit more than that.  D To answer the earlier poster, unless the executive was talking aboutB introducing TWO new IA-64 designs in 2004, Montecito was stated asB going to include some Alpha technology.  I have not a clue what he" meant, nor even whether he had :-)     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679.   ------------------------------   Date: 7 Mar 2002 18:27:18 GMTr( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: Itanium troubles 0 Message-ID: <a68be6$fp4$1@pegasus.csx.cam.ac.uk>  , In article <3C868FC7.5C55EC3F@videotron.ca>,/ JF Mezei  <jfmezei.spamnot@videotron.ca> wrote:  >e  >Not sure if this is valid, BUT: >iO >If one compares the ratio of spec / clock, one might get an idea of efficiencye	 >of chip.t >e& >hp733		0.78   (peak specmark per mhz) >hp800		0.76 >intel 1700	0.39 >intel 2200	0.35 >alpha 833	0.94e >alpha 1000	0.96 >ibm 1300	0.90 >tM >Is it correct to state that such a metric gives some insight on what happens7  >when you raise the clock rate ?  B As other people have said, "no".  Think vector systems - which are2 very like an extreme EPIC in some of these senses.  O >So if the IA64 gets a just process shrink allowing comparable clock speeds, ite4 >will still be performing less than alpha and power. >sN >So if McKinley is to become competitive, it really needs to be more than just( >process shrink and greater clock speed.   Yes, that is correct.k  O >What puzzles me is that, although this sample is too small to draw any serious.L >conclusions, both intel chips see a lower spec/mhz rating as the clock rateA >icreases. Alpha managed to increase it between the 833 and 1000.k  B I'll bet that the chipset was different.  The ES45 has an upgraded) motherboard (over the ES40), for example.-     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679    ------------------------------   Date: 7 Mar 2002 18:31:30 GMTc( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: Itanium troublese0 Message-ID: <a68bm2$g8g$1@pegasus.csx.cam.ac.uk>  * In article <3C869F64.B351F8E8@us.ibm.com>,% Stuecheli  <jeffas@us.ibm.com> wrote:t >sP >Funny thing, Mckinley is not a process shrink.  It is more of a redesign.  FromR >my understanding talking with Intel folks it went like this.  Late in the projectO >Itanic was mapped to the new .18 process.  It was not really optimized for the:Q >.18 process.  The Itanic design was then shipped to hp for a redesign.  They areuQ >both in a .18 process.  This redesign  included the following changes, (from the>+ >02 ISSCC presentation, on the hp web site)s  @ Oh, yeah?  They were at best spinning you a line.  Unless things? were RADICALLY different from the roadmap that Intel originallyo> announced, that is precisely what did not happen.  I am pretty+ certain that the following is roughly true:e  ; The Itanic was designed by 500 engineers each with 2 years' > experience and the McKinley by 50 each with 20.  The former is? entirely Intel's pigeon, and the latter did not get passed overs< to Intel (from HP) until well after most validation had been
 completed.     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679t   ------------------------------   Date: 7 Mar 2002 18:56:50 GMTe( From: nmm1@cus.cam.ac.uk (Nick Maclaren) Subject: Re: Itanium troubles30 Message-ID: <a68d5i$htm$1@pegasus.csx.cam.ac.uk>  0 In article <a68bm2$g8g$1@pegasus.csx.cam.ac.uk>,) Nick Maclaren <nmm1@cus.cam.ac.uk> wrote:t > < >The Itanic was designed by 500 engineers each with 2 years'? >experience and the McKinley by 50 each with 20.  The former is @ >entirely Intel's pigeon, and the latter did not get passed over= >to Intel (from HP) until well after most validation had been- >completed.-  @ I forgot to say that the above remark about engineers was a jokeC made by HP to Intel, and not a statement of fact.  But it has a lot.@ of truth in it, from what I have heard from engineers who worked- on the Itanic and left (for various reasons).      Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679    ------------------------------  # Date: Thu, 07 Mar 2002 00:33:22 GMT02 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)0 Subject: Re: LIB$CVT_FROM_INTERNAL_TIME - year??2 Message-ID: <mzyh8.669$fL6.10271@news.cpqcorp.net>  _ In article <3C84A836.D9D7C678@herald.ox.ac.uk>, Peter Harding <harding@herald.ox.ac.uk> writes: I :Am I missing something glaringly obvious, or is it rather hard to figuret :out what year it is with this?s  1   Um, sys$numtim or sys$numutc might be easier.  t  J   With just a one-line question, the ability to suggest other options and I   alternatives is difficult at best.  sys$fao or lib$format_date_time or aH   lib$date_time or the gazillions of DTSS calls built into OpenVMS V7.3 G   and also into earlier versions with DECnet-Plus installed or the many0@   C time-related calls, for instance, might well be useful here.  J   When posting, the OpenVMS version and platform -- and some background onJ   the question -- can be useful.  One- or two-line questions are often andJ   surprisingly difficult questions to answer -- or to get an answer to the   question that was intended...e  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, 06 Mar 2002 23:35:33 GMTi2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)4 Subject: Re: Microsoft Curries Favor With Undergrads2 Message-ID: <9Jxh8.666$fL6.10333@news.cpqcorp.net>  b In article <3c7fbf31$1@news.si.com>, "Brian Tillman" <tillman_brian@notnoone.notnohow.com> writes:G :>   Microsoft on Thursday released software development tools aimed atoF :>   college-level computer science students, hoping it will produce aE :>   fresh crop of software programmers loyal to its Windows and .Net  :>   technology. :e: :At one time, Digital knew this.  Compaq just says, "Huh?"  D   Compaq has provided the commercial tool-set for OpenVMS under the B   educational program (www.openvmsedu.com) and under the hobbyist 4   program (www.montagar.com/hobbyist).  Free, too.    A   What would you have us do here -- what additional steps can we eA   provide here that we could afford to provide and that we could u=   profit (indirectly, obviously) from -- that is not already e   available?     Specific suggestions welcome.l  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: Thu, 07 Mar 2002 16:13:59 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>e4 Subject: Re: Microsoft Curries Favor With Undergrads; Message-ID: <01KF32CGUM8Y8ZM0W4@sysdev.deutsche-boerse.com>e  C >   What would you have us do here -- what additional steps can we hC >   provide here that we could afford to provide and that we could  ? >   profit (indirectly, obviously) from -- that is not already n >   available?  D Legal use of multi-user licenses with the base license from the new @ educational-license programme.  OK, one can combine it with the I multi-user license from DECcampus, IF one has DECcampus.  The problem is  I that DECcampus only makes financial sense if you have a relatively large gB number of machines.  As it stands now, an educational institution F STARTING OUT (is that a permissable thought?!) with VMS---which would D not buy enough machines to make DECcampus attractive right away---isF stuck in a situation which makes a VMS machine look like a single-userF billybox: a rather bad situation for someone wanting to demonstrate toA the bean counters the economics (TCO etc) of multi-user systems. <  H A lot of institutions (all I've worked at, in fact :-( ) used to be VMS E places but are now unix or even Windows (what always gets my goat is FI that the SAME GUYS who said "we need to move away from VMS because it is eG proprietary" (bullshit anyway) a few years later said "we need to move dI to Windows because it is industry-standard" (and, of course, at least as iG proprietary as VMS)).  Usually, there are a few VMS boxes left used by -E the die-hards.  Since these are of course a bit long in the tooth, ite@ creates the false impression (but I've seen Ph.D.'s who believe I it---what matters isn't the truth, but what the bean counters believe is e? the truth) that VMS machines are somehow intrinsically not verye? efficient, which leads to the die-hard greybeard happy with hissG MicroVAX II being made into a laughing stock since it is not as fast as E a brand new Pentium IV 2 GHz or whatever.  So much for the chance of cG convincing management to invest in new machines.  The same morons will lF probably compare the cost of that MicroVAX II (and adjust upwards for " inflation) to that new PC as well.  H Of course, another reason (and there is a chicken-and-egg problem here) H is that students who develop an interest in VMS will be told that it is F bad for their job prospects.  Again, truth doesn't matter; it is what I people are told.  How about an ad in a typical computer magazine showing eH a student thinking about his future---perhaps a typical college unix or ? Windows type thinking of a cubicle in some dotcom fly-by-night aI operation, and a VMS type thinking about a Real Job at a Real Company in sE the Real World which uses VMS because no other platform delivers the dG required performance (which is A LOT more than just raw speed---though l3 of course Alpha has been the leader there as well).c  G Hell, when universities were told "NT on ALPHA is the future", it sold oH some NT boxes.  Marketing in the educational regime CAN be done.  If it C can be done for falsehoods with NT, why not for the truth with VMS?B   ------------------------------  % Date: Thu, 07 Mar 2002 08:49:17 -0800i' From: David Mathog <mathog@caltech.edu> 4 Subject: Re: Microsoft Curries Favor With Undergrads+ Message-ID: <3C879A0D.3B7BEC8F@caltech.edu>m   Hoff Hoffman wrote:t > d > In article <3c7fbf31$1@news.si.com>, "Brian Tillman" <tillman_brian@notnoone.notnohow.com> writes:I > :>   Microsoft on Thursday released software development tools aimed atyH > :>   college-level computer science students, hoping it will produce aG > :>   fresh crop of software programmers loyal to its Windows and .Neto > :>   technology. > : < > :At one time, Digital knew this.  Compaq just says, "Huh?" > E >   Compaq has provided the commercial tool-set for OpenVMS under thesC >   educational program (www.openvmsedu.com) and under the hobbyist-4 >   program (www.montagar.com/hobbyist).  Free, too.  M Not the media.  Last time I looked there was no link from the EDU site to thebP montagar site.  Also no mention that the only way to get affordable media was to buylN it from montagar.  This wouldn't be a problem if the Q joined the modern world and hadtF software downloadable over the net.  (Ie, like your competitors do.)     > B >   What would you have us do here -- what additional steps can weB >   provide here that we could afford to provide and that we could0                          ^^^^^^^^^^^^^^^^^^^^^^^> >   profit (indirectly, obviously) from -- that is not already >   available?  I That's the key phrase.  Compaq must consider the CSLG/ESL a profit center2 because otherwiseSO it would cancel both and modify the new Education license to replace them - andn make itoP crystal clear that VMS may be used on academic sites with multiuser licenses for
 real work.N (Nonprofit - but real, ie research and/or administration.)  One way to do that would beK to make all Educational licenses unlimited user licenses.  The ESL would bed replaced byeM software download over the net. I suspect that one person could implement thel	 technicala< side of this in under a day, since all it would require are:  O  1.  put the ISO cd images for VMS boot disks on an FTP site and put the backupa/        sets for the software in the same place. 4  2.  change the user units on the education license.M  3.  Add one line to the new edu web site.  "The new EDU license replaces thes old ESL/CSLG andL        may be used for all of the same purposes that the old programs were."O  4.  Put a notice on the old CSLG/ESL site announcing its demise and linking toe the new one.8  5.  Rewrite the new EDU license to reflect the changes.  O Since I've (we've) been requesting these same changes for years and nothing hasj happened oneO can only conclude that the Q (at some level) is not in the least bit interestedl in this marketJ or in the damage done to the long term prospects for VMS which result from students having no exposure to it.a  N Moreover, this entire debate is ironically only an academic argument since VMS isG already deader than a doornail in academia.  In the end that's the besto	 indicatoraM of the success (sic) of Compaq's current academic programs.  Just for yucks -T when wasO the last time DEC or the Q sold a VMS system to a new user in academia - and bye new user, I meanK somebody with no prior experience with VMS.  1992 maybe???  I rest my case.g   Regards,   David Mathog mathog@caltech.edu   ------------------------------  # Date: Thu, 07 Mar 2002 16:43:54 GMT # From: "John Smith" <a@nonymous.com>P4 Subject: Re: Microsoft Curries Favor With Undergrads1 Message-ID: <eNMh8.36522$xG.26764@news2.bloor.is>t  F "Phillip Helbig" <HELBPHI@sysdev.deutsche-boerse.com> wrote in message5 news:01KF32CGUM8Y8ZM0W4@sysdev.deutsche-boerse.com...nD > >   What would you have us do here -- what additional steps can weD > >   provide here that we could afford to provide and that we could@ > >   profit (indirectly, obviously) from -- that is not already > >   available? >1E > Legal use of multi-user licenses with the base license from the new-A > educational-license programme.  OK, one can combine it with the3J > multi-user license from DECcampus, IF one has DECcampus.  The problem isJ > that DECcampus only makes financial sense if you have a relatively largeC > number of machines.  As it stands now, an educational institutionnG > STARTING OUT (is that a permissable thought?!) with VMS---which would F > not buy enough machines to make DECcampus attractive right away---isH > stuck in a situation which makes a VMS machine look like a single-userH > billybox: a rather bad situation for someone wanting to demonstrate toB > the bean counters the economics (TCO etc) of multi-user systems. >hI > A lot of institutions (all I've worked at, in fact :-( ) used to be VMSaF > places but are now unix or even Windows (what always gets my goat isJ > that the SAME GUYS who said "we need to move away from VMS because it isH > proprietary" (bullshit anyway) a few years later said "we need to moveJ > to Windows because it is industry-standard" (and, of course, at least asH > proprietary as VMS)).  Usually, there are a few VMS boxes left used byG > the die-hards.  Since these are of course a bit long in the tooth, iteA > creates the false impression (but I've seen Ph.D.'s who believetJ > it---what matters isn't the truth, but what the bean counters believe isA > the truth) that VMS machines are somehow intrinsically not veryvA > efficient, which leads to the die-hard greybeard happy with his I > MicroVAX II being made into a laughing stock since it is not as fast asoF > a brand new Pentium IV 2 GHz or whatever.  So much for the chance ofH > convincing management to invest in new machines.  The same morons willG > probably compare the cost of that MicroVAX II (and adjust upwards fore$ > inflation) to that new PC as well. > I > Of course, another reason (and there is a chicken-and-egg problem here)dI > is that students who develop an interest in VMS will be told that it iscG > bad for their job prospects.  Again, truth doesn't matter; it is what:J > people are told.  How about an ad in a typical computer magazine showingI > a student thinking about his future---perhaps a typical college unix orc@ > Windows type thinking of a cubicle in some dotcom fly-by-nightJ > operation, and a VMS type thinking about a Real Job at a Real Company inF > the Real World which uses VMS because no other platform delivers theH > required performance (which is A LOT more than just raw speed---though5 > of course Alpha has been the leader there as well).e >hH > Hell, when universities were told "NT on ALPHA is the future", it soldI > some NT boxes.  Marketing in the educational regime CAN be done.  If ittE > can be done for falsehoods with NT, why not for the truth with VMS?7 >w  I Because like other regime's, if you dissent you are put up against a wallt	 and shot.-  J OpenVMS is not something that can be sold/marketed one-on-one anymore - itB used to be possible. Unfortunately what is required is a long-termF commitment to marketing, dispelling the lies and misinformation in theB marketplace, and that hasn't been Digital/Compaq's strongest area.  I And even if you get beyond all that, there's the attitude that Windows oruK unix is 'good enough', especially when it comes to educational institutionsnK talking about the placement percentages of their graduates in jobs. WindowsrL & unix placements would be nearly 100% - what percentage of grads who wantedE to pursue VMS careers do you think would find jobs in the stream theye? desired (think a bit beyond your own institution for a moment)?i  I I don't know what Compaq's policy is about giving hardware to educationaldK institutions is, but if I were an administrator a university I'd be saying, I 'You want me to bring that now-obsolete Alpha in here and run an o/s that1A has been killed (Tru64) or an o/s whose future is unknown at bestaI (OVMS)...fine, give me everything for free - hardware, software, support,c# maintenance, and then we can talk.'t    K tongue-in-cheek  -- Part of the reason why there are so many Windows & unix H jobs available is that they are crap operating systems to administer andJ need hordes of worker bees to keep them running, whereas VMS tends not to.H Of course the answer to this is to create OVMS skills demand go up is byL making VMS break more often. For this we have to count on VMS Engineering to come through for us.  L Perverse isn't it, how a well engineered product is so good that nobody everA gives it a second thought -- including the company that sells it.e   ------------------------------  + Date: Thu, 07 Mar 2002 16:23:07 +0100 (MET)o9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>rA Subject: Re: Microsoft Curries Favor With Undergrads (correction)s; Message-ID: <01KF335O4TVC8ZM0W4@sysdev.deutsche-boerse.com>d  J > A lot of institutions (all I've worked at, in fact :-( ) used to be VMS            ^i           |s           |I           |)           |f*           +--- academic and/or educational   ------------------------------  % Date: Thu, 07 Mar 2002 09:33:17 +0100 , From: Georgi Kozinakov <kozinakov@mt.net.mk>+ Subject: MicroVAX 3100/30 to connect to ISP+( Message-ID: <3C8725CD.45F939C@mt.net.mk>  & --------------6D4263E2703A4438A2084FBE* Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bitR  7 My ISP provides only PPP for dial-up modem connections..A How should I set-up MicroVAX 3100/30 (OpenVMS 7.2.1, TCP/IP 5) tokG a phone line to dial ISP via PPP (I've reserved one static IP address)?0A How to setup a SMTP, POP3 to send/receive mails to/from internet?e- How about spam-ing - is VMS Mail aware of it? E I have also WASD http server on the mVAX running (stil testing it ...iG it's ok when browsing it locally), I want to put it on the internet andh0 if possible to enable telnet using sometimes ...F In the meantime I'm trying to set a PC/Win2000 server (192.168.0.1 andA the mentioned above static IP address when dialing to ISP, PPP noe problemn from Win2000srv) between theE internet and the private intranet (where the mVAX is now with privatecF IP 192.168.0.4). I want to set up NAT in RRAS service on the W2000srv.G What should I say in this case to mVAX (tcpip>set route 0.0.0.0 gateway- 192.168.0.1) ?    & --------------6D4263E2703A4438A2084FBE) Content-Type: text/html; charset=us-asciia Content-Transfer-Encoding: 7bitB   <HTML>L <FONT SIZE=-1>My ISP provides only PPP for dial-up modem connections.</FONT>F <BR><FONT SIZE=-1>How should I set-up MicroVAX 3100/30 (OpenVMS 7.2.1, TCP/IP 5) to</FONT>sL <BR><FONT SIZE=-1>a phone line to dial ISP via PPP (I've reserved one static IP address)?</FONT>sI <BR><FONT SIZE=-1>How to setup a SMTP, POP3 to send/receive mails to/fromh internet?</FONT>F <BR><FONT SIZE=-1>How about spam-ing - is VMS Mail aware of it?</FONT>H <BR><FONT SIZE=-1>I have also WASD http server on the mVAX running (stil testing it ...</FONT> H <BR><FONT SIZE=-1>it's ok when browsing it locally), I want to put it on the internet and</FONT>mI <BR><FONT SIZE=-1>if possible to enable telnet using sometimes ...</FONT> G <BR><FONT SIZE=-1>In the meantime I'm trying to set a PC/Win2000 server  (192.168.0.1 and</FONT>uG <BR><FONT SIZE=-1>the mentioned above static IP address when dialing tot ISP, PPP no problem</FONT>5 <BR><FONT SIZE=-1>from Win2000srv) between the</FONT>eF <BR><FONT SIZE=-1>internet and the private intranet (where the mVAX is now with private</FONT>iG <BR><FONT SIZE=-1>IP 192.168.0.4). I want to set up NAT in RRAS service- on the W2000srv.</FONT> I <BR><FONT SIZE=-1>What should I say in this case to mVAX (tcpip>set routep% 0.0.0.0 gateway 192.168.0.1) ?</FONT>d& <BR><FONT SIZE=-1></FONT>&nbsp;</HTML>  ( --------------6D4263E2703A4438A2084FBE--   ------------------------------  % Date: Thu, 07 Mar 2002 04:01:23 -0500r- From: JF Mezei <jfmezei.spamnot@videotron.ca>6/ Subject: Re: MicroVAX 3100/30 to connect to ISPf, Message-ID: <3C872C62.16ABDDA4@videotron.ca>  , Georgi Kozinakov wrote in illegal HTML text: > 9 > My ISP provides only PPP for dial-up modem connections.   G What TCPIP stack do you have on your VAX ?  Is this a hobbyist system ?   I When you dial in, you usually get a different IP everytime. So it becomesn quite hard to do this.  V The TCPIP product from Digital/Compaq/HP dooes not support PPP on VAX (only on Alpha).P But there are other stacks (notably from Process software) which may support it.  M With a dialup connection, VMS is not very well suited because VMS is designedu as a server not as a client.  6  I want to set up NAT in RRAS service on the W2000srv.I > What should I say in this case to mVAX (tcpip>set route 0.0.0.0 gateway  > 192.168.0.1) ?    F the gateway should be your windows server IP. You may also wany to add" /DEFAULT to the set route command.   ------------------------------   Date: 7 Mar 2002 09:22:35 GMTe) From: leslie@clio.rice.edu (Jerry Leslie)-/ Subject: Re: MicroVAX 3100/30 to connect to ISP-' Message-ID: <a67bgr$ftd$1@joe.rice.edu>n  - Georgi Kozinakov (kozinakov@mt.net.mk) wrote:z :d9 : My ISP provides only PPP for dial-up modem connections.aC : How should I set-up MicroVAX 3100/30 (OpenVMS 7.2.1, TCP/IP 5) touI : a phone line to dial ISP via PPP (I've reserved one static IP address)?e :gF The easiest way to connect VMS systems to the internet is a router/hub1 or router/switch device, which are available for:e     o 56K analog modem   o ISDN
   o cable/DSLi  @ 3COM makes a  router/hub unit that connects to the internet via + an analog modem (URL wrapped to two lines):   &    http://www.3com.com/products/en_US/:    detail.jsp?tab=prodspec&sku=3C886A-US&pathtype=purchaseK    3Com OfficeConnect 56K LAN Modem  (3C886A-US) - Product Specificationsc       "Product SpecificationsD  E      Number of computers for 56K V.90 service: Up to 4; up to 25 withb      additional hubs  2      Voice ports: 1 pass-through (RJ-11 connector).      Interface/WAN: 1 analog (RJ-11 connector)  C      Interfaces/local: 4 10BASE-T Ethernet ports (RJ-45 connectors)a  I      Operating system requirements/Ethernet ports: Computer with 10BASE-Ts5      network interface card and TCP/IP protocol suiteg  E      Minimum requirements/modem: Computer with 10BASE-T port, and 56Ke/      V.90 service from a local service providero  I      Minimum requirements/included software: Windows 2000/98/95/NT 4.0 PCw      with a CD-ROM drive  9      Protocols supported: TCP/IP, UDP/IP, PPP, DHCP, DNS,mE      PAP/CHAP/MS-CHAP, Intelligent NAT; PPTP, L2TP, NetBIOS filterings  =      Speed capability: 56 Kbps V.90 downloads, 33.6 Kbps V.34d      modem-to-modem throughput  ?      Management: Browser-based installation, configuration, ande,      management using a standard Web browser        Height: 4 cm (1.6 in)      Width: 22 cm (8.7 in)      Depth: 13.8 cm (5.5 in)      System Requirements  G    Compatible 56K V.90 service from a local service provider; a WindowsaC    2000/98/95/ NT 4.0 PC with a CD-ROM drive is required to use thel    included software  ?    To connect via Ethernet port: Computer with 10BASE-T networko+    interface card and TCP/IP protocol suitee      Package Contentsv        LAN modem        AC power cable and adaptern        RJ-11 to RJ-11 cable   "      RJ-45 to RJ-45 Ethernet cable  >      CD that includes custom Web browser installation software        Getting started guide      Please Note  ?      Optional Product: OfficeConnect Wall Mounting Kit(3C16765)d  G      Maximum speed in the U.S. and Canada limited to 53K for downloads, I      31K for uploads. Actual speeds will vary. Requires compatible analog )      phone line and server equipment...."i  0 IIRC, Multinet also makes similar routers/hubs.   @ Since you're in Macedonia, you should check with 3COM, Multinet,1 about any differences in your local phone system.V  4 --Jerry Leslie     (my opinions are strictly my own)   ------------------------------  % Date: Thu, 07 Mar 2002 10:53:50 +0100e, From: Georgi Kozinakov <kozinakov@mt.net.mk>/ Subject: Re: MicroVAX 3100/30 to connect to ISPn) Message-ID: <3C8738AE.91B416E2@mt.net.mk>u  & I use Digital (Compaq) TCPIP services.   JF Mezei wrote:   . > Georgi Kozinakov wrote in illegal HTML text: > > ; > > My ISP provides only PPP for dial-up modem connections.  > I > What TCPIP stack do you have on your VAX ?  Is this a hobbyist system ?  >2K > When you dial in, you usually get a different IP everytime. So it becomeso > quite hard to do this. >gX > The TCPIP product from Digital/Compaq/HP dooes not support PPP on VAX (only on Alpha).R > But there are other stacks (notably from Process software) which may support it. >eO > With a dialup connection, VMS is not very well suited because VMS is designed  > as a server not as a client. > 8 >  I want to set up NAT in RRAS service on the W2000srv.K > > What should I say in this case to mVAX (tcpip>set route 0.0.0.0 gateway  > > 192.168.0.1) ? > H > the gateway should be your windows server IP. You may also wany to add$ > /DEFAULT to the set route command.   ------------------------------  % Date: Thu, 07 Mar 2002 10:57:09 +0100 , From: Georgi Kozinakov <kozinakov@mt.net.mk>/ Subject: Re: MicroVAX 3100/30 to connect to ISPs) Message-ID: <3C873975.15ACE2F5@mt.net.mk>r  
 Thanks Jerry.e* What about the Win2000 server as a router? Georgi   Jerry Leslie wrote:g  / > Georgi Kozinakov (kozinakov@mt.net.mk) wrote:  > :n; > : My ISP provides only PPP for dial-up modem connections.iE > : How should I set-up MicroVAX 3100/30 (OpenVMS 7.2.1, TCP/IP 5) to K > : a phone line to dial ISP via PPP (I've reserved one static IP address)?e > :tH > The easiest way to connect VMS systems to the internet is a router/hub3 > or router/switch device, which are available for:  >  >   o 56K analog modem
 >   o ISDN >   o cable/DSLs >rA > 3COM makes a  router/hub unit that connects to the internet via - > an analog modem (URL wrapped to two lines):p >e( >    http://www.3com.com/products/en_US/< >    detail.jsp?tab=prodspec&sku=3C886A-US&pathtype=purchaseM >    3Com OfficeConnect 56K LAN Modem  (3C886A-US) - Product Specificationsa >o >     "Product Specifications  >cG >      Number of computers for 56K V.90 service: Up to 4; up to 25 withe >      additional hubs >e4 >      Voice ports: 1 pass-through (RJ-11 connector)0 >      Interface/WAN: 1 analog (RJ-11 connector) > E >      Interfaces/local: 4 10BASE-T Ethernet ports (RJ-45 connectors)a >TK >      Operating system requirements/Ethernet ports: Computer with 10BASE-T 7 >      network interface card and TCP/IP protocol suite  >dG >      Minimum requirements/modem: Computer with 10BASE-T port, and 56Kl1 >      V.90 service from a local service providera >pK >      Minimum requirements/included software: Windows 2000/98/95/NT 4.0 PCi >      with a CD-ROM drive >n; >      Protocols supported: TCP/IP, UDP/IP, PPP, DHCP, DNS,eG >      PAP/CHAP/MS-CHAP, Intelligent NAT; PPTP, L2TP, NetBIOS filteringe >o? >      Speed capability: 56 Kbps V.90 downloads, 33.6 Kbps V.34e  >      modem-to-modem throughput >tA >      Management: Browser-based installation, configuration, and . >      management using a standard Web browser >h >      Height: 4 cm (1.6 in) >      Width: 22 cm (8.7 in) >      Depth: 13.8 cm (5.5 in) >e >    System Requirements >dI >    Compatible 56K V.90 service from a local service provider; a WindowsfE >    2000/98/95/ NT 4.0 PC with a CD-ROM drive is required to use thek >    included software >mA >    To connect via Ethernet port: Computer with 10BASE-T networke- >    interface card and TCP/IP protocol suiter >c >    Package Contents? >  >      LAN modem >e! >      AC power cable and adaptere >e >      RJ-11 to RJ-11 cable- >-$ >      RJ-45 to RJ-45 Ethernet cable > @ >      CD that includes custom Web browser installation software >s >      Getting started guide >      Please Note >6A >      Optional Product: OfficeConnect Wall Mounting Kit(3C16765)o >eI >      Maximum speed in the U.S. and Canada limited to 53K for downloads,.K >      31K for uploads. Actual speeds will vary. Requires compatible analog + >      phone line and server equipment...."e >d1 > IIRC, Multinet also makes similar routers/hubs.) >oB > Since you're in Macedonia, you should check with 3COM, Multinet,3 > about any differences in your local phone system.  >c6 > --Jerry Leslie     (my opinions are strictly my own)   ------------------------------   Date: 7 Mar 2002 10:15:49 GMTh) From: leslie@clio.rice.edu (Jerry Leslie)i/ Subject: Re: MicroVAX 3100/30 to connect to ISPu' Message-ID: <a67ekl$k3q$1@joe.rice.edu>d  - Georgi Kozinakov (kozinakov@mt.net.mk) wrote:  : Thanks Jerry.h   You're welcome.c  , : What about the Win2000 server as a router?  0 I think JF's answer on that question is correct.   ------------------------------  + Date: Thu, 07 Mar 2002 12:55:14 +0100 (MET)s9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>c/ Subject: Re: MicroVAX 3100/30 to connect to ISPt; Message-ID: <01KF2VTXNQDY8ZM0W4@sysdev.deutsche-boerse.com>t  9 > My ISP provides only PPP for dial-up modem connections.dC > How should I set-up MicroVAX 3100/30 (OpenVMS 7.2.1, TCP/IP 5) toeI > a phone line to dial ISP via PPP (I've reserved one static IP address)?fC > How to setup a SMTP, POP3 to send/receive mails to/from internet?s/ > How about spam-ing - is VMS Mail aware of it?p  F VAXman has a page about "how to put a hobbyist system on the net".  I D don't see a link to it from http://www.tmesis.com, though.  Perhaps F Brian could post that link here.  (I knew there was a link to it from F another page, but that other page seems to be no longer available (it F was moving, and I bookmarked the old location)---does anyone have the - URL for Phil Ottewell's (hi Phil!) new site?)p  9 Where are you in Macedonia?  I'll be visiting there soon.s   ------------------------------  # Date: Thu, 07 Mar 2002 12:05:03 GMTS From: system@SendSpamHere.ORGv/ Subject: Re: MicroVAX 3100/30 to connect to ISP 0 Message-ID: <00A0A92E.DBEF6282@SendSpamHere.ORG>  S In article <a67bgr$ftd$1@joe.rice.edu>, leslie@clio.rice.edu (Jerry Leslie) writes:f. >Georgi Kozinakov (kozinakov@mt.net.mk) wrote: >:: >: My ISP provides only PPP for dial-up modem connections.D >: How should I set-up MicroVAX 3100/30 (OpenVMS 7.2.1, TCP/IP 5) toJ >: a phone line to dial ISP via PPP (I've reserved one static IP address)? >:G >The easiest way to connect VMS systems to the internet is a router/hub-2 >or router/switch device, which are available for: >i >  o 56K analog modemo	 >  o ISDNl >  o cable/DSL >tA >3COM makes a  router/hub unit that connects to the internet via  , >an analog modem (URL wrapped to two lines): >p' >   http://www.3com.com/products/en_US/n; >   detail.jsp?tab=prodspec&sku=3C886A-US&pathtype=purchaseuL >   3Com OfficeConnect 56K LAN Modem  (3C886A-US) - Product Specifications >e >    "Product Specifications >,F >     Number of computers for 56K V.90 service: Up to 4; up to 25 with >     additional hubsd > 3 >     Voice ports: 1 pass-through (RJ-11 connector)S/ >     Interface/WAN: 1 analog (RJ-11 connector)s > D >     Interfaces/local: 4 10BASE-T Ethernet ports (RJ-45 connectors) >aJ >     Operating system requirements/Ethernet ports: Computer with 10BASE-T6 >     network interface card and TCP/IP protocol suite >tF >     Minimum requirements/modem: Computer with 10BASE-T port, and 56K0 >     V.90 service from a local service provider > J >     Minimum requirements/included software: Windows 2000/98/95/NT 4.0 PC >     with a CD-ROM drivel >h: >     Protocols supported: TCP/IP, UDP/IP, PPP, DHCP, DNS,F >     PAP/CHAP/MS-CHAP, Intelligent NAT; PPTP, L2TP, NetBIOS filtering >t> >     Speed capability: 56 Kbps V.90 downloads, 33.6 Kbps V.34 >     modem-to-modem throughputw >d@ >     Management: Browser-based installation, configuration, and- >     management using a standard Web browserw >  >     Height: 4 cm (1.6 in)n >     Width: 22 cm (8.7 in)n >     Depth: 13.8 cm (5.5 in)  >i >   System Requirementsd > H >   Compatible 56K V.90 service from a local service provider; a WindowsD >   2000/98/95/ NT 4.0 PC with a CD-ROM drive is required to use the >   included software, >v@ >   To connect via Ethernet port: Computer with 10BASE-T network, >   interface card and TCP/IP protocol suite >  >   Package Contents >y >     LAN modemy >n  >     AC power cable and adapter >t >     RJ-11 to RJ-11 cable >t# >     RJ-45 to RJ-45 Ethernet cablee >e? >     CD that includes custom Web browser installation software  >w >     Getting started guides >     Please Notei >y@ >     Optional Product: OfficeConnect Wall Mounting Kit(3C16765) >eH >     Maximum speed in the U.S. and Canada limited to 53K for downloads,J >     31K for uploads. Actual speeds will vary. Requires compatible analog* >     phone line and server equipment...." >-1 >IIRC, Multinet also makes similar routers/hubs. 0 > A >Since you're in Macedonia, you should check with 3COM, Multinet, 2 >about any differences in your local phone system. >o5 >--Jerry Leslie     (my opinions are strictly my own)s  B For more information concerning the connection of OpenVMS machines4 to the internet see: http://www.tmesis.com/internet/   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM             -J   "And of course, I'm a genius, so people are naturally drawn to my fiery I   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbes    ------------------------------  % Date: Thu, 07 Mar 2002 11:26:46 +00002. From: Dougie Sharp <dougie.sharp@motorola.com>' Subject: MicroVAX Operating Temperaturey, Message-ID: <3C874E75.DD3BB3DA@motorola.com>  , This is a multi-part message in MIME format.& --------------E2F6108852D5CA2D69D5A648* Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bita  F Can anyone tell me the operating temperature thresholds for a MicroVAX 3600.   F Most MicroVAX systems seem to be around 10-40 C (50-104 F) but I can't3 find any specific information relating to the 3600.I      & --------------E2F6108852D5CA2D69D5A648- Content-Type: text/x-vcard; charset=us-ascii;e  name="dougie.sharp.vcf" Content-Transfer-Encoding: 7bit1* Content-Description: Card for Dougie Sharp  Content-Disposition: attachment;  filename="dougie.sharp.vcf"   begin:vcard  n:Sharp;Dougie tel;fax:+44 (0)1355 267200 tel;work:+44 (0)1355 35 5562 x-mozilla-html:FALSE5 org:Motorola Global Infrastructure Solutions;EMEA GIS  version:2.1t( email;internet:Dougie.Sharp@motorola.com8 title:SCO Operations Manager / EMEA VMS Technical Leaderp adr;quoted-printable:;;Kelvin Industrial Estate=0D=0AColvilles Road;East Kilbride;Glasgow;G75 0TG;United Kingdom fn:Dougie Sharpr	 end:vcardd  ( --------------E2F6108852D5CA2D69D5A648--   ------------------------------  % Date: Thu, 07 Mar 2002 12:26:47 +0000B( From: Nic Clews <sendspamhere@127.0.0.1>+ Subject: Re: MicroVAX Operating Temperature?) Message-ID: <3C875C87.ECD4CDD5@127.0.0.1>S   Dougie Sharp wrote:v > H > Can anyone tell me the operating temperature thresholds for a MicroVAX > 3600.a > H > Most MicroVAX systems seem to be around 10-40 C (50-104 F) but I can't5 > find any specific information relating to the 3600.-  G This information is in the systems and options, this link may hopefully/ take you there  H http://www.compaq.com/products/quickspecs/soc_archives/SOC_Archives.html  A I think it also has thermal output information as well to aid air2( conditioning requirements specification.  B From personal experience it is the disk drives that go first. Some@ systems have internal monitoring and will switch themselves off.   HTHF -- I( Regards, Nic Clews CSC Computer Sciences nclews at csc dot comi   ------------------------------  % Date: Thu, 07 Mar 2002 14:24:26 +0000 % From: Alan Greig <a.greig@virgin.net>T+ Subject: Re: MicroVAX Operating Temperaturei8 Message-ID: <alte8uc5uucmvc2e9bilih9c1t8l42189n@4ax.com>  F On Thu, 07 Mar 2002 12:26:47 +0000, Nic Clews <sendspamhere@127.0.0.1> wrote:   >8C >From personal experience it is the disk drives that go first. Some2A >systems have internal monitoring and will switch themselves off.u  F At a DEC User show London Wembley November 1989, I was running a standC for FACTS Software and we had a couple of VAXStation 3100s down for E the show. On Day 0 during load-in we could not boot either system andPD called field service. Diagnosis: Too Cold! There had been an extremeE cold-snap and we were near some main loading bays all with doors widen open.   A IIRC an error was produced during self-test (maybe disk related).tD After borrowing some heaters everything worked fine. Memory says the  temp was not far above freezing.   >p >HTH   -- Alan   ------------------------------  $ Date: Thu, 7 Mar 2002 10:07:00 -0500* From: WILLIAM WEBB <WWEBB1@email.usps.gov>+ Subject: RE: MicroVAX Operating Temperature - Message-ID: <0033000055534486000002L062*@MHS>e  ; =0AIf it was *almost* freezing you *were* operating outsideu of acceptable conditions:P  @ http://www.whiteice.com/~williamwebb/ChapD/DOC-D-3.html#tabled-4  8 Table D-4   System Operating and Nonoperating Conditions  H -----------------------------------------------------------------------=	 ---------t Operating Conditions  H -----------------------------------------------------------------------=	 ---------y6 Temperature range 10=B7C to 40=B7C (50=B7F to 104=B7F)   :^)h   WWWebb   -----Original Message-----/ From: Info-VAX-Request@Mvb.Saic.Com at INTERNETv& Sent: Thursday, March 07, 2002 9:51 AMB To: Webb, William W Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET+ Subject: RE: MicroVAX Operating Temperatureh    F On Thu, 07 Mar 2002 12:26:47 +0000, Nic Clews <sendspamhere@127.0.0.1> wrote:   >oC >From personal experience it is the disk drives that go first. SomeLA >systems have internal monitoring and will switch themselves off.   F At a DEC User show London Wembley November 1989, I was running a standC for FACTS Software and we had a couple of VAXStation 3100s down for E the show. On Day 0 during load-in we could not boot either system andcD called field service. Diagnosis: Too Cold! There had been an extremeE cold-snap and we were near some main loading bays all with doors wideo open.)  A IIRC an error was produced during self-test (maybe disk related).sD After borrowing some heaters everything worked fine. Memory says the  temp was not far above freezing.   >  >HTH   -- Alan=    ------------------------------  % Date: Thu, 07 Mar 2002 16:28:04 +0000 % From: Alan Greig <a.greig@virgin.net>r+ Subject: Re: MicroVAX Operating Temperaturen8 Message-ID: <g55f8uccbcoh5gb3aop38j3jn28r277pq8@4ax.com>  / On Thu, 7 Mar 2002 10:07:00 -0500, WILLIAM WEBB  <WWEBB1@email.usps.gov> wrote:   >A9 >If it was *almost* freezing you *were* operating outside  >of acceptable conditions:   Yes, I know that!3  A >http://www.whiteice.com/~williamwebb/ChapD/DOC-D-3.html#tabled-4o >r9 >Table D-4   System Operating and Nonoperating Conditionss > Q >--------------------------------------------------------------------------------  >Operating Conditionsi > Q >--------------------------------------------------------------------------------v/ >Temperature range 10C to 40C (50F to 104F)  >R >:^) >i >WWWebbd >  >-----Original Message-----e0 >From: Info-VAX-Request@Mvb.Saic.Com at INTERNET' >Sent: Thursday, March 07, 2002 9:51 AM C >To: Webb, William W Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET , >Subject: RE: MicroVAX Operating Temperature >J > G >On Thu, 07 Mar 2002 12:26:47 +0000, Nic Clews <sendspamhere@127.0.0.1>t >wrote:o >r >>D >>From personal experience it is the disk drives that go first. SomeB >>systems have internal monitoring and will switch themselves off. > G >At a DEC User show London Wembley November 1989, I was running a stand D >for FACTS Software and we had a couple of VAXStation 3100s down forF >the show. On Day 0 during load-in we could not boot either system andE >called field service. Diagnosis: Too Cold! There had been an extreme F >cold-snap and we were near some main loading bays all with doors wide >open. > B >IIRC an error was produced during self-test (maybe disk related).E >After borrowing some heaters everything worked fine. Memory says then! >temp was not far above freezing.: >: >> >>HTH1   -- Alan   ------------------------------  # Date: Wed, 06 Mar 2002 23:12:05 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)6 Subject: Re: Need help on Alpha:  Calling C from MACRO2 Message-ID: <9nxh8.664$fL6.10259@news.cpqcorp.net>  : In article <3C7DE7E1.1070207@tsoft-inc.com>, David Froble  <davef@tsoft-inc.com> writes:n ..P :The key thing to remember on Alpha is that you don't really have access to the : :machine, or at least some things like AP, SP, and such...  H   Alpha argument passing techniques and mechanisms differ markedly from H   those of VAX, and variable argument passing on Alpha usually involves F   the (hidden) assistance of the compiler -- if you want to implement D   this stuff, you really need to read and follow the details of the G   Alpha calling standard -- these details are available in the OpenVMS o   Calling Standard manual.  D   Most of this stuff is entirely transparent to even the programmer.     That said...    G   Anything that directly believes it understands the details of the VAXaD   call stack can make for an interesting port -- the fetching of theJ   arguments via mumble(AP) is obviously simulated by the Macro32 compiler,J   of course -- but the underlying use of registers, procedure descriptors,H   and (if needed) the homing of the argument list is something that you F   will likely need to know about if you are constructing the argument G   list and function call yourself -- if you're rolling your own version    of lib$callg or such.   F   As for part of the original discussion involving C, I would stronglyF   encourage porting from VAX C to Compaq C on VAX *first*, then movingF   to Compaq C on OpenVMS Alpha.  (The C RTL manual does -- did -- listD   this "va_start_1" macro as a feature specific to VAX -- this stuffG   uses the VAX C varargs.h mechanisms, and all of that should be moved wB   to stdarg.h -- this is all part of moving from VAX C to ANSI C.)  G   I'm guessing that there might be some rather "tolerant" settings for OG   the C compilation operations, and this might not be such a good idea. H   (Why?  Well, amid all the noise that the newer compilers can and oftenB   do generate around type mismatches can lurk some actual bugs...)    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: 7 Mar 2002 03:23:55 -0800 ) From: P.Young@unsw.EDU.AU (Patrick Young)o< Subject: Re: Netscape annoyance with OpenVMS CGI executables= Message-ID: <55f85d77.0203070323.26c31f0c@posting.google.com>   a JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3C86F943.B64D08FF@videotron.ca>...P   E > One way to debug this is to use a kermit script that sends the postnB > transaction with the right parameters and captures the response.  B Not being big on boxes not running OpenVMS (or Tru64), I made up aF file from the one you provided with the appropriate request, $ type(d); it from a DECterm and pasted into a telnet/port=80 and got:)  , HTTP/1.0 200 CGI script output data follows. MIME-version: 1.0e Server: OSU/3.3a Content-type: text/html # Date: Thu, 07 Mar 2002 10:56:21 GMTR   <HTML> <HEAD>K    <META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">:/    <META NAME="Author" CONTENT="Patrick Young">c,    <META NAME="GENERATOR" CONTENT="VMS EDT"> ....  L Which is correct (the web server inserting the Date: field within the outputK of my program). You have given me another idea though - the user might haveC& somehow set text/html to save to disk?  I But I'm not sure why, in that case, deleting the user profile in Netscape < and creating/using a new one does not remove such a setting.  H I've got a staff member at work with a laptop that has the problem whichH is being brought in tommorow. It is actually the first I've been able toK study -  others belonging to students - or reinstalled by PC support staff.tG I'll try to gather as much information as possible if I can't figure its out.   Many thanks.   ------------------------------  # Date: Thu, 07 Mar 2002 14:28:06 GMT-L From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")< Subject: Re: Netscape annoyance with OpenVMS CGI executables8 Message-ID: <00A0A929.B4BC1D30@SSRL04.SLAC.STANFORD.EDU>  i In article <55f85d77.0203070323.26c31f0c@posting.google.com>, P.Young@unsw.EDU.AU (Patrick Young) writes:   C >Not being big on boxes not running OpenVMS (or Tru64), I made up asG >file from the one you provided with the appropriate request, $ type(d)-< >it from a DECterm and pasted into a telnet/port=80 and got:  @ Kermit runs on VMS, although of course what you did worked fine.   >j- >HTTP/1.0 200 CGI script output data follows.e >MIME-version: 1.0 >Server: OSU/3.3au >Content-type: text/html$ >Date: Thu, 07 Mar 2002 10:56:21 GMT >c  N Shouldn't there be a blank line after the Content-type: header?  I don't thinkO it's properly recognized unless there are two linefeeds after it, although thisoD is the sort of niggly thing that _might_ vary with Netscape version.  O When you try the laptop of the user who's having a problem, also try it on some = external page that the problem has not been reported against.        ><HTML>i ><HEAD>oL >   <META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">0 >   <META NAME="Author" CONTENT="Patrick Young">- >   <META NAME="GENERATOR" CONTENT="VMS EDT">   C I'm waiting to see one of these that says "TECO"; mine say "TPU".  t  J Just curious; can the HTTP_EQUIV for the content type header actually takeJ effect?  The client can't read META tags if it doesn't know it has HTML to read.    -- Alan   O ===============================================================================-0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056 M  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210 O ===============================================================================7   ------------------------------  % Date: Thu, 07 Mar 2002 10:43:50 -0500s1 From: Michael Austin <maustin@firstdbasource.com> < Subject: Re: Netscape annoyance with OpenVMS CGI executables1 Message-ID: <3C878AB6.6DF44E5@firstdbasource.com>a   This is a SIMPLE fix.   + remove the .EXE or .COM from the href link.t   Patrick Young wrote: > > > I have a _REALLY_ annoying problem with certain instances of; > Netscape 4.* on PCs. This might have also been _reported_s) > with M$ Explorer but not actually seen.C > A > The problem is not that common (quite uncommon really), however > > the only solution seems to be to totally remove Netscape andH > re-install. I've never seen the problem with UNIX or OpenVMS browsers. > F > On the affected PCs: <form method="post" action="/htbin/my_cgi.exe">I > will cause the browser to prompt to save my_cgi.exe to disk even though H > the method is "post". In this particular application the web server is( > OSU 3.3a running on OpenVMS 7.3 Alpha. > G > I can't see any obvious settings in the browser that specify files ofmG > type .exe to be saved to disk. Not that file types are easily read ina5 > "Edit" "Preferences..." "Navigator" "Applications".u >  > Has anyone experienced this? >  > Many thanks.   --   Regards,  7 Michael Austin            Registered Linux User #261163,7 First DBA Source, Inc.    http://www.firstdbasource.comt Sr. Consultant 704-947-1089 (Office)i 704-236-4377 (Mobile)a   ------------------------------  # Date: Thu, 07 Mar 2002 16:56:29 GMTt# From: "John Smith" <a@nonymous.com>r% Subject: OpenVMS jobs available - FYIa0 Message-ID: <1ZMh8.36526$xG.1279@news2.bloor.is>   www.directemployers.com L This site is a co-op that takes you directly to the employer's own web site.  * I read about this site in Information Week5 http://www.informationweek.com/story/IWK20020222S0044C      L Enter VMS and OpenVMS as the search terms as separate queries as they return different result sets.  K VMS returns about 160 positions available, whereas OpenVMS returns about 21k! positions available, as of today.   L Seems like it's mostly the military contractors who are looking for people -L understandable - they use the best to produce the best, and failure isn't an option.i   ------------------------------   Date: 7 Mar 2002 04:08:35 -0800y# From: k_moehr@yahoo.de (Karl Moehr)c/ Subject: Re: Pathworks with Windows 2000 clientp= Message-ID: <72e3dad6.0203070408.3c9c9d83@posting.google.com>    Hi,   E Thanks for all inputs. As a fact I am running Pathworks ver. 6.00 und & must talk to my boss about an upgrage.   Regards, Karl   ------------------------------   Date: 7 Mar 2002 01:23:03 -0800u, From: tony.cheung@asiayeah.com (Tony Cheung)F Subject: Re: provide remote administration facility on a C/C++ process< Message-ID: <f9dc0a5a.0203070123.27146cd@posting.google.com>  q tony.cheung@asiayeah.com (Tony Cheung) wrote in message news:<f9dc0a5a.0202050851.508e1c14@posting.google.com>...t	 > Hi All,u > E > I'm building a C/C++ server and I would like to build another C/C++eH > client that could send commands and receive responses from the server.G > The server is a single-threaded process that basically looping around- > a select() call. > @ > I know I could implement my own protocol using TCP/IP sockets.E > However, I would like to take advantage if there's any higher level-G > classes/tools available. I would simply need a RPC-like interface, soID > that the client would issue a Remote Procedure Call on the server. > 2 > Should I go for RPC on OpenVMS or anything else? >  > Thank you very much. >  > Tony   Hi All,t  F Eventually, I have given up the idea of using RPC and simply implement@ my own simple protocol for sending request and reply. I am usingD asynchronous non-blocking berkely sockets. If anyone's interested, I would be able to share.    Tony Cheungi   ------------------------------  $ Date: Wed, 6 Mar 2002 10:44:02 -0800= From: "Lee Courtney" <lcourtney-removethisfilter-@mvista.com>nY Subject: Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of  Compaq Ac0. Message-ID: <u8fci8rplm88b@corp.supernews.com>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3C8596BB.802FDEC2@videotron.ca... > John Smith wrote:: > >pI > > Bet you that OpenVMS goes into maintenance-only mode within 6 months.  >uL > HP won't announce VMS's maintenance mode until the personal and commercialL > wintel stuff becomes profitable enough to sustain the company without VMS.  J You give HP too much credit. Their recent behavior in killing the MPE cashI cow does not bode well for VMS. MPE was profitable (albeit not generatingrK nearly the revenue of HP-UX), but that didn't stop HP from pulling the plug F w/o any migration plan in place for users. The sheer stupidity of this decision boggles the mind.   Lee Courtney   ------------------------------  % Date: Thu, 07 Mar 2002 09:23:56 +0000t% From: Alan Greig <a.greig@virgin.net>rY Subject: Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of Compaq Acqw8 Message-ID: <gvbe8uc1srjia6thmulcaovontddhno7gb@4ax.com>  F On Wed, 06 Mar 2002 22:38:15 GMT, "Bill Todd" <billtodd@metrocast.net> wrote:  M >The fact that a fair amount of the Tru64 business (certainly a great deal ofvJ >the business that otherwise felt 'Unix is Unix', regardless of the brand)L >was due to Alpha's superior performance didn't save Alpha - but you *could*C >attribute this to the utter incompetence of Compaq's management in C >understanding how superior Alpha was, and would remain, to Itanic.:  D Again from the financial analyst conference Curly (I believe it was)A referred briefly to the recent Compaq Alpha based "supercomputer"lD sales saying something like "We've won our fair share of these bids.F In fact you could say we've won more than our fair share". He actuallyA sounded as if he couldn't understand why people kept buying these B pesky Alpha chips when clearly they should just buy loads of Intel chips.  D Can you imagine for a split second Bill Gates standing up and saying8 "We sell more than our fair share of operating systems".     -- Alan   ------------------------------  # Date: Thu, 07 Mar 2002 14:41:46 GMTn# From: "John Smith" <a@nonymous.com>pY Subject: Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of Compaq Acq 0 Message-ID: <K_Kh8.35654$xG.9086@news2.bloor.is>  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3C86E928.8347D1F0@videotron.ca...E > > Where the F! do I find staff who WANT to work on OVMS these days?. >>K > I am willing to work VMS for food, but of course, would prefer a job that0+ > provides retraining in marketable skills.e    I See what I mean. And you are someone who takes an active interest in VMS.i  L Seems to me that being a SAP or data warehouse consultant is the best payingH thing to do these days. Have a smattering of unix skills and know Oracle  and/or DB2. Mostly o/s agnostic.   ------------------------------  # Date: Thu, 07 Mar 2002 14:44:49 GMT-# From: "John Smith" <a@nonymous.com>-Y Subject: Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of Compaq AcqO1 Message-ID: <B1Lh8.35683$xG.22742@news2.bloor.is>i  2 "Alan Greig" <a.greig@virgin.net> wrote in message2 news:gvbe8uc1srjia6thmulcaovontddhno7gb@4ax.com...H > On Wed, 06 Mar 2002 22:38:15 GMT, "Bill Todd" <billtodd@metrocast.net> > wrote: >2L > >The fact that a fair amount of the Tru64 business (certainly a great deal ofL > >the business that otherwise felt 'Unix is Unix', regardless of the brand)F > >was due to Alpha's superior performance didn't save Alpha - but you *could*eE > >attribute this to the utter incompetence of Compaq's management in E > >understanding how superior Alpha was, and would remain, to Itanic.  >cF > Again from the financial analyst conference Curly (I believe it was)C > referred briefly to the recent Compaq Alpha based "supercomputer"lF > sales saying something like "We've won our fair share of these bids.H > In fact you could say we've won more than our fair share". He actuallyC > sounded as if he couldn't understand why people kept buying these D > pesky Alpha chips when clearly they should just buy loads of Intel > chips. >2F > Can you imagine for a split second Bill Gates standing up and saying: > "We sell more than our fair share of operating systems". >p  L To paraphrase the movie title, 'Sleepless in Seattle' we get....'Clueless in Houston'  F (hope you aren't in Houston this week Alan. I know you are traveling.)   ------------------------------  % Date: Thu, 07 Mar 2002 16:22:24 +0000.% From: Alan Greig <a.greig@virgin.net>.Y Subject: Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of Compaq Acq-8 Message-ID: <4p4f8u8oeu09e9efoubko81vb6gu74vmcj@4ax.com>  F On Thu, 07 Mar 2002 14:44:49 GMT, "John Smith" <a@nonymous.com> wrote:   >-M >To paraphrase the movie title, 'Sleepless in Seattle' we get....'Clueless in	 >Houston'1 >FG >(hope you aren't in Houston this week Alan. I know you are traveling.)-  B I'm definitely in Dallas from Sunday and might just make a trip to/ Houston depending on how I can schedule things.V  E Houston is the HQ of the division of the multinational I work for ande= Dallas is where most of our US outsourced data processing andp networking is done.w     -- Alan   ------------------------------  % Date: Thu, 07 Mar 2002 08:22:38 -0800y' From: David Mathog <mathog@caltech.edu>wY Subject: Re: R.I.P. OpenVMS - ISS Recommends That H-P Holders Vote in Favor of Compaq Acq + Message-ID: <3C8793CE.EF861A3D@caltech.edu>l   John Smith wrote:   H > 4) By not focusing on the good side of their business to the extent itB > deserves, they create opportunities for every competitor to say,  P above and beyond what you say, they seem to have this bizarre delusion that theyP can remain #1 in technical computing without investing in the Alpha.  Talk about notrL knowing your marketplace!!!  Technical computing has no loyalty whatsoever - whateverP brings the most bang for the buck to the table wins, and since Compaq eliminated any O significant future Alpha development it's only a matter of time, and likely notc muchL time, before somebody else blows by them and  their HTC market share goes to zero.cP Ok, if by some miracle the Itanium ever delivers on its hype then maybe they can hangN onto some part of the market with Itanium boxes - but they'll have to share it withG Dell, IBM, and anybody else who wants to crank out similar machinery.  .  N Using somebody else's "standard" CPU is at best a mixed blessing.  In the case9 of the Itanium it may well turn out to be a double curse.i   Regards,   David Mathog mathog@caltech.edu   ------------------------------   Date: 7 Mar 2002 01:13:02 -0800a, From: tony.cheung@asiayeah.com (Tony Cheung). Subject: Re: Real-time scheduling with sockets= Message-ID: <f9dc0a5a.0203070113.520c9b66@posting.google.com>f  ^ "Richard Brodie" <R.Brodie@rl.ac.uk> wrote in message news:<a64rbm$15a4@newton.cc.rl.ac.uk>...; > "Tony Cheung" <tony.cheung@asiayeah.com> wrote in messagec9 > news:f9dc0a5a.0203060201.5adbc9f7@posting.google.com...r > H > > Is it the behavior of the OpenVMS? I am already using v7.3. Is thereI > > any way for a thread to yield control whenever it blocks on a socket?  > M > Last time I tried, it worked fine if you linked with upcalls enabled. Checke > that you are doing that.   Thanks! After linking withC /THREADS_ENABLE=(UPCALLS,MULTIPLE_KERNEL_THREADS). When a thread is C blocked on an I/O operation, a context switch happens. When the I/OnF operation is finished, the higher priority thread immediately preempts% (as I am using real-time scheduling).m  C I've read the POSIX THREAD guide for the OpenVMS and the guide only"? mentions about "UPCALLS" inside the Appendix. I really hope thesC OpenVMS documentations are easier to read, especially for someone's E more familar with Unix. OpenVMS is certainly very educational though.    Anyway, thanks again!i   Tony Cheungh   ------------------------------  # Date: Wed, 06 Mar 2002 22:22:29 GMTg2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: RWAST2 Message-ID: <FEwh8.658$fL6.10368@news.cpqcorp.net>  q In article <20020306171949.79865.qmail@web20209.mail.yahoo.com>, Fabio Cardoso <fabiopenvms@yahoo.com.br> writes:- :I think it is : ( :RESOURCE WAIT AST (ASSYNC. SYSTEM TRAP) :  :Hoff ? Are you there?    C   Nope.  Sorry, I don't read everything posted here and didn't see -   this particular posting.  :-)a     :Fabio2 :--- Malcolm Davies <malcolm.davies@cw.com> wrote:7 :> I work with ACMS on VMS systems, we often see server 5 :> processes in RWAST - I think this is READ WAIT ASTR3 :> or READ WRITE AST, usually get this if a processe4 :> dies,and another process is waiting on a reply to1 :> its AST messages. Trying to kill RWAST processr6 :> leaves it in suspended state and a reboot is needed3 :> to clear it. Similar to RWMBX (same as above butp :> Mailbox instead of AST)  G   This situation involves waiting for an AST entry to become available. D   I'd start with a check of the process quotas -- make sure that youD   have quotas appropriate for your use -- and I'd check for ECOs andG   upgrades to the product(s) involved.  ACMS and OpenVMS, in this case.rH   RWAST (resource wait for AST) turning into a hard hang after a $delprcI   can imply that there is a lost I/O (or lost AST) somewhere.  That said,<H   if the process is sufficiently wedged, things can get very interesting   in the process run-down.   :> Mal. 4 :>   "Richard Wolff" <rwolff@selkirk.bc.ca> wrote in
 :> message6 :> news:20020306T085717Z_5E1800050000@selkirk.bc.ca...7 :>   I can't find where in the manuals is a descriptiont- :> of the various states a process can be in. - :>   I have a process that is in RWAST state.u    C   Most of the detailed discussions of the process scheduling statesgE   are in the IDSM (Internals and Data Structures Manual), and not in mE   the OpenVMS documentation set.  If you are asking these questions, i3   well, it is time to acquire and to skim the IDSM.o    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: Thu, 07 Mar 2002 10:28:35 GMTg1 From: CSABA  HARANGOZO   <csabah@zipworld.com.au> * Subject: Re: Standalone backup question...3 Message-ID: <nhHh8.92$k6.7486@nasal.pacific.net.au>g  8 Michael Worsley <michael.worsley@praxis-cs.co.uk> wrote:? > Mike Rechtman <michael.rechtman@digital.com> wrote in messagea' > news:3C86132D.11938C7B@digital.com...a >> Michael Worsley wrote:t >> >= >> > Warren Spencer <wspencer@ap.nospam.org> wrote in message 6 >> > news:91C880267warrenspencer1977@209.249.90.100...A >> > > michael.worsley@praxis-cs.co.uk (Michael Worsley) wrote int0 >> > > <a62t3f$b5fsd$1@ID-65806.news.dfncis.de>: >> > > >> > <snip the stuff I wrote>l >> >L >> > > It's not clear what you want to have happen here.  The first tape hasM >> > > filled up, it tells you (%BACKUP-I-READYWRITE), rewinds the first tapehL >> > > (whirrs quietly to itself), spits it out ("Operate lever" light comes > on),) >> > > and you get to load the next tape.n >> >I >> >.... Typing Y in response to the READYWRITE prompt is an excersize ino >> > futility :-)U >>F >> Have you tried entering "YES" (no quotes) instead of just "Y"? IIRCJ >> there were some console-level things that did not accept abbreviations.  H > Y, y, YES and yes all tried, and all with precisely the same effect...  > 	Back in time, way back, I used to use the /INIT qualifier for! 	standalone image backups, like :a  4 	$BACKUP/IMAGE/REW/INIT/LOG  disk:  tape:saveset.bck  < 	It worked well... One could also use the /BLOCK=... switch, 	depends. Worth a try. 	Hope this helps...i 						Cheers,    Csaba  I    ----------------------------------------------------------------------oE    * Csaba I. Harangozo     |    'To err is human', said the hedgehog,E    * csabah@zipworld.com.au |           as he dismounted a wirebrush.=I    ----------------------------------------------------------------------a;    EARTH::AUSTRALIA:[SYDNEY]HARANGOZO.CSABA;1, delete? [N]:a   ------------------------------  # Date: Wed, 06 Mar 2002 22:53:54 GMTC2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)0 Subject: Re: Steps for installation OVMS 6.2 1H32 Message-ID: <66xh8.663$fL6.10356@news.cpqcorp.net>  c In article <a5krrj$49m$1@news2.ipartners.pl>, "Aleksander Koodziejczyk" <olekk@kki.net.pl> writes:TE :I'll be install OVMS 6.2, I never to do it before, it is difficult ?l% :Can you tell me any steps, to do it.   -   OpenVMS Alpha V6.2-1H3 is ancient history. n  .   Please use current OpenVMS Alpha software.    *   Please use OpenVMS Alpha V7.2-2 or V7.3.  G   If you are forced to install OpenVMS Alpha V6.2-1H3 onto the system, pH   boot the V6.2 distribution disk and follow the installation procedure.8   Then use VMSINSTAL to apply the V6.2-1H3 distribution.  F   If you are installing on an Alpha system which requires the V6.2-1H3G   release (the FAQ has pointers to the minimum version requirements for H   various platforms), your first and best approach is to install V7.2-2 G   other more current version.   If you must use ancient software, then nB   you must either use a "remastered" distribution kit for V6.2-1H3F   (QA-MT1AG-H8), or you must install V6.2 and then VMSINSTAL V6.2-1H3 D   on another Alpha system and transfer the disk over to this system.  E   The V6.2-1H3 release is sufficiently ancient -- any appearance of a-D   theme here is quite deliberate :-) -- that it uses different toolsF   and different installation mechanisms, and will thus differ from theC   available documentation on the website.  In particular, you must sA   boot and use PCSI to install the base release -- as is the casesC   with V6.1 and later -- but you must use VMSINSTAL to install the a@   V6.2-1H3 update and NOT (as is done on V7.1-2 and later) PCSI.  G   Current documentation for current releases is on the OpenVMS website.tF   (OpenVMS releases as ancient as V6.2-1H3 predate the availability of   HTML-format documentation.),  G   If this is a hobbyist system, please see the hobbyist pointers in theo   OpenVMS FAQ.    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: Thu, 7 Mar 2002 16:38:30 +01000 From: "Paul Janssen" <paul.janssen@intrinsic.be>
 Subject: SWCC-6 Message-ID: <3c8788a0$0$33504$ba620e4c@news.skynet.be>  F Does anyone have any experience with SWCC and the the KZPAC/KZPSC RAID controllers?  H It seems that the version posted on the public compaq site does not work with these controllers.oK The message at the client side reports a "No agent running on the specified5 host".H As this message comes up very quickly and any entry there gives the same result, I doubt if it is what it says it is.e  J By the way I tried the classing things, like pinging, checking in the help files, internet,etc.   Thanks for any clue...   -- Paul Janssen paul.janssen@intrinsic.bew Intrinsic Consulting nv  Houwaartstraat 58I 3210 Linden  Tel +32 (0)16 62.27.43 Fax +32 (0)16 62.23.10 GSM +32 (0)475 201.201   ------------------------------  % Date: Thu, 07 Mar 2002 16:47:18 +0100h9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>r Subject: Re: SWCCI' Message-ID: <3C878B86.27AE5ED6@aaa.com>   9 SWCC works with the HS* line of external SCSI controlers.l0 RA200, HSZ22/40/50/70/80, HSD or HSJ controlers.  8 I don't think that it work the the internal KZPA* series of controlers.   Jan-Erik Sderholm.o     Paul Janssen wrote:  > H > Does anyone have any experience with SWCC and the the KZPAC/KZPSC RAID > controllers? > J > It seems that the version posted on the public compaq site does not work > with these controllers.nM > The message at the client side reports a "No agent running on the specifiedf > host".J > As this message comes up very quickly and any entry there gives the same > result, I doubt if it is > what it says it is.. > L > By the way I tried the classing things, like pinging, checking in the help > files, internet,etc. >  > Thanks for any clue... >  > -- > Paul Janssen > paul.janssen@intrinsic.beo > Intrinsic Consulting nvl > Houwaartstraat 58a
 > 3210 Lindenu > Tel +32 (0)16 62.27.43 > Fax +32 (0)16 62.23.10 > GSM +32 (0)475 201.201   ------------------------------   Date: 7 Mar 2002 03:04:35 -0800a% From: spi@equicon.de (Joerg Spilling)e+ Subject: Re: TCPIP$FTP hangs in MUTEX statet= Message-ID: <dfc30fc1.0203070304.145cae77@posting.google.com>o   THANKS ALL!e  C I think, that's it. I have also had idea that this is a quota/limit-E problem and set the process parameters for the TCPIP$FTP account. ButrA I had no chance to try it (the cluster stands in Warszawa and I'm ; currently not connected to it). So I will try it next week.    Cheers Joerg8  j spi@equicon.de (Joerg Spilling) wrote in message news:<dfc30fc1.0203060316.73c5cd7d@posting.google.com>...@ > Hi folks, I need some ideas or help for the following problem: > D > I'm running OpenVMS 7.3 on as DS20e, TCP/IP V5.1 ECO 3 and I think7 > that I have most of the recent 7.3 patches installed.r > F > In my application, I have many PC's connecting to the server via FTPH > and download some files. After some times, the FTP process goes in theG > MUTEX state (all what I can do is reboot?). Starting and stopping FTPtE > don't work (because of MUTEX state). I'm currently not able to findm9 > out on what MUTEX the process hangs. Any ideas or help?k > F > Perhaps anyone can tell me, which newer patches are really required? >  > Cheers > Jrg >  > " > Here are some hopefully outputs: >  > ! > US2|...[EQSPED.SRC] show systemeB > OpenVMS V7.3  on node VENUS2   6-MAR-2002 11:51:23.95  Uptime  0
 > 02:04:44H >   Pid    Process Name    State  Pri      I/O       CPU       Page flts >  PagesH > 21200401 SWAPPER         HIB     16        0   0 00:00:04.03         0 >      0
 > <<snip>>H > 21200416 SMHANDLER       HIB      6       52   0 00:00:00.02       154 >    150H > 21200418 NETACP          HIB      9       47   0 00:00:00.04        53 >     91H > 21200419 TCPIP$INETACP   HIB     10      116   0 00:00:00.02       177 >    125H > 2120041A EVL             HIB      6       60   0 00:00:00.00       121 >    105  NdH > 2120041B REMACP          HIB      9       14   0 00:00:00.00        35 >     21H > 2120041C TCPIP$FTP_1     MUTEX   10    10072   0 00:00:00.58       685 >    303  NeH > 2120041D TCPIP$PWIP_ACP  HIB      9       57   0 00:00:00.05       134 >    156H > 2120041E SMTP_VENUS2_01  HIB      4      658   0 00:00:00.11       432 >    213H > 2120041F TNT_SERVER      HIB      6      319   0 00:00:00.18       641 >    944H > 21200421 TNT12120041F    LEF      2       68   0 00:00:00.04       674 >    227  SMH > 21200422 DCE$DCED        HIB      8      276   0 00:00:00.08       640 >    590H > 21200424 LATACP          HIB     14        8   0 00:00:00.01        74 >     57H > 21200427 TOMCAT_8007     HIB      6     2596   0 00:00:02.80     10570
 >   2436 MH > 2120042B REGISTRY_SERVER HIB      8      101   0 00:00:00.07       225
 >    277 MH > 2120043B M2DBIO$FTPWATCH LEF      5    17773   0 00:00:04.50     16465 >     77H > 2120043C Watch_VENUS2    LEF      6     5995   0 00:00:01.42     10291 >    103H > 21200442 TCPIP$FTPC00013 HIB      6     5318   0 00:00:01.38       267 >    239  NAH > 2120044E TCPIP$FTPC00020 HIB      6    10182   0 00:00:02.99       304 >    279  NeH > 2120044F TCPIP$FTPC00017 HIB      6     4961   0 00:00:01.58       270 >    241  N H > 21200452 TCPIP$FTPC00023 HIB      6     4967   0 00:00:01.41       268 >    239  NyH > 21200455 TCPIP$FTPC00024 HIB      6    11868   0 00:00:02.90       267 >    237  NeH > 21200458 TCPIP$FTPC00029 HIB      6     6622   0 00:00:01.57       268 >    241  NrH > 2120045C TCPIP$FTPC0002B HIB      6     6888   0 00:00:01.76       258 >    237  NsH > 21200464 TCPIP$FTPC00052 HIB      6       69   0 00:00:00.00       222 >    202  N H > 21200465 TCPIP$FTPC00050 HIB      6       67   0 00:00:00.05       222 >    202  N H > 21200466 TCPIP$FTPC00053 HIB      6       66   0 00:00:00.01       222 >    202  NlH > 21200467 TCPIP$FTPC00059 HIB      6       66   0 00:00:00.02       222 >    202  NeH > 21200468 TCPIP$FTPC0005A HIB      6       67   0 00:00:00.02       222 >    202  NaH > 2120046C DECW$TE_046C    LEF      5    12429   0 00:00:01.36       921 >    671H > 2120046D _FTA2:          COM      4     7244   0 00:00:01.22      8969 >   2048H > 21200470 SPI             CUR      4     1097   0 00:00:00.32       858 >    196( > US2|...[EQSPED.SRC] tcpip show version > A >   Compaq TCP/IP Services for OpenVMS Alpha Version V5.1 - ECO 3eB >   on a COMPAQ AlphaServer DS20E 666 MHz running OpenVMS V7.3     > 2 > US2|...[EQSPED.SRC] tcpip show service ftp /full >  > Service: FTP/ >                            State:     EnabledIA > Port:               21     Protocol:  TCP             Address: m	 > 0.0.0.0lA > Inactivity:          5     User_name: TCPIP$FTP       Process: o > TCPIP$FTP E > Limit:              80     Active:     16             Peak:      29  >   . > File:         TCPIP$SYSTEM:TCPIP$FTP_RUN.COM > Flags:        None >    > Socket Opts:  Rcheck ScheckS2 >  Receive:            0     Send:               0 >   H > Log Opts:     Acpt Actv Dactv Conn Error Exit Logi Logo Mdfy Rjct TimO > Addr: >  File:        SYS$SYSDEVICE:[TCPIP$FTP]TCPIP$FTP_RUN.LOG >   
 > Security >  Reject msg:  not definedv >  Accept host: 0.0.0.0  >  Accept netw: 0.0.0.0  > * > US2|...[EQSPED.SRC] product show history > = > ----------------------------------- ----------- -----------h > --------------------F > PRODUCT                             KIT TYPE    OPERATION   DATE AND > TIME= > ----------------------------------- ----------- -----------, > --------------------= > DEC AXPVMS ALP_DWEURO V1.1-1        Full LP     Install    t > 14-JAN-2002 09:39:52= > DEC AXPVMS TCPIP_ECO V5.1-153       Patch       Install      > 14-JAN-2002 09:37:55= > DEC AXPVMS VMS73_LIBRTL V1.0        Patch       Install    t > 14-JAN-2002 09:34:19= > DEC AXPVMS VMS73_UPDATE V1.0        Patch       Install      > 14-JAN-2002 09:34:11= > DEC AXPVMS VMS73_MAIL V1.0          Patch       Install    l > 05-OCT-2001 11:07:40= > DEC AXPVMS VMS73_RMS V1.0           Patch       Install    I > 05-OCT-2001 11:07:40= > DEC AXPVMS VMS73_SYSBOOT V1.0       Patch       Install    a > 05-OCT-2001 11:07:40= > DEC AXPVMS FASTVM130 V1.3-BETA1     Full LP     Install    : > 25-SEP-2001 16:00:35= > CPQ AXPVMS CSWS_JAVA V1.0           Full LP     Install    d > 25-SEP-2001 13:07:17= > CPQ AXPVMS CSWS V1.1                Full LP     Install    R > 25-SEP-2001 12:22:27= > CPQ AXPVMS CSWS V1.1                Full LP     Remove     k > 25-SEP-2001 12:21:40= > CPQ AXPVMS CSWS V1.1                Full LP     Install    3 > 25-SEP-2001 12:20:14= > DEC AXPVMS DWMOTIF V1.2-6           Full LP     Install      > 21-AUG-2001 18:10:04= > DEC AXPVMS VMS73_DRIVER V1.0        Patch       Install    c > 21-AUG-2001 10:17:34= > DEC AXPVMS VMS73_LAN V1.0           Patch       Install    a > 21-AUG-2001 10:17:07= > DEC AXPVMS VMS73_LINKER V1.0        Patch       Install    = > 21-AUG-2001 10:16:36= > DEC AXPVMS VMS73_SHADOWING V1.0     Patch       Install    n > 21-AUG-2001 10:16:04= > DEC AXPVMS VMS73_SYS V2.0           Patch       Install    m > 21-AUG-2001 10:15:19= > DEC AXPVMS DCOM V1.1-B              Full LP     Install    D > 21-AUG-2001 09:29:07= > DEC AXPVMS JAVA130 V1.3-1           Full LP     Install    x > 21-AUG-2001 09:01:24= > DEC AXPVMS DECNET_PHASE_IV V7.3     Full LP     Install    ' > 26-JUL-2001 09:38:46= > DEC AXPVMS DECNET_OSI V7.3          Full LP     Remove     h > 26-JUL-2001 09:38:22= > DEC AXPVMS NS_NAV_EXPORT V3.3       Full LP     Install    2 > 24-APR-2001 09:47:58= > DEC AXPVMS DECNET_OSI V7.3          Full LP     Install      > 17-APR-2001 10:33:28= > DEC AXPVMS DWMOTIF V1.2-6           Full LP     Install    q > 12-APR-2001 15:57:55= > DEC AXPVMS OPENVMS V7.3             Platform    Install    i > 12-APR-2001 15:57:55= > DEC AXPVMS TCPIP V5.1-15            Full LP     Install      > 12-APR-2001 15:57:55= > DEC AXPVMS VMS V7.3                 Oper System Install      > 12-APR-2001 15:57:55= > ----------------------------------- ----------- -----------r > -------------------- >  e > 28 items found >  > Joerg Spilling > EQUIcon Software GmbH Jena	 > Germany    ------------------------------  # Date: Wed, 06 Mar 2002 22:41:51 GMTs2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)" Subject: Re: UNMESSAGE for Alpha ?2 Message-ID: <PWwh8.660$fL6.10277@news.cpqcorp.net>  x In article <JfinIcRWICie@eisner.encompasserve.org>, clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) writes:, :Is there a version of UNMESSAGE for Alpha ? ..G :[UNMESSAGE is a utility to convert a VMS message file back to a sourcet :message file].     H   About half an hour (worst case) permits one to put together DCL to do H   this...  This task is little more than a DCL loop containing f$messageI   and some WRITE statements to reformat the results to a file, of course.s    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: Thu, 07 Mar 2002 09:29:56 -0500n* From: John Reagan <john.reagan@compaq.com> Subject: Re: VAX Pascal ) Message-ID: <3C877964.7080900@compaq.com>r   Chris Sharman wrote: >  > I've got to ask:M > why don't you (and the generated starlet etc) now use the builtin datatypes  > such as UNSIGNED16 ?F > Is it back-compatibility, portability, habit, or some other reason ?  	 1) Habit.-A 2) I wrote that little snipet of code many years ago (before the Q invention of UNSIGNED16)     -- - John Reagan-' Compaq Pascal/{A|I}MACRO Project Leader-   ------------------------------  # Date: Thu, 07 Mar 2002 01:00:43 GMTf2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: vax to tcp/ip2 Message-ID: <%Yyh8.671$fL6.10342@news.cpqcorp.net>  g In article <3C85D7AA.5BF7BA29@netcom.com.tn>, oussama hassairi <oussama.hassairi@netcom.com.tn> writes:e  ? :can you help me in vax  integration to a TCP/IP switched  LAN? ? :how can IP users of a switched lan communicate with vax6000 or  :microvax3100?      VAX is hardware, not software.  F   OpenVMS VAX -- one of various operating systems that operate on VAX G   hardware, and also the subject of this newsgroup so I'll assume that >H   it is what you meant -- does have support for TCP/IP communications.    I   There are several commercial IP stacks available and at least one free lJ   stack (CMU).  Among the commercial stacks is the Compaq TCP/IP Services I   product.  (The TCP/IP services stack is licensed with most NAS license eD   PAKs and separately with a UCX license PAK, and thus many OpenVMS $   VAX systems have licenses for it.)  H   Once you have an IP stack installed and configured -- and to learn howD   to configure it, please acquire and read the documentation for theH   stack and the version -- you will have to work out the connection fromD   the switch to the particular VAX system.  The specifics can and do@   vary by OpenVMS version, by IP stack, and by IP stack version.  G   To continue the discussion, we need to know the IP stack, the OpenVMS K   VAX version -- or the name of the operating system, if it is not OpenVMS eJ   VAX -- and details of the VAX 6000 and MicroVAX 3100 platforms.  (These L   two VAX boxes are old enough that they typically have 10 megabit Ethernet L   LAN connections, not the more common 100 megabit (fast) and gigabit links.I   This means you need to find out if your switch can provide or negotiateaG   to 10 megabit LAN links.)  You will also want to indicate if you can -J   upgrade the OpenVMS VAX version, as more recent releases of OpenVMS have$   newer versions of TCP/IP Services.  G   Please also avail yourself of the OpenVMS FAQ and the TCP/IP Services3G   documentation -- the FAQ has pointers to the Compaq documentation for:E   OpenVMS, and the TCP/IP Services documentation is at the same site.n    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: 7 Mar 2002 09:00:56 -0800a From: ohm62@hotmail.com (OHM)p  Subject: VMS DEBUG V7.2 bugs me!= Message-ID: <9d337b47.0203070900.48badda4@posting.google.com>c  	 Hi folks,n  < VMS Alpha V7.2 DEBUG - I got lots of small problems with it!  " Here are two of the most annoying:  C 1_ With a multithreaded application, apparently reaching a deadlockv situation...  B I try to get a picture of the lock down pattern, using the phtread command; Here is the output:   DBG> pthread mutex -lmB Mutex  Name                      State Owner  Pri Type     Waiters (+Count): ------ ------------------------- ----- ------ --- -------- --------------------B     39 <anonymous>               Lock             Normal   5, 3, 4<     43 <anonymous>               Lock       3     Recurs   6  F Thread 3 is waiting for a mutex #43 visibly owned by thread 3... Fine.F  Threads 5, 3 and 4 are apparently waiting for a mutex #39 ... that is? not owned by any thread!!! How that possible?   I guess that is  nonsensical debugger display.o  ? 2_ Attaching to a running process which image carries tracebackiD information, I could not get to the source, whether I use SET SOURCEE with or without /LATEST and with or without /MODULE doesn't help: thenF debugger keeps bugging me, complaining about the absence of the source file.  Using a SPAWN DIR path2F where path is what SHOW SOURCE displays, I can actually see the source file.sB I did a SET MODULE upfront.  I checked that the image is tracebackA using ANALYZE/IMAGE. I checked that the object file carries debug-D info. using ANALYZE/OBJ.  And using DBG> SHOW MODULE myModule, I can> see the module listed with a yes in the symbols column...  ButD EXAMINE/SOURCE <address>  just doesn't work!  Okay, the drive is DFSF mounted, but this works on some other VMS versions, and anyways, I end= up in the same situation after copying the sources in a localr9 directory and redirecting the SET SOURCE to that place...C   What am I doing wrong?  C BTW, I tried the 7.2X version with no more luck.  My application ishB built with DEC C V6.4-008 and DEC C++ V6.3-020.  VMS Alpha V7.2 is patched all the way up.u   Thanks,   	    -- Ol.    ------------------------------  $ Date: Thu, 7 Mar 2002 11:10:22 +01007 From: "Dijk, Jeroen van" <Jeroen.vanDijk@Getronics.com>g; Subject: RE: What VMS version runs on a Digital Server 3000eO Message-ID: <2795B75EF003D311801A00A0C906B511011C6D61@cucexec.gbc.getronics.nl>h  ? The reason why VMS don't want to install on a NTFS volumes has h? every thing to do with the diference in security level between b WNT & 0S/390 and VMS.g  J Personaly I don't believe that you could install OS/390 on a NTFS volume, N but if you can install linux in VM under OS/390, so NT could also be possible.  Y Personaly I work in shop supports and have several platforms, and only 3 VMS specialists j and around 20 VMS boxes.    0 Because the DR tests on VMS boxes are so simple.Q I will try a little bit more complexed so I going to learn and do HP9000 and AIX.l  m BTW if someone in Holland is an expert in HP9000, VMS and maybe some other OS and is looking for a job ..... l  n? > there's no reason to keep whatever was left on the NT4 disk.   > VMS can read= > NTFS volumes thru a layered product but not by itself, nor d > can it dual boot	 > from an ; > NT4 system disk. Not that I'd particularly want that BTW.i? > The problem was just that VMS did not want to install on the   > NTFS formatted > disk.v@ > Once I did that manually (via the $$$ prompt on the Alpha/VMS  > distribution > CD)s? > all went well. Don't know  why but the box now boots VMS. In i > an all OS/390h	 > and WNTi< > shop I felt a little alone without a VMS system nearby.... >  > Hans > B > Dijk, Jeroen van <Jeroen.vanDijk@Getronics.com> wrote in message@ > news:2795B75EF003D311801A00A0C906B511011C6D5D@cucexec.gbc.getr
 > onics.nl...S= > > Remove the NT4 disk from the system or backup it, if you   > want to preserve > it.  > > F > > I haven't heard of if VMS can read NTFS volumes, but if you want a > dualboot system.; > > Just leave it on DKA0 and install VMS on an other disk.A > >  > > >n > > > Jeff,  > > >a@ > > > the Alpha now boots from the AXP/VMS V7.3 installation CD.A > > > I tried to boot after each step and step 3, set srm_boot ontD > > > did the trick. It apparently creates a new variable and allows > > > VMS to boot from CD.; > > > The next question is: how can I format the disk as a g > Files-11 volume.A > > > DKA0 now has NT4 on it and is (probably) formatted as NTFS.s > > >d
 > > > Hans > > > 2 > > > Jeff Campbell <jcampbell@ins-msi.com> wrote:? > > > >The DS 3300 is a "white box" AlphaServer 800 hobbled to o > run NT only. > > > >h) > > > >To run VMS on it do the following;h > > > >P6 > > > >  1) Change the machine to use the SRM console. > > > >AF > > > >  2) Do a  >>> show conf  command to see the machine's firmware > > > >     versions. Go toa > > > >n> > > > >         http://ftp.digital.com/pub/DEC/Alpha/firmware/ > > > >f9 > > > >     If the machine is not at the latest firmware e > version download! > > > >     a copy and update it.c > > > >e7 > > > >  3) Then at the >>> prompt, type the following:s > > > >u" > > > >        >>> set os_type vmsG > > > >        >>> cat nvram       (to see what, if anything, is in it)3 > > > >        >>> edit nvram % > > > >            10 set srm_boot onI > > > >            20 el > > > >        >>> initS > > > > & > > > >   4) Install VMS and have fun! > > > >0C > > > >Change the line numbers, 10 and 20 above, if your nvram file24 > > > >happens to have other commands in it already. > > > > A > > > >The edit command above runs a simple line numbered editor.95 > > > >Type a ? in it to see a list of it's commands:  > > > >  > > > >        >>> edit nvram  > > > >            ? > > > >0 > > > >Jeff Campbell > > > >n8wxs@arrl.netM > > > >  > > > >Hans Vlems wrote: > > > >>? > > > >> The label on the back shows the following information:1 > > > >> > > > >> Model: FR-K7F2W-WA3 > > > >> PN: 3300 6400AI > > > >> Series: P8800 > > > >>? > > > >> The system is a white PC style box. It is _not_ a DEC   > 3000 machineC > > > >> (the ones that look remotely like a VAX 3100). Could it be  > > > a Jensen?0+ > > > >> The ARC console identification is:2 > > > >>- > > > >> Digital Server 3000 model 3300 6400A  > > > >> Digital Alpha 21164 > > > >>C > > > >> The machine used to boot NT4, but its ARC console supportsT > > > also VMS and	 > > > OSF8> > > > >> modes. All I want is to be able to run AXP/VMS on it.@ > > > >> The error message reported was the result of >>> B -fl  > 0,0 dka500B > > > >> where dka500 is the CD drive. It does boot into SYSBOOT>, > > > no problem withH > > > >> that.? > > > >> The fact that the system supports both ARC and VMS/OSF  > > > console modes58 > > > >> made me think that VMS might run on the system. > > > >>
 > > > >> Hans8 > > > >>< > > > >> rdeininger@mindspring.com (Robert Deininger) wrote:< > > > >> >In article <3c840244.1214705283@news.wcc.govt.nz>, > > > rob.buxton@wcc.govt.nz > > > >> wrote:0
 > > > >> >G > > > >> >>Not sure if it's the same, we got a couple of DEC 3000s here0. > > > >> >>2 DEC 3000 600M and a DEC 3000 300L$ > > > >> >>The latter is running 7.3& > > > >> >>The 600Ms are running 7.2-1 > > > >> >>PG > > > >> >>Palcode Version on the 300L is 5.54 and on the 600Ms is 5.56  > > > >> >>P" > > > >> >>Maybe a Firmware thing?
 > > > >> >C > > > >> >The above are conventional DEC 3000 server or workstation  > > > systems.  VMS A > > > >> >has supported them since the flood.  There haven't been7 > > > major changes  > > > toA > > > >> >the firmware for years.  These systems will load either0C > > > >> >SYS$CPU_ROUTINES_0702.EXE (any of the 3000-300 family) or 8 > > > >> >SYS$CPU_ROUTINES_0402.EXE (any of the others).
 > > > >> >
 > > > >> >9 > > > >> >>On Mon, 04 Mar 2002 15:49:21 GMT, "Hans Vlems"0 > > > <hvlems@zfree.co.nz> > > > >> >>wrote:: > > > >> >>  > > > >> >>>> > > > >> >>>I found an old Dig. Server 3000 and tried to boot  > VMS 7.3 on it.- > > > >> >>>It failed with the error message:04 > > > >> >>>Unable to load SYS$CPU_ROUTINES_E505.EXE. > > > >> >>>The VMS PALcode version is V1.20-38 > > > >> >>>The >>>show version command returns V5.4-113
 > > > >> >: > > > >> >This is something completely different.  I don't > > > recognize the name> > > > >> >"Digital Server 3000" as being a VMS-capable system. > > > (Which does notI
 > > > meanC > > > >> >there is no such beast.)  Certainly, if it is looking for C > > > >> >SYS$CPU_ROUTINES_E505.EXE, it is NOT a Pelican, Flamingo,  > > > or Sandpiper5 > > > >> >class DEC 3000 system.  No relation at all. 
 > > > >> > > > > >> >(This web page
 > > > >> >F > > http://www.phys.ufl.edu/~prescott/linux/alpha/dec3000-sysinfo.html@ > > >> >summarizes the DEC 3000 family pretty nicely, including  > the formerly& > > >> >top-secret system code names.) > > >> >? > > >> >On my systems, there's no SYS$CPU_ROUTINES_E505.EXE in G > the [SYS$LDR]y@ > > >> >directory, which is the root problem.  The console sets 
 > up the data > > > >> >structures that lets VMS figure this out.  There's no  > intelligence in- > > VMS- > > >> >in this area.  > > >> >> > > >> >Please be VERY careful with system names.  If you can  > find the label,D > > postG > > >> >exactly what the model name is.  Since you seem to have console-	 > access,-7 > > >> >tell us the exact system name from the console.l > > >> >; > > >> >This _may_ be one of the crippled systems that was a > tweeked to make VMS < > > >> >unbootable.  The "E505" part of the file name looks  > suspicious, since  > > I4@ > > >> >don't think the most significant bit of that hex number  > is ever set on@ > > >> >"official" VMS-supported systems.  But my memory may be  > faultly.  Some > > of= > > >> >these systems shipped with VMS PALcode, but VMS does   > not support them.  > > >> >8 > > >> >>>Is there any hope to boot, say, VMS 6.2 on it? > > >> >? > > >> >If it's similar enough to a supported system, it could   > likely be made > > to, > > >> >work.  Licensing might be a problem. > > >> >? > > >> >If you find the detailed system specs, and have access  
 > to a source1> > > >> >listings kit, and can write your own platform support  > and maybe some > > bootE > > >> >drivers, then I'm certain you could boot VMS on it.  But thisV > paragraph H > > >> >likely needs multiple smileys.  You're looking for hobbyist, not > > >> >obsession, right?  > > >> > > > >> >  -- RobertA > > >> > > >> http://www.zfree.co.nzI > >1 > >  > >a > > http://www.zfree.co.nz >  >    ------------------------------  % Date: Thu, 07 Mar 2002 08:11:28 -0500 5 From: David Beatty <David.Beatty@qwertysasasdfgh.com> Y Subject: Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startu 2 Message-ID: <RmWHPA3wBtCLfXsWvshFM2b9Lgrz@4ax.com>  <     I remember I had fiddled around with group logical names: way back in V5.5-2.  The one feature I noted is there must4 be at least one process with group membership in the6 group in question for the logical name table to exist.  ?     It appears that V7.2-2 and V7.3 work the same way.  You can ! perform a SHOW LOGICAL LNM$GROUP*2& /TABLE=LNM$SYSTEM_DIRECTORY to verify.  1     ... and as others have noted, you can use the9, /TABLE=LNM$GROUP_nnnnnn qualifier to specify; the group, where nnnnnn is the left zero-filled octal valueS for the group.   David R. Beatty   , On Wed, 6 Mar 2002 20:37:46 -0000, "Malcolm", <malcolmix@neverness.freeserve.co.uk> wrote:  7 >"Patrick Young" <P.Young@unsw.EDU.AU> wrote in message 8 >news:55f85d77.0203012349.59b9b99e@posting.google.com...; >> Kilgallen@SpamCop.net (Larry Kilgallen) wrote in message 0 >news:<7p9b8KjweUJY@eisner.encompasserve.org>... >> >
 >> > $ RUNF >SYS$SYSTEM:LOGINOUT/INPUT=NL:/OUTPUT=NL:/ERROR=NL:/UIC=[37774,177776] >>I >> Didn't think of that - That is _WAY_ cool! - missed a group of 3 of use >since >> 1989. >>G >> X run/uic=[12345,177776]/err=nl:/out=nl:/inp=nl: sys$system:loginout-@ >> %RUN-S-PROC_ID, identification of created process is 00000177! >> X show lo/tab=lnm$group_012345) >> (LNM$GROUP_012345)E >rE >...but why isn't there a DEFINE /GROUP=[group name or group number],cM >requiring (say) SYSNAM and CMKRNL privileges and creating the group table ifE >it doesn't exist?M >Nice and consistent, and wouldn't break any existing DCL (plain /GROUP couldaH >default to user's current group and require the appropriate privileges;/ >/GROUP=nnnn would require heavier privileges).n >o& >Another one for the next release? ;-) >a	 >-Malcolme   ------------------------------  # Date: Thu, 07 Mar 2002 14:21:03 GMTo, From: "Frank Sapienza" <sapienza@noesys.com>Y Subject: Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startuo9 Message-ID: <jHKh8.16486$gQ.4130926@news02.optonline.net>p  D That may be the logic wherein OpenVMS creates the name table itself.C However, anyone can create a name table that has a name in the form 8 LNM$GROUP_nnnnnn.  As long as they have SYSPRV, that is.   Franke  B "David Beatty" <David.Beatty@qwertysasasdfgh.com> wrote in message, news:RmWHPA3wBtCLfXsWvshFM2b9Lgrz@4ax.com... >3> >     I remember I had fiddled around with group logical names< > way back in V5.5-2.  The one feature I noted is there must6 > be at least one process with group membership in the8 > group in question for the logical name table to exist. > A >     It appears that V7.2-2 and V7.3 work the same way.  You cany# > perform a SHOW LOGICAL LNM$GROUP*U( > /TABLE=LNM$SYSTEM_DIRECTORY to verify. >s3 >     ... and as others have noted, you can use the-. > /TABLE=LNM$GROUP_nnnnnn qualifier to specify= > the group, where nnnnnn is the left zero-filled octal value: > for the group. >e > David R. Beatty1 >6   ------------------------------  % Date: Thu, 07 Mar 2002 09:42:21 -0500u5 From: David Beatty <David.Beatty@qwertysasasdfgh.com>eY Subject: Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startua2 Message-ID: <w3uHPOoPAUFeHkO2=8ErLr3vliZf@4ax.com>  F ... and it would appear in both V7.2-2 and V7.3 that once the table is  D created, it is not deleted if there are no users in the group, which* is very good news for the original poster!   David   2 On Thu, 07 Mar 2002 14:21:03 GMT, "Frank Sapienza" <sapienza@noesys.com> wrote:  E >That may be the logic wherein OpenVMS creates the name table itself..D >However, anyone can create a name table that has a name in the form9 >LNM$GROUP_nnnnnn.  As long as they have SYSPRV, that is.c >e >Frank >aC >"David Beatty" <David.Beatty@qwertysasasdfgh.com> wrote in messagey- >news:RmWHPA3wBtCLfXsWvshFM2b9Lgrz@4ax.com...  >>? >>     I remember I had fiddled around with group logical namesu= >> way back in V5.5-2.  The one feature I noted is there mustc7 >> be at least one process with group membership in thet9 >> group in question for the logical name table to exist.p >>B >>     It appears that V7.2-2 and V7.3 work the same way.  You can$ >> perform a SHOW LOGICAL LNM$GROUP*) >> /TABLE=LNM$SYSTEM_DIRECTORY to verify.p >>4 >>     ... and as others have noted, you can use the/ >> /TABLE=LNM$GROUP_nnnnnn qualifier to specifyo> >> the group, where nnnnnn is the left zero-filled octal value >> for the group.h >> >> David R. Beatty >> >g >t   ------------------------------   Date: 7 Mar 2002 16:11:38 +0100 * From: eplan@kapsch.net (Peter LANGSTOEGER)Y Subject: Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startul* Message-ID: <3c87832a$1@news.kapsch.co.at>  h In article <jHKh8.16486$gQ.4130926@news02.optonline.net>, "Frank Sapienza" <sapienza@noesys.com> writes:E >That may be the logic wherein OpenVMS creates the name table itself. D >However, anyone can create a name table that has a name in the form9 >LNM$GROUP_nnnnnn.  As long as they have SYSPRV, that is.t  ? But how to create this table in KERNEL mode (without hacking) ?>   -- -< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111 2651; Network and OpenVMS system manager  Fax.    +43 1 81111 888-< KAPSCH AG      Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  $ Date: Thu, 7 Mar 2002 12:12:39 -0500, From: "Frank Sapienza" <sapienza@noesys.com>Y Subject: Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startus, Message-ID: <a6870d0279g@enews1.newsguy.com>  D Being the realist that your signature claims you to be, I'm sure you? understand that there are times when you can't have everything.t  + This would appear to be one of those times.w     Franku :-)t  7 "Peter LANGSTOEGER" <eplan@kapsch.net> wrote in messaget$ news:3c87832a$1@news.kapsch.co.at...L > In article <jHKh8.16486$gQ.4130926@news02.optonline.net>, "Frank Sapienza" <sapienza@noesys.com> writes:-G > >That may be the logic wherein OpenVMS creates the name table itself.mF > >However, anyone can create a name table that has a name in the form; > >LNM$GROUP_nnnnnn.  As long as they have SYSPRV, that is.r > A > But how to create this table in KERNEL mode (without hacking) ?# >. > --> > Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111 2651= > Network and OpenVMS system manager  Fax.    +43 1 81111 888n> > KAPSCH AG      Wagenseilgasse 1     E-mail  eplan@kapsch.netJ > A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------   Date: 7 Mar 2002 19:37:52 +0100m* From: eplan@kapsch.net (Peter LANGSTOEGER)Y Subject: Re: Why no DEFINE /GROUP=nnnn? (was re: define "group-logicals" at system startu * Message-ID: <3c87b380$1@news.kapsch.co.at>  [ In article <a6870d0279g@enews1.newsguy.com>, "Frank Sapienza" <sapienza@noesys.com> writes:fE >Being the realist that your signature claims you to be, I'm sure youE@ >understand that there are times when you can't have everything. >u, >This would appear to be one of those times.  L I think you misunderstood. This my statement was only a rhetorical question.  J As long as CREATE/NAME_TABLE has no /KERNEL qualifier the whole discussionL with DEFINE/TABLE=LNM$GROUP* (or why the table gets cleared when a user logsL in - because it was a /EXECUTIVE table and when the first user logs in, thenJ VMS creates an empty [kernel,no_alias] table and deletes the [exec] table)I is useless. The only working approach was stated on the beginning of this-G thread with a workaround like SUBMIT/USER (or RUN/UIC). And this we allr& know (or should know) for years now...   -- c< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111 2651; Network and OpenVMS system manager  Fax.    +43 1 81111 888e< KAPSCH AG      Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------   End of INFO-VAX 2002.130 ************************390 on a NTFS volume, N but if you can install linux in VM under OS/390, so NT could also be possible.  Y Personaly I work in @B[    AB[    BB[    CB[    DB[    EB[    FB[    GB[    HB[    IB[    JB[    KB[    LB[    MB[    NB[    OB[    PB[    QB[    RB[    SB[    TB[    UB[    VB[    WB[    XB[    YB[    ZB[    [B[    \B[    ]B[    ^B[    _B[    `B[    aB[    bB[    cB[    dB[    eB[    fB[    gB[    hB[    iB[    jB[    kB[    lB[    mB[    nB[    oB[    pB[    qB[    rB[    sB[    tB[    uB[    vB[    wB[    xB[    yB[    zB[    {B[    |B[    }B[    ~B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[    B[     C[    C[    C[    C[    C[    C[    C[    C[    C[    	C[    
C[    C[    C[    
C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[     C[    !C[    "C[    #C[    $C[    %C[    &C[    'C[    (C[    )C[    *C[    +C[    ,C[    -C[    .C[    /C[    0C[    1C[    2C[    3C[    4C[    5C[    6C[    7C[    8C[    9C[    :C[    ;C[    <C[    =C[    >C[    ?C[    @C[    AC[    BC[    CC[    DC[    EC[    FC[    GC[    HC[    IC[    JC[    KC[    LC[    MC[    NC[    OC[    PC[    QC[    RC[    SC[    TC[    UC[    VC[    WC[    XC[    YC[    ZC[    [C[    \C[    ]C[    ^C[    _C[    `C[    aC[    bC[    cC[    dC[    eC[    fC[    gC[    hC[    iC[    jC[    kC[    lC[    mC[    nC[    oC[    pC[    qC[    rC[    sC[    tC[    uC[    vC[    wC[    xC[    yC[    zC[    {C[    |C[    }C[    ~C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    C[    