1 INFO-VAX	Wed, 17 Sep 2003	Volume 2003 : Issue 516       Contents: Re: " USP for OpenVMS " + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch 0 audio/multimedia program collection via anon ftp+ Re: Backup foreground/background operations + Re: Backup foreground/background operations + Re: Backup foreground/background operations % Re: Choosing between ASTs and Threads ' Re: CSwing and OpenVMS v7.3-1, Again... 
 Data Recovery  Re: Data Recovery 2 Re: Difference Intel EFI and Phoenix cME FirstWare" Re: HELP: Using CLI with C program# how to add a printer (HP laser jet) ' Re: how to add a printer (HP laser jet) ' Re: how to add a printer (HP laser jet) ' Re: how to add a printer (HP laser jet) ' Re: how to add a printer (HP laser jet) ' Re: how to add a printer (HP laser jet) ' Re: how to add a printer (HP laser jet) ' Re: how to add a printer (HP laser jet) - Re: How to use shareable image to share data? - Re: How to use shareable image to share data? - Re: How to use shareable image to share data? - Re: How to use shareable image to share data? 1 Re: HP extends lead in open storage area networks 1 Re: HP extends lead in open storage area networks 1 Re: HP extends lead in open storage area networks P Re: I have a fix!!!  Re: Anyone else having problems with VMS and HSG80-based ar  Re: IA64 to be dual core by 2005  Re: IA64 to be dual core by 2005  Re: IA64 to be dual core by 2005  Re: IA64 to be dual core by 2005  Re: IA64 to be dual core by 2005  Re: IA64 to be dual core by 2005  Re: IA64 to be dual core by 2005  Re: IA64 to be dual core by 2005  Re: IA64 to be dual core by 2005 Re: KZPBA Question Re: Memory for VS 4000-xx  MySQL 4.0.15 for OpenVMS Re: Newbie question about NTP  Re: Newbie question about NTP  Re: Newbie question about NTP  Re: Newbie question about NTP 3 Re: OpenVMS 7.3-1 drowning HSG80s with I/O requests K Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster L Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster?, OpenVMS hobbyyest - free to join or not ????0 Re: OpenVMS hobbyyest - free to join or not ????0 Re: OpenVMS hobbyyest - free to join or not ????0 RE: OpenVMS hobbyyest - free to join or not ????0 Re: OpenVMS hobbyyest - free to join or not ????0 Re: OpenVMS hobbyyest - free to join or not ????0 Re: OpenVMS hobbyyest - free to join or not ????0 OT: security flaw in Solaris and Trusted Solaris- Re: OT: Verisign messes with DSN root servers - Re: OT: Verisign messes with DSN root servers - Re: OT: Verisign messes with DSN root servers - Re: OT: Verisign messes with DSN root servers - Re: OT: Verisign messes with DSN root servers  Re: padding a csv file Re: padding a csv file/ PHP library containing PREG_MATCH, ... commands 3 Re: PHP library containing PREG_MATCH, ... commands 3 Re: Porting from UNIX to OpenVMS for a real newbie. 3 Re: Porting from UNIX to OpenVMS for a real newbie. # Re: Question on running GnuPG 1.2.0 $ Re: Remove Pathworks/Advanced Server$ Re: Remove Pathworks/Advanced Server' Re: Remove satellite node from cluster? ' Re: Remove satellite node from cluster? , Re: Solaris Security (was: OpenVMS Security)' TCP/IP Services for VMS - sloppy output + Re: TCP/IP Services for VMS - sloppy output + Re: TCP/IP Services for VMS - sloppy output + Re: TCP/IP Services for VMS - sloppy output + Re: TCP/IP Services for VMS - sloppy output + Re: TCP/IP Services for VMS - sloppy output G There is no Joy in Sunville -- mighty Billy has struck out [on his own] K Re: There is no Joy in Sunville -- mighty Billy has struck out [on his own] ; Very strange Monitor (and possible VMS scheduler) behaviour # We need your Ds20 and DS20e Systems  WWIDMGR - HSG80 -> EVA Re: WWIDMGR - HSG80 -> EVA Re: WWIDMGR - HSG80 -> EVA Re: WWIDMGR - HSG80 -> EVA  F ----------------------------------------------------------------------  + Date: Wed, 17 Sep 2003 05:10:49 -0700 (PDT) . From: Fabio Cardoso <fabiopenvms@yahoo.com.br>  Subject: Re: " USP for OpenVMS "@ Message-ID: <20030917121049.97280.qmail@web20210.mail.yahoo.com>   Valentin   As your email rejected ...   (..)  So how did you implemented that.& I never needed to use NCS so I really & dont know how it works. I  was reading% the documentatiom and If I understood  it generates a library ?     Regards    FC  + --- Valentin Likoum <me@privacy.net> wrote:  > Fabio Cardoso wrote: > > Richard  > > 8 > > Is there any other way to encode this data in OVMS?  > [snip] > < >    Yes, certanly. NCS! We use it to encode/decode UTF8/16 = > data. It's ability to decode UTF16 was a real surprise for  ? > me. Contact me directly (decode e-mail in the bottom) if you  8 > need some help on it. Our conversion routines are for < > russian code page only, but it's not a big deal to figure 1 > out the principle and remake it for portuguese.  >  > --   > Best regards,  >   Valentin+ >   valentin.likoum at ncc dot volga dot ru  >      =====  ========================== Fbio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazil  fabiopenvms@yahoo.com.br ==========================  " __________________________________ Do you Yahoo!?? Yahoo! SiteBuilder - Free, easy-to-use web site design software  http://sitebuilder.yahoo.com   ------------------------------    Date: 17 Sep 2003 05:47:00 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) 4 Subject: Re: "HP breakup on the way" - Merrill Lynch3 Message-ID: <Nc9MYfmzHRI0@eisner.encompasserve.org>   \ In article <87vfrs9gfr.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:   > Can I add a `me too' :)   A No, you can't, since you offer no context regarding that to which = your "me too" would pertain.  Please do not presume either of   4 	a. Everybody else's newsreader is the same as yours2 	b. Everybody else will read your post right after* 			the one to which you reply (rather than 			the next day)   ------------------------------    Date: 17 Sep 2003 07:09:02 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 4 Subject: Re: "HP breakup on the way" - Merrill Lynch3 Message-ID: <LyggA8QoXEk3@eisner.encompasserve.org>   V In article <3F673C13.35ED58E5@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes: > Bob Koehler wrote:I >>    Fairly early in the merger process HP publcially said they couldn't G >>    actually yet influence Compaq corporate decisions, but they liked F >>    the decision to port VMS to IA64.  That's note quite the same as >>    ignoring it. > # > Interesting revisionist history.    F    That's fact, not revisionism.  Now I'll just have to put you in the    killfile next to Andrew.    ------------------------------    Date: 17 Sep 2003 13:10:25 +0200C From: vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann) 9 Subject: audio/multimedia program collection via anon ftp - Message-ID: <3f684121$1@news.uni-konstanz.de>    Hi,   G Here is a collection of multimedia applications for OpenVMS/AXP V7.3 or  higher.    [.BIN_AXP] binaries  [.ZIP_FILES] sources   Description   C CDDA2WAV.EXE CD-ripper                    (CDRTOOLS_201A18_VMS.ZIP) B CDRECORD.EXE CD-recorder                 (CDRTOOLS_201A18_VMS.ZIP)8 LAME.EXE     Wav to Mp3 Converter        (LAME-3_70.ZIP)9 MKISOFS.EXE  ISO-Filesystem Program  (CDRECORD_1_8_1.ZIP) E MPG123.EXE   Mp3-Player                        (MPG123-0_59S-MH4.ZIP)    Have fun Eberhard  / mailto: Eberhard.Heuser-Hofmann@uni-konstanz.de   . ftp v36.chemie.uni-konstanz.de/multimedia_vms/   ------------------------------    Date: 17 Sep 2003 06:49:53 -0700- From: djesys@earthlink.net (David J Dachtera) 4 Subject: Re: Backup foreground/background operations< Message-ID: <66a00d01.0309170549.6b76715@posting.google.com>  _ Wayne Sewell <wayne@tachysoft.com> wrote in message news:<00A25EE5.273C42EB.1@tachysoft.com>...  > [snip]
 > Rudolph, > L > You might look into the TAPESYS product from Software Partners (sp32.com),M > which will do exactly what you want.  TAPESYS provides a wrapper around VMS P > backup called SYSBAK, and being able to have multiple backups to the same tape? > drive in progress simultaneously is one of the main features.   E I'm told that SLS began life as TAPESYS. Do you know if this is true?   B If so, do you know how much SLS and TAPESYS have diverged over the years?  E ...ad do you know if TAPESYS provides any support for StorageTek L700 % libraries and ACSLS (ala SLS + DCSC)?   I > The SYSBAK definition file allows specification of a list of backups to M > perform, such as all of the disks in your cluster.  SYSBAK will do these in O > sequence, using the same drive, but it does *not* wait for recording passes.  M > Each individual backup is performed by a separate process.  Once the master Q > process determines that the recording (or delete) pass has started and the tape P > drive is no longer in use, the next backup is started in a new process.  ThereO > is no rewinding and no tape movement because the tape is already at the end.    C I know that in SLS the actual BACKUP runs in the grand-child of the 0 parent process. Is this how TAPESYS works, also?  F If so, do you know if SLS does BACKUP and /RECORD asynchronously also?  F I once wrote a technique for using EDT to massage BACKUP/LISTING files> into files lists that can be fed to DFU to record backup dates< separately from the BACKUP process. Naturally, it's all DCL.  Q > Thus the individual backup processes are running in parallel, one writing while O > one or more are recording.  The master process monitors the recording backups N > and marks them complete when the recording pass is finished.  Obviously, theF > overall job does not finish until all recording passes are complete.  ? ...which is what my "asynchronous /RECORD" technique does. Each > volume(-set) is porcessed in a separate batch job. You can use? /JOB_LIMITs to run as few or as many as the situation requires.    ---  David J Dachtera dba DJE Systems    ------------------------------  % Date: Wed, 17 Sep 2003 10:24:06 -0500 ( From: Wayne Sewell <wayne@tachysoft.com>4 Subject: Re: Backup foreground/background operations0 Message-ID: <00A2608E.49E340F1.15@tachysoft.com>   > F >I'm told that SLS began life as TAPESYS. Do you know if this is true?  J Yes.  SLS started as a snapshot of an extremely early version of tapesys. ; Since then, the two products have gone their separate ways.    > C >If so, do you know how much SLS and TAPESYS have diverged over the  >years?     H The snapshot was many years ago.  By now, the two products are radicallyM different, though some concepts remain the same, such as the existance of the   tapmgrdb and tapmgrrq processes.      J Tapesys 6.1 included *major* changes such as support for ICC and TCP/IP asK transports, and redesign of the master and history databases.  There are so + many changes that it's like a new product.    K The main databse was streamlined and many integrity checks were added.  The K fields of the old-format database, which I assume SLS still uses, were just N strings and you could enter pretty much anything you wanted in any field, suchK as "My Moose Has a Hernia" for a media type.  Customers used to trash their L databases on a regular basis.  Locations, media types, pools, and containersJ are now keywords that have to be explicitly defined before use.  Also, you> can't delete a location or media type if any reel is using it.  N The history database was *completely* redesigned.  The old format could simplyL not work with the radically different characteristics of ODS-5.  The historyG files are now hierarchical in nature, much like the physical disks they 
 represent.    F >...ad do you know if TAPESYS provides any support for StorageTek L700& >libraries and ACSLS (ala SLS + DCSC)?  J tapesys has support for DCSC.  There is a companion product called JukeboxI Manager (JB), which I believe is included with a tapesys licence.  JB has K support for MRU, so theoretically any jukebox that is MRU-compatible can be N used.  There is also a GK interface.  JB used to have a separate database, butM the current versions of the two products are more tightly integrated.  The JB F database can now be encapsulated inside the tapesys database, i.e. theJ jukebox records are extensions to the tapesys location records, and the jbO volume records are extension to the tapesys reel records.  Therefore, it is not 9 necessary to coordinate the two databases as in the past.    > D >I know that in SLS the actual BACKUP runs in the grand-child of the1 >parent process. Is this how TAPESYS works, also?   M Yes, the master backup process creates two subprocesses for each backup.  One K process runs sysbak.exe, which coordinates mounts and access to the tapesys K database.  The other runs vmsbu, which is a wrapper around vms backup.  The J master process monitors the process pairs and does not exit until all have
 completed.   > G >If so, do you know if SLS does BACKUP and /RECORD asynchronously also?  >   N Probably.  This setup goes back a long way and probably predates the snapshot.   Wayne O =============================================================================== N Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html    O =============================================================================== N Butler:"Gentlemen!"  Curly(as he and other Stooges look around):"Who came in?"   ------------------------------  # Date: Wed, 17 Sep 2003 15:12:53 GMT 4 From: brad@.gateway.2wire.net (Bradford J. Hamilton)4 Subject: Re: Backup foreground/background operations/ Message-ID: <VR_9b.487909$o%2.217361@sccrnsc02>   l In article <66a00d01.0309170549.6b76715@posting.google.com>, djesys@earthlink.net (David J Dachtera) writes:` !Wayne Sewell <wayne@tachysoft.com> wrote in message news:<00A25EE5.273C42EB.1@tachysoft.com>... !snip!D !I know that in SLS the actual BACKUP runs in the grand-child of the1 !parent process. Is this how TAPESYS works, also?  ! G !If so, do you know if SLS does BACKUP and /RECORD asynchronously also?   J I'm not Wayne, but in fact, SLS does as you describe - the /RECORD pass isF another spawned subprocess, which runs while the "next" backup begins.   !snip! !--- !David J Dachtera  !dba DJE Systems  J __________________________________________________________________________A Bradford J. Hamilton                    "All opinions are my own" K bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-'  0                                          with @"   ------------------------------  % Date: Wed, 17 Sep 2003 05:41:21 -0400 * From: "Bill Todd" <billtodd@metrocast.net>. Subject: Re: Choosing between ASTs and Threads2 Message-ID: <oqycnRR3-IXZsfWiXTWJig@metrocast.net>  7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in message # news:3F679D86.FB6EF39D@istop.com... A > When you take VMS clustering into consideration, then having an  application L > architecture that is multi-process based instead of multi-threaded results inJ > greater capabilities and potential in terms of tolerance to failures and1 > automatic failover when one node fails etc etc.   K Bullshit.  Until you've exhausted the capabilities of the processors on one J node, you're leaving performance on the table (and substituting inter-nodeD communication overheads for the efficient in-process synchronizationE mechanisms).  If you want failure tolerance, you can simply fail that J multi-threaded process over to another cluster node (but run only one copy> of it at any given time in the cluster, avoiding the increasedJ synchronization overheads).  If you also want greater aggregate throughputF than the processors on one node can support, you can run multiple suchK multi-threaded processes concurrently on multiple nodes - but at that point 1 your synchronization latencies take a major jump.    - bill   ------------------------------  # Date: Wed, 17 Sep 2003 11:54:21 GMT - From: "John E. Malmberg" <wb8tyw@qsl.network> 0 Subject: Re: CSwing and OpenVMS v7.3-1, Again...: Message-ID: <NXX9b.358$iT4.227576@news1.news.adelphia.net>   Richard L. Dyson wrote: H > I believe that the problems that I had read about in the past and thenL > experienced first hand with the old CSwing (v3.7.6) program and the v7.3-1 > filesystem have gone away. > K > I recently installed from scratch an AlphaServer 2000 with v7.3-1 and all L > the ECOs and TCPIP v5.3, etc.  I happened to accidentally copy over an oldG > copy of CSwing from another AlphaServer and was using it to read and  
 > discoverD > ODS-5 volumes, etc. when I realized that was not supposed to work! > J > I can only guess that the recent RMS ECO might be the difference?  Does  > anyoneK > else know or if they are aware of the CSwing problems can they check too? L > I seem to recall I couldn't do directory tree "walks" (the S function) forJ > ODS-2 or ODS-5.  There were some other problems too, but I can't recall  > them > just right now.  >  > Does anyone else remember?  8 There are a lot of small programming errors in the code.  C In the exit code, the cleanup handler is not working because it is  C calling a routine with the incorrect number of parameters, and not  E checking the status code.  OpenVMS is cleaning up after it though so   that bug does not show.   G There are many assumptions hard coded about parsing filenames that are  5 not true for for ODS-5 and may not be true for ODS-2.   G It is improperly modifying a context variable in the places that it is  F using IO$_ACCESS instead of LIB$FIND_FILE to get a list of files, and ' not properly checking the status codes.   I As it is a very complex program, it is quite possible that some of these  G bugs will only show up if you have files with names or directory trees   that it can not handle.    -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------   Date: 17 Sep 2003 12:33:43 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Data Recovery9 Message-ID: <bk9kb6$qbqsa$1@ID-135708.news.uni-berlin.de>   D I have a query that is only peripherally OT, but I still figure this( is a good place to get the right answer.  G I have just had a Professor stop by.  He is looking for a Data Recovery I Company that can recover the data from a crashed laptop disk.  Obviously, H cost matters as at some point the cost may exceed the value of the data,C but his first concern is the veracity of the company.  Is there any E company people here may have dealt with that they would recommend?  I D have never had to have it done so I have no personal experience withF any company, but I figured there was bound to be someone in this group who has.  B You can just email any responses as I am sure the group in general doesn't really care.   Thanks.    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  # Date: Wed, 17 Sep 2003 13:23:51 GMT # From: "John Smith" <a@nonymous.com>  Subject: Re: Data RecoveryG Message-ID: <HfZ9b.52916$DZ.42887@news04.bloor.is.net.cable.rogers.com>    On-Trac http://www.ontrack.com/  &  CBL  http://www.cbltech.com/  / Both good and have been around for a long time.        Bill Gunshannon wrote:F > I have a query that is only peripherally OT, but I still figure this* > is a good place to get the right answer. > @ > I have just had a Professor stop by.  He is looking for a DataB > Recovery Company that can recover the data from a crashed laptopE > disk.  Obviously, cost matters as at some point the cost may exceed E > the value of the data, but his first concern is the veracity of the E > company.  Is there any company people here may have dealt with that F > they would recommend?  I have never had to have it done so I have no > personal experience withH > any company, but I figured there was bound to be someone in this group
 > who has. > D > You can just email any responses as I am sure the group in general > doesn't really care. > 	 > Thanks.  >  > bill   ------------------------------  # Date: Wed, 17 Sep 2003 12:58:25 GMT 9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> ; Subject: Re: Difference Intel EFI and Phoenix cME FirstWare 1 Message-ID: <RTY9b.4920$4n.4141@news.cpqcorp.net>   > "Godfried van Loo" <tochmaarniet@hotmail.com> wrote in message7 news:ef7d6766.0309142344.4f02bfa9@posting.google.com...  > Dear all,  > F > Does anybody know the exact difference between Intel EFI and PhoenixH > cME FirstWare? EFI is a replacement for BIOS, but is it true that whatF > EFI will be able to do, Phoenix cME FirstWare already is capable to?4 > What will be a better sollution, EFI or FirstWare?G > EFI is now on all Itanium systems, but will it be on other systems as  > well?  > @ > Anybody any experiences with EFI and/or Phoenix cME FirstWare? >   ? Here is a pointer that might help answer questions about EFI...   3 http://www.intel.com/technology/framework/index.htm    ------------------------------    Date: 17 Sep 2003 07:06:28 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) + Subject: Re: HELP: Using CLI with C program 3 Message-ID: <gUDaQ220ag+v@eisner.encompasserve.org>   w In article <bk72gu$pqkr6$1@ID-56200.news.uni-berlin.de>, "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de> writes:  > G > It starts to work, once you set input.dsc$w_length to inputLen before E > handing the descriptor to cli$dcl_parse, as suggested by Bagbourne.   I    Yes, but only because the input string just happens to be shorter than E    what cli$dcl_parse allows, not because the code was writtento make     sure of it.   ------------------------------  + Date: Wed, 17 Sep 2003 08:56:13 +0100 (BST) = From: =?iso-8859-1?q?amit=20sawhney?= <thick_guy_9@yahoo.com> , Subject: how to add a printer (HP laser jet)@ Message-ID: <20030917075613.57431.qmail@web41106.mail.yahoo.com>   Guys, 4 I am supposed to add a printer to a VMS machine. The6 user said the printer was \\ABC_XYC\EFG_UVW which is a HP laser jet 5j.  5 I have identified some steps. Are they correct? Where ) do I specify the printer path? (\\abc...)   - Here are some of the steps I have identified:  $ run sys$MANAGER:LATCP  LATCP>SHOW PORT  LATCP>CREATE PORT LTA10: /LOG 
 LATCP>EXIT  6 $!NOW, THE PORT IS CREATED. SET UP DEVICE & DEFINE IT., $ SET TEMINAL -_$LTA10:/DEVICE=LTA10/PAGE=66 /NOBROADCAST/PERMANENT $ SET DEVICE* LTA10:/SPOOLED=(LTA10$PRINT,SYS$SYSDEVICE)   $!CREATE THE PRINT QUEUES.) $INITIALIZE/QUEUE/START/DEVICE=TERMINAL - ) _$ /AUTOSTART_ON=(HUDSON::LTA10$PRINT:) -  _$8 /RECORD_BLOCKING/BLOCK_LIMIT=600/CHARACTERISTICS=(EAST)- _$5 /SEPARATE=(NOBURST,NOTRAILER,NOFLAG,RESET=ANSI$RESET)  -  _$6 /DEFAULT=(NOFEED,NOBURST,FLAG=ONE,NOTRAILER,FORM=MEMO) -  _$ /PROCESSOR=LATSYM      LTA10    $ START/QUEUE LTA10   , where do I specify the printer path \\abc...   PLEASE HELP..... thanks in advance  Amit Sawhney   ===== 	 Sincerely  Amit S  H ________________________________________________________________________E Want to chat instantly with your online friends?  Get the FREE Yahoo! + Messenger http://mail.messenger.yahoo.co.uk    ------------------------------  % Date: Wed, 17 Sep 2003 11:31:13 +0300 ; From: "Oleksii Krykun" <krikun@do.not.spam.academy.kiev.ua> 0 Subject: Re: how to add a printer (HP laser jet)Q Message-ID: <1037270357C4D411A1C900A0C9D4BFCBDC6F4D@hqnts40div01.academy.kiev.ua>   I Not enough information. Which network operation system? (PATHWORKS, Samba  etc...)   7 "amit sawhney" <thick_guy_9@yahoo.com> wrote in message : news:20030917075613.57431.qmail@web41106.mail.yahoo.com... > Guys, 6 > I am supposed to add a printer to a VMS machine. The8 > user said the printer was \\ABC_XYC\EFG_UVW which is a > HP laser jet 5j. > 7 > I have identified some steps. Are they correct? Where + > do I specify the printer path? (\\abc...)  > / > Here are some of the steps I have identified:  > $ run sys$MANAGER:LATCP  > LATCP>SHOW PORT  > LATCP>CREATE PORT LTA10: /LOG  > LATCP>EXIT > 8 > $!NOW, THE PORT IS CREATED. SET UP DEVICE & DEFINE IT.. > $ SET TEMINAL -_$LTA10:/DEVICE=LTA10/PAGE=66 > /NOBROADCAST/PERMANENT > $ SET DEVICE, > LTA10:/SPOOLED=(LTA10$PRINT,SYS$SYSDEVICE) >  > $!CREATE THE PRINT QUEUES.+ > $INITIALIZE/QUEUE/START/DEVICE=TERMINAL - + > _$ /AUTOSTART_ON=(HUDSON::LTA10$PRINT:) -  > _$: > /RECORD_BLOCKING/BLOCK_LIMIT=600/CHARACTERISTICS=(EAST)- > _$7 > /SEPARATE=(NOBURST,NOTRAILER,NOFLAG,RESET=ANSI$RESET)  > -  > _$8 > /DEFAULT=(NOFEED,NOBURST,FLAG=ONE,NOTRAILER,FORM=MEMO) > - ! > _$ /PROCESSOR=LATSYM      LTA10  >  > $ START/QUEUE LTA10  > . > where do I specify the printer path \\abc... >    ------------------------------  % Date: Wed, 17 Sep 2003 11:42:04 +0100 ( From: "John Travell" <john@jomatech.com>0 Subject: Re: how to add a printer (HP laser jet)9 Message-ID: <bk9dt1$qlqks$1@ID-120847.news.uni-berlin.de>   9 > "amit sawhney" <thick_guy_9@yahoo.com> wrote in message < > news:20030917075613.57431.qmail@web41106.mail.yahoo.com...	 > > Guys, 8 > > I am supposed to add a printer to a VMS machine. The: > > user said the printer was \\ABC_XYC\EFG_UVW which is a > > HP laser jet 5j. > > 9 > > I have identified some steps. Are they correct? Where - > > do I specify the printer path? (\\abc...)  > > 1 > > Here are some of the steps I have identified:  > > $ run sys$MANAGER:LATCP  > > LATCP>SHOW PORT ! > > LATCP>CREATE PORT LTA10: /LOG  > > LATCP>EXIT > > : > > $!NOW, THE PORT IS CREATED. SET UP DEVICE & DEFINE IT.0 > > $ SET TEMINAL -_$LTA10:/DEVICE=LTA10/PAGE=66 > > /NOBROADCAST/PERMANENT > > $ SET DEVICE. > > LTA10:/SPOOLED=(LTA10$PRINT,SYS$SYSDEVICE) > >  > > $!CREATE THE PRINT QUEUES.- > > $INITIALIZE/QUEUE/START/DEVICE=TERMINAL - - > > _$ /AUTOSTART_ON=(HUDSON::LTA10$PRINT:) -  > > _$< > > /RECORD_BLOCKING/BLOCK_LIMIT=600/CHARACTERISTICS=(EAST)- > > _$9 > > /SEPARATE=(NOBURST,NOTRAILER,NOFLAG,RESET=ANSI$RESET)  > > -  > > _$: > > /DEFAULT=(NOFEED,NOBURST,FLAG=ONE,NOTRAILER,FORM=MEMO) > > - # > > _$ /PROCESSOR=LATSYM      LTA10  > >  > > $ START/QUEUE LTA10  > > 0 > > where do I specify the printer path \\abc...  I The printer is connected to a PC. The path information "\\abc" gives that  away. J If I remember correctly there was a discussion recently about using an LPDG driver on the PC to export access to the printer to systems that cannot ! directly access microsoft shares. L Note, Samba or Pathworks can make selected VMS connected resources availableD to microsoft machines, but not, AFAIK, allow VMS to access resources* resident elsewhere on a microsoft network.K (it would be nice to be able to mount an Ms share as a foreign disk and use $ VMS backup to preserve the contents)     -- John Travell" Independent VMS crashdump analyst. john- at - jomatech - dot - com  +44-(0)23-92552229 http://www.jomatech.com/           --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.518 / Virus Database: 316 - Release Date: 11/09/2003    ------------------------------  + Date: Wed, 17 Sep 2003 05:09:27 -0700 (PDT) . From: Fabio Cardoso <fabiopenvms@yahoo.com.br>0 Subject: Re: how to add a printer (HP laser jet)@ Message-ID: <20030917120927.50660.qmail@web20201.mail.yahoo.com>  , Is not time to the HP Printers Group develop. built-in drivers for HP Printers (at least) ? 0 Would be good a JetDirect Admin integration with the Queue Manager !    Just another idea !   	 Spread...      Fabio C.> --- Oleksii Krykun <krikun@do.not.spam.academy.kiev.ua> wrote:K > Not enough information. Which network operation system? (PATHWORKS, Samba 	 > etc...)  > 9 > "amit sawhney" <thick_guy_9@yahoo.com> wrote in message < > news:20030917075613.57431.qmail@web41106.mail.yahoo.com...	 > > Guys, 8 > > I am supposed to add a printer to a VMS machine. The: > > user said the printer was \\ABC_XYC\EFG_UVW which is a > > HP laser jet 5j. > > 9 > > I have identified some steps. Are they correct? Where - > > do I specify the printer path? (\\abc...)  > > 1 > > Here are some of the steps I have identified:  > > $ run sys$MANAGER:LATCP  > > LATCP>SHOW PORT ! > > LATCP>CREATE PORT LTA10: /LOG  > > LATCP>EXIT > > : > > $!NOW, THE PORT IS CREATED. SET UP DEVICE & DEFINE IT.0 > > $ SET TEMINAL -_$LTA10:/DEVICE=LTA10/PAGE=66 > > /NOBROADCAST/PERMANENT > > $ SET DEVICE. > > LTA10:/SPOOLED=(LTA10$PRINT,SYS$SYSDEVICE) > >  > > $!CREATE THE PRINT QUEUES.- > > $INITIALIZE/QUEUE/START/DEVICE=TERMINAL - - > > _$ /AUTOSTART_ON=(HUDSON::LTA10$PRINT:) -  > > _$< > > /RECORD_BLOCKING/BLOCK_LIMIT=600/CHARACTERISTICS=(EAST)- > > _$9 > > /SEPARATE=(NOBURST,NOTRAILER,NOFLAG,RESET=ANSI$RESET)  > > -  > > _$: > > /DEFAULT=(NOFEED,NOBURST,FLAG=ONE,NOTRAILER,FORM=MEMO) > > - # > > _$ /PROCESSOR=LATSYM      LTA10  > >  > > $ START/QUEUE LTA10  > > 0 > > where do I specify the printer path \\abc... > >  >  >      =====  ========================== Fbio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazil  fabiopenvms@yahoo.com.br ==========================  " __________________________________ Do you Yahoo!?? Yahoo! SiteBuilder - Free, easy-to-use web site design software  http://sitebuilder.yahoo.com   ------------------------------  # Date: Wed, 17 Sep 2003 12:31:45 GMT + From: Ken Robinson <sendspamhere@rbnsn.com> 0 Subject: Re: how to add a printer (HP laser jet)@ Message-ID: <013fe525f0d4f2cb9cabf868ffe488fb@news.teranews.com>  - amit sawhney <thick_guy_9@yahoo.com> wrote in C news:20030917075613.57431.qmail@web41106.mail.yahoo.com (in part):     > Guys, 6 > I am supposed to add a printer to a VMS machine. The8 > user said the printer was \\ABC_XYC\EFG_UVW which is a > HP laser jet 5j. >   H [I posted this earlier, but it never showed up. Apologies if it appears  twice ... ken]  E That is the designation on the PC for the printer which is not a good H address for VMS. Does the printer have an IP address? Or is it connected* via a printserver that has an IP address?   C If it is connected via a Printserver, does the Printserver run LAT?   D You didn't say what version of VMS you're running. If you're running2 TCP/IP, what version (and stack) are you running?   C The easiest way to connect HP Laserjets to VMS is with DCPS (if the F printer supports PostScript) via either IP or LAT. If the printer doesH not support Postscript, use the LPR/LPQ print services or use the telnetD print symbiont if the printer supports print via port 9100 or 2100.   ) Tell us some details and we can help you.    Ken Robinson" kenrbnsn1 (at) patmedia (dot) net    ------------------------------    Date: 17 Sep 2003 07:59:57 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 0 Subject: Re: how to add a printer (HP laser jet)3 Message-ID: <cLujcqyka8zh@eisner.encompasserve.org>    In article <20030917075613.57431.qmail@web41106.mail.yahoo.com>, =?iso-8859-1?q?amit=20sawhney?= <thick_guy_9@yahoo.com> writes: > Guys, 6 > I am supposed to add a printer to a VMS machine. The8 > user said the printer was \\ABC_XYC\EFG_UVW which is a > HP laser jet 5j. >   E    That's a Microsoft style path name which allows a client to access B    a printer known as EFG_EVW via the server ABC_XYC.  It's been a?    long time since I used Pathworks and I don't know much about G    Samba, but someone out there might have a client for VMS that allows      it to make such a connection.     G    But do you really want your VMS printouts going through a PC server?   F    The 5j is sitting on the network and probably supports both streamsD    and LPD printing, either of which would allow you to print to theF    5j directly from the VMS system.  That's how we do it.  You'll needF    the IP address of the printer.  If you're goind to use streams, theB    port is probably 9100, if your going to use LPD you'll need the%    printer's internal LDP queue name.      E    If the printer has only been used as a PC printer, the IP name and E    LPD queue name may not be set up.  Also it's possible to alter the G    port number.  Most of this can be determined from the front panel of G    most HP printers, or you can download software for a Mac or PC which '    will allow you to query the printer.    ------------------------------  # Date: Wed, 17 Sep 2003 14:45:34 GMT * From: Paul Anderson <paul.anderson@hp.com>0 Subject: Re: how to add a printer (HP laser jet)5 Message-ID: <170920031045341665%paul.anderson@hp.com>e  ? In article <cLujcqyka8zh@eisner.encompasserve.org>, Bob KoehlerH0 <koehler@eisner.nospam.encompasserve.org> wrote:  D > If you're goind to use streams, the port is probably 9100, if yourE > going to use LPD you'll need the printer's internal LDP queue name.f  B Make sure TCP/IP is enabled on the printer.  You can check this by= printing a configuration page from the printer's front panel.d  E The port _is_ 9100 for a raw socket port.  If you use LPD instead, nor  internal queue name is required.  F In article <20030917120927.50660.qmail@web20201.mail.yahoo.com>, Fabio) Cardoso <fabiopenvms@yahoo.com.br> wrote:j  F > Is not time to the HP Printers Group develop built-in drivers for HPC > Printers (at least) ? Would be good a JetDirect Admin integrationa > with the Queue Manager !  C The closest thing to built-in drivers for OpenVMS is DCPS.  I'm notPD sure what you mean by integration between the JetDirect card and the> queue manager.  The JetDirect card gets its print jobs via theE protocols it supports (for OpenVMS, that's raw IP, LPD and AppleTalk) C and there would be nothing further to be gained from any additionala integration with the card.   Paul   -- t  Paul Anderson   OpenVMS Engineering    Hewlett-Packard CompanyA   ------------------------------  % Date: Wed, 17 Sep 2003 11:23:45 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>0 Subject: Re: how to add a printer (HP laser jet)) Message-ID: <3F687C7F.731C3881@istop.com>o  L If your printer has LAT support, then you need two commands (in your example you only had one).  % LATCP> CREATE PORT LTA10:/APPLICATION + LATCP> SET PORT LTA10/NODE=LAT_0011223344550  Q The digits in the node name are essentially the ethernet address of your printer.    You can then test this with:   COPY myfile.ps LTA10:e  H If your printer does not support lat, then you need to provide a list of supported protocols.  L (If your printer is attached to some wintel PC, you need to provide the list3 of protocols your wintel PC supports for printing).n  9 Out of the box, VMS supports serial, LAT, telnet, LPD/LPRs  C However, if you want full functionality on VMS with DCPS which willoK automatically translate text files into postscript etc (on the fly), then IsM think that serial and LAT are the only supported ones for now. (Paul Anderson ! can probably correct me on this).n   ------------------------------    Date: 17 Sep 2003 01:21:15 -0700# From: dooleys@snowy.net.au (dooley)r6 Subject: Re: How to use shareable image to share data?= Message-ID: <1ca82fc6.0309170021.167b71b2@posting.google.com>s  j tadamsmar@yahoo.com (Tom Adams) wrote in message news:<ea44f5a1.0309161228.713de53f@posting.google.com>...A > I am trying to set up a shareable image for sharing a data area 	 > between H > processes.  I have an old app that was developed on the Vax and is notG > working on the Alpha (7.2-1). It uses shareable images to communicateeD > data.  I set up a simple test case and it is also not working, not > sharing data.t9 from: http://h18009.www1.hp.com/fortran/migrating-va.htmlt@ The VAX compiler gives COMMON blocks the SHR (shareable) programC section (PSECT) attribute while the AXP compiler defaults to NOSHR.   2 The linker has also changed so you may have to use8 SYMBOL_VECTOR=(name=PSECT) and SYMBOL_VECTOR=(name=DATA)  B Alignment will probably also be an issue, use /MAP in both compileC and link to check that you and the compiler agree on the alignment.n Phil   ------------------------------    Date: 17 Sep 2003 07:32:17 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)r6 Subject: Re: How to use shareable image to share data?3 Message-ID: <hR+BmkqvQGdy@eisner.encompasserve.org>>  U In article <3F677B31.AA20425@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes:  > Tom Adams wrote:E >> I have demonstrated in various ways that the executable X does noteE >> share the common /TEST/ when I run the executable in two differentm
 >> processes.R >> U >> What am I doing wrong?D > N > Isn't that how shareable images are supposed to behave ? the code is common, > but each has its own data.  F    Not quite.  What is shared and what is private depends on the PSECT    attributes.  G > Shouldn't you be using global sections instead of a shareable image ?a  G    Both are possible, and I wouldn't change the existing technique just D    to do a VAX to Alpha port, but I've found global sections presentD    much less of a maintenance and privilege headache.  I was able toE    dramatically improve the maintenance of a system and remove CMKRNL6G    privileges from lots of accounts when I made that change to a systemh    I inherited.     e   ------------------------------    Date: 17 Sep 2003 07:42:41 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)i6 Subject: Re: How to use shareable image to share data?3 Message-ID: <ti9a3wIpO8ry@eisner.encompasserve.org>t  e In article <ea44f5a1.0309161228.713de53f@posting.google.com>, tadamsmar@yahoo.com (Tom Adams) writes:g  C    Alpha changed the PSECT attribute for Fortran COMMON blocks from=@    SHR to NOSHR, and requires a symbol_vector option to make the-    PSECT name visible in the shareable image.   @    To combine what others have said to a single, working answer:   > 	LINK/SHAR C.FOR  B    I hope that's a typo, you didn't really link the .FOR, did you?  D    You need the following options for this link on Alpha (I put them
    in c.opt):i         psect=test,shr        symbol_vector=(test=psect)   > and X.OPT: >    clognam/share  +    For this link you need to add an option:          psect=test,shr   Then you get what you want:e   $ run xt           0o           5o $ run xi           5            5i   ------------------------------  % Date: Wed, 17 Sep 2003 08:49:41 -0700b+ From: "Barry Treahy, Jr." <Treahy@MMaz.com>t6 Subject: Re: How to use shareable image to share data?' Message-ID: <3F688295.6020401@MMaz.com>e   JF Mezei wrote::   >Tom Adams wrote:e >  n >nD >>I have demonstrated in various ways that the executable X does notD >>share the common /TEST/ when I run the executable in two different >>processes. >> >>What am I doing wrong? >>     >> >iM >Isn't that how shareable images are supposed to behave ? the code is common,i >but each has its own data.i >yF >Shouldn't you be using global sections instead of a shareable image ? >n >  t >eG What he is attempting to do, I believe, is a shared global section for s, common data, between various applications...   Barrys     -- .  > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                            ------------------------------    Date: 17 Sep 2003 01:04:03 -0500+ From: young_r@encompasserve.org (Rob Young)s: Subject: Re: HP extends lead in open storage area networks3 Message-ID: <Cwt1C8g0+NaY@eisner.encompasserve.org>   V In article <3F679DF8.55672CA9@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes:O > And Steve Jobs, in a BBC interview, said that there were only 2 profitable PC I > businesses: Dell because they are high volume and efficient producer ofEH > commodity stuff, and Apple because of their innovation. All others areV > struggling and must subsidize their PC operations with profits from other divisions.    A 	He says that because he can, sounds good and is awe-inspiring.  eE 	Apple made $9 million this last quarter.  Pop the champagne corks.  aF 	Apple lost money the prior 3 quarters.  As if Apple isn't struggling.  @ 	Dell made $811 million this quarter, $819 million last quarter,@ 	$758 million the prior quarter, $677 million the quarter before
 	that one.  4 	Yeah, Apple and Dell.  Just like Yugo and Mercedes.   				Rob.   ------------------------------    Date: 17 Sep 2003 01:09:07 -0500+ From: young_r@encompasserve.org (Rob Young) : Subject: Re: HP extends lead in open storage area networks3 Message-ID: <eHTrJhwPGL5F@eisner.encompasserve.org>b  a In article <Cwt1C8g0+NaY@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes:eX > In article <3F679DF8.55672CA9@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes:P >> And Steve Jobs, in a BBC interview, said that there were only 2 profitable PCJ >> businesses: Dell because they are high volume and efficient producer ofI >> commodity stuff, and Apple because of their innovation. All others aretW >> struggling and must subsidize their PC operations with profits from other divisions.K >  > C > 	He says that because he can, sounds good and is awe-inspiring.  fG > 	Apple made $9 million this last quarter.  Pop the champagne corks.  tH > 	Apple lost money the prior 3 quarters.  As if Apple isn't struggling. >    	Past quarter:  I > 	Dell made $811 million this [past] quarter, $819 million last quarter,aB > 	$758 million the prior quarter, $677 million the quarter before > 	that one. > 6 > 	Yeah, Apple and Dell.  Just like Yugo and Mercedes. > 	 > 				Roba >    ------------------------------  # Date: Wed, 17 Sep 2003 14:50:49 GMTd* From: Paul Anderson <paul.anderson@hp.com>: Subject: Re: HP extends lead in open storage area networks5 Message-ID: <170920031050480547%paul.anderson@hp.com>   = In article <Cwt1C8g0+NaY@eisner.encompasserve.org>, Rob YoungK" <young_r@encompasserve.org> wrote:  5 > Yeah, Apple and Dell.  Just like Yugo and Mercedes.   / Mercedes and Ford would be a better comparison.o  8 $ (PROFITS .NE. QUALITY) .AND. (PROFITS .NE. INNOVATION)   Paul   -- m  Paul Anderson   OpenVMS Engineeringn   Hewlett-Packard Companyv   ------------------------------  % Date: Wed, 17 Sep 2003 21:38:45 +1000i: From: "David McKenzie" <david.mckenzie@paradigm-shift.biz>Y Subject: Re: I have a fix!!!  Re: Anyone else having problems with VMS and HSG80-based arrC Message-ID: <3f6847ba$0$95050$c30e37c6@lon-reader.news.telstra.net>J  J No, I mean that there were various flavours of kgpsa and there ealier ones did not work under high load.a  D In particular lp7000's displayed bad behavious and lp8000's did not.   -- g David McKenzie  7 David.Mckenzie@paradigm-shift.dot.biz  remove the "dot"l OpenVMS IT Privacy and Law   http://www.paradigm-shift.bizr1 "Scott Vieth" <svieth@wi.rr.com> wrote in messaget7 news:5a85bce2.0309131145.507a2362@posting.google.com...pG > "David McKenzie" <david.mckenzie@paradigm-shift.biz> wrote in messagee? news:<3f62dad8$0$95043$c30e37c6@lon-reader.news.telstra.net>...yB > > I have also seen this sort of problem with out of date KGPSA's > >a > > -- m > > David McKenzie > >o; > > David.Mckenzie@paradigm-shift.dot.biz  remove the "dot"n > > OpenVMS IT Privacy and Law >oF > Do you mean firmware that it out-of-date?  I am very careful to keepG > my ES40 up-to-date whenever new AlphaServer firmware is announced.  I.H > let the firmware update upgrade *all* of the pieces in the system that > it can upgrade.  > H > I'll double-check the firmware version this afternoon when the ES40 is, > rebooted to install the 731-MSA1000 patch. > 	 > Thanks,d >u > -Scott   ------------------------------    Date: 17 Sep 2003 01:38:30 -0500+ From: young_r@encompasserve.org (Rob Young)e) Subject: Re: IA64 to be dual core by 2005e3 Message-ID: <s40V+i+z97GL@eisner.encompasserve.org>   c In article <Pine.LNX.4.56.0309161906170.6107@jaipur>, Ryan Moore <rmoore@rmoore.dyndns.org> writes:*& > On Tue, 16 Sep 2003, JF Mezei wrote: >> question: >>N >> what does dual core mean from the point of view of the operating system and >> scheduling ?c >>P >> Does it appear as a single (faster) chip to the OS, or does the OS see 2 cpus! >> which it can use for SMP etc ?S > K > Assuming it's the same sort of technology as the new P4 "hyperthreading", L > then it would show up as two CPUs on one piece of silicon.  In the case ofL > P4 "hyperthreading", there are two sets of execution units, but some otherK > components are shared.  Thus, the CPU isn't as fast as two complete CPUs.n > K > The more componenets that are shared between the cores, the less speed-upw > that can be accomplished.n > K > VMS would see two CPUs and would do SMP scheduling.  There would probably 0 > need to be support in the OS to do it, though. >   5 	2 CPUs seen as one.  Something I speculated on here:i  j http://www.realworldtech.com/forums/index.cfm?action=detail&PostNum=1396&Thread=36&roomID=11&entryID=17658  A 	And the same info quoted from news.com back to comp.os.vms (thish	 	article:   d http://groups.google.com/groups?selm=d7791aa1.0301181143.23c3574c%40posting.google.com&output=gplain  ( From: bob@instantwhip.com (Bob Ceculski) Newsgroups: comp.os.vmst0 Subject: Itanium ahead of itself, not behind ...  Date: 18 Jan 2003 11:43:39 -0800    C Besides enhancing performance, Intel may use its dual-core chips to E undercut IBM, said Kevin Krewell, senior editor at the Microprocessor E Report. In larger servers, Oracle and other software vendors charge aT4 licensing fee for every processor in a given server.  F To date, IBM has said that the Power4, although a single chip, has twoB processors. Hence, software customers have to buy two licenses forG each Power4 chip. >>> INTEL IS ALREADY INDICATING THAT IT WILL CONSIDERtB MONTECITO A SINGLE PROCESSOR, REQUIRING ONLY ONE SOFTWARE LICENSE, KREWELL SAID. <<<=                   F "I think they are going to sell it that way to make it cost effective"E to switch to Itanium from other servers with different chips, Krewell3 said.5     	Yesterday:d  ) http://www.theinquirer.net/?article=11579u  K The big announcement is Vanderpool, or Vanderpool Technology (VT), which is%L Intel's way of putting VMWare functionality on a chip. The time for hardwareG partitioning is near, or at least the time for more announcements aboutyN hardware partitioning is near, they didn't put a time frame on this technology) family.  [How about 2005 with Monticeto?]U  O In short, it allows you to run two virtual CPUs on a single CPU, basically like J having a second PC in the box. This is not HT technology, which emulates aL second CPU, the demo they showed had one 'computer' playing a Simpsons clip,M and the other playing a game, rebooting, and then installing drivers. NothingsE that you can't do now in software, the take home message is that this_D technology will decrease, or possibly eliminate the overhead of this+ increasingly important corporate mainstay. y   ---   1 	The advantages are huge as I speculated earlier:S  j http://www.realworldtech.com/forums/index.cfm?action=detail&PostNum=1396&Thread=36&roomID=11&entryID=17658   There is a huge incentive.  [ http://www.intel.com/technology/itj/2002/volume06issue01/art03_specprecomp/p05_hw_vs_sw.htmr  K "Initial results indicate that the prefetching performed by the speculative.K threads can achieve significant speed-ups on an in-order processor, ranging 3 from 16% to 104%, on pointer-intensive benchmarks."   4 Imagine if it sped up Oracle/MSQL 40-80% per Itanium$ CPU without increasing the external  CPU count thereby keepinga/ down expensive licensing costs. Just a thought.a   ---a  
 	Speculation:a  ? 	Oracle WILL be redoing their licensing.  However, those $40000a? 	per-CPU licensing costs that you pay for today on 8, 9i and 11a< 	will become a WHOLE lot better with Monticeto.  The ROI may> 	be: swap out your current box for Monticeto and save money inA 	the process.  Seems that will be the case (for the common case -l? 	i.e. running out of horsepower in that 4-cpu box - get a 4-cpur5 	Montecito which provides 50-70% higher performance).r   				Robt   ------------------------------    Date: 17 Sep 2003 01:56:15 -0500+ From: young_r@encompasserve.org (Rob Young)e) Subject: Re: IA64 to be dual core by 2005h3 Message-ID: <3WJFt5TDrt$n@eisner.encompasserve.org>o  a In article <s40V+i+z97GL@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes:-    
 > 	Yesterday:  > + > http://www.theinquirer.net/?article=11579h > M > The big announcement is Vanderpool, or Vanderpool Technology (VT), which is.N > Intel's way of putting VMWare functionality on a chip. The time for hardwareI > partitioning is near, or at least the time for more announcements about P > hardware partitioning is near, they didn't put a time frame on this technology+ > family.  [How about 2005 with Monticeto?]e > Q > In short, it allows you to run two virtual CPUs on a single CPU, basically likeTL > having a second PC in the box. This is not HT technology, which emulates aN > second CPU, the demo they showed had one 'computer' playing a Simpsons clip,O > and the other playing a game, rebooting, and then installing drivers. NothingsG > that you can't do now in software, the take home message is that thiseF > technology will decrease, or possibly eliminate the overhead of this- > increasingly important corporate mainstay.   >   = 	Vanderpool technology is a ways off so how Montecito appearsr- 	as a single CPU is still a bit of a mystery.u  M http://asia.reuters.com/newsArticle.jhtml?type=technologyNews&storyID=3454786r  I Vanderpool, which will be available in the next five years, will create arJ partition inside the chip to allow it to be used for different programs onN different machines. For instance, Otellini demonstrated a chip with VanderpoolN technology that allowed him to watch a TV program on a TV while another person was playing a video game.     C 	Montecito uses an "arbiter" but how that works or what it does is rD 	speculation (maybe it is still doing speculative computation - but A 	that is a guess and a high risk of being totally wrong and mayber: 	not nearly the time to verify and stuff it into silicon):  1 http://www.eetimes.com/semi/news/OEG20030210S0007   O Then, in 2005, Intel will unveil the dual-core Montecito. Each core will have aaL processor and L3 cache in the same package. Intel has also developed what isM calls an "arbiter" bus technology, which will manage two cores as one systemse bus. l  M The "arbiter" itself appears to be an internal bus that will interface to theuH overall systems bus. The "arbiter" technology will enable a processor to: support twice the cache over previous devices, Modi said.   K Little else was disclosed about the "arbiter" but the technology is key for J high-end servers. "The fundamental driver is the cache," he said. "You get8 significantly better latencies with more on-die cache."   K The "arbiter" will support the dual-core Montecito, but Intel dropped hintsoO that it may be required to handle multi-core processors. "We are not ruling outh, multi-core systems in the future," he said.    ---i  9 	Maybe the whole business of appearing as a single CPU isi" 	wrong as Bill pointed out before.   				Roby   ------------------------------  # Date: Wed, 17 Sep 2003 13:05:29 GMTn9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>-) Subject: Re: IA64 to be dual core by 2005c1 Message-ID: <t_Y9b.4921$Qu.3630@news.cpqcorp.net>f  7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messagen# news:3F6785FD.D135B863@istop.com...e > question:- >-I > what does dual core mean from the point of view of the operating systemh andt > scheduling ? > J > Does it appear as a single (faster) chip to the OS, or does the OS see 2 cpus  > which it can use for SMP etc ?  I Dual core looks exactly like 2 CPUs in a single CPU socket.  So it allowse@ 2-socket systems to become 4-way, 4-socket to become 8-way, etc.   ------------------------------  # Date: Wed, 17 Sep 2003 13:05:42 GMTa9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> ) Subject: Re: IA64 to be dual core by 2005a1 Message-ID: <G_Y9b.4922$Mu.3632@news.cpqcorp.net>o  8 "Ryan Moore" <rmoore@rmoore.dyndns.org> wrote in message/ news:Pine.LNX.4.56.0309161906170.6107@jaipur...e& > On Tue, 16 Sep 2003, JF Mezei wrote:
 > > question:  > >nK > > what does dual core mean from the point of view of the operating systemr ande > > scheduling ? > >oL > > Does it appear as a single (faster) chip to the OS, or does the OS see 2 cpus" > > which it can use for SMP etc ? >gK > Assuming it's the same sort of technology as the new P4 "hyperthreading",uL > then it would show up as two CPUs on one piece of silicon.  In the case ofL > P4 "hyperthreading", there are two sets of execution units, but some otherK > components are shared.  Thus, the CPU isn't as fast as two complete CPUs.  > K > The more componenets that are shared between the cores, the less speed-up  > that can be accomplished.i >hK > VMS would see two CPUs and would do SMP scheduling.  There would probablye0 > need to be support in the OS to do it, though. >x  K Look mom, no hands!  Dual core for the most part will be transparent to them& OS, and just doubles the CPU capacity.   ------------------------------  % Date: Wed, 17 Sep 2003 11:09:34 -0400h* From: JF Mezei <jfmezei.spamnot@istop.com>) Subject: Re: IA64 to be dual core by 2005b) Message-ID: <3F68792D.6DCF75E6@istop.com>    Rob Young wrote:I > each Power4 chip. >>> INTEL IS ALREADY INDICATING THAT IT WILL CONSIDER D > MONTECITO A SINGLE PROCESSOR, REQUIRING ONLY ONE SOFTWARE LICENSE, > KREWELL SAID. <<</  N OK then. As a single processor, what is the advantage of a dual core ? If theyK insist on appearing as a single CPU to the OS, shouldn't they then focus onLN adding more execution pipelines (or whatever the terminology is) ? What is theN point of a dual core if the second core cannot act independantly as a separate CPU ?t  H > "I think they are going to sell it that way to make it cost effective"G > to switch to Itanium from other servers with different chips, Krewell$ > said.,  N I think it is easy for Intel to talk like this now since their dual core stuffJ is so far away. It may be much harder for them to be cost competitive if aM single CPU costs a lot more because it has a dual core but you're not allowed  to use the second core.   N > Intel's way of putting VMWare functionality on a chip. The time for hardwareI > partitioning is near, or at least the time for more announcements about.P > hardware partitioning is near, they didn't put a time frame on this technology+ > family.  [How about 2005 with Monticeto?]_  L Well, this is no surprise since existing chips/systems such as Alpha alreadyL have the ability to be building blocks of systems that support partitioning.M So Intel has to add that functionality to its  IA64 if IA64 is to survive and  replace Alpha and PaRisc.O  L However, where partitioning woudl be needed is really in the wintel world soN that you could run multiple instances of Windows on a single big box, allowingG you to minimize the hardware and still have one instance of windows perI6 application to minimise the impact of windows hanging.   ------------------------------  % Date: Wed, 17 Sep 2003 11:27:12 -0400R* From: JF Mezei <jfmezei.spamnot@istop.com>) Subject: Re: IA64 to be dual core by 2005F) Message-ID: <3F687D4E.5847BEBD@istop.com>M   Fred Kleinsorge wrote:M > Look mom, no hands!  Dual core for the most part will be transparent to thet( > OS, and just doubles the CPU capacity.  K Are you talking about Intel's specific plans for IA64, or is that a general_> statement applicable to existing platforms such as Power etc ?   ------------------------------  % Date: Wed, 17 Sep 2003 11:29:36 -0400m* From: JF Mezei <jfmezei.spamnot@istop.com>) Subject: Re: IA64 to be dual core by 20051( Message-ID: <3F687DDD.D0A685A@istop.com>  " Fred Kleinsorge wrote in one post:L >Look mom, no hands!  Dual core for the most part will be transparent to the' >OS, and just doubles the CPU capacity.y    & Fred Kleinsorge wrote in another post:K > Dual core looks exactly like 2 CPUs in a single CPU socket.  So it allowscB > 2-socket systems to become 4-way, 4-socket to become 8-way, etc.    N OK, which is which ? Will the OS see a single CPU or will it see multiple CPUs allowing SMP etc etc ?   ------------------------------    Date: 17 Sep 2003 12:16:07 -0500+ From: young_r@encompasserve.org (Rob Young)e) Subject: Re: IA64 to be dual core by 2005?3 Message-ID: <3WILMlYtkTrm@eisner.encompasserve.org>   V In article <3F68792D.6DCF75E6@istop.com>, JF Mezei <jfmezei.spamnot@istop.com> writes: > Rob Young wrote:J >> each Power4 chip. >>> INTEL IS ALREADY INDICATING THAT IT WILL CONSIDERE >> MONTECITO A SINGLE PROCESSOR, REQUIRING ONLY ONE SOFTWARE LICENSE,I >> KREWELL SAID. <<< > P > OK then. As a single processor, what is the advantage of a dual core ? If theyM > insist on appearing as a single CPU to the OS, shouldn't they then focus onCP > adding more execution pipelines (or whatever the terminology is) ? What is theP > point of a dual core if the second core cannot act independantly as a separate > CPU ?E >    	<Backpedling mode>R  > 	As Bill pointed out in realworldtech.com when this discussion@ 	came up a few weeks ago ... Krewell may have it wrong.  I can't@ 	find definitive evidence of what Montecito's "arbiter" does, if= 	Montecito does indeed have Speculative Precomputation or anye1 	mechanism to "appear as a single CPU" to the OS.0  I >> "I think they are going to sell it that way to make it cost effective" H >> to switch to Itanium from other servers with different chips, Krewell >> said. > P > I think it is easy for Intel to talk like this now since their dual core stuffL > is so far away. It may be much harder for them to be cost competitive if aO > single CPU costs a lot more because it has a dual core but you're not allowedb > to use the second core.s >   D 	It isn't far away at all.  2005 is 15 months away - maybe 19 months  	if they ship Montecito Q2 2005.  O >> Intel's way of putting VMWare functionality on a chip. The time for hardware J >> partitioning is near, or at least the time for more announcements aboutQ >> hardware partitioning is near, they didn't put a time frame on this technology , >> family.  [How about 2005 with Monticeto?] > N > Well, this is no surprise since existing chips/systems such as Alpha alreadyN > have the ability to be building blocks of systems that support partitioning.O > So Intel has to add that functionality to its  IA64 if IA64 is to survive andD > replace Alpha and PaRisc.N  > 	VMWare or Vanderpool appears to be across CPU lines.  Perhaps5 	Intel steps up and explains it or that info shows upI 	on the WibblyWeb.  N > However, where partitioning woudl be needed is really in the wintel world soP > that you could run multiple instances of Windows on a single big box, allowingI > you to minimize the hardware and still have one instance of windows perh8 > application to minimise the impact of windows hanging.  ; 	It was late and I was winging on - on reinspection I thinkb; 	Vanderpool technology is LPARs without the LPAR - meaning s< 	virtual CPUs carved out of 1 or more cores, if I am reading 	this correctly:  8 http://www.internetnews.com/dev-news/article.php/3078291  E Intel developers also are working on a separate technology code-namedsN "Vanderpool," a new performance enhancement that enables multiple, independentO software environments in a single PC. As Otellini described it, this technologyyO enables users to partition their home computers just as a network administratorsM might partition a mainframe-class system, increasing reliability and speedinge. the ability to recover from computer crashes.   ( 	So a number of technologies are baking. 		: 		1)  Multiple cores that appear as one - Maybe Montecito?) 			Advantage:  Decreased licensing costs.>  ; 		2)  Vanderpool technology - single CPU appear as multiplee? 		    (or virtual CPUs or systems?) such that processes run in   			a separate domain.u8 			Advantage:  Application crash - just that application% 			goes down, others continue to run..  @ 	Questions abound.  How does an OS take advantage of Vanderpool? 	Is it transparent?m   				RobS   ------------------------------  # Date: Wed, 17 Sep 2003 17:32:06 GMTi9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>.) Subject: Re: IA64 to be dual core by 2005 0 Message-ID: <qU0ab.4955$QI.936@news.cpqcorp.net>  K Get a grip.  I said the same thing.  Dual core to SW looks like 2 CPUs.  By J doubling the CPU capacity, I mean that the same number of "sockets" allowsE you to effectively have twice the CPU count - doubling a machines CPUv	 capacity.d  C Now, how Intel views the *chip* is something different.  From theirEG pricing/licensing/etc pov, they may view the part as a single CPU chip.   K How an OS will treat it is something different.  *As* may be how an OS does * license checks -- which is a policy issue.    7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messageR" news:3F687DDD.D0A685A@istop.com...$ > Fred Kleinsorge wrote in one post:J > >Look mom, no hands!  Dual core for the most part will be transparent to theL) > >OS, and just doubles the CPU capacity.  >S >R( > Fred Kleinsorge wrote in another post:F > > Dual core looks exactly like 2 CPUs in a single CPU socket.  So it allowsD > > 2-socket systems to become 4-way, 4-socket to become 8-way, etc. >  >1K > OK, which is which ? Will the OS see a single CPU or will it see multiple  CPUs > allowing SMP etc etc ?   ------------------------------  + Date: Wed, 17 Sep 2003 16:48:40 +0000 (UTC)a' From: "Harry" <Harry.uk@btinternet.com>= Subject: Re: KZPBA Question_0 Message-ID: <bka398$6bq$1@sparta.btinternet.com>  L I think its unsupported.... You can try and unhook the internal disk and seeK what you get. Then try the same for the externals. That will work, but bothmJ together is a no go. I can't remember if it was just wide drives or mixingL wides with narrow external drives. Another controller is the best way to go. Harrym    & <healyzh@aracnet.com> wrote in message& news:bk8e4i11i3p@enews2.newsguy.com...J > On a DEC PWS 433au, can you have disks attached to both the Internal andK > External Wide connections?  I seem to recall that there is a problem withs
 > doing this.j >5K > Is this simply something that is unsupported like using both the InternaleH > Narrow and Wide connectors, or is this something that just plain won't work.s >tB > I can see the disk on the internal connector, but can't INIT it: >t > $ init dka0: testn# > %INIT-F-MEDOFL, medium is offlinef > $s >n > Zane >a >i   ------------------------------  % Date: Wed, 17 Sep 2003 09:26:18 -0400V& From: "Island" <dbturner@islandco.com>" Subject: Re: Memory for VS 4000-xx/ Message-ID: <vmgocbgpdbbb8b@news.supernews.com>e   AlainV   Vous vivez encore !!!!     -- u David B Turner Island Computers US Corporationr 2700 Gregory St., Suite 180c Savannah GA 31404P Tel: 912 447 6622  Fax: 912 201 0402n Email: dbturner@hpaq.net http://www.hpaq.netr    1 "Brokline" <brokline@wanadoo.fr> wrote in messagep, news:bk7brq$m0i$2@news-reader3.wanadoo.fr...D > je suis le dernier digital broker en France je vais vous faire une  > proposition sur cette mmoire. >s
 > Salutationsv< > Alain Hod / www.digital-equipment.com/ tel 01 72 74 00 13 >  >) >e@ > "BAVAY Marc" <mbavay@slb.com> a crit dans le message de news: > 3F672F29.4B28AA4E@slb.com... > >a	 > > hellom > >i9 > > I am searching for 16Mo memory boards for VS 4000-90A:B > > possible references could be KTV-4000-16, MS44-CA, 54-19103-?? > >rJ > > anybody know a place on the web, or a broker who sells used stuff like > that ? > > what could be the price ?h > >l+ > > thank you very much for any informationa > > marc >  >    ------------------------------  % Date: Wed, 17 Sep 2003 17:26:44 +0200' From: jf.pieronne@laposte.netu! Subject: MySQL 4.0.15 for OpenVMSi2 Message-ID: <bk9ufl$as3$1@news-reader4.wanadoo.fr>   Hi all,   D the latest stable MySQL release 4.0.15 is now available for OpenVMS.A Can be download from http://www.pi-net.dyndns.org/anonymous/kits/h or from one of the two mirrors:t% http://erebus.homeip.net/mirror/kits/o# http://www.vsm.com.au/ftp/jfp/kits/t  M Please download from the mirrors, they are much faster than the primary site.'   Thanks,     
 Jean-Franois    ------------------------------  # Date: Wed, 17 Sep 2003 10:15:56 GMTt) From: "Stefano Borselli" <ibbo@libero.it> & Subject: Re: Newbie question about NTP: Message-ID: <wvW9b.322559$Ny5.10208825@twister2.libero.it>   My version of  TCP/IP is:e  >   Digital TCP/IP Services for OpenVMS VAX Version V4.2 - ECO 4(   on a VAX 7000-740 running OpenVMS V7.1  ! and in my system don't exist ntp.P% I try to dowload it from www.ntp.org.e   Thanks to all.   Stefano.   ------------------------------  % Date: Wed, 17 Sep 2003 14:02:20 +0200 $ From: "MikeR" <rechtman@tzora.co.il>& Subject: Re: Newbie question about NTP9 Message-ID: <bk9eu7$q1tfd$1@ID-103225.news.uni-berlin.de>t  4 "Stefano Borselli" <ibbo@libero.it> wrote in message4 news:wvW9b.322559$Ny5.10208825@twister2.libero.it... > My version of  TCP/IP is:r >s@ >   Digital TCP/IP Services for OpenVMS VAX Version V4.2 - ECO 4* >   on a VAX 7000-740 running OpenVMS V7.1 >a# > and in my system don't exist ntp.t' > I try to dowload it from www.ntp.org.n >u > Thanks to all. >p
 > Stefano. >p@ IIRC NTP on OpenVMS TCPIP only became possible circa V5.1 or 5.3     Mike   ------------------------------  # Date: Wed, 17 Sep 2003 12:44:53 GMT04 From: brad@.gateway.2wire.net (Bradford J. Hamilton)& Subject: Re: Newbie question about NTP/ Message-ID: <9HY9b.372722$Oz4.149361@rwcrnsc54>r  ` In article <bk9eu7$q1tfd$1@ID-103225.news.uni-berlin.de>, "MikeR" <rechtman@tzora.co.il> writes: !n5 !"Stefano Borselli" <ibbo@libero.it> wrote in messagen5 !news:wvW9b.322559$Ny5.10208825@twister2.libero.it...o !> My version of  TCP/IP is: !>A !>   Digital TCP/IP Services for OpenVMS VAX Version V4.2 - ECO 4 + !>   on a VAX 7000-740 running OpenVMS V7.1w !>$ !> and in my system don't exist ntp.( !> I try to dowload it from www.ntp.org. !> !> Thanks to all.  !> !> Stefano.  !>A !IIRC NTP on OpenVMS TCPIP only became possible circa V5.1 or 5.3  !p   It goes back as far as UCX 4.0:-  
 $ uc sho vers1  6   Digital TCP/IP Services for OpenVMS VAX Version V4.0)   on a VAX 4000-600A running OpenVMS V6.2n   $ @sys$manager:ucx$configD .3 .7 ..  " (Option 11 under the SERVER menu):  K         Digital TCP/IP Services for OpenVMS SERVER Components Configuratione Menu           Configuration options:  .                  1  -  BIND           Disabled.                  2  -  BOOTP          Disabled.                  3  -  TFTP           Disabled-                  4  -  FTP            Enablede.                  5  -  LPR/LPD        Disabled.                  6  -  NFS            Disabled.                  7  -  PC-NFS         Disabled-                  8  -  PORTMAPPER     Enableds-                  9  -  TELNET/RLOGIN  Enabled .                 10  -  SNMP           Disabled.                 11  -  NTP            Disabled.                 12  -  METRIC         Disabled  /                  A  -  Configure options 1 - 12,                  [E] -  Exit menu    1 To the OP: just enable the service, and then edit C sys$specific:[ucx$ntp]ucx$ntp.template.  Save the resulting file as F ucx$ntp.conf, and start the service with @sys$startup:ucx$ntpd_startup   !j !Mike: !w !   J __________________________________________________________________________A Bradford J. Hamilton                    "All opinions are my own"sK bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-'  0                                          with @"   ------------------------------  # Date: Wed, 17 Sep 2003 16:42:10 GMT   From: Rob Brown <brown@gmcl.com>& Subject: Re: Newbie question about NTPL Message-ID: <Pine.LNX.4.44.0309171040030.16396-100000@localhost.localdomain>  $ On Wed, 17 Sep 2003, it was written:  ! > It goes back as far as UCX 4.0:a   It also exists in UCX 3.3:   > $ ucx show versR >h4 >   DEC TCP/IP Services for OpenVMS VAX Version V3.30 >   on a VAXstation 4000-90 running OpenVMS V6.2  < It is item 11 on the server menu in @sys$manager:ucx$config.     -- O  / Rob Brown                        brown@gmcl.comoA G. Michaels Consulting Ltd.      (866)438-2101 (voice) toll free!a6 Edmonton                         (780)438-9343 (voice)4                                  (780)437-3367 (FAX)1                                  http://gmcl.com/r   ------------------------------  % Date: Wed, 17 Sep 2003 12:45:45 +0800e, From: Paul Repacholi <prep@prep.synonet.com>< Subject: Re: OpenVMS 7.3-1 drowning HSG80s with I/O requests- Message-ID: <87r82g9g5i.fsf@prep.synonet.com>r  & svieth@wi.rr.com (Scott Vieth) writes:  E > Today, the Storage division admitted that VMS 7.3-1 is too powerful  > for mere HSG80 controllers.m  $ So, when do you fire up the ES47s :)   -- g< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 17 Sep 2003 10:14:59 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)gT Subject: Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster= Message-ID: <cf15391e.0309170914.40ffd0c3@posting.google.com>h  | Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote in message news:<01L0Q7Y2MLKIAOQMVC@sysdev.deutsche-boerse.com>...- > JF Mezei <jfmezei.spamnot@istop.com> wrote:rJ > > node A has a connection to ethernet, and a separate connection to someJ > > financial network. node B has a connection to ethernet, and a separateJ > > connection to some financial network (normally down, used as backup).  > > D > > Node A blows its connection to ethernet, but it maintains quorum( > > (essentially shutting down node B).  > > G > > The logical thing to do is to boot node B into production to resumet > > financial transactions.  > > L > > HOWEVER, during the time it takes to do this, Node A is still capable ofJ > > receiving financial transactions from its link that isn'ty affected by > > ethernet.  > > G > > When node B goes into production, they realise that the link to the J > > financial network is still active by node A, so they shut down node A,L > > enabling node B to connect to financial network. But as a result, you'veH > > had a gap where node A has processed transactions to its disks only.L > > They realise that those transactions are not on B. And they realise thatL > > since they will continue the day processing on B, B will have the latest > > copy of the drives.  > > H > > So when A is fixed and is re-integrated into the cluster, B's drivesJ > > will be used as source, wiping out transactions that had been receivedJ > > between the time the ethernet card blew and the time A was shut down.  > K > OK, one can come up with some configuration which is badly designed, but aG > anyone actually doing financial transactions (or anything else) with i2 > such a non-robust configuration has my sympathy.  C The basic problem in the design in JF's scenario is that node A hasbD been given more votes than Node B, so the decision of whether A or BC continue in the event of any loss of communications between A and B @ has been pre-determined, and that choice is Node A.  It would beE better to either have a tie-breaking node help decide which of A or B E is best to continue, or else make the votes equal, forcing a human to-B make the decision and allow only one of the two nodes to continue.  @ An important thing JF's scenario demonstrates is that the QuorumD scheme, when properly set up and used, protects not only the data on@ disks/shadowsets within the cluster, but is also used to protect? shared resources external to the cluster, including things likea external data feed flows.r   ------------------------------  % Date: Wed, 17 Sep 2003 08:16:40 -0400 . From: "Rob Lyons" <rob.lyons@resilientsys.com>U Subject: Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster?i+ Message-ID: <bk9jb5$n7f$1@bob.news.rcn.net>H  6 Although I am currently an independent VMS consultant,7 in my previous life I was the lead support engineer fore: VMS clusters.  Jeffrey, if you need a hand or want someone) to review your final design, let me know.a   --	 Rob Lyonsh Resilient Systems, Inc   ------------------------------  % Date: Wed, 17 Sep 2003 14:44:20 +0100e0 From: "Dr. David Kirkby" <drkirkby@ntlworld.com>5 Subject: OpenVMS hobbyyest - free to join or not ????-, Message-ID: <3F686534.DB8D17DD@ntlworld.com>   Hi,mE 	I'm contemplating trying to get some OpenVMS media for my Dec Alpha,iG although have not decided yet whether to or not - mainly because of therH cost of buying what while probably need to be two new disks, fitting theD disks, re-installing tru64 etc etc, but also because of the price ofF OpenVMS for hobbyists. I'm confused by the pricing structure of that.   F At an FAQ at http://www.openvmshobbyist.com/hobbyist_faq/index.html it says::  C "Q3: What is the process for getting the PAKS  (PAK=OpenVMS ProductiB Authorization Keys) so I can run OpenVMS and the layered products?   A3: The process is as follows:G 1) Join and receive a membership number from a participating Encompass,tC or DECUS, or recognized User Group world wide. (This is free at thed basic member level)"  H So I thought all I needed to do is join (which will cost me nothing) and+ buy the media, which I can get at $30 from:P, http://www.montagar.com/hobbyist/mount.html   E Well that is what I thought until I tried to join DECUS, where I finduH that it is now called "HP-Interex.org" Clicking aroud their site I find:  ? http://www.hp-interex.org/site/shop/shopsublistnonmem.asp?id=91-  C where I discover that a hobbyist membership, which will allow me to H apply for a hobbyist license, costs 52.88  (UK pounds), which is aroundH $85. So it is not free as the FAQ says. So the total cost of being aboutE to get a set of media and install some osftware is not $30, but about  $115.    Am I missing something here? e   -- lA "The day Microsoft makes something that doesn't suck is probably l> the day they start making vacuum cleaners." -Ernst Jan Plugge.   Dr. David Kirkby,) Senior Research Fellow,n Department of Medical Physics, University College London," 11-20 Capper St, London, WC1E 6JA., Website: http://www.medphys.ucl.ac.uk/~davek- Author of 'atlc' http://atlc.sourceforge.net/.   ------------------------------  + Date: Wed, 17 Sep 2003 16:29:39 +0200 (MET)H9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>r9 Subject: Re: OpenVMS hobbyyest - free to join or not ????o; Message-ID: <01L0S087E9KKAOQMVC@sysdev.deutsche-boerse.com>r  F > I'm contemplating trying to get some OpenVMS media for my Dec Alpha,I > although have not decided yet whether to or not - mainly because of the J > cost of buying what while probably need to be two new disks, fitting th= erF > disks, re-installing tru64 etc etc, but also because of the price ofJ > OpenVMS for hobbyists. I'm confused by the pricing structure of that.=20 >=20H > At an FAQ at http://www.openvmshobbyist.com/hobbyist_faq/index.html it > says:s >=20G > "Q3: What is the process for getting the PAKS  (PAK=3DOpenVMS ProductsD > Authorization Keys) so I can run OpenVMS and the layered products? >=20  > A3: The process is as follows:I > 1) Join and receive a membership number from a participating Encompass, E > or DECUS, or recognized User Group world wide. (This is free at thee > basic member level)" >=20J > So I thought all I needed to do is join (which will cost me nothing) an= dc- > buy the media, which I can get at $30 from:a0 > http://www.montagar.com/hobbyist/mount.html=20 >=20G > Well that is what I thought until I tried to join DECUS, where I find-J > that it is now called "HP-Interex.org" Clicking aroud their site I find= :p >=20C > http://www.hp-interex.org/site/shop/shopsublistnonmem.asp?id=3D91g >=20E > where I discover that a hobbyist membership, which will allow me to3J > apply for a hobbyist license, costs =A352.88  (UK pounds), which is aro= undeJ > $85. So it is not free as the FAQ says. So the total cost of being abou= trG > to get a set of media and install some osftware is not $30, but about>
 > $115.=20  I I don't know any details, but with some DECUS chapters there is a free=20iJ "associate membership" or something like that which, however, is enough=20' to allow one to get a hobbyist license.   3 Even without it, $115 isn't that much for VMS.  :-)t  C Note that you don't have to buy the media from Montagar.  Under thesH terms of the hobbyist license, you can borrow the media from a "real" CDB distribution and install the stuff from there.  (I would recommendG having your own OS CD---doesn't have to be the latest version---just soMF you can boot from that if you have to.)  IIRC, there are many productsK (compilers etc) which are covered by the hobbyist license but not on the=20.# compact distribution from Montagar.    ------------------------------  % Date: Wed, 17 Sep 2003 10:39:33 -0400r< From: "Peter Weaver" <WeaverConsultingServices@sympatico.ca>9 Subject: Re: OpenVMS hobbyyest - free to join or not ????/9 Message-ID: <bk9rn6$qoh89$1@ID-141708.news.uni-berlin.de>%   Dr. David Kirkby wrote:= >...9 > where I discover that a hobbyist membership, which willo allow me tou: > apply for a hobbyist license, costs 52.88  (UK pounds), which is= > around $85. So it is not free as the FAQ says. So the totals cost ofe= > being about to get a set of media and install some osftwaret is not > $30, but about $115. >...  < The last time I looked Encompass in the US would let anybody6 in the world join as an Associate member for free. See< http://www.encompassus.org/membership/join.html for details.: As an Associate member you can get the PAKs. If you do not; want to buy the media then you can borrow the official (HP)T4 media from anybody who is willing to lend out a set.   -- N Peter Weaver Weaver Consulting Services Inc.  Canadian VAR for CHARON-VAXe www.weaverconsulting.caa   ------------------------------  % Date: Wed, 17 Sep 2003 07:30:52 -0700i# From: "Tom Linden" <tom@kednos.com> 9 Subject: RE: OpenVMS hobbyyest - free to join or not ????79 Message-ID: <CIEJLCMNHNNDLLOOGNJIKENLHPAA.tom@kednos.com>h   >-----Original Message----- A >From: Phillip Helbig [mailto:HELBPHI@sysdev.deutsche-boerse.com]e, >Sent: Wednesday, September 17, 2003 7:30 AM >To: Info-VAX@Mvb.Saic.Com: >Subject: Re: OpenVMS hobbyyest - free to join or not ???? >n >iG >> I'm contemplating trying to get some OpenVMS media for my Dec Alpha,aJ >> although have not decided yet whether to or not - mainly because of theK >> cost of buying what while probably need to be two new disks, fitting theUG >> disks, re-installing tru64 etc etc, but also because of the price ofhH >> OpenVMS for hobbyists. I'm confused by the pricing structure of that. >>I >> At an FAQ at http://www.openvmshobbyist.com/hobbyist_faq/index.html ito >> says: >>F >> "Q3: What is the process for getting the PAKS  (PAK=OpenVMS ProductE >> Authorization Keys) so I can run OpenVMS and the layered products?d >>! >> A3: The process is as follows:cJ >> 1) Join and receive a membership number from a participating Encompass,F >> or DECUS, or recognized User Group world wide. (This is free at the >> basic member level)"t >>K >> So I thought all I needed to do is join (which will cost me nothing) andg. >> buy the media, which I can get at $30 from:. >> http://www.montagar.com/hobbyist/mount.html >>H >> Well that is what I thought until I tried to join DECUS, where I findK >> that it is now called "HP-Interex.org" Clicking aroud their site I find:r >>B >> http://www.hp-interex.org/site/shop/shopsublistnonmem.asp?id=91 >>F >> where I discover that a hobbyist membership, which will allow me toK >> apply for a hobbyist license, costs 52.88  (UK pounds), which is around K >> $85. So it is not free as the FAQ says. So the total cost of being aboutyH >> to get a set of media and install some osftware is not $30, but about >> $115. >tG >I don't know any details, but with some DECUS chapters there is a freeoH >"associate membership" or something like that which, however, is enough( >to allow one to get a hobbyist license. >-4 >Even without it, $115 isn't that much for VMS.  :-) >rD >Note that you don't have to buy the media from Montagar.  Under theI >terms of the hobbyist license, you can borrow the media from a "real" CDnC >distribution and install the stuff from there.  (I would recommendvH >having your own OS CD---doesn't have to be the latest version---just soG >you can boot from that if you have to.)  IIRC, there are many productseI >(compilers etc) which are covered by the hobbyist license but not on thel$ >compact distribution from Montagar.  H And if you have enough bandwidth, you could even download from somewhereD -rw-rw-r--   1 root     ftp      221384250 Aug 26 19:55 ovms0731.zipE -rw-rw-r--   1 root     ftp      183218103 Aug 26 19:56 vaxvms073.zipt" I guess that amounts to borrowing. >  >---' >Incoming mail is certified Virus Free.c; >Checked by AVG anti-virus system (http://www.grisoft.com).aA >Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003i >o ---i& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003   ------------------------------  % Date: Wed, 17 Sep 2003 16:39:03 +0100i0 From: "Dr. David Kirkby" <drkirkby@ntlworld.com>9 Subject: Re: OpenVMS hobbyyest - free to join or not ????a, Message-ID: <3F688017.CCD88FD1@ntlworld.com>   Tom Linden wrote:.  J > And if you have enough bandwidth, you could even download from somewhereF > -rw-rw-r--   1 root     ftp      221384250 Aug 26 19:55 ovms0731.zipG > -rw-rw-r--   1 root     ftp      183218103 Aug 26 19:56 vaxvms073.zip $ > I guess that amounts to borrowing.  C I did a google on those files but did not come up with anything. SoT where could I borrow them from?e   -- aA "The day Microsoft makes something that doesn't suck is probably t> the day they start making vacuum cleaners." -Ernst Jan Plugge.   Dr. David Kirkby,p Senior Research Fellow,a Department of Medical Physics, University College London," 11-20 Capper St, London, WC1E 6JA., Website: http://www.medphys.ucl.ac.uk/~davek- Author of 'atlc' http://atlc.sourceforge.net/I   ------------------------------  % Date: Wed, 17 Sep 2003 16:51:30 +0100 0 From: "Dr. David Kirkby" <drkirkby@ntlworld.com>9 Subject: Re: OpenVMS hobbyyest - free to join or not ????w, Message-ID: <3F688302.F3239D51@ntlworld.com>   Peter Weaver wrote:a >  > Dr. David Kirkby wrote:a > >...; > > where I discover that a hobbyist membership, which willg
 > allow me to < > > apply for a hobbyist license, costs 52.88  (UK pounds),
 > which is? > > around $85. So it is not free as the FAQ says. So the totaly	 > cost of ? > > being about to get a set of media and install some osftwared > is not > > $30, but about $115. > >... > > > The last time I looked Encompass in the US would let anybody8 > in the world join as an Associate member for free. See> > http://www.encompassus.org/membership/join.html for details.< > As an Associate member you can get the PAKs. If you do not= > want to buy the media then you can borrow the official (HP)i6 > media from anybody who is willing to lend out a set.  D Cheers, I made an application. Whether or not they accept me I don'tB know. There was no option to put in a country, so I put "UK" under state. y  F Given I know of nobody with a set of media, where is the best place toG obtain them? The $30 deal I see did not seem too bad, but I gather thatpH is for a slim kit. In which case I might rather spend a bit more and getF some more. All I guess I really want is a C compiler, but depending on price I might get a bit more.   F I've still not decided whether or not I will go to OpenVMS. The hassleH of re-configuring my Dec Alpha is more of a concern than the cost of theG software. It has two SCSI disks, both of which are in use by Tru64 so I-C would need to replace both with a larger disk to have all the Tru64 C stuff, then get another disk for OpenVMS. That seems quite a bit of6G work, to basically port a single application. I might still take the HPnD testdrive route, if I try to port it, but at least I'll give it some	 thought. .   -- 1A "The day Microsoft makes something that doesn't suck is probably  > the day they start making vacuum cleaners." -Ernst Jan Plugge.   Dr. David Kirkby,i Senior Research Fellow,e Department of Medical Physics, University College London," 11-20 Capper St, London, WC1E 6JA., Website: http://www.medphys.ucl.ac.uk/~davek- Author of 'atlc' http://atlc.sourceforge.net/e   ------------------------------  % Date: Wed, 17 Sep 2003 17:03:42 +0100e0 From: "Dr. David Kirkby" <drkirkby@ntlworld.com>9 Subject: Re: OpenVMS hobbyyest - free to join or not ???? , Message-ID: <3F6885DE.1656691F@ntlworld.com>   Phillip Helbig wrote:s  H > I don't know any details, but with some DECUS chapters there is a freeI > "associate membership" or something like that which, however, is enoughu) > to allow one to get a hobbyist license.e > 5 > Even without it, $115 isn't that much for VMS.  :-)o  F True, but to be honest I'm more interested in UNIX so I'm not overseenG to spend much at all to just port an application. I might be wrong, but-> I don't think I'll be using it much. So the cost and hassle ofH reconfiguring my Dec Alpha is probably more of an issue than the cost of media. 6  E > Note that you don't have to buy the media from Montagar.  Under theNJ > terms of the hobbyist license, you can borrow the media from a "real" CDD > distribution and install the stuff from there.  (I would recommendI > having your own OS CD---doesn't have to be the latest version---just soyH > you can boot from that if you have to.)  IIRC, there are many productsJ > (compilers etc) which are covered by the hobbyist license but not on the% > compact distribution from Montagar.e  E Where is a decent place to buy a set of media? I thought the MontagareF kit included the C compiler and it included international carriage, soG was not a bad price. However, if I do decide to install the OS, I would F not mind spending a bit more on media. That is going to be quite smallB compared to the amount I'll need to spend on upgrading the disks.   G That said is it possible to download the OpenVMS OS and C compiler fromyE anywhere ? I have a large bandwidth connection at work, so can easilyg download CD images.   F What is the requirements for booting OpenVMS from a CD now? One of theG documents I found said you need a CD capable of 512 bytes/sector, where G I'm sure I read someone else that IDE CD's were supported. I do have antC external SCSI CD that will do 512 bytes/sector, so it is not a hugeeE hassle, but since the Alpha has an internal CD (which I'm 99% sure istC IDE), I'd rather boot from that of course. I also added a DVD driveyD (which I'm 100% sure is IDE), although it has never had a single DVD anywhere near it.   E I don't tend to use the 600 MHz Dec Alpha much, as it is considerablydC slower than the Sun Ultra 80 I have (4 x 450 MHz), or the HP C3000, D which despite its modest 400 MHz clock speed is faster than a singleE processor on the Sun. However, when all 4 CPUs are in use on the Sun,-$ that is the fastest machine I have.      -- 7A "The day Microsoft makes something that doesn't suck is probably :> the day they start making vacuum cleaners." -Ernst Jan Plugge.   Dr. David Kirkby,e Senior Research Fellow,m Department of Medical Physics, University College London," 11-20 Capper St, London, WC1E 6JA., Website: http://www.medphys.ucl.ac.uk/~davek- Author of 'atlc' http://atlc.sourceforge.net/_   ------------------------------  # Date: Wed, 17 Sep 2003 13:55:39 GMT-# From: "John Smith" <a@nonymous.com>,9 Subject: OT: security flaw in Solaris and Trusted SolarisoF Message-ID: <vJZ9b.53112$DZ.3825@news04.bloor.is.net.cable.rogers.com>  3 http://www.eweek.com/article2/0,4149,1269850,00.aspo  , Solaris Flaw Leaves Machines Open to Attacks   September 16, 2003 By  Dennis FisherO    H There is a serious security flaw in several versions of both Solaris andL Trusted Solaris that make it possible for virtually any remote or local userH to gain root privileges on a vulnerable machine. There is also a workingE exploit for this vulnerability circulating in the security community.l  L The problem lies in the Solstice AdminSuite, a set of tools Sun MicrosystemsE Inc. includes with the operating system that allows administrators to L perform remote administration tasks. The tool set uses the sadmind daemon toE execute these tasks. The daemon by default uses a weak authenticationiC scheme, which allows an attacker to send a series of special RemoteeI Procedure Call (RPC) packets to the daemon and forge a client's identity,OK according to an advisory on the flaw published Tuesday by iDefense Inc., in< Reston, Va.t  I Once this is accomplished, the attacker can do whatever he chooses on the  compromised machine.  K The sadmind daemon is installed by default on most default installations of E Solaris. The issue affects versions 7, 8 and 9 of Solaris, as well asdE Trusted Solaris 7 and 8, on both the Sparc and x86 platforms. TrustedPC Solaris is the hardened version of Sun's flagship operating system.t  J Sun, based in Santa Clara, Calif., does not plan to issue a patch for thisL vulnerability. However, the company has published a security advisory, which includes a workaround.  H IDefense officials recommend placing inbound filters on TCP and UDP port* 111, which is used by the Sun RPC service.    = Copyright (c) 2003 Ziff Davis Media Inc. All Rights Reserved.h   ------------------------------    Date: 17 Sep 2003 07:51:10 -0700( From: Javier Henderson <javier@KJSL.COM>6 Subject: Re: OT: Verisign messes with DSN root servers- Message-ID: <86llsntqn5.fsf@skylane.kjsl.com>s  , JF Mezei <jfmezei.spamnot@istop.com> writes:  I > added information: seems that Verisign have temporarily restorer propereP > behaviour for the .com tld, but .net still has a wildcard response pointing to > verisign's own IP address. >  > P > I have heard discussions about hacking BIND such that any response pointing to6 > verisign's IP address will yield a "site not found". > O > If this issies isn't resolved, any chances that such a hack might come to VMSs > any time soon ?m > O > Consider the whole RBL issue. RBL have become moot since any and all .com andg/ > .net domain names now translate to something.a  - RBL's list IP addresses, not domains, though.o   -jav   ------------------------------  % Date: Wed, 17 Sep 2003 16:16:08 +0100t* From: "Richard Brodie" <R.Brodie@rl.ac.uk>6 Subject: Re: OT: Verisign messes with DSN root servers, Message-ID: <bk9trq$15nm@newton.cc.rl.ac.uk>  ] "Javier Henderson" <javier@KJSL.COM> wrote in message news:86llsntqn5.fsf@skylane.kjsl.com...l  Q > > Consider the whole RBL issue. RBL have become moot since any and all .com andg1 > > .net domain names now translate to something.e >d/ > RBL's list IP addresses, not domains, though.c  J Yes. Although some deceased RBLs now resolve everything to Verisign's box,R and simple 'is this address registered in the DNS' checks get incorrect responses.   ------------------------------  % Date: Wed, 17 Sep 2003 11:38:58 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>6 Subject: Re: OT: Verisign messes with DSN root servers) Message-ID: <3F68800F.39A90BAA@istop.com>-  D For those who are against the verisign move, (the .coms are also nowI wildcarded so that any domain returns a valid IP address), you can sign a  petition at:  ' http://www.petitiononline.com/icanndns/m   ------------------------------  % Date: Wed, 17 Sep 2003 11:58:09 -0400l* From: JF Mezei <jfmezei.spamnot@istop.com>6 Subject: Re: OT: Verisign messes with DSN root servers) Message-ID: <3F68848C.328611E2@istop.com>	  8 Some discussions have begun on an icann sponsored forum:    http://forum.icann.org/offtopic/   ------------------------------  % Date: Wed, 17 Sep 2003 13:31:23 -0400	) From: Jfmezei <jfmezei.spamnot@istop.com> 6 Subject: Re: OT: Verisign messes with DSN root servers) Message-ID: <3F689A60.C1B5F53B@istop.com>n  . Another reason to leave javascript disabled...  , http://www.circleid.com/article/260_0_1_0_C/  J Verisign has hired a company to monitor individuals who "stumble" on theirU sitefinder web site. They use javascript to set and monitor cookies to analyse usage.z   ------------------------------    Date: 17 Sep 2003 06:26:17 -0700- From: djesys@earthlink.net (David J Dachtera)r Subject: Re: padding a csv file < Message-ID: <66a00d01.0309170526.378d678@posting.google.com>  x pa@it.singer-friedlander.com (Piyush Avichal) wrote in message news:<eb55ac2a.0309160118.37c83817@posting.google.com>... > Hello, > K > I have a csv file with variable length fields. I would like to be able toiJ > sort this file, but the vms sort function can not handle variable lengthM > fields. Does anyone know how I could pad the fields to make it looked fixedc0 > length and thus making it easy to sort in vms. > 
 > Many Thanks. >  > Piyush   Use DCL:  %  1. Isolate the elements of the list:KB     $ vbl = F$ELEMENT( element_pointer, delimiter, source_string )  B  2. Use F$FAO() to write out fixed-length fields, as other posters have suggested.   : See HELP Lexicals for F$ELEMENT() and F$FAO() information.  ) See HELP for OPEN, READ, WRITE and CLOSE.o  
 Helpful hint:oE To avoid some problems with files created by DCL, use this technique:.   $ COPY NLA0: filespecw# $ OPEN/APPEND logical_name filespecr  5 ...when creating your output file. The result will be C sequential/variable instead of the sequential/VFC you get with just0 OPEN/WRITE.m   ---y David J Dachtera dba DJE Systemsc   ------------------------------  # Date: Wed, 17 Sep 2003 14:59:36 GMT 6 From: "Craig A. Berry" <craigberry@mac.com.spamfooler> Subject: Re: padding a csv fileg@ Message-ID: <ca4791372b986a62fe06cdd29dd34baa@news.teranews.com>  A > pa@it.singer-friedlander.com (Piyush Avichal) wrote in message  ; > news:<eb55ac2a.0309160118.37c83817@posting.google.com>...   M > > I have a csv file with variable length fields. I would like to be able tobL > > sort this file, but the vms sort function can not handle variable lengthO > > fields. Does anyone know how I could pad the fields to make it looked fixeda2 > > length and thus making it easy to sort in vms.  D The GNV package includes GNU sort, so the easiest thing might be to E just use that.  Since the name conflicts with the name of the native nD sort utility, you'll either have to define a foreign command symbol  point to it:   $ gsort :== $gnu:[bin]sort.exe   or run it directly:l   $ mcr gnu:[bin]sort --help   ------------------------------    Date: 17 Sep 2003 02:40:37 -0700; From: Dominique.GEORGES@dom-auralann.be (Dominique GEORGES)M8 Subject: PHP library containing PREG_MATCH, ... commands= Message-ID: <9f6f3333.0309170140.79e8a97d@posting.google.com>i   Hi,n  D I need to use preg_split(), preg_replace(), ... PHP commands, but it3 seems they are not bundled in the OpenVMS PHP kit !n  ? I found the sources on http://www.pcre.org, but was not able tor= compile it, ... (I'm NOT a prorammer, just a system manager).a  % Is there anybody who already did it ?d   Thanks,e	 Dominique    ------------------------------  # Date: Wed, 17 Sep 2003 14:23:06 GMTn5 From: Matthew Doremus <Matthew.Doremus@hp.com.NoSpam> < Subject: Re: PHP library containing PREG_MATCH, ... commands, Message-ID: <3F686E4A.8090801@hp.com.NoSpam>  
 Hi Dominique,   H    Are you referring to the CSWS_PHP kit for OpenVMS ?  If so, the PCRE F extension is available as a loadable extension and not bundled as you F mention.  To enable the PCRE extension for all PHP scripts you should ? edit PHP_ROOT:[000000]PHP.INI and uncomment the following line  G  ";extension=php_pcre.exe" or add a dl call in the specific scripts to - load it (i.e. dl("php_pcre")).   Thanks,6      Matt    Dominique GEORGES wrote:   >Hi, >:E >I need to use preg_split(), preg_replace(), ... PHP commands, but itd4 >seems they are not bundled in the OpenVMS PHP kit ! >.@ >I found the sources on http://www.pcre.org, but was not able to> >compile it, ... (I'm NOT a prorammer, just a system manager). >t& >Is there anybody who already did it ? >t >Thanks,
 >Dominique >f   ------------------------------  % Date: Wed, 17 Sep 2003 07:11:54 +0100 0 From: "Dr. David Kirkby" <drkirkby@ntlworld.com>< Subject: Re: Porting from UNIX to OpenVMS for a real newbie.+ Message-ID: <3F67FB2A.A6F4EA6@ntlworld.com>r   David Mathog wrote:w >  > On 12 Sep 2003 18:05:52 -0700w1 > drkirkby@ntlworld.com (Dr. David Kirkby) wrote:v > D > > The problem is the application uses the normal (in UNIX circles) > > autoconf/automake. > ) > I've done a _LOT_ of this sort of port.y  > Good, it's nice to know of someone that has done this before.   9 > By far the easiest way to do so is to first build it on A > Unix and log everything that "make" does to a file.  If you canz= > find a Tru64 machine with DEC, um, Compaq, um, HP compilers = > that would give you the best starting point.  For instance:e >  >  make | tee make_did_this.come  @ Luckily I have a 64-bit Dec Alpha with Tru64 5.1B and the Compaq
 compiler.   = > Then move the .com file to VMS and open it in a wide editor2: > (nedit works great) and go through doing global replaces7 > for the unix bits, converting them to VMS.  Typically + > there are only 3 or 4 types of "cc" lines   % Cheers that looks fairly easy to do.     > >cB > > -DPACKAGE_VERSION=\"4.4.1\" -DPACKAGE_STRING=\"atlc\ 4.4.1\" \F > > -DPACKAGE_BUGREPORT=\"drkirkby@ntlworld.com\" -DPACKAGE=\"atlc\" \6 > > -DVERSION=\"4.4.1\" -DHAVE_LIBM=1 -DSTDC_HEADERS=1 > > -DHAVE_SYS_TYPES_H=1 \ > # > etc. etc. for way too many lines.a > 2 > If you really need this many  defines you should9 > instead have all modules include a single blah_config.hi > which has: > ! > #define PACKAGE_VERSION "4.4.1"5% > #define PACKAGE_STRING "atlc 4.4.1" 3 > #define PACKAGE_BUGREPORT "drkirkby@ntlworld.com"  > #define HAVE_WHATEVER 1r  n: > etc.  These strings look like they are all invariant for9 > the platform so no point having them all on the commandl > line.o   Well not quite, as all the o #define HAVE_WHATEVER's  are platform specific. T  H I've changed autoconf so there is now an automatically produces a headerF file that has all this in, so my header file now looks like this, with2 some bits that are not platform specific, such as:   #define PACKAGE "atlc"1 #define PACKAGE_BUGREPORT "drkirkby@ntlworld.com"  #define PACKAGE_NAME "atlc"/# #define PACKAGE_STRING "atlc 4.4.5"e   and those that are such as     #define HAVE_CPUCOUNTSET 1 #define HAVE_CPUSETCREATE 1  #define HAVE_CPU_GET_MAX 1 #define HAVE_CPU_GET_NUM 1  G etc. It is a bit annoying that autoconf/automake don't allow one to put4D the platform specific stuff in one file an the non platform specific stuff in another.     = > This would improve matters on Unix as well, where I imagine > > the logs of your build procedures are essentially unreadable' > due to all of these extraneous lines.b   Yes, I have done that.    " > You might also want to refer to: > 4 >   http://saf.bio.caltech.edu/unix_to_vms_notes.txt  F Thanks, I took a quick look at that. The main issue that I will hit isD the fact I've used fseek() several times and ftell() to find out howD long the file is. I do need to be able to seek to specific bits of aG file. It would not appear to be possible to do that using the suggestedrG replacements of fgetpos() and fsetpos(), since those two functions lack_1 some of the functionality of fseek() and ftell().d   --  A "The day Microsoft makes something that doesn't suck is probably l> the day they start making vacuum cleaners." -Ernst Jan Plugge.   Dr. David Kirkby,  Senior Research Fellow,  Department of Medical Physics, University College London," 11-20 Capper St, London, WC1E 6JA., Website: http://www.medphys.ucl.ac.uk/~davek- Author of 'atlc' http://atlc.sourceforge.net/0   ------------------------------  % Date: Wed, 17 Sep 2003 16:05:32 +0100e0 From: "Dr. David Kirkby" <drkirkby@ntlworld.com>< Subject: Re: Porting from UNIX to OpenVMS for a real newbie., Message-ID: <3F68783C.38DFC9EC@ntlworld.com>   Brad McCusker wrote:  M > Well, there is the hobbyist program, if you qualify for that.  I don't know.+ > the details, but others in this group do.   F I'm sure I would be eligable as not for commerical gain. The software  http://atlc.sourceforge.net/ is GPL'ed.    H But the OpenVMS binaries can be downloaded for free (I'd need to buy theH media), and I might (confused by this so see my other post), need to pay+ for membership of a club to get a license. 4C I'd need to replace both disk in the Alpha (neither on their own isO sufficient for my tru64 needs). G Overall, I suspect I'd need to spend quite a bit of time/effort getting.F my Dec Alpha in a position where it could dual boot OpenVMS and Tru64.# I'm tempted to not want to bother.    e > >0H > > However, I am aware of the OpenVMS systems at the HP testdrive site:$ > > http://www.testdrive.compaq.com/ > J > This will get you to OpenVMS V7.3-2 field test, the latest Alpha versionL > available, and, for this project you will want to use the latest.  I wouldL > not recommend using the Itanium version yet.  Port it to alpha first, then > go to Itanium, if you want. L > Problem is, the test drive systems don't have GNV (I don't think they do).J > I would urge you to try to do this with GNV (as someone else suggested),L > which brings me back to trying to use the hobbyist program to get your own > VMS system up and running. > E > Regardless of how you go, I am very interested in how you make out.v  G You might hear. I'll try to change a few things on the current code nows" I'm aware of a few of the issues.  -- pA "The day Microsoft makes something that doesn't suck is probably a> the day they start making vacuum cleaners." -Ernst Jan Plugge.   Dr. David Kirkby,t Senior Research Fellow,  Department of Medical Physics, University College London," 11-20 Capper St, London, WC1E 6JA., Website: http://www.medphys.ucl.ac.uk/~davek- Author of 'atlc' http://atlc.sourceforge.net/    ------------------------------  % Date: Wed, 17 Sep 2003 09:30:04 +0100s* From: Nic Clews <sendspamhere@[127.0.0.1]>, Subject: Re: Question on running GnuPG 1.2.0' Message-ID: <bk95v8$itd$1@lore.csc.com>n   healyzh@aracnet.com wrote: >  ...tL > Overall, PGP 2.6.3ia-multi06 seems preferable to GnuPG 1.2.0.  What reallyK > stinks about GnuPG (besides the above privilige problem) is the followingv > from the README:L > "For encryption/decryption & making/verifying signatures, only 'Stream LF'" > format is supported on OpenVMS."L > This means that you need to convert files to "Stream LF" before encrypting > them.r  B I just checked the multi05 I pulled the objects for and tried, andF failed to work with signatures, my keying files are stream_lf, so thatH isn't the problem... FWIW the keyrings were created at PGP 2.6.2 at 1024 bits.   > However, are you saying (as it appears to read) that if you're9 encrypting a file, it also has to be in stream_lf format?e   -- h? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencesf nclews at csc dot comy   ------------------------------  % Date: Wed, 17 Sep 2003 17:17:41 +0100n% From: David Gray <police@spamcop.net>f- Subject: Re: Remove Pathworks/Advanced Server 8 Message-ID: <i02hmvgc9mtoge4nq7v8se3kgbp6ts1egq@4ax.com>   Cheers for that.     	Dave   	 >Hi Dave,o >hM >You want to run $ @sys$update:pwrk$deinstal.com  (that's one L in deinstal).  >  >Best of luck, >i   ------------------------------  % Date: Wed, 17 Sep 2003 17:31:54 +0100r% From: David Gray <police@spamcop.net> - Subject: Re: Remove Pathworks/Advanced Serverq8 Message-ID: <7a2hmvk8k1cajrqhu9chengjctg8v9im45@4ax.com>  	 Hi Brad, w    Firstly thanks for your input.    C I installed V7.3A  without any errors and only started getting themeD whilst trying to configure Advanced Server.  Too many error messages% to list them all but here are a few. p     SPIKE] admin LANGroup\\SPIKE> logon Username: grayd 	 Password:a> 17-SEP-2003 17:12:30.37 0000051E:0063BFA0 %ERR, Could not load1 Transport library: SYS$SHARE:PWRK$NBLIB_TCPIP.EXEb/ 17-SEP-2003 17:12:30.51 0000051E:0063BFA0 -ERR,:; "%SYSTEM-F-PROTINSTALL, protected images must be installed"t> 17-SEP-2003 17:12:30.55 0000051E:0063BFA0 %ERR, Could not load1 Transport library: SYS$SHARE:PWRK$NBLIB_NETBEUI.E, XE/ 17-SEP-2003 17:12:30.55 0000051E:0063BFA0 -ERR, ; "%SYSTEM-F-PROTINSTALL, protected images must be installed"o< You were logged on, but have not been validated by a server.E Therefore, you may not have permission to use some network resources.e    4 Running the config utility  @SYS$UPDATE:PWRK$CONFIG    Gives this error.    Upgrading SAM databases...  e  o/ %PWRK-F-SAMCHECK, error upgrading SAM databasesD5  PWRK-F-CONFIGERR, severe error during configuration.t  m= Error getting this server's role from the existing databases.e= Make sure the file server or another utility is not running  a  C If it looks like I have missed something obvious then please let mel know.  4   Cheers,u 	Dave. c      3 On Mon, 15 Sep 2003 21:10:23 -0400, "Brad McCusker"y& <brad.mccuskerNosp@Mcompaq.com> wrote:   >Dave -t >/G >"PEN"'s suggestion of pwrk$deinstal is of course correct.  But, I'd besL >surprised if de-installing V6 is going to solve your problem.  Of course, IL >don't know what your problem is, so, why listen to me?  Installing Advanced0 >Server right over the top of PATHWORKS is "OK". >rJ >When you install V7.3A, please be sure you install V7.3A-ECO1 (its a full0 >kit, you can install it directly, no pre-reqs). >tJ >Let us know the problems you are seeing, maybe we can save you some time. >i >Brad McCusker >OpenVMS Engineering >Nashua NH USA >i3 >"David Gray" <police@spamcop.net> wrote in message 3 >news:d0ibmv4cktj0javmdkvp9c94ap7simvvkl@4ax.com...l >> Greetings all," >>E >> Is it possible to remove *all* trace of Pathworks /Advanced Serverh >> from OpenVMS 7.3-1? >>H >> Basically I'm trying to install Advanced Server ( V7.3A) over the topF >> of Pathworks (V6.1).   V6.1 was installed but never configured, andF >> all the errors I'm getting are leading me to the  conclusion that a >> clean install might be best.l >>
 >> Cheers, >> Dave. >u   ------------------------------    Date: 17 Sep 2003 02:19:32 -0700A From: dieter.rossbach@gmx.de (=?ISO-8859-1?Q?Dieter__Ro=DFbach?=)h0 Subject: Re: Remove satellite node from cluster?= Message-ID: <e1d40caf.0309170119.2f414481@posting.google.com>o  D > I can't just delete the files in [sys10...] and [sys010...] can I?? > Aren't there a lot of aliases?  Should I do a SET FILE/REMOVEo > on all those files?u  / Before you delete any files in [sys10...], do a   & $ set file/remove [sys10]syscommon.dir  < This is a link to sys$sysdevice:[vms$common] and a Delete onB [sys10...] without removing that as some unwanted side effects :-)   Regards8   Dieter   ------------------------------    Date: 17 Sep 2003 17:36:07 +0200' From: huber@mppmu.mpg.de (Joseph Huber)h0 Subject: Re: Remove satellite node from cluster?+ Message-ID: <TpHwcRDrqrVX@vms.mppmu.mpg.de>l   In article <e1d40caf.0309170119.2f414481@posting.google.com>, dieter.rossbach@gmx.de (=?ISO-8859-1?Q?Dieter__Ro=DFbach?=) writes:ME >> I can't just delete the files in [sys10...] and [sys010...] can I?0@ >> Aren't there a lot of aliases?  Should I do a SET FILE/REMOVE >> on all those files? > 1 > Before you delete any files in [sys10...], do a=( > $ set file/remove [sys10]syscommon.dir> > This is a link to sys$sysdevice:[vms$common] and a Delete onD > [sys10...] without removing that as some unwanted side effects :-)  7 And there is the official and recommended way to do it:r  B    @SYS$MANAGER:CLUSTER_CONFIG            if defined in DECnet/MOP or o=    @SYS$MANAGER:CLUSTER_CONFIG_LAN        if defined in LANcpn  a -- c>    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/   ------------------------------  % Date: Thu, 18 Sep 2003 00:20:11 +1000i: From: "David McKenzie" <david.mckenzie@paradigm-shift.biz>5 Subject: Re: Solaris Security (was: OpenVMS Security) C Message-ID: <3f686d91$0$95047$c30e37c6@lon-reader.news.telstra.net>?  ' >Non existant when they should be thereo >ort
 >Incorrect  L Incorrect is interesting. If that is really so, then why not legal recourse?  K I am always interested in the fees for lawyering in such actions, shamelesst troll   C IMHO it would make a company tortiously liable if if responded withs  knowingly incorrect information.  F It would constitute knowling breaching a duty of care that caused loss   More lawsuits I sayi  D Of course if it were not true, that opens up other possibilities :-)     -- b David McKenzie  7 David.Mckenzie@paradigm-shift.dot.biz  remove the "dot"o OpenVMS IT Privacy and Law   http://www.paradigm-shift.biz K "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com>D; wrote in message news:bjphqj$bqe$1@new-usenet.uk.sun.com...c Hoff Hoffman wrote:wG >   As a comparsion to the five year old CERT reports cited, as well asfJ >   the the Bind 8 bug found in early TCP/IP Services V5.3 releases though0 >   not in older releases nor in current ECOs... >eH >   91 matches for Solaris among the CERT Advisories -- not VulerabilityF >   Notes, or other areas -- just advisories.   Twenty breaches in theG >   first page of the display, from the month of July 2003 -- the other 7 >   five reports on the first page are newer than July.0 >e   I dont think you get it do you.k  D Regardless of how many CERT advisories there are that effect Solaris? making a comparison between those and the ones that Compaq have : fessed up to for OpenVMS is completely pointless because a@ significant proportion of the OpenVMS responses are demonstrably either:n  & Non existant when they should be there or	 Incorrect   7 How many other CERTS was OpenVMS actually vunerable to.t< I don't know and neither do your customers and by definition; CERT is actually the last place where you are going to findl out.  A A lot of CERTS are actually reported by vendors Sun among them in ? order to alert customers to vunerabilites that exist and shouldo be patched for.   : However Compaq/Digital never appeared to have followed the> strategy of reporting their vunerabilites for OpenVMS in patch reports.  S http://ftp1.support.compaq.com/public/vms/axp/v7.1-2/vms712_dw_mot_mup-v0100.READMEe= Why for example is this vunerability not listed in a specificj0 DECwindows advisory reported to CERT by Compaq ?  > As I said earlier using CERT as a source of comfort for people= who think that OpenVMS is bullet proof is a BS excercise thatS$ is best left to the marketing folks.   RegardsI Andrew HarrisonC9 >   The next twenty five look to be heavily in July, too.? > * >   Bind bugs reported in CA-1999-14, too. >  >nF > CERT Advisory CA-2001-15 Buffer Overflow In Sun Solaris in.lpd Print DaemonK > .. CERT  Advisory CA-2001-15 Buffer Overflow In Sun Solaris in.lpd Printo$ Daemon ... Solaris 2.6 for SPARC ...G > http://www.cert.org/advisories/CA-2001-15.html - 14.8KB - solaris: 21f > 28 Jul 03  > K > CERT Advisory CA-2002-34 Buffer Overflow in Solaris X Window Font ServiceeH > .. CERT  Advisory CA-2002-34 Buffer Overflow in Solaris X Window Font< Service ... Sun Microsystems Solaris 2.5.1 (Sparc/Intel) ...G > http://www.cert.org/advisories/CA-2002-34.html - 17.8KB - solaris: 21> > 28 Jul 03e >i? > CERT Advisory CA-1993-19 Solaris System Startup Vulnerability L > .. CERT  Advisory CA-1993-19 Solaris System Startup Vulnerability ... theJ system startup scripts on Solaris 2.x and Solaris x86 systems. The changes
 described ...nG > http://www.cert.org/advisories/CA-1993-19.html - 10.8KB - solaris: 17$ > 02 Sep 03h >:3 > CERT Advisory CA-1995-09 Solaris ps VulnerabilityhL > .. CERT  Advisory CA-1995-09 Solaris ps Vulnerability ... A vulnerabilityI exists in Solaris systems that allows a race condition to be exploited to:< gain root access. The essential problem is that the ps(1 ...G > http://www.cert.org/advisories/CA-1995-09.html - 18.8KB - solaris: 17  > 28 Jul 03a >b > Find SimilarE > CERT Advisory CA-1996-15 Vulnerability in Solaris 2.5 KCMS programssJ > .. CERT  Advisory CA-1996-15 Vulnerability in Solaris 2.5 KCMS programsK ... contact information has been replaced with CERT/CC contact information.n@ ... Note that this vulnerability also affects Solaris 2.5.1. ...G > http://www.cert.org/advisories/CA-1996-15.html - 13.1KB - solaris: 17o > 28 Jul 03a >bJ > http://www.cert.org/advisories/ CA-93.09a.SunOS.expreserve.vulnerability? > .. 1, 1993 REVISION NOTICE: SunOS/Solaris /usr/lib/expreserverK Vulnerability ... 4.1.2, 4.1.3, Solaris 2.0 (SunOS 5.0), Solaris 2.1 (SunOSC 5.1 ...rL > http://www.cert.org/advisories/ CA-93.09a.SunOS.expreserve.vulnerability - 4.8KB - solaris: 17v > 16 Mar 01a >t= > CERT Advisory CA-1996-16 Vulnerability in Solaris admintoolIJ > .. CERT  Advisory CA-1996-16 Vulnerability in Solaris admintool ... hasH received a report of a vulnerability in the Sun Microsystems Solaris 2.xF distribution involving the program admintool. This program is used ...G > http://www.cert.org/advisories/CA-1996-16.html - 13.7KB - solaris: 13w > 28 Jul 03  > 8 > CERT Advisory CA-1996-17 Vulnerability in Solaris vold@ > .. known to be present in Solaris 2.4 and Solaris 2.5. SolarisK distributions prior ... 2.4 and Solaris 2.5. Solaris distributions prior tor! Solaris 2.4 are also expected ... G > http://www.cert.org/advisories/CA-1996-17.html - 15.1KB - solaris: 13e > 28 Jul 03y >G. > CERT Advisory CA-2001-11 sadmind/IIS Worm...J >  ... Systems running unpatched versions of Solaris up to, and including,B Solaris 7 ... worm exploits a vulnerability in Solaris systems and" subsequently installs software ...G > http://www.cert.org/advisories/CA-2001-11.html - 16.6KB  - solaris: 9s >-' > ---------------------------- #includeo' <rtfaq.h> -----------------------------u4 >     For additional, please see the OpenVMS FAQ --  www.hp.com/go/openvms/faqt, >  --------------------------- pure personal# opinion ---------------------------1G >         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com  >i   ------------------------------    Date: 17 Sep 2003 02:20:52 -0700? From: Roy.Omond@BlueBubble.UK.Com (Roy.Omond@BlueBubble.UK.Com)m0 Subject: TCP/IP Services for VMS - sloppy output= Message-ID: <d98660c6.0309170120.53dcd016@posting.google.com>i   Gentle colleagues,  J anyone else come across this sloppy programming in TCP/IP Services for VMS3 (version v5.3 ECO 2 as far back as at least v5.1) ?.    ,    $ spawn/output=x.x tcpip show device/full    $ edit/edt x.xt  %    Note all these embedded <CR><LF>'sa      $ search x.x error1  ;    You basically get every "line", which is pretty useless.u  L I can come up with a workaround (see below), but I consider this a bug(let).   (Workaround)      $ create/fdl=sys$input x.x 	    recordi    format stream     $ define/user sys$output x.x.    $ tcpip show device/fullw  $ This produces the output I'd expect.  
 Comments ?  	 Roy Omond8 Blue Bubble Ltd.   ------------------------------  + Date: Wed, 17 Sep 2003 14:42:38 +0200 (MET)t9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>e4 Subject: Re: TCP/IP Services for VMS - sloppy output; Message-ID: <01L0RWKO9AN6AMK0HT@sysdev.deutsche-boerse.com>a  . >    $ spawn/output=x.x tcpip show device/full >    $ edit/edt x.xp > ' >    Note all these embedded <CR><LF>'so >  >    $ search x.x error  > = >    You basically get every "line", which is pretty useless.   9 Have you tried MIME?  What an example of bad programming.w   ------------------------------  % Date: Wed, 17 Sep 2003 11:15:32 -0400o* From: JF Mezei <jfmezei.spamnot@istop.com>4 Subject: Re: TCP/IP Services for VMS - sloppy output( Message-ID: <3F687A92.1A76F08@istop.com>  $ "Roy.Omond@BlueBubble.UK.Com" wrote:/ >>    $ spawn/output=x.x tcpip show device/fullo >    $ edit/edt x.xa > ' >    Note all these embedded <CR><LF>'sr  K Yep, this was discussed previously. Ruins the ability to pipe the output ofpN TCPIP commands and then use serach on that output. Those commands send a wholeI big chunk of text at a time, and this big chunk doesn't fit into the PipeAM buffers so a subsequent search command on the pipe will get truncated chunks.,   ------------------------------  % Date: Wed, 17 Sep 2003 11:24:44 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>4 Subject: Re: TCP/IP Services for VMS - sloppy output( Message-ID: <3F687CBA.17D553A@istop.com>   Phillip Helbig wrote: ; > Have you tried MIME?  What an example of bad programming.a  K In fairness, LATCP is also extremely bad with overuse of screen control fora1 what should have been a simple line mode utility.k   ------------------------------  + Date: Wed, 17 Sep 2003 18:23:00 +0200 (MET)e9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>L4 Subject: Re: TCP/IP Services for VMS - sloppy output; Message-ID: <01L0S49WIOYEAOQMVC@sysdev.deutsche-boerse.com>t  = > > Have you tried MIME?  What an example of bad programming.e > M > In fairness, LATCP is also extremely bad with overuse of screen control foru3 > what should have been a simple line mode utility.?  I True.  I haven't actually parsed the output.  Is it "intelligent" screen  & control, or more in the style of MIME?   ------------------------------  % Date: Wed, 17 Sep 2003 12:54:22 -0400e) From: Jfmezei <jfmezei.spamnot@istop.com>w4 Subject: Re: TCP/IP Services for VMS - sloppy output) Message-ID: <3F6891B5.517B2536@istop.com>r   Phillip Helbig wrote:mO > > In fairness, LATCP is also extremely bad with overuse of screen control fora5 > > what should have been a simple line mode utility.b > J > True.  I haven't actually parsed the output.  Is it "intelligent" screen( > control, or more in the style of MIME?  " LATCP is worse offender than MIME.  D However, in fairness, one can define LCP :== "$LATCP" and then issueF individual commands from the $ prompt without having to go through itsI cumbersome screen handling. But I know that the help from within LATCP is'/ pretty well useless due to the screen handling.    ------------------------------    Date: 17 Sep 2003 09:30:34 -07001 From: keithparris_NOSPAM@yahoo.com (Keith Parris)iP Subject: There is no Joy in Sunville -- mighty Billy has struck out [on his own]= Message-ID: <cf15391e.0309170830.1eb73ef3@posting.google.com>d  > "Sun announced on September 9, 2003, that co-founder and ChiefF Scientist Bill Joy is leaving the company. As chief scientist, Joy has? been responsible for leading Sun's technical strategy since theaE company was founded in 1982. Now Sun is at the most critical juncture @ in its history, facing declining revenues and market share while: coming under attack from the shift to industry standards."  ; CNet says at http://news.com.com/2100-1012_3-5073205.html :rE "'I have decided the time is now right for me to move on to different % challenges,' Joy said in a statement./ ...r@ But some customers might perceive Joy's resignation as a sign ofF overall financial troubles within the company. 'It's just like when EdF Zander (Sun's former president and chief operating officer) left. SomeB see that as rats deserting a sinking ship,' Plummer said. 'Joy has8 been in on every innovative idea since Sun was founded.'  C That, in turn, could further hurt Sun's already rocky reputation in F the marketplace. 'A lot of (Sun) customers are looking for a reason to" pick someone else,' Plummer said."  E It's no fun to work at Sun anymore, apparently.  In any case, there's ) certain to be no Joy in Sun's future. :-)e   ------------------------------  # Date: Wed, 17 Sep 2003 17:33:15 GMT 9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> T Subject: Re: There is no Joy in Sunville -- mighty Billy has struck out [on his own]1 Message-ID: <vV0ab.4956$ZQ.3709@news.cpqcorp.net>t  L Wanna bet that they are about to kill Sparc and move to a merchant chip like the AMD?    > "Keith Parris" <keithparris_NOSPAM@yahoo.com> wrote in message7 news:cf15391e.0309170830.1eb73ef3@posting.google.com...-@ > "Sun announced on September 9, 2003, that co-founder and ChiefH > Scientist Bill Joy is leaving the company. As chief scientist, Joy hasA > been responsible for leading Sun's technical strategy since theoG > company was founded in 1982. Now Sun is at the most critical juncturewB > in its history, facing declining revenues and market share while< > coming under attack from the shift to industry standards." >o= > CNet says at http://news.com.com/2100-1012_3-5073205.html :iG > "'I have decided the time is now right for me to move on to different ' > challenges,' Joy said in a statement.  > ...sB > But some customers might perceive Joy's resignation as a sign ofH > overall financial troubles within the company. 'It's just like when EdH > Zander (Sun's former president and chief operating officer) left. SomeD > see that as rats deserting a sinking ship,' Plummer said. 'Joy has: > been in on every innovative idea since Sun was founded.' >cE > That, in turn, could further hurt Sun's already rocky reputation inmH > the marketplace. 'A lot of (Sun) customers are looking for a reason to$ > pick someone else,' Plummer said." >lG > It's no fun to work at Sun anymore, apparently.  In any case, there'sa+ > certain to be no Joy in Sun's future. :-)c   ------------------------------    Date: 17 Sep 2003 06:04:02 -0700? From: Roy.Omond@BlueBubble.UK.Com (Roy.Omond@BlueBubble.UK.Com)eD Subject: Very strange Monitor (and possible VMS scheduler) behaviour< Message-ID: <d98660c6.0309170504.f021543@posting.google.com>   Gentle VMS colleagues,  C apologies for the length of the following (I have made every effortf# to cut its size down to a minimum).   I I am completely baffled by the behaviour exhibited at one of my (biggest)w customers' site:  	 Scenario:   F non-clustered ES40, VMS 7.3-1, 4 x 833 MHz processors, 2 Gbyte memory,* 2 x Gigabit Ethernet adaptors fairly busy.  E This is a static and dedicated workload, processing a fairly constant-E stream of heavy Gigabit Ethernet traffic (buffered I/O rate is prettyaF much a constant ca. 9,000 I/O's per second).  The processing processesD are all running at realtime priorities (16, 17 and 18), and are hardA affinitised, RDX_RDC1 to CPU 1, the rest between CPU 2 and CPU 3.sC CPU 0, as expected, is running at ca. 20% Interrupt State, and next  to nothing else.  ? I've included below some monitor output that has me stumped for @ an explanation.  Observe that the the Monitor Modes output shows@ a reasonably constant Kernel Mode + User Mode, both of which are= attributable to running processes (cf. Interrupt State and MPy? Sync which cannot).  The Monitor Process/TopCPU output seems to @ cycle (over something like 70-80 seconds) between showing fairly? high individual process utilization and a ludicrous almost zerog@ utilization (note that the workload is pretty constant as can beE seen from Monitor Process/TopBIO (which I could supply if necessary). 7 I cannot explain this phenomenon at all to my customer.O  C In addition, I can "watch" an individual process (e.g. the RDX_RDC1oC which is the only one affinitised to CPU 1) either in SDA or simplygA in Show Process/Continuous, and I can see some strange behaviour.e; The process will be running fine, clocking up CPU time at a > commensurate rate with the number of buffered IO's, then everyB 70-80 seconds or so, the buffered I/O rate is the same, but no CPU@ time is clocking up - it's almost as if the VMS scheduler is not4 adding any CPU time to any process for a long while.  # Pagefaulting is totally negligible.-  A Additional notes:  quantum end scheduler processing does not playf; a part here (since the processes are in the realtime zone).t. SYSGEN IOTA is set to 0 (makes no difference).  @ Any help with a plausible explanation would be most appreciated, (Hoff ?)  	 Roy Omondd Blue Bubble Ltd.  A --- Monitor output (Monitor Modes, Process/TopCPU/Interval=1) ---   3                             OpenVMS Monitor Utility-3             +-----+         TIME IN PROCESSOR MODES ,             | CUR |              on node XXX3             +-----+         16-SEP-2003 14:58:23.69sC Combined for  4 CPUs     0         100       200       300      400sC                          + - - - - + - - - - + - - - - + - - - - -+f  Interrupt State      32 |***-  MP Synchronization   15 |*    Kernel Mode          62 |******!  User Mode            75 |*******j/  Idle Time           217 |*********************sC                          + - - - - + - - - - + - - - - + - - - - -+v  3                             OpenVMS Monitor Utilityi3                              TOP CPU TIME PROCESSESo,                                  on node XXX3                             16-SEP-2003 14:58:25.02oC                          0         25        50        75       100eC                          + - - - - + - - - - + - - - - + - - - - -+h#  00000253  RDX_ILNM  26  **********l  00000255  RDX_ILSM   8  ***  00000251  RDX_RDC1   5  **.  00000257  RDX_RR01   5  **d  0000025B  RDX_RR03   3  *  00000266  RDX_RR08   3  *  000001E4  RDX_RR07   2n  0000026C  RDX_RR11   2,  3                             OpenVMS Monitor UtilityI3             +-----+         TIME IN PROCESSOR MODES ,             | CUR |              on node XXX3             +-----+         16-SEP-2003 14:58:26.35eC Combined for  4 CPUs     0         100       200       300      400tC                          + - - - - + - - - - + - - - - + - - - - -+   Interrupt State      33 |***   MP Synchronization   17 |*i  Kernel Mode          59 |***** !  User Mode            74 |*******d/  Idle Time           217 |********************* C                          + - - - - + - - - - + - - - - + - - - - -+   3                             OpenVMS Monitor Utilityo3                              TOP CPU TIME PROCESSESs,                                  on node XXX3                             16-SEP-2003 14:58:27.69iC                          0         25        50        75       100 C                          + - - - - + - - - - + - - - - + - - - - -+   00000253  RDX_ILNM  14  *****  00000257  RDX_RR01   3  *  00000255  RDX_ILSM   3  *  0000025B  RDX_RR03   2a  00000251  RDX_RDC1   2m  00000266  RDX_RR08   2c  0000026E  RDX_RR12   2g  0000026C  RDX_RR11   29  3                             OpenVMS Monitor Utilityg3             +-----+         TIME IN PROCESSOR MODES ,             | CUR |              on node XXX3             +-----+         16-SEP-2003 14:58:29.02lC Combined for  4 CPUs     0         100       200       300      400 C                          + - - - - + - - - - + - - - - + - - - - -+s  Interrupt State      32 |***e  MP Synchronization   19 |*e   Kernel Mode          60 |******   User Mode            68 |******0  Idle Time           221 |**********************C                          + - - - - + - - - - + - - - - + - - - - -+h  3                             OpenVMS Monitor Utilitya3                              TOP CPU TIME PROCESSES ,                                  on node XXX3                             16-SEP-2003 14:58:30.35-C                          0         25        50        75       100oC                          + - - - - + - - - - + - - - - + - - - - -+R  00000253  RDX_ILNM   8  ***  0000026C  RDX_RR11   20  00000259  RDX_RR02   2c  0000026A  RDX_RR10   1   0000025D  RDX_RR04   1p  0000025B  RDX_RR03   1   0000026E  RDX_RR12   1t  00000268  RDX_RR09   1H  3                             OpenVMS Monitor Utility 3             +-----+         TIME IN PROCESSOR MODESa,             | CUR |              on node XXX3             +-----+         16-SEP-2003 14:58:31.69 C Combined for  4 CPUs     0         100       200       300      400nC                          + - - - - + - - - - + - - - - + - - - - -+:  Interrupt State      32 |***T  MP Synchronization   14 |*   Kernel Mode          59 |***** !  User Mode            75 |*******r0  Idle Time           221 |**********************C                          + - - - - + - - - - + - - - - + - - - - -+d  3                             OpenVMS Monitor Utilityl3                              TOP CPU TIME PROCESSESN,                                  on node XXX3                             16-SEP-2003 14:58:33.02cC                          0         25        50        75       100 C                          + - - - - + - - - - + - - - - + - - - - -+e  00000251  RDX_RDC1  11  ****b  00000255  RDX_ILSM  11  ****d  0000025D  RDX_RR04  10  ****.  0000025F  RDX_RR05   8  ***  00000253  RDX_ILNM   6  **a  00000261  RDX_RR06   4  *  000002A2  Roy Omond  1   000001E4  RDX_RR07   1   3                             OpenVMS Monitor Utilityn3             +-----+         TIME IN PROCESSOR MODESe,             | CUR |              on node XXX3             +-----+         16-SEP-2003 14:58:34.35iC Combined for  4 CPUs     0         100       200       300      4004C                          + - - - - + - - - - + - - - - + - - - - -+5  Interrupt State      33 |***.  MP Synchronization   14 |*v   Kernel Mode          61 |******!  User Mode            75 |*******c0  Idle Time           220 |**********************C                          + - - - - + - - - - + - - - - + - - - - -+   3                             OpenVMS Monitor Utilityo3                              TOP CPU TIME PROCESSESn,                                  on node XXX3                             16-SEP-2003 14:58:35.69pC                          0         25        50        75       100eC                          + - - - - + - - - - + - - - - + - - - - -+ $  00000255  RDX_ILSM  28  ***********#  00000251  RDX_RDC1  27  **********l  00000261  RDX_RR06  15  ******i  00000253  RDX_ILNM  12  ****   0000025D  RDX_RR04  12  ****s  0000025F  RDX_RR05  11  ****n  000001E4  RDX_RR07   6  **   00000266  RDX_RR08   3  *  3                             OpenVMS Monitor Utility 3             +-----+         TIME IN PROCESSOR MODEST,             | CUR |              on node XXX3             +-----+         16-SEP-2003 14:58:37.02 C Combined for  4 CPUs     0         100       200       300      400PC                          + - - - - + - - - - + - - - - + - - - - -+l  Interrupt State      20 |**  MP Synchronization   11 |*d  Kernel Mode          42 |****  User Mode            47 |****6  Idle Time           281 |****************************C                          + - - - - + - - - - + - - - - + - - - - -+d  3                             OpenVMS Monitor Utilityl3                              TOP CPU TIME PROCESSESp,                                  on node XXX3                             16-SEP-2003 14:58:38.21EC                          0         25        50        75       100oC                          + - - - - + - - - - + - - - - + - - - - -+y  00000253  RDX_ILNM   3  *  000002A2  Roy Omond  2t  00000259  RDX_RR02   2c  0000026A  RDX_RR10   1K  00000268  RDX_RR09   1G  00000276  RDX_MCD1@  0000026C  RDX_RR11n  00000266  RDX_RR08/   ------------------------------  % Date: Wed, 17 Sep 2003 10:33:49 -0400d& From: "Island" <dbturner@islandco.com>, Subject: We need your Ds20 and DS20e Systems/ Message-ID: <vmgsau9mcc62f0@news.supernews.com>y  / Please call or email with what you have to sell    Thanks     -- i David B Turner Island Computers US Corporation  2700 Gregory St., Suite 180s Savannah GA 31404e Tel: 912 447 6622  Fax: 912 201 0402  Email: dbturner@hpaq.net http://www.hpaq.netl   ------------------------------  + Date: Wed, 17 Sep 2003 10:54:14 +0000 (UTC) + From: "Craig Cooke" <storm7@btinternet.com>  Subject: WWIDMGR - HSG80 -> EVA_0 Message-ID: <bk9egm$55l$1@sparta.btinternet.com>   Hi,T  H Last night we moved our data from an HSG80 to an EVA - the systems are aL cluster consisting of an ES45 and an ES40 - we did WWIDMGR -quickset -udid 1& on the ES45 - after an init we can seeK $1$DAG1 - (our boot disk) - so we can boot the ES45 - we do the same on the(C ES40 (WWIDMGR -quickset -udid 2 as $1$DGA2 is the boot disk on thatnL system) - we are informed of the disks that will be availabe after an INIT -J DGA2 is mentioned - we do an INIT and voila - it's gone - SHOW DEV doesn'tD see the disk at all - anyone had any similar problems - am I missing something ?i  I A quick search of GOOGLE tells me we are not the only site to suffer this 2 problem - buit I cannot seeem to find a resolution    ? Any help appreciated (usually in the form of Nic Clewes :-)   )-   Thanks.e   -- Regards0   Craig Cooke0   ------------------------------    Date: 17 Sep 2003 07:49:34 -0500+ From: young_r@encompasserve.org (Rob Young) # Subject: Re: WWIDMGR - HSG80 -> EVAl3 Message-ID: <Us7iD37jpdFs@eisner.encompasserve.org>e  ^ In article <bk9egm$55l$1@sparta.btinternet.com>, "Craig Cooke" <storm7@btinternet.com> writes: > Hi,  > J > Last night we moved our data from an HSG80 to an EVA - the systems are aN > cluster consisting of an ES45 and an ES40 - we did WWIDMGR -quickset -udid 1( > on the ES45 - after an init we can seeM > $1$DAG1 - (our boot disk) - so we can boot the ES45 - we do the same on thepE > ES40 (WWIDMGR -quickset -udid 2 as $1$DGA2 is the boot disk on thatpN > system) - we are informed of the disks that will be availabe after an INIT -L > DGA2 is mentioned - we do an INIT and voila - it's gone - SHOW DEV doesn'tF > see the disk at all - anyone had any similar problems - am I missing
 > something ?  > K > A quick search of GOOGLE tells me we are not the only site to suffer thist4 > problem - buit I cannot seeem to find a resolution >  > A > Any help appreciated (usually in the form of Nic Clewes :-)   )a >   = 	It's not a problem.  It is a feature.  You can only quickset   	ONE device.  From a prior post:  c http://groups.google.com/groups?selm=dwMv169VmRao%40eisner.encompasserve.org&oe=UTF-8&output=gplaine  + From: young_r@encompasserve.org (Rob Young). Newsgroups: comp.os.vmsr# Subject: Re: Fiber channel questione  Date: 15 Apr 2003 20:47:04 -0500  7 	You can only quickset one device.  You have 4 slots so B 	Bart is right.  The slots in use and how to do more than one are 4 	all in the wwidmgr manual found on the firmware CD.   	Here is an example of a show:  < 	[This URL is stale.  You will either have to muck around on= 	HP web finding it or pull out a firmware CD and pull it off]b  . http://www.keltec.co.uk/whitepaper/Wwidmgr.pdf 	n 	[Manual page 30]r   P08>>>wwidmgr -show ev  @ wwid0 13 1 WWID:01000010:6000-1fe1-0000-04d0-0009-8170-1057-002b wwid1, wwid2  wwid3S  7    ! here is wwid0 in use wwid1 through wwid3 are free./  < N1 50001fe1000074b1    ! N1 in use , N2 through N4 are free. N3 N4   P08>>>wwidmgr -show port [0] 1000-0000-c920-ca720 [1] 2006-0060-6900-0953n [2] 20fc-0060-6900-0953u [3] 5000-1fe1-0000-74a1n [4] 5000-1fe1-0000-74b1 (N1) [5] 5000-1fe1-0000-74c2m [6] 5000-1fe1-0000-74b2   - After the -quickset, one device is reachable.s, The ev display shows one disk wwid stored in, the non-volatile ev wwid0, and one port path$ is stored in the non-volatile ev N1.  ? 	The examples that follow show how to use set port and set wwidt? 	to make more than one device available from console, see pagess 	31-33.a  H > With a properly configured SAN you have four access paths to the disk.   	Using which hardware?  ;')e  4 > Two controllers (HSG80 redundant/multi-path pair):   	Oh, that hardware!e   ---t   				Rob    ------------------------------    Date: 17 Sep 2003 07:48:55 -0700) From: oliver.steeples@compaq.com (Oliver) # Subject: Re: WWIDMGR - HSG80 -> EVAe= Message-ID: <e5029990.0309170648.1304dabc@posting.google.com>    Hi,eC   there are a few things you can try.  A wwidmgr -show ev and checks& there are some values in WWIDx and Nx.  D The best bet is to do a quickset and note the 6000-xxxx wwid of dga2F (well the disk with an identifier of 2, hopefully this has been set onC the EVA).  Then do a wwidmgr -show wwid and find the disk WWID.  Itt  will have an item number in [x].  # Then do a wwidmgr -quickset -item xe   This 'should' work.    Regards,
     Oliver   ------------------------------  % Date: Wed, 17 Sep 2003 16:39:36 +01003! From: "Craig Cooke" <cs@dabs.com><# Subject: Re: WWIDMGR - HSG80 -> EVAm5 Message-ID: <uf85mHTfDHA.2252@juno.intranet.dabs.com>c  6 "Craig Cooke" <storm7@btinternet.com> wrote in message* news:bk9egm$55l$1@sparta.btinternet.com... > Hi,o >aJ > Last night we moved our data from an HSG80 to an EVA - the systems are aL > cluster consisting of an ES45 and an ES40 - we did WWIDMGR -quickset -udid 1P( > on the ES45 - after an init we can seeI > $1$DAG1 - (our boot disk) - so we can boot the ES45 - we do the same onl the E > ES40 (WWIDMGR -quickset -udid 2 as $1$DGA2 is the boot disk on that7G > system) - we are informed of the disks that will be availabe after ang INIT -L > DGA2 is mentioned - we do an INIT and voila - it's gone - SHOW DEV doesn'tF > see the disk at all - anyone had any similar problems - am I missing
 > something ?5 >0K > A quick search of GOOGLE tells me we are not the only site to suffer this 4 > problem - buit I cannot seeem to find a resolution >e > A > Any help appreciated (usually in the form of Nic Clewes :-)   )s > 	 > Thanks.  >  > --	 > RegardsS >U
 > Craig CookeI >  >i  H All fixed now - the problem was that the card in the ES40 was running atG 1Gb - the port on the SAN switch was set to auto-negotiate - but did'ntiI auto-negotiate down to 1Gb - we've inserted a new 2Gb card and all is nows working o.k.   Craig.   ------------------------------   End of INFO-VAX 2003.516 ************************