1 INFO-VAX	Wed, 05 Mar 2003	Volume 2003 : Issue 126       Contents: %UAF-E-PWDSYNTAX9 ?= Install patch vms731_sys_v0300 > v0200 and ACRTL_v010. = Re: ?= Install patch vms731_sys_v0300 > v0200 and ACRTL_v010. 1 Re: About VMS on Itanium in the French "Press"...  Alpha doesn't boot& Alpha Station 200 4/233 Memory Upgrade* Re: Alpha Station 200 4/233 Memory Upgrade* Re: Alpha Station 200 4/233 Memory Upgrade Re: C-Kermit 8.0.208 cdrom 	 Re: cdrom  Re: CLI question for HP reps Re: CLI question for HP reps Re: CLI question for HP reps DCL COPY Re: DCL COPY Re: DCL COPY Re: DCL COPY  Re: disk queue length in monitor  Re: disk queue length in monitor6 Re: Editor wars - Was: Re: Upcoming VMS improvements ?6 Re: Editor wars - Was: Re: Upcoming VMS improvements ?6 RE: Editor wars - Was: Re: Upcoming VMS improvements ?6 Re: Editor wars - Was: Re: Upcoming VMS improvements ?6 Re: Editor wars - Was: Re: Upcoming VMS improvements ? Fibre Channel and SCS  Re: Fibre Channel and SCS  Re: Fibre Channel and SCS  Re: Fibre Channel and SCS  Re: Fibre Channel and SCS ; Re: hobbiest RDB/Sql*Services security (unrelated question) ; Re: hobbiest RDB/Sql*Services security (unrelated question) B Re: Ideas for implementing fork() on VMS ?, was: Re: fork() on VMS Re: Internet attacks/ Re: Linus trashes Itanium - he's scared of VMS! / Re: Linus trashes Itanium - he's scared of VMS! / Re: Linus trashes Itanium - he's scared of VMS! / Re: Linus trashes Itanium - he's scared of VMS! / Re: Linus trashes Itanium - he's scared of VMS!  Re: Mailbox and their logicals Re: Mailbox and their logicals MS Virus Re: MS Virus Re: MS Virus Re: MS VirusP Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha Retai( OT:  National Moratorium to Stop the War Re: Questions on SSH for VMS, Re: running autogen on alternate system disk Re: SCS/DECnet over IP3 Re: SCSI cluster disk thrashing between connections 4 Re: Sending OPCOM or Security Audit events via email4 Re: Sending OPCOM or Security Audit events via email4 Re: Sending OPCOM or Security Audit events via email7 Re: strange behaviour of SET ENVIRONMENT/NODE in SYSMAN 7 Re: strange behaviour of SET ENVIRONMENT/NODE in SYSMAN  Re: unix Re: unix Re: unix/ Re: Upgrade to OPEN VMS 7.3-1 - Patches needed? / Re: Upgrade to OPEN VMS 7.3-1 - Patches needed? < Re: VAXELN vs Posix Threads (Was: DECthreads problem on VAX)< Re: VAXELN vs Posix Threads (Was: DECthreads problem on VAX) Re: VMS Backup solutions Re: VMS Backup solutions Re: VMS Backup solutions Re: VMS Backup solutions Re: VMS Backup solutions Re: VMS Backup solutions! [OT] ex-Etrade CEO house for sale  Re: [OT}:to Sue   F ----------------------------------------------------------------------  $ Date: Wed, 5 Mar 2003 09:42:15 +0100, From: "Rainer Giese" <waste.not@welcome.net> Subject: %UAF-E-PWDSYNTAX 6 Message-ID: <b44d9u$1r3jf7$1@ID-138444.news.dfncis.de>  I I've made MCR AUTHORIZE MODIFY username /PASSWORD=password in a procedure H and found, that invalid passwort-characters produce  an %UAF-E-PWDSYNTAXE error message, but no error status code and no ON ERROR trap. And the H message also not appears, when redirecting SYS$OUTPUT and SYS$ERROR to a file.    Anyone knows a reason ?    Regards  Rainer Giese   ------------------------------  % Date: Wed, 05 Mar 2003 16:05:37 +0100 . From: "Hans M. Aus" <aus@vim.uni-wuerzburg.de>B Subject: ?= Install patch vms731_sys_v0300 > v0200 and ACRTL_v010.B Message-ID: <aus-2E1D8D.16053705032003@wrzx08.rz.uni-wuerzburg.de>  F ?= Procedure for installing new patch vms731_sys_v0300 over v0200 and  ACRTL_v010.   4 vms731_sys is a prerequisite for vms731_acrtl_v0100.  ? What procedure do I need to follow in order to install the new  : vms731_sys patch on a system that already has the patches 2 vms731_sys_v0200 and vms731_acrtl_v0100 installed?  G Do I need to remove both vms731_sys_v0200 and vms731_acrtl_v0100 first? : Then install vms731_sys_v0300, followed by a reinstall of  vms731_acrtl_v0100?    --  B Cheers, Hans M. Aus, Wuerzburg, Germany,  aus@vim.uni-wuerzburg.de   ------------------------------  $ Date: Wed, 5 Mar 2003 11:20:29 -0500 From: norm.raphael@metso.comF Subject: Re: ?= Install patch vms731_sys_v0300 > v0200 and ACRTL_v010.? Message-ID: <OFD687299A.F4C429A5-ON85256CE0.0058A1D8@metso.com>    With _no_ authority...  @ Since ACRTL V1 shares nothing in SYS V2 or SYS V3, it seems that& just installing SYS V3 should be fine.I If you are paranoid, you could reinstall ACRTL V1 afterwards, but is does  not seem necessary.   F There is no way to remove an ECO with PCSI, so "remove" is a dangerous thing to attempt.   F From:  "Hans M. Aus" <aus@vim.uni-wuerzburg.de> on 03/05/2003 10:05 AM  : Please respond to "Hans M. Aus" <aus@vim.uni-wuerzburg.de>   To:    Info-VAX@Mvb.Saic.Com cc:   E Subject:    ?= Install patch vms731_sys_v0300 > v0200 and ACRTL_v010.     E ?= Procedure for installing new patch vms731_sys_v0300 over v0200 and  ACRTL_v010.   4 vms731_sys is a prerequisite for vms731_acrtl_v0100.  > What procedure do I need to follow in order to install the new9 vms731_sys patch on a system that already has the patches 2 vms731_sys_v0200 and vms731_acrtl_v0100 installed?  G Do I need to remove both vms731_sys_v0200 and vms731_acrtl_v0100 first? 9 Then install vms731_sys_v0300, followed by a reinstall of  vms731_acrtl_v0100?    --B Cheers, Hans M. Aus, Wuerzburg, Germany,  aus@vim.uni-wuerzburg.de   ------------------------------  $ Date: Wed, 5 Mar 2003 13:31:57 +0100) From: "remy.younes" <remy.younes@free.fr> : Subject: Re: About VMS on Itanium in the French "Press"...3 Message-ID: <3e65ee5a$0$2695$626a54ce@news.free.fr>   L It's the line 14, also known as 'Meteor'. Was the first full automatic metroJ line in the world (no driver on board, and so on). The Control Center is aE cluster of 3 Alpha 4000, with more or less 10 alpha-500 workstations.   C I used to be the system engineer responsible for the system/network  installation in 1998.     J "JF Mezei" <jfmezei.spamnot@vl.videotron.ca> a crit dans le message news:$ 3E638A23.8BA1B014@vl.videotron.ca... > Philippe Bocher wrote:I > >     the good Url is http://www.01net.com/article/201922.html, I don't  know9 > > why I've tried to use minilien for a so short URL ;-(  > 8 > I tried it again, but this time it was OK on netscape. > J > Interesting that the article makes its sound as if the port is complete. > J > Have the engineers fixed the "DIR" command to show the current directory at5 > the top yet as it does on the real version of VMS ?  > I > What I found most interesting is that they reveal that one of the metro  lines L > in PARIS uses VMS to runthe trains. That could be one heck of a good story for , > VMS marketing or even the VMS newsletters.   ------------------------------  % Date: Wed, 05 Mar 2003 14:25:06 +0100 6 From: Horst Drechsel <ai05@sternwarte.uni-erlangen.de> Subject: Alpha doesn't boot < Message-ID: <00A1C6AB.7B7C4C66.9@sternwarte.uni-erlangen.de>   Hi all,   >    one of our VMS (7.2-1) cluster members - AlphaPC164 500 MHz: (chip 21164A) doesn't boot anymore. It comes to hang right- at the beginning of boot (over EWA0 adapter):       Try MOP boot     ...........=    MOP booting sequence starts normally, server receives boot      request, says ...   ;    %%%%%%%%%%%  OPCOM   5-MAR-2003 12:58:26.91  %%%%%%%%%%% #    Message from user DECNET on A400 +    DECnet event 0.3, automatic line service 2    From node 1.400 (A400),  5-MAR-2003 12:58:22.867    Circuit EWA-0, Load, Successful, Node = 1.16 (LUCKY) T    File = A400$DKA0:<SYS15.>, Operating system, Ethernet address = 08-00-2B-E6-9B-5C      client says ...      (normal messages up to:)     ...&    Setting affinity to the primary CPU    Jumping to bootstrap mode  0         >>>>>>>>>>  and here it hangs  <<<<<<<<<    E    The problem appeared after we had a crash of the main power supply A on this node, which was replaced, so that we could regularly boot @ once again after the repair and use the system for a short time.G Before returning to normal operation we shut down again to exchange the G noisy CPU cooling fan. After replacement we tried to boot again, but no % way (see the above described result).   D    To exclude a not properly functioning network card, we plugged itH into a different PCI slot. We also tried to boot from the OS CD, but got exactly the same behavior.  G    Maybe someone has experienced a similar case and could tell us where 3 to locate the problem (damaged CPU or main board?)?   
    Thanks,             Horst       --M  **************************************************************************** )   Horst Drechsel                          L   Dr. Remeis Observatory                 drechsel@sternwarte.uni-erlangen.deL   Astronomical Institute                             Phone: +49-951-95222-15L   University Erlangen-Nuernberg                        Fax: +49-951-95222-22*   Sternwartstr.7, D-96049 Bamberg, GermanyM  ****************************************************************************    ------------------------------  $ Date: Wed, 5 Mar 2003 07:05:26 -0700. From: "David D Miller" <ddmiller@raytheon.com>/ Subject: Alpha Station 200 4/233 Memory Upgrade F Message-ID: <OFD80C3F53.E8C099F2-ON07256CE0.004CC2FD@rsc.raytheon.com>   Folks   J I'd like to upgrade my Alpha Station 200 memory from 80 MB to <more>MB.  ID need help -- any experience/suggestions out there?  FYI, 80MB is 96%G utilized on a hobby 7.3-1 machine with minimum TCPIP support and Motif.    Questions that occur to me are:    1. Max size? 2. Upgrade in pairs?5 3. What kind of memory is it anyway?  Commodity SIMM? 9 4. Any only line discussion or documents I can reference?    dave.    ------------------------------  $ Date: Wed, 5 Mar 2003 09:42:29 -06002 From: "Stuart Johnson" <ssj152 AT charter DOT net>3 Subject: Re: Alpha Station 200 4/233 Memory Upgrade / Message-ID: <v6c6o5dobugc2e@corp.supernews.com>   9 "David D Miller" <ddmiller@raytheon.com> wrote in message @ news:OFD80C3F53.E8C099F2-ON07256CE0.004CC2FD@rsc.raytheon.com... > Folks  > L > I'd like to upgrade my Alpha Station 200 memory from 80 MB to <more>MB.  IF > need help -- any experience/suggestions out there?  FYI, 80MB is 96%I > utilized on a hobby 7.3-1 machine with minimum TCPIP support and Motif.  > ! > Questions that occur to me are:  >  > 1. Max size? > 2. Upgrade in pairs?7 > 3. What kind of memory is it anyway?  Commodity SIMM? ; > 4. Any only line discussion or documents I can reference?  >  > dave.  >  > K David, I have a similar system, a 200 4/100 and I bought commodity ram that G works fine. It needs to be real (true) parity, fast page mode, 60-70 ns D speed. They are installed in pairs on most Alpha's. I bought mine at0 www.kahlon.com - you might look at their models:  5 KAH4X36-60   - 16 mb currently $9 (2 of these = 32mb) 5 KAH8X36-60   - 32mb currently $16 (2 of these = 64mb) 6 KAH16X36-60 - 64 mb currently $32 (2 of these = 128mb)  G You can also use their "select by manufacturer" section and see several H models of ram listed for this exact computer under "Digital". Prices areJ different because they are selling kits of TWO on this page. They show 384( mb as the maximum supported by this box.  " KDEC77C -  32 mb, currently $16.00" KDEC77D -  64 mb, currently $29.50# KAH1208 -  128 mb, currently $56.00   7 As you can see, they are slightly cheaper here in kits.   I I do not work with or have any relationship with Kahlon other than having K bought memory from them several times and never having had any problem with K them or their products. I've also used this memory in a Multia (UDB) and an  AlphaServer 1000A.   Regards, Stuart Johnson   ------------------------------  $ Date: Wed, 5 Mar 2003 09:37:35 -0700. From: "David D Miller" <ddmiller@raytheon.com>3 Subject: Re: Alpha Station 200 4/233 Memory Upgrade F Message-ID: <OF5CB38DBD.18968948-ON07256CE0.005B263E@rsc.raytheon.com>  B Many thanks to Mr Schweda and Mr Johnson for their excellent tips.   dave.          Folks   J I'd like to upgrade my Alpha Station 200 memory from 80 MB to <more>MB.  ID need help -- any experience/suggestions out there?  FYI, 80MB is 96%G utilized on a hobby 7.3-1 machine with minimum TCPIP support and Motif.    Questions that occur to me are:    1. Max size? 2. Upgrade in pairs?5 3. What kind of memory is it anyway?  Commodity SIMM? 9 4. Any only line discussion or documents I can reference?    dave.    ------------------------------  % Date: Wed, 05 Mar 2003 09:25:53 -0500 ! From: Jim Agnew <jpagnew@vcu.edu>  Subject: Re: C-Kermit 8.0.208 ' Message-ID: <3E6608F1.C8779A85@vcu.edu>   A hhmm... I'd *never* thought of it that way. sounds faintly pretty 
 reasonable...    jim   * Alan Winston - SSRL Admin Cmptg Mgr wrote: > d > In article <3E64D4B1.1E5A5CC3@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes: > >Frank da Cruz wrote: L > >> Eased but not lifted.  This is why we are allowed to put the source outO > >> in the open, but not the binaries.  Previously we couldn't put the sources  > >> up either.  > > E > >Wow, talk about politicians really being so clueless. What sort of O > >justification was given for not distributing binaries, but allowing source ?  > O > I don't know the answer, but I know an argument: Source is considered speech, + > as in freedom of speech; binaries aren't.  > 	 > -- Alan  > Q > =============================================================================== 2 >  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUO >  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056 O >  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025 Q > ===============================================================================    --  F "4,000 years ago I made a mistake."  Elrond Half-Elven, in "Fellowship of the Ring"   ------------------------------   Date: 4 Mar 2003 23:44:17 -0800 " From: asmeda@supereva.it (Giorgio) Subject: cdrom< Message-ID: <609321cb.0303042344.886dbeb@posting.google.com>  B how to mount for read cdrom iso or joljet on vax 4000vls on vms5.5 thank you .    ------------------------------  $ Date: Wed, 5 Mar 2003 13:56:05 +0100E From: "Eberhard Heuser-Hofmann" <vaxinf@chclu.chemie.uni-konstanz.de>  Subject: Re: cdrom0 Message-ID: <000701c2e316$97489fe0$fe7aa8c0@WS1>   ----- Original Message -----$ From: "Giorgio" <asmeda@supereva.it> To: <Info-VAX@Mvb.Saic.Com> ' Sent: Wednesday, March 05, 2003 8:44 AM  Subject: cdrom    D > how to mount for read cdrom iso or joljet on vax 4000vls on vms5.5
 > thank you .   ! decus has a program called cdrom:   H CD-ROM access routines for VAX/VMS                      29 November 1989H ------------------------------------------------------------------------    1 The programs in this directory were developed by:   $      United States Geological Survey      Branch of Astro Geology      2255 N. Gemini Drive       Flagstaff, AZ  86001   F for the purpose of reading the CD-ROMs produced in the ISO 9660 formatA standard (or in the "High Sierra" standard) on a VAX/VMS computer @ system.  Permission has been obtained from the U.S.G.S. to placeC these programs on the VAX SIG tape.  The programs are in the public ? domain, and are provided for use with the following disclaimer:   =         "Although these programs have been used by the United 9         States Geological Survey no warranty expressed or :         implied is made by the USGS as to the accuracy and7         functioning of the programs and related program :         materials, and no responsibility is assumed by the&         USGS in connection therewith."  K The programs were produced to read the Voyager CD-ROMs to be read on VAXes.     K * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *  *   H The programs allow a file on an ISO 9660 format CD-ROM to be opened, andG 512 byte blocks of data read from the file relative to the beginning of 	 the file.   J The original files from the U.S.G.S., as downloaded from the SPAN network,( are   CDVMS2.TXT;2   and   CDVMS3.TXT;1.  H All other files were extracted from these .TXT files, with the exceptionG of the B2B.MAR program which was not in the original files and was sent C in hardcopy by U.S.G.S. and rekeyed.  All programs work with either F ISO 9660 or "High Sierra" format CD-ROMs.  Main programs provided are:  1      CDDIR     Do a directory listing of a CD-ROM :      CDCOPY    Copy a file from a CD-ROM to another device6      CDTYPE    Type the contents of a file on a CD-ROM    F The original version of CDDIR (CDDIR.ORIGINAL) was modified at ICDD toE default to a 3 character file extension since the LENOSP routine used G in CDDIR was missing.  The original version of CDOPEN (CDOPEN.ORIGINAL) E was modified at ICDD to correctly find and report the contents of the I extended attribute record (XAR).  The following subroutines are provided:   C      CDOPEN    Find and open a CD-ROM file, returning the offset of $                the start of the fileD      CDREAD    Read 512 byte block records relative to the beginning                of the file"      CDCLOSE   Close a CD-ROM fileC      CDSTAND   Determine if the CD-ROM is ISO 9660 or "High Sierra" B      CDATT     Obtain extended attribute record (XAR); CDATT is an(                ENTRY point within CDOPEN      B2B       Move bytes     7 All executables were produced on a VAX/VMS V5.2 system.     K * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *  *   ? These routines have been made available to the VAX SIG tape by:         Mark A. Holomany 4      JCPDS-International Centre for Diffraction Data      1601 Park Lane "      Swarthmore, PA  19081  U.S.A.      (215) 328-9403   I The disclaimer that appears at the beginning of this file applies also to H Mark A. Holomany and the JCPDS-International Centre for Diffraction Data (also known as ICDD).   L These routines were described in the session "CDROM Use on a VAX/VMS System"D (VA076) given at the Fall 1989 DECUS Symposium in Anaheim by Mark A.	 Holomany.  The session was taped.  K Feel free to contact me at the above number should you have any comments on C these programs, or if you have any questions on CD-ROMs in general.    [End of AAAREADME.TXT]   eberhard   ------------------------------  $ Date: Wed, 5 Mar 2003 05:27:42 -0500  From: John Santos <JOHN@egh.com>% Subject: Re: CLI question for HP reps / Message-ID: <1030305050721.26948C@Ives.egh.com>   , On Tue, 4 Mar 2003, Stanley F. Quayle wrote:  * > On 4 Mar 2003 at 21:57, Dirk Munk wrote:S > > So my question is why can't we get a 'real' VMS commandline interface to these  
 > > products.  > @ > The idea is to easily port these Unix tools to VMS.  To add a G > different interface would require some programming.  And it wouldn't  E > be useful to the Unix people, anyway (might even get in their way).  > E > This might be a great open-source project for someone to pick up...     > As in design a standardized syntax to map DCL-style parameters" and qualifiers into Unix switches.  8 write a program that uses mapping tables written in that@ syntax to map a set of DCL-style parameters and qualifiers into 8 Uniz-style switches and generate appropriate CDL for it.  = For each new unix tool (e.g, GPG or wgrep or whatever), write = the mapping table (a text file) describing the mapping in the  standardized syntax.  ; Run the program on that file, producing a .CLD file for the : DCL-style syntax.  The .CLD points its image either at the= program itself, or at a companion program that is part of the A same package (trade-offs being the advantage of having everything A in one program, or having a smaller and simpler executable at run ; time, with the risk of getting it out of synch with the 1st ; program.)  (It might be easier to have the program create a ? custom converter program for each mapping table, if the parsing ? and mapping rules get too complicated to execute efficiently at 
 run time.)  ? Compile the .CLD and add the new command(S) to system DCLTABLES A or the user's current table (user preference) with $ SET COMMAND.   : When the user executes the DCL command for the ported Unix6 tool, it starts up the converter program, which parses; the DCL command, converts it to the equivalent Unix command ; by using the mapping table, and starts up the Unix program, & passing the mapped command line to it.  ? Is this what you had in mind or something completely different?    > --Stan Quayle  > Quayle Consulting Inc. >    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  $ Date: Wed, 5 Mar 2003 14:32:04 -00003 From: "Tom Wade" <t.wade@vms.eurokom.removespam.ie> % Subject: Re: CLI question for HP reps * Message-ID: <b451ab$kar$1@kermit.esat.net>  - "John Santos" <JOHN@egh.com> wrote in message ) news:1030305050721.26948C@Ives.egh.com...   @ > As in design a standardized syntax to map DCL-style parameters$ > and qualifiers into Unix switches.  H I absolutely *HATE* Unix style arguments, and it is a great pity that HP inflict this on us.   : > write a program that uses mapping tables written in thatA > syntax to map a set of DCL-style parameters and qualifiers into : > Uniz-style switches and generate appropriate CDL for it.  I I've been writing precisely such a tool, and it's nearly finished (mainly ) the documentation).  It works as follows:   2 * You provide a .CLD file defining the VMS syntax.@ * You provide a mapping file that maps from VMS syntax into Unix hieroglyphics.J * You compile this mapping file using a generator program which produces C- source, which you run through the C compiler. H * You add a single call to the main program immediately after it starts,J that passes argc and argv to the translator routine.  This routine returnsH the Unix syntax that the rest of the program requires (minimal change to existing code). K * You relink the executable with the extra .OBJ files to produce a new EXE. L * If you don't have the original sources, it can activate the old image withI the Unix equivalent arguments (not as efficient, as it requires two image 
 activations).   D Optionally, the original Unix syntax is also accepted by the parser.  < I hope to have it ready for submission to the next freeware.  D --------------------------------------------------------------------< Tom Wade             | EMail: tee dot wade at eurokom dot ie? EuroKom              | X400:  g=tom;s=wade;o=eurokom;p=eurokom; / Unit A2              |        a=eirmail400;c=ie / Nutgrove Office Park | Tel:   +353 (1) 296-9696 / Rathfarnham          | Fax:   +353 (1) 296-9697 < Dublin 14            | Disclaimer:  This is not a disclaimerC Ireland              | Tip:   "Friends don't let friends do Unix !"    ------------------------------   Date: 5 Mar 2003 09:56:02 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) % Subject: Re: CLI question for HP reps 3 Message-ID: <xsNnUe9MiX2t@eisner.encompasserve.org>   R In article <1030305050721.26948C@Ives.egh.com>, John Santos <JOHN@egh.com> writes: > ? > For each new unix tool (e.g, GPG or wgrep or whatever), write ? > the mapping table (a text file) describing the mapping in the  > standardized syntax.  E    There's the problem.  Figuring out what -r stands for on each tool     is time consuming.    ------------------------------  $ Date: Wed, 5 Mar 2003 11:26:04 +0100# From: "Andrew Clark" <cap@mail.com>  Subject: DCL COPY % Message-ID: <3e65d0bc$1@news.post.ch>    Hi All,   K Is there a programing interface to the DCL COPY command? The system library J has Rename_File and Delete_File but seems to stop there. Having to Spawn aF DCL command seems like a large overhead if I just want to copy a file.   Andrew   ------------------------------  # Date: Wed, 05 Mar 2003 10:54:50 GMT L From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr") Subject: Re: DCL COPY 6 Message-ID: <00A1C64B.0DFA6DA5@SSRL.SLAC.STANFORD.EDU>  K In article <3e65d0bc$1@news.post.ch>, "Andrew Clark" <cap@mail.com> writes:  >Hi All, > L >Is there a programing interface to the DCL COPY command? The system libraryK >has Rename_File and Delete_File but seems to stop there. Having to Spawn a G >DCL command seems like a large overhead if I just want to copy a file.  >   M Nope, there is no such interface as far as I know.  All is not lost, however.   0 If you go to google, select groups, and type in    comp.os.vms lib$copy    O and select the first thread that comes up, you'll be rewarded with a discussion  on this very topic from 1994.   * Arne suggest using callable Convert, with    =      PROGRAM Z. =      CALL CONV$PASS_FILES('LOGIN.COM','Z.Z') =      CALL CONV$PASS_OPTIONS  =      CALL CONV$CONVERT
 =      END =    as an example.  M Carl Lydick contributed NOTLIB_COPY, which is basically the routine you want.   L Chris Chiesa put in a nice MACRO-32 program to link to the same routine thatJ the COPY command links to and do image activation.  This might make a nice8 project to port; I wouldn't guarantee it works on Alpha.  " Maybe this should go in the FAQ.     Oh wait, it's already there.  D           __________________________________________________________8           10.11  How do I do a file copy from a program?  H                    There are several options available for copying filesG                    from within a program. Obvious choices include using G                    lib$spawn(),  system(), sys$sndjbc() or sys$creprc() J                    to invoke a DCL COPY command. Other common alternativesF                    include using the callable convert routines and theE                    BACKUP application programming interface (V7.1 and                     later).   -- Alan   O =============================================================================== 0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056 M  Paper mail to: SSRL -- SLAC BIN 99, 2575 Sand Hill Rd, Menlo Park CA   94025 O ===============================================================================    ------------------------------   Date: 5 Mar 2003 13:01:31 +0100 ' From: huber@mppmu.mpg.de (Joseph Huber)  Subject: Re: DCL COPY + Message-ID: <J8ldwWhPW3Os@vms.mppmu.mpg.de>   K In article <3e65d0bc$1@news.post.ch>, "Andrew Clark" <cap@mail.com> writes: 	 > Hi All,  > M > Is there a programing interface to the DCL COPY command? The system library L > has Rename_File and Delete_File but seems to stop there. Having to Spawn aH > DCL command seems like a large overhead if I just want to copy a file. >   ( Look into the help for CONV$_Routines  .  4 The simpliest sequence for a 1-to-one copy would be:  :       CALL CONV$PASS_FILES(inputfilename , outputfilename)       CALL CONV$PASS_OPTIONS       CALL CONV$CONVERT     --  N Joseph "Sepp" Huber   mailto:joseph.huber@web.de   http://www.huber-joseph.de/   ------------------------------   Date: 5 Mar 2003 09:29:24 -0700 1 From: nothome@spammers.are.scum (Malcolm Dunnett)  Subject: Re: DCL COPY - Message-ID: <Lw8$zVPorNSm@malvm7.mala.bc.ca.>   , In article <J8ldwWhPW3Os@vms.mppmu.mpg.de>, .      huber@mppmu.mpg.de (Joseph Huber) writes: > * > Look into the help for CONV$_Routines  . > 6 > The simpliest sequence for a 1-to-one copy would be: > < >       CALL CONV$PASS_FILES(inputfilename , outputfilename) >       CALL CONV$PASS_OPTIONS >       CALL CONV$CONVERT  >     :    I believe that convert always does a conversion on the ? file, as opposed to a straight copy. For simple file structures 8 the two are pretty much the same - though convert may be9 slower as it is doing a record-by-record copy rather than 9 a block copy. For indexed files ( particularly those with 7 lots of keys ) convert can be a whole lot slower than a  block mode copy would be.   8   I'd recommend looking into using callable backup to do3 the copy, or write a routine to copy the file using 4 RMS block IO ( sample below, not guaranteed to meet  your needs ):    .TITLE	COPY_FILE .IDENT  /V2.0/ ; $ ;	COPY AN RMS FILE - USING BLOCK I/O ;  ;	MODULE		:	COPY_FILE  ;	AUTHOR		:	M.DUNNETT  ;	DATE		:	FEB 12/82  ;    $FABDEF  $RABDEF   5 	.PSECT	$LOCAL PIC,USR,CON,REL,LCL,NOSHR,NOEXE,RD,WRT    	; 	;	DEFINE FABS 	;   INPFAB:		$FAB	FAC = <BIO,GET>     OUTFAB:		$FAB	FAC = <BIO,PUT> ,- 			FOP = <SUP,CBT>   	; 	;	DEFINE RABS 	;	    INPRAB:		$RAB	FAB = INPFAB,-
 			BKT = 0 ,-  			RAC = SEQ,- 			UBF = BUFFER_SPACE ,- 			USZ = 16384   OUTRAB:		$RAB	FAB = OUTFAB ,-  			RAC = SEQ,-
 			BKT = 0 ,-  			RBF = BUFFER_SPACE    	; 	;	DEFINE BUFFER AREA  	;   BUFFER_SPACE:	.BLKB	16384   . 	.PSECT $CODE,PIC,CON,REL,LCL,SHR,EXE,RD,NOWRT   ; % ;	STAT = COPY_FILE(INP.FLE$,OUT.FLE$)  ;    	.ENTRY COPY_FILE,^M<R2,R3,R4>   	MOVAL	INPFAB,R2 	MOVAL	OUTFAB,R3  % 	MOVL	4(AP),R4		; GET INPUT FILE NAME  	MOVB	(R4),FAB$B_FNS(R2) 	MOVL	4(R4),FAB$L_FNA(R2)   & 	MOVL	8(AP),R4		; GET OUTPUT FILE NAME 	MOVB	(R4),FAB$B_FNS(R3) 	MOVL	4(R4),FAB$L_FNA(R3)   ! 	$OPEN	FAB=(R2)	; OPEN INPUT FILE   $ 	BLBS	R0,1$		; GIVE UP IF OPEN FAILS 	RET  6 1$:	$CONNECT RAB=INPRAB	; CONNECT RECORD ACCESS STREAM   	BLBS	R0,2$  	RET   2$:	; : 	;	COPY ALL RELEVANT FAB FIELDS INTO OUTFAB IN PREPARATION 	;	FOR THE $CREATE 	;  7 	MOVL	FAB$L_ALQ(R2),FAB$L_ALQ(R3)	; ALLOCATION QUANTITY / 	MOVB	FAB$B_BKS(R2),FAB$B_BKS(R3)	; BUCKET SIZE . 	MOVW	FAB$W_BLS(R2),FAB$W_BLS(R3)	; BLOCK SIZE8 	MOVW	FAB$W_DEQ(R2),FAB$W_DEQ(R3)	; DFLT EXTEND QUANTITY3 	MOVB	FAB$B_FSZ(R2),FAB$B_FSZ(R3)	; FIXED AREA SIZE 5 	MOVL	FAB$L_MRN(R2),FAB$L_MRN(R3)	; MAX RECORD NUMBER 3 	MOVW	FAB$W_MRS(R2),FAB$W_MRS(R3)	; MAX RECORD SIZE 0 	MOVB	FAB$B_ORG(R2),FAB$B_ORG(R3)	; ORGANIZATION5 	MOVB	FAB$B_RAT(R2),FAB$B_RAT(R3)	; RECORD ATTRIBUTES 1 	MOVB	FAB$B_RFM(R2),FAB$B_RFM(R3)	; RECORD FORMAT     & 	$CREATE	FAB=(R3)	; CREATE OUTPUT FILE   	BLBS	R0,3$  	RET  0 3$:	$CONNECT  RAB=OUTRAB	; CONNECT OUTPUT STREAM   	BLBS	R0,4$  	RET   4$:	MOVAL	INPRAB,R2  	MOVAL	OUTRAB,R3   5$:	;  	;	COPY THE FILE NOW 	;  ' 	$READ	RAB=(R2)	; READ A BUNCH OF STUFF  	BLBC	R0,6$   5 	MOVW	RAB$W_RSZ(R2),RAB$W_RSZ(R3)  ; COPY RECORD SIZE   * 	$WRITE	RAB=(R3)	; WRITE IT TO OUTPUT FILE   	BLBS	R0,5$  	RET   	  6$:	$CLOSE	FAB=INPFAB  	$CLOSE	FAB=OUTFAB   	RET   	.END    ------------------------------  $ Date: Wed, 5 Mar 2003 10:03:42 -0000! From: "Craig Cooke" <cs@dabs.com> ) Subject: Re: disk queue length in monitor 5 Message-ID: <vQgoA6v4CHA.4716@juno.intranet.dabs.com>   0 "dooley" <dooleys@snowy.net.au> wrote in message7 news:1ca82fc6.0303042135.5f8e4fc1@posting.google.com... 9 > when running monitor disk /item=queue on OpenVMS V7.2-1 4 > one of my DSA disks is showing rather large values1 > the other disks are OK - values between 0 and 4 D > on dsa201 the current queue length seems to like a figure of 43880 > (but moves around a little) ; > Since there was little activity on this disk at the time, D > I suspect this figure is spurious - or should I be more concerned? > Phil   Hi Phil   I The article below describes some of what you are experiencing - I seem to L recall it happened on one of the systems I worked in the past - although not on Shadowed disks but FC disks.   F http://h18000.www1.hp.com/support/askkcs/hpcg/1_0_1063190_2794503.html   Regards    Craig Cooke    ------------------------------  $ Date: Wed, 5 Mar 2003 11:13:32 +01000 From: "labadie" <en_trajectant_a_mort@127.0.0.1>) Subject: Re: disk queue length in monitor , Message-ID: <3e65cf55$1@usenet01.boi.hp.com>  0 "dooley" <dooleys@snowy.net.au> wrote in message7 news:1ca82fc6.0303042135.5f8e4fc1@posting.google.com... 9 > when running monitor disk /item=queue on OpenVMS V7.2-1 4 > one of my DSA disks is showing rather large values1 > the other disks are OK - values between 0 and 4 D > on dsa201 the current queue length seems to like a figure of 43880 > (but moves around a little) ; > Since there was little activity on this disk at the time, D > I suspect this figure is spurious - or should I be more concerned? > Phil   Hello   " In the text of VMS721_UPDATE-V0300 I read  L o  As a result, I/O monitoring tools, such as MONITOR_DISK/ITEM=QUEUE_LENGT,D      would report erroneous and increasing values for some multipath devices,>      even when there was actually no active I/O on the device.  /      Images Affected:  [SYS$LDR]IO_ROUTINES.EXE 5                        [SYS$LDR]IO_ROUTINES[_MON].EXE -                        [SYS$LDR]MULTIPATH.EXE 1                        [SYS$LDR]MULTIPATH_MON.EXE /                        [SYS$LDR]IO_ROUTINES.STB 5                        [SYS$LDR]IO_ROUTINES[_MON].STB -                        [SYS$LDR]MULTIPATH.STB 1                        [SYS$LDR]MULTIPATH_MON.STB    Put this patch, it will fix it.    Regards    Grard   ------------------------------  % Date: Wed, 05 Mar 2003 09:28:20 -0500 ! From: Jim Agnew <jpagnew@vcu.edu> ? Subject: Re: Editor wars - Was: Re: Upcoming VMS improvements ? ' Message-ID: <3E660984.B177F3D9@vcu.edu>   G yup... it helps to be able to modify your playthings at work...  I have G a combo batch job and xv setup that grabs the latest sat pixes and make E them a background on a vaxstation..  least it gives the illusion of a  window in that room... ;-)   jim    "David J. Dachtera" wrote: >  > Jim Agnew wrote: > > L > > i almost went there, practiced vi on a dos laptop while reading o'reilly1 > > in a playhouse type place while babysitting..  > >  > > talk about self-torture... > >  > > and rotten coffee..  > ) > Hey - when ya got nuttin' else, y'know?  > J > Even had to deal with a green-screen Hazeltine-emulating tube. Not a badG > keyboard (just not good!). I missed my VT220, and the VT102 and VT100 1 > tubes we'd had a the site before the UN*X shop.  > H > If anyone knows where "Montagar" comes from, another character in that> > same movie says, "Sometimes, I get tired of bein' so old"... >  > -- > David J. Dachtera  > dba DJE Systems  > http://www.djesys.com/ > * > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/    --  F "4,000 years ago I made a mistake."  Elrond Half-Elven, in "Fellowship of the Ring"   ------------------------------  % Date: Wed, 05 Mar 2003 09:28:59 -0500 ! From: Jim Agnew <jpagnew@vcu.edu> ? Subject: Re: Editor wars - Was: Re: Upcoming VMS improvements ? ' Message-ID: <3E6609AB.1370A3A6@vcu.edu>   ; Son Of Stopgap!!!! EDI... man, that brings back memories.      I got one, remember KED????    jim    Paul Sture wrote:  > Y > In article <01C2E193.40E2C240@sulfer.icius.com>, Shane Smith <ssmith@icius.com> writes: I > > VI was still a mess last time I was forced to use it (and I /do/ mean  > > forced). > > J > > I am a TPU user who respects EDT, and even uses it occasionally. But IJ > > still have a soft spot for something called TED on a PDP-11 in the midI > > 80s. It was a pure line editor, and could only go one way through the I > > file, down. If you went past the line you wanted, you had to exit and I > > start again. You /couldn't/ forget to save occasionally. :-) See? You * > > can find something good in any editor. > >  > B > Which reminds me of EDI, which came to VMS from RSX (AFAIK) as a > compatibility mode program.  > E > That was also a pure line editor, which (IIRC) read in a multi-line E > buffer at a time. You could go back within the buffer, but once you H > have moved to the next buffer, couldn't go back to the previous buffer  > without saving the whole file. > 5 > Yes, my very first editor war was with EDI fans :-)  > + > Anyone remember SOS, whilst we are at it?  > -- > Paul Sture   --  F "4,000 years ago I made a mistake."  Elrond Half-Elven, in "Fellowship of the Ring"   ------------------------------   Date: 5 Mar 2003 09:43:54 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ? Subject: RE: Editor wars - Was: Re: Upcoming VMS improvements ? 3 Message-ID: <RZ$G$IWko6vT@eisner.encompasserve.org>   U In article <0KF256bMNdli@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes:  > + > Anyone remember SOS, whilst we are at it?   @    Yep.  Used it a lot on a DECSYSTEM 10 when I couldn't find my@    docs on TECO.  Used it for a while on my first 11/780 when weA    discovered that EDT knew an ADM-3 wasn't a real terminal.  Had C    a hard time weaning people of it when we went to VMS 3.0 and EDT G    became the default editor.  I could tell when peple were editting in H    SOS from the increased CPU load, all of which was compatability mode.   ------------------------------   Date: 5 Mar 2003 09:45:18 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ? Subject: Re: Editor wars - Was: Re: Upcoming VMS improvements ? 3 Message-ID: <q0ESd6i9tr2Z@eisner.encompasserve.org>   K In article <3E6609AB.1370A3A6@vcu.edu>, Jim Agnew <jpagnew@vcu.edu> writes: = > Son Of Stopgap!!!! EDI... man, that brings back memories.    >  > I got one, remember KED????  > E    On RSX-11M before we got EDT.  The heritage was obvious, EDT was a     welcome improvement.   '    But did you ever use Columbia's SED?    ------------------------------  % Date: Wed, 05 Mar 2003 11:31:06 -0500 ! From: Jim Agnew <jpagnew@vcu.edu> ? Subject: Re: Editor wars - Was: Re: Upcoming VMS improvements ? ' Message-ID: <3E66264A.4C58504B@vcu.edu>   F nooo.. not really. after KED, linemode was history... However, when weH started using visual editors, I started losing the ability to hold 1,000E line FORTRAN programs in memory to edit 'em..  not that that was bad,  but interesting...   jim    Bob Koehler wrote: > M > In article <3E6609AB.1370A3A6@vcu.edu>, Jim Agnew <jpagnew@vcu.edu> writes: = > > Son Of Stopgap!!!! EDI... man, that brings back memories.  > >  > > I got one, remember KED????  > > G >    On RSX-11M before we got EDT.  The heritage was obvious, EDT was a  >    welcome improvement.  > ) >    But did you ever use Columbia's SED?    --  F "4,000 years ago I made a mistake."  Elrond Half-Elven, in "Fellowship of the Ring"   ------------------------------  # Date: Wed, 05 Mar 2003 13:01:17 GMT # From: "John N." <JNixon@cfl.rr.com>  Subject: Fibre Channel and SCS: Message-ID: <xym9a.1204$ge3.12494@twister.tampabay.rr.com>  4 Does FibreChannel support SCS?  Are there any plans?E If my Alpha systems are connected by a single LAN interconnect, and a J FibreChannel interconnect, are the nodes subject to CLUEXIT if the network1 segment becomes unavailable for a period of time?    ------------------------------  # Date: Wed, 05 Mar 2003 13:27:54 GMT # From: "John N." <JNixon@cfl.rr.com> " Subject: Re: Fibre Channel and SCS: Message-ID: <uXm9a.1207$ge3.13680@twister.tampabay.rr.com>  J As a followup to my own question, I found an article that said SCS supportF was a "future" for gigabit ethernet.  I assume this was an old articleI because we are using that feature now.  There was no date on the article. / Can anyone tell  me when this became supported?   . "John N." <JNixon@cfl.rr.com> wrote in message4 news:xym9a.1204$ge3.12494@twister.tampabay.rr.com...6 > Does FibreChannel support SCS?  Are there any plans?G > If my Alpha systems are connected by a single LAN interconnect, and a L > FibreChannel interconnect, are the nodes subject to CLUEXIT if the network3 > segment becomes unavailable for a period of time?  >  >  >    ------------------------------  $ Date: Wed, 5 Mar 2003 09:48:31 -05000 From: "Alan Boyles" <alan.boyles@mindspring.com>" Subject: Re: Fibre Channel and SCS/ Message-ID: <v6c3t366g2b083@corp.supernews.com>   L Currently Fiber Channel cannot be used as a cluster interconnect, but I hearK that's changing.  Gigabit E/net is now a supported Cluster Interconnect and  has been for a while.    Alan. "John N." <JNixon@cfl.rr.com> wrote in message4 news:xym9a.1204$ge3.12494@twister.tampabay.rr.com...6 > Does FibreChannel support SCS?  Are there any plans?G > If my Alpha systems are connected by a single LAN interconnect, and a L > FibreChannel interconnect, are the nodes subject to CLUEXIT if the network3 > segment becomes unavailable for a period of time?  >  >    ------------------------------  % Date: Wed, 05 Mar 2003 10:35:12 -0500 % From: "Mark D. Jilson" <jilly@hp.com> " Subject: Re: Fibre Channel and SCS& Message-ID: <3E661930.598E15BD@hp.com>  C Fibre Channel does not have SCS at this time.  The biggest sticking F point has been a lack of agreement between vendors/standards as to howB host to host communication would work.  Last I knew there were 2-3@ different proposals being pushed.  Once this gets sorted out andE controllers get delivered that support it then SCS may be made to use D FC.  As for gigabit, yes this has SCS support.  Also you are correctH that in your current setup you will have a CLUEXIT bugcheck if the nodes; lose network connectivity for more than RECNXINTERVAL time.    "John N." wrote: > 6 > Does FibreChannel support SCS?  Are there any plans?G > If my Alpha systems are connected by a single LAN interconnect, and a L > FibreChannel interconnect, are the nodes subject to CLUEXIT if the network3 > segment becomes unavailable for a period of time?    --  C Jilly	- Working from Home in the Chemung River Valley - Waverly, NY 0 	- jilly@clarityconnect.com			- Brett Bodine fan+ 	- Mark.Jilson@hp.com				- since 1975 or so  	- http://www.jilly.baka.com   ------------------------------   Date: 5 Mar 2003 09:38:19 -0600 + From: young_r@encompasserve.org (Rob Young) " Subject: Re: Fibre Channel and SCS3 Message-ID: <7mmaJ$rzGdvg@eisner.encompasserve.org>   b In article <v6c3t366g2b083@corp.supernews.com>, "Alan Boyles" <alan.boyles@mindspring.com> writes:N > Currently Fiber Channel cannot be used as a cluster interconnect, but I hearM > that's changing.  Gigabit E/net is now a supported Cluster Interconnect and  > has been for a while.  >   G 	SCS over fibre was presented to me as a "great thing to have" but the  E 	performance isn't stellar (I guess shouldn't be either as you think  G 	about the protocols involved, how many layers?).  As mentioned before, B 	SCS over fibre for DT "mostly" and that is borne out on slide 11 > 	of the latest roadmap and it is a wide box on that powerpoint 	(form mid-2004 to 2006).    				Rob    ------------------------------  # Date: Wed, 05 Mar 2003 16:26:33 GMT ! From: JBloggs@acme..spamless..com D Subject: Re: hobbiest RDB/Sql*Services security (unrelated question)8 Message-ID: <n88c6vsqbofaiepdghgh9ras6ap5ek26nl@4ax.com>  F On Tue, 4 Mar 2003 21:20:54 +0100, "Hans Vlems" <hvlems@iae.nl> wrote:   > 2 >"Chris G." <ccc_crg@yahoo.com> schreef in bericht8 >news:5437ff2f.0303041045.459e1fea@posting.google.com...E >> I've just completed installing RDB and Sql/Services on my hobbiest H >> Alpha. Sql/Services appears to be running on a couple of IP ports but( >> UCX SHOW SERVICES does not list them. >>D >> I only want SQL/Services to accept connections from the localhost0 >> 127.0.0.1.  IS there an equivalent of UCX SET! >> SERVICE/ACCEPT=HOST=127.0.0.1?  >> >> many thanks! - ChrisG >  >Chris > 5 >where did you get the rdb and sql distribution kits? < >Are the licenses covered by the hobbyist PAK from Montagar? >  >Hans Vlems   > I have downloaded some RDB kits from the OTN site awhile back;F I was only able to see the links to the actual kits,  when using MSIE. (not Mozilla, not Opera, etc.)  < But I was able to get the kits via LYNX, once i figured out 4 the proper URLs.   (you do have to sign up with OTN, for a username and password).    ------------------------------  $ Date: Wed, 5 Mar 2003 18:03:04 +0100$ From: Michael Unger <unger@decus.de>D Subject: Re: hobbiest RDB/Sql*Services security (unrelated question)* Message-ID: <00A1C6C9.EE9887D6.9@decus.de>  $ <JBloggs@acme..spamless..com> wrote:  H > On Tue, 4 Mar 2003 21:20:54 +0100, "Hans Vlems" <hvlems@iae.nl> wrote: >  > > 4 > >"Chris G." <ccc_crg@yahoo.com> schreef in bericht: > >news:5437ff2f.0303041045.459e1fea@posting.google.com...G > >> I've just completed installing RDB and Sql/Services on my hobbiest J > >> Alpha. Sql/Services appears to be running on a couple of IP ports but* > >> UCX SHOW SERVICES does not list them. > >>F > >> I only want SQL/Services to accept connections from the localhost2 > >> 127.0.0.1.  IS there an equivalent of UCX SET# > >> SERVICE/ACCEPT=HOST=127.0.0.1?  > >> > >> many thanks! - ChrisG > >  > >Chris > > 7 > >where did you get the rdb and sql distribution kits? > > >Are the licenses covered by the hobbyist PAK from Montagar? > > 
 > >Hans Vlems  > @ > I have downloaded some RDB kits from the OTN site awhile back;H > I was only able to see the links to the actual kits,  when using MSIE.  > (not Mozilla, not Opera, etc.) > > > But I was able to get the kits via LYNX, once i figured out 6 > the proper URLs.   (you do have to sign up with OTN, > for a username and password).  >   F Do you still have the links to the kits? I visited the OTN site a few F hours ago and found it full of JavaScript, i.e., unusable for secured E browsers. (I was using Mozilla 1.2.1 DE with JavaScript disabled for   obvious reasons.)    Michael    ------------------------------  # Date: Wed, 05 Mar 2003 09:14:13 GMT ) From: P.Lj <plj@NOSPAMbyron.ext.telia.se> K Subject: Re: Ideas for implementing fork() on VMS ?, was: Re: fork() on VMS 8 Message-ID: <1ufb6vgepgj6medfoerei07vkg4rrian2p@4ax.com>   On 4 Mar 2003 06:29:49 -0600, C clubley@remove_me.eisner.decus.org-Earth.UFP (Simon Clubley) wrote:   r >In article <cf15391e.0303031657.20e66c98@posting.google.com>, keithparris_NOSPAM@yahoo.com (Keith Parris) writes: >>  E >> My apologies.  After looking into this more carefully, it appears: G >> (1) The old VMS Integrated POSIX (VIP) product had a fork() function G >> that could fork a (constrained) VMS process (but that product's been " >> unsupported for some years now)D >> (2) The initial DII-COE releases (7.2-6Cn) don't have fork() (see8 >> http://openvms.compaq.com/solutions/government/coe/). >>  " >> (1) was what I was thinking of. >>  5 >> So this fork() is a future feature on the Roadmap.  >  >Thanks for the feedback Keith.  >  >Next question:  > K >Does anybody have _any_ ideas about how to implement fork() on current VMS , >versions, ignoring any performance issues ? >  >Simon.   4 Isn't there a vfork() in Porting-Library (jackets) ?   /P.Lj    ------------------------------  % Date: Wed, 05 Mar 2003 09:31:13 -0500 ! From: Jim Agnew <jpagnew@vcu.edu>  Subject: Re: Internet attacks ' Message-ID: <3E660A31.E41BA306@vcu.edu>   C eewwww...  cruel...  I love it...  I inadvertenly (sp?) got to be a G "honeypot" for spammers for a while, until the load drove my little ole E mv3900 into the ground...  It was a bother, but my slight comfort was = all that mail being wasted...  I deleted them with a smile...    "John E. Malmberg" wrote:  >  > Jim Agnew wrote:I > > thanks... i've disusered for now those accounts of mine that match... # > > luckily, they were low-usage...  > / > You do not have a captive account for "root"?  > & > write "Welcome to LINUX for the Z80" > Loop1:  >    Read password with no echo. >    if timeout then stop/id 0 >    if rnd(mostofthetime) then " >          if retries expired then$ >              logout with mismatch. >           else >               goto loop1	 >    else G >          write "your password has expired, please enter new password:  > LOOP2:; >          read new password with no echo and short timeout $ >          if timeout then stop/id 0 > E >           read confirmation password with no echo and short timeout % >           if timeout then stop/id 0 D >           if rnd(most of the time) or (new_pass <> conf_pass) then1 >                write "Password does not match." / >                if newtries expired goto loop1  >                goto loop2  >           if rnd(x) then9 >                 write "Password has already been used." 0 >                 if newtries expired goto loop1 >                 goto loop2 > / > ! Someone was persitent and made it this far? . > read with short timeout and prompt with "# " > stop/id 0  > F > Unfortunately my home ISP does not allow me to have any server ports > open to the Internet.  >  > -John  > wb8tyw@qsl.network > Personal Opinion Only    --  F "4,000 years ago I made a mistake."  Elrond Half-Elven, in "Fellowship of the Ring"   ------------------------------  $ Date: Wed, 5 Mar 2003 02:15:18 -0500* From: "Bill Todd" <billtodd@metrocast.net>8 Subject: Re: Linus trashes Itanium - he's scared of VMS!2 Message-ID: <w2CdnTuqJZ-WOfijXTWc3A@metrocast.net>  K "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com> 9 wrote in message news:3E64D05C.9010001@nospamn.sun.com...  >  >  > Bill Todd wrote:2 > > "Dan Allen" <dallen@nist.gov> wrote in message8 > > news:JFEPKAPBPMDFDBOIANGDIEPICNAA.dallen@nist.gov... > >  > >> > >>>-----Original Message----- , > >>>From: Andrew Harrison SUNUK Consultancy3 > >>>[mailto:Andrew_No.Harrison_No@nospamn.sun.com] * > >>>Sent: Monday, March 03, 2003 10:34 AM > >>>To: Info-VAX@Mvb.Saic.Com= > >>>Subject: Re: Linus trashes Itanium - he's scared of VMS!  > >> > >  > > ...  > >  > > = > >>>Of course part of the cost differential is because Xeons ; > >>>cost Intel a lot less to build than IPF, with a 25% of $ > >>>the die size in a given process > >> > > L > > Your credibility is thin enough in this forum already without the aid ofL > > such blatantly erroneous statements.  Itanic2's chip area is 421 mm^2 in its L > > 180 nm process, while P4's is 140 mm^2 in 130 nm and with only 512 KB ofK > > cache vs. Itanic2's 3 MB.  That would put P4 over 200 mm^2 in Itanic2's  180 ? > > nm process, and close to 300 mm^2 in that process in a Xeon 
 configuration K > > with 2 MB cache - in other words, between 50% and 70% of Itanic2's area  in > > the same process.  > >  >  > E > Humm according to chipgeek P4 is 131 mm^2 in 130 nanometer not 140.   K I was using one of Paul DeMone's numbers - but close enough in either case.    > 7 > And P4/Xeon have a 512K cache not 2 MB thats Xeon-MP.   F Which is why I noted above that in the same process the range would beJ between about 50% (for the 512 KB Xeon) and 70% (for the 2 MB Xeon) of theJ Itanic2's chip area:  still very different (even just at 50%) from the 25%
 you asserted.   4 > Xeon-MP cache sizes are largely irrelevant because> > the P4/Xeon with 512K cache outperforms Itanium 2 with 3 MB.  K Not in, e.g., TPC-C:  the 512 KB Xeon is significantly behind, and even the F (slower-clocked) Xeon with 2 MB of on-chip cache doesn't *quite* catchK Itanic2 there.  So both Xeon versions are relevant to a general performance  comparison.    > 8 > But you are right Xeon-MP has a 2 MB onchip cache with9 > a total of 108 million transistors, 55 million of which 6 > are the 2 MB cache. With more for the 512K L2 cache. > @ > However I wan't refering to Xeon-MP which is currently clocked* > at 2 GHz not 2.8 or 3 as P4 or Xeon are. > > > And anyway the discussion about relative cache sizes betweenC > IPF and P4/Xeon are pretty irrelevant because they have differing - > bus bandwidth/latencies and code densities.   K Since most customers buy systems rather than CPUs (that's one reason Sun is B still in business), the entire package is pertinent when comparing3 performance (e.g., see the TPC-C discussion above).    > 3 > IPF code density is arround half that of P4/Xeon.   G Given competent I-cache behavior (which my impression is both platforms F have) code density isn't the bugaboo that it was on something like the PDP-11.    > 2 > Itanium 2 has a 6.4 GB/s 128 bit 400 MHz bus but< > it wastes 30% of this bus bandwidth on average for SPECint: > because of discarded prefetches. Interesting because the= > extensive use of feedback directed optimisation for SPECint > > should reduce the wasted prefetches if so then the effective0 > rate for less micro type apps would be higher.  B You won't catch me defending Itanic's efficiency - just correcting: misstatements about it (whether flattering or derogatory).   >  > Xeon is in order   I don't think so.   +  and its 533 Mhz system bus has a bandwidth < > of 4.3 GB/s giving it roughly the same effective bandwidth	 > as IPF.  > < > Xeon also has a lower measured memory latency than IPF, 92= > nanoseconds with the 533 MHz Bus when compared with Itanium < > 2 and the ZX1 chipset which does 110 (both systems running% > the lmbench on a Linux 2.4 kernel).   G 92 ns is excellent, especially considering that Itanic2's *real* memory F latency with the zx1 chipset is not 110 (or 112, which I think was theH original official number) ns but instead over 150 (154?) ns:  turned outH that the original measurement used a stride which the system was able toJ recognize and prefetch (of course, it's also possible that the Xeon number! you saw was similarly incorrect).    > @ > These factors plus the code density issues make relative cache3 > size discusions in largely academic in this case.   I No, they don't:  more cache is important for some activities (TPC-C likes % it, for example), less so for others.    > ? >  From a fabrication standpoint the amount of onchip cache and B > the number of transistors it consumes is less important than the > core and how it is laid out.  L That's only marginally true for McKinley, where the cache takes up about 43%K of the total chip area.  The situation will be reversed with Madison, where E the cache will consume about 57% of the chip area (and even more with B Madison II if it increases the cache by another 50% as scheduled).  !  Cache is denser than the rest of > > the chip as any die photos will illustrate it also has lower > defect rate. > A > Itanium 2's die area is ~65% core and 35% Cache with 75% of the ? > total metal route being core and 25% cache (with 3 MB cache).  > B > Its the fabrication of the core which has the greatest influence( > on the move to a newer faster process. > > > The P4's are built using 12 inch wafers giving you about 470 > dies per wafer.  > B > Itanium 2 is built using 8 inch wafers which should give you ~70< > per wafer or 15% of the P4/s per wafer. On top of that the@ > defect rate for Itanium is likely to be higher than P4 because% > of the relatively larger core area.   K That's all very interesting, but has absolutely no bearing on your original L misstatement (which was that "Xeons cost Intel a lot less to build than IPF,L *with a 25% of the die size in a given process*" [emphasis added in case youC had forgotten:  your original quote is still right up there above].      > D > In the circumstances it isn't difficult to understand why P4/Xeons( > will be rather cheaper than Itanium 2. > . > I hope this corrects any missunderstandings.  H AFAICT it corrected nothing, but did provide some additional interestingE information (assuming that it's more factual than the assertion which  prompted my original response).    - bill   ------------------------------  % Date: Wed, 05 Mar 2003 12:21:03 +0000 ' From: Andrew Harrison SUNUK Consultancy 8 Subject: Re: Linus trashes Itanium - he's scared of VMS!. Message-ID: <3E65EBAF.9030805@nospamn.sun.com>   Bill Todd wrote:M > "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com> ; > wrote in message news:3E64D05C.9010001@nospamn.sun.com...  >  >> >>Bill Todd wrote: >>1 >>>"Dan Allen" <dallen@nist.gov> wrote in message 7 >>>news:JFEPKAPBPMDFDBOIANGDIEPICNAA.dallen@nist.gov...  >>>  >>>  >>>>>-----Original Message----- , >>>>>From: Andrew Harrison SUNUK Consultancy3 >>>>>[mailto:Andrew_No.Harrison_No@nospamn.sun.com] * >>>>>Sent: Monday, March 03, 2003 10:34 AM >>>>>To: Info-VAX@Mvb.Saic.Com= >>>>>Subject: Re: Linus trashes Itanium - he's scared of VMS!  >>>> >>>... >>>  >>>  >>> = >>>>>Of course part of the cost differential is because Xeons ; >>>>>cost Intel a lot less to build than IPF, with a 25% of $ >>>>>the die size in a given process >>>>K >>>Your credibility is thin enough in this forum already without the aid of K >>>such blatantly erroneous statements.  Itanic2's chip area is 421 mm^2 in  >> > its  > K >>>180 nm process, while P4's is 140 mm^2 in 130 nm and with only 512 KB of J >>>cache vs. Itanic2's 3 MB.  That would put P4 over 200 mm^2 in Itanic2's >> > 180  > > >>>nm process, and close to 300 mm^2 in that process in a Xeon >> > configuration  > J >>>with 2 MB cache - in other words, between 50% and 70% of Itanic2's area >> > in >  >>>the same process. >>>  >> >>E >>Humm according to chipgeek P4 is 131 mm^2 in 130 nanometer not 140.  >  > M > I was using one of Paul DeMone's numbers - but close enough in either case.  >  > 7 >>And P4/Xeon have a 512K cache not 2 MB thats Xeon-MP.  >  > H > Which is why I noted above that in the same process the range would beL > between about 50% (for the 512 KB Xeon) and 70% (for the 2 MB Xeon) of theL > Itanic2's chip area:  still very different (even just at 50%) from the 25% > you asserted.  >  > 4 >>Xeon-MP cache sizes are largely irrelevant because> >>the P4/Xeon with 512K cache outperforms Itanium 2 with 3 MB. >  > M > Not in, e.g., TPC-C:  the 512 KB Xeon is significantly behind, and even the H > (slower-clocked) Xeon with 2 MB of on-chip cache doesn't *quite* catchM > Itanic2 there.  So both Xeon versions are relevant to a general performance 
 > comparison.  >   ) I wasn't refering to TPC-C but to SPECint  but you are right.   77K TPM for a 4 CPU Xeon MP  87K TPM for a 4 way Itanium 2    Both with MS-SQLserver  . The most up to date Xeon number is only with a, 1 P 2.4 GHz box and it turns in 18K TPM. One( would expect the 2.8/3 GHz systems to be faster.   1 But then who uses TPC-C as a performance measure.   1 Point is that on Integer micro benchmarks Xeon is 5 competitive against a similarly configured Itanium 2.    > 8 >>But you are right Xeon-MP has a 2 MB onchip cache with9 >>a total of 108 million transistors, 55 million of which 6 >>are the 2 MB cache. With more for the 512K L2 cache. >>@ >>However I wan't refering to Xeon-MP which is currently clocked* >>at 2 GHz not 2.8 or 3 as P4 or Xeon are. >>> >>And anyway the discussion about relative cache sizes betweenC >>IPF and P4/Xeon are pretty irrelevant because they have differing - >>bus bandwidth/latencies and code densities.  >  > M > Since most customers buy systems rather than CPUs (that's one reason Sun is D > still in business), the entire package is pertinent when comparing5 > performance (e.g., see the TPC-C discussion above).  >  > 3 >>IPF code density is arround half that of P4/Xeon.  >  > I > Given competent I-cache behavior (which my impression is both platforms H > have) code density isn't the bugaboo that it was on something like the	 > PDP-11.  >  > 2 >>Itanium 2 has a 6.4 GB/s 128 bit 400 MHz bus but< >>it wastes 30% of this bus bandwidth on average for SPECint: >>because of discarded prefetches. Interesting because the= >>extensive use of feedback directed optimisation for SPECint > >>should reduce the wasted prefetches if so then the effective0 >>rate for less micro type apps would be higher. >  > D > You won't catch me defending Itanic's efficiency - just correcting< > misstatements about it (whether flattering or derogatory). >  >  >>Xeon is in order >  >  > I don't think so.  >   4 It doesn't support out of order execution does it ??   > I > 92 ns is excellent, especially considering that Itanic2's *real* memory H > latency with the zx1 chipset is not 110 (or 112, which I think was theJ > original official number) ns but instead over 150 (154?) ns:  turned outJ > that the original measurement used a stride which the system was able toL > recognize and prefetch (of course, it's also possible that the Xeon number# > you saw was similarly incorrect).  >   E 110.6 is the number HP still have up on their web site and I have not G found any references that suggest that this number is incorrect, though  if it is then HP would  I http://www.hp.com/products1/itanium/performance/architecture/lmbench.html   9 I cannot comment on the Xeon number which is available at   ; http://www.linuxhardware.org/Features/02/09/26/172240.shtml    > @ >>These factors plus the code density issues make relative cache3 >>size discusions in largely academic in this case.  >  > K > No, they don't:  more cache is important for some activities (TPC-C likes ' > it, for example), less so for others.  >   : Of course it is but the comparison is only really relevant< if you are comparing CPU X with Y MB of cache and CPU X with8 Z MB of cache and assuming that both CPU's have the same memory subsystem.   < Comparing Xeon and Itanium cache sizes is irrelevant because: code densities are different and they have different buses' with different latancy characteristics.    >  > M > That's all very interesting, but has absolutely no bearing on your original N > misstatement (which was that "Xeons cost Intel a lot less to build than IPF,N > *with a 25% of the die size in a given process*" [emphasis added in case youE > had forgotten:  your original quote is still right up there above].  >     D Well no not quite, implimenting Itanium 2 in a 130 nanometer processD should decrease the die size to ~314 mm^2. This in the 12 inch waferA process currently used for P4/Xeon would result in ~195 Itanium 2 ? dies per wafer. Or ~40% of the yield given the same defect rate  per die as Xeon.  - BTW gate length is also a factor in die size.   @ So I guess we were both wrong with your 50-70% estimate being atB the lower end more accurate than my 25% estimate and at the higher end much less accurate.   F Of course one you add in the higher expected defect rate for Itanium 2A you easily get to a 25% Itanium 2 yield at the same process size.   = I don't know where this puts your an my relative credibility.    ------------------------------   Date: 5 Mar 2003 09:03:43 -0600 + From: young_r@encompasserve.org (Rob Young) 8 Subject: Re: Linus trashes Itanium - he's scared of VMS!3 Message-ID: <bKV8otYUV+hH@eisner.encompasserve.org>   X In article <3E65EBAF.9030805@nospamn.sun.com>, Andrew Harrison SUNUK Consultancy writes: >    > : > [Xeon] doesn't support out of order execution does it ?? >   U http://www.intel.com/technology/itj/2002/volume06issue01/art01_hyper/p01_abstract.htm   P "This paper describes the Hyper-Threading Technology architecture, and discussesJ the microarchitecture details of Intel's first implementation on the IntelM Xeon processor family. Hyper-Threading Technology is an important addition to M Intel's enterprise product line and will be integrated into a wide variety of  products. "   ] http://www.intel.com/technology/itj/2002/volume06issue01/art01_hyper/p06_execution_engine.htm    OUT-OF-ORDER EXECUTION ENGINE F The out-of-order execution engine consists of the allocation, registerN renaming, scheduling, and execution functions, as shown in Figure 6. This partK of the machine re-orders instructions and executes them as quickly as their ? inputs are ready, without regard to the original program order.    				Rob    ------------------------------  % Date: Wed, 05 Mar 2003 16:00:21 +0000 ' From: Andrew Harrison SUNUK Consultancy 8 Subject: Re: Linus trashes Itanium - he's scared of VMS!. Message-ID: <3E661F15.6080108@nospamn.sun.com>   Rob Young wrote:Z > In article <3E65EBAF.9030805@nospamn.sun.com>, Andrew Harrison SUNUK Consultancy writes: >  > : >>[Xeon] doesn't support out of order execution does it ?? >> >  > W > http://www.intel.com/technology/itj/2002/volume06issue01/art01_hyper/p01_abstract.htm  > R > "This paper describes the Hyper-Threading Technology architecture, and discussesL > the microarchitecture details of Intel's first implementation on the IntelO > Xeon processor family. Hyper-Threading Technology is an important addition to O > Intel's enterprise product line and will be integrated into a wide variety of 
 > products. "  > _ > http://www.intel.com/technology/itj/2002/volume06issue01/art01_hyper/p06_execution_engine.htm  >   > OUT-OF-ORDER EXECUTION ENGINE H > The out-of-order execution engine consists of the allocation, registerP > renaming, scheduling, and execution functions, as shown in Figure 6. This partM > of the machine re-orders instructions and executes them as quickly as their A > inputs are ready, without regard to the original program order.  >   > Slightly different. Hyperthreading works at an OS thread level? so for things like SPECint and SPECfp which are single threaded > and single task it won't make any difference. In fact there is= no indication that it has been enabled and it is optional you  can turn it on or off.   IPF is always out of order.    Regards  Andrew Harrison    ------------------------------   Date: 5 Mar 2003 12:29:53 -0600 + From: young_r@encompasserve.org (Rob Young) 8 Subject: Re: Linus trashes Itanium - he's scared of VMS!3 Message-ID: <5gubJtp+FcY8@eisner.encompasserve.org>   X In article <3E661F15.6080108@nospamn.sun.com>, Andrew Harrison SUNUK Consultancy writes: >  >  > Rob Young wrote:[ >> In article <3E65EBAF.9030805@nospamn.sun.com>, Andrew Harrison SUNUK Consultancy writes:  >>   >>  ; >>>[Xeon] doesn't support out of order execution does it ??  >>>  >>   >>  X >> http://www.intel.com/technology/itj/2002/volume06issue01/art01_hyper/p01_abstract.htm >>  S >> "This paper describes the Hyper-Threading Technology architecture, and discusses M >> the microarchitecture details of Intel's first implementation on the Intel P >> Xeon processor family. Hyper-Threading Technology is an important addition toP >> Intel's enterprise product line and will be integrated into a wide variety of >> products. " >>  ` >> http://www.intel.com/technology/itj/2002/volume06issue01/art01_hyper/p06_execution_engine.htm >>  ! >> OUT-OF-ORDER EXECUTION ENGINE  I >> The out-of-order execution engine consists of the allocation, register Q >> renaming, scheduling, and execution functions, as shown in Figure 6. This part N >> of the machine re-orders instructions and executes them as quickly as theirB >> inputs are ready, without regard to the original program order. >>   > @ > Slightly different. Hyperthreading works at an OS thread levelA > so for things like SPECint and SPECfp which are single threaded @ > and single task it won't make any difference. In fact there is? > no indication that it has been enabled and it is optional you  > can turn it on or off. >   9 	It may be different in a marketing sense but it is still > 	executing instructions out of order.  Click on the link, look; 	at the image.  Check out the last stage that reorders the  ; 	instructions on retiring them.  So calling it out-of-order % 	execution is accurate.   OOO is OOO.   ? 	Your question was "Xeon doesn't support out of order execution + 	does it?"  And the answer is, yes it does.    > IPF is always out of order.   = 	IPF is in-order.  Here is a reference to a top hitter Google 6 	that talks about techniques to going beyond in-order:  = http://www.computer.org/proceedings/hpca/1525/15250187abs.htm   B "The performance of in-order execution Itanium(tm) processors ..."   [snip]  G "For a select set of memory-intensive programs, an in-order SMT Itanium O processor using speculative precomputation can achieve per formance improvement N (92%)comparable to that of an out of-order design (87%). Applying both OOO andK SP yields a total performance improvement of 141%over the baseline in order M machine. OOO tends to be effective in prefetching for L1 misses;whereas SP is - primarily good at covering L2 and L3 misses."   ; 	Maybe Emer and crew are aware of this work and are working C 	with these researchers and others for a future IPF processor.  All 1 	speculation, who knows what they are working on.    				Rob   B Men with walkie-talkie                  I'm home again to you babeC Men with flashlights waving             You know it makes me wonder G Up upon the tower                       Sittin' in the quiet slipstream > The clock reads daylight savin'         Rollin' in the thunder  .                                 -- Neil Young    ------------------------------  % Date: Wed, 05 Mar 2003 17:13:26 +0000 + From: John Laird <john@laird-towers.org.uk> ' Subject: Re: Mailbox and their logicals 8 Message-ID: <lrbc6v8bf4nktjtgkomva65j1tlj1pbcoq@4ax.com>  , On Tue, 04 Mar 2003 22:37:57 -0400, JF Mezei( <jfmezei.spamnot@vl.videotron.ca> wrote:  X >Process A creates a mailbox with a group or system logical "CHOCOLATE_MAILBOX" created. > J >Process B assigns a channel to "CHOCOLATE_MAILBOX" and sends it chunks of >delicious data. >  >Process A dies. > O >At this point, I know  that the actual MBAnnn: device continues to exist since & >B still has a channel assigned to it. > I >But will the logical name still exist as well, or does that die when the  >creator of the mailbox dies ? >  > K >In other words, if I restart A after it dies, am I garanteed that  $CREMBX M >will see the existing mailbox and do a $ASSIGN instead of trying to create a  >totally new one ?  K Yes, the logical name is alive as long as the mailbox is.  I would guess it H is associated with the device in the same way that logical names created9 during disk mounts are associated with the volume/device.      	John    ------------------------------   Date: 5 Mar 2003 11:39:25 -0600  From: briggs@encompasserve.org' Subject: Re: Mailbox and their logicals 3 Message-ID: <OYB4E+rwFSvA@eisner.encompasserve.org>   b In article <3E656300.192E5BC6@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:Y > Process A creates a mailbox with a group or system logical "CHOCOLATE_MAILBOX" created.  > K > Process B assigns a channel to "CHOCOLATE_MAILBOX" and sends it chunks of  > delicious data.  >  > Process A dies.  > P > At this point, I know  that the actual MBAnnn: device continues to exist since' > B still has a channel assigned to it.  > J > But will the logical name still exist as well, or does that die when the > creator of the mailbox dies ?   F Looking in the 5.5 Internals and Data Structures manual, page 1078, itF appears that there is some black magic involved.  There is a link fromB the logical name to the mailbox UCB (translation index number -127@ contains this link) and from the mailbox UCB to the logical name (UCB$L_MB_LOGADR).  H It seems clear that if the logical name is deleted ($ DEASSIGN, $DELLNM,C or job logical name table deletion at job exit), translation number D -127 is queried and , if non-null, the pointed-to UCB$L_MB_LOGADR isF cleared.  And at UCB deletion (last channel deassigned to temporary orE permanent, marked for delete mailbox) UCB$L_MB_LOGADR is queried and, A if non-null, the logical pointed-to logical name name is removed.   L > In other words, if I restart A after it dies, am I garanteed that  $CREMBXN > will see the existing mailbox and do a $ASSIGN instead of trying to create a > totally new one ?   G As long as you haven't managed to lose the job logical name table, this  would seem to be a safe bet.   	John Briggs   ------------------------------  $ Date: Wed, 5 Mar 2003 06:39:03 -0800# From: "Tom Linden" <tom@kednos.com>  Subject: MS Virus 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIMENGGLAA.tom@kednos.com>   D There was a post the other day about a virus masquerading as a patchH from Microsoft.  Just got it here.  If I don't recognize mail, I read it< with vms mail, otherwise Outlook.  Here's what it looks like   MAIL> read 205'     #205         5-MAR-2003 05:10:49.38  MAIL+ From:   SMTP%"mzrllgg-fsqlkyfb@licqihq.com" $ To:     "MS Consumer"@hughestele.com CC:  Subj:   Internet Security Pack   Return-Path: amo@vsnl.net 4 Received: from mums1.hughestele.com (202.149.208.91)<          by freja.kednos.com (V5.1-15H, OpenVMS V7.3 Alpha);,         Wed, 5 Mar 2003 05:10:38 -0800 (PST)) Received: from UEbvJBA ([202.149.221.85]) K  by smtp1.hughestele.com (iPlanet Messaging Server 5.2 (built Feb 21 2002)) L  with SMTP id <0HBA001JJ0ELW9@smtp1.hughestele.com> for tom@kednos.com; Wed,!  05 Mar 2003 18:12:39 +0530 (IST) + Date: Wed, 05 Mar 2003 18:07:50 +0530 (IST) > Date-warning: Date header was inserted by smtp1.hughestele.comF From: Microsoft Network Public Services <mzrllgg-fsqlkyfb@LicQiHq.com> Subject: Internet Security Pack   To: "MS Consumer"@hughestele.com1 Message-id: <0HBA001JK0ELW9@smtp1.hughestele.com>  MIME-version: 1.0  Content-type: multipart/mixed;/ boundary="Boundary_(ID_BNEp6dy434hEAiWoElXmRw)"   K I'm sure it has long since been reported, but could someone please post the ! address where forward complaints?   C Also, I am curious how to decipher this,  am I in a mailing list at 
 hughestele called "MS Consumer"  ?  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.449 / Virus Database: 251 - Release Date: 1/27/2003   ------------------------------  # Date: Wed, 05 Mar 2003 17:09:16 GMT " From:   VAXman-  @SendSpamHere.ORG Subject: Re: MS Virus 0 Message-ID: <00A1C698.818A41A5@SendSpamHere.ORG>  _ In article <CIEJLCMNHNNDLLOOGNJIMENGGLAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes: E >There was a post the other day about a virus masquerading as a patch I >from Microsoft.  Just got it here.  If I don't recognize mail, I read it = >with vms mail, otherwise Outlook.  Here's what it looks like   E There's a difference between a Micro$oft patch and a Micro$oft virus?    --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  $ Date: Wed, 5 Mar 2003 18:35:21 +0100$ From: Michael Unger <unger@decus.de> Subject: Re: MS Virus + Message-ID: <00A1C6CE.70F4D1AF.35@decus.de>   # <VAXman-  @SendSpamHere.ORG> wrote:   a > In article <CIEJLCMNHNNDLLOOGNJIMENGGLAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes: G > >There was a post the other day about a virus masquerading as a patch K > >from Microsoft.  Just got it here.  If I don't recognize mail, I read it ? > >with vms mail, otherwise Outlook.  Here's what it looks like  > G > There's a difference between a Micro$oft patch and a Micro$oft virus?   D Indeed. It is the very same as between a M$ feature and a M$ bug ...   Michael    ------------------------------  % Date: Wed, 05 Mar 2003 11:01:08 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com>  Subject: Re: MS Virus ' Message-ID: <3E663B64.3080204@MMaz.com>    VAXman-@SendSpamHere.ORG wrote:   ` >In article <CIEJLCMNHNNDLLOOGNJIMENGGLAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes: >    > F >>There was a post the other day about a virus masquerading as a patch >>     >>J >>from Microsoft.  Just got it here.  If I don't recognize mail, I read it >    > > >>with vms mail, otherwise Outlook.  Here's what it looks like >>     >> > F >There's a difference between a Micro$oft patch and a Micro$oft virus? >    > I Certainly, you should know better.  One you have to pay for the pleasure  7 of, the second is the gift that lives on forever... :-)    Barry    --    @ Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO   A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028    ------------------------------  % Date: Wed, 05 Mar 2003 16:24:11 +0000 ' From: Andrew Harrison SUNUK Consultancy Y Subject: Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha Retai . Message-ID: <3E6624AB.1070602@nospamn.sun.com>   jlsue wrote:G > On Thu, 27 Feb 2003 16:55:28 +0000, Andrew Harrison SUNUK Consultancy 0 > <Andrew_No.Harrison_No@nospamn.sun.com> wrote: >  >  >> >>jlsue wrote: >>H >>>On Wed, 26 Feb 2003 14:01:13 +0000, Andrew Harrison SUNUK Consultancy1 >>><Andrew_No.Harrison_No@nospamn.sun.com> wrote:  >>>  >> > @ >>>What?  Is this claim so extraordinary that it requires proof? >>>  >>: >>Yes because your own engineers say that there are issues8 >>your own benchmark results illustrate these issues andA >>customers on this newsgroup have benchmarked their applications % >>and discovered them for themselves.  >  > D > I see, so now you expect some documentation that I have never seenA > something.  Wow, your argument must be really weak to make such 
 > demands. >   C Hang on are you really that clueless, I have provided documentation , you are the person you has provided nothing.  < A TPC-C full disclosure is documentation, as it you Kingston7 marketing benchmark release as are the postings on this 6 group made by people who found that GS160/320's didn't= go as fast as they were expecting because of the NUMA issues.   6 So what is your next excuse because that was terrible.  F > But you keep overstating the benchmarks and the engineers statementsG > by referencing them out of context.  So far, I've only seen that some G > applications of specific types have been found to be more affected by  > the numa effect than others. >   D Look do you have any idea how something like Oracle is architected ?  : Well let me enlighten you, its a bunch of processes with a0 global shared memory area ever heard of the SGA.  # Let me ask you what does oracle do.    Heck it does transactions.  C Now what remid me how did your customer describe their application.   = Let me paraphrase, it was a bunch of processes using a global * shared memory pool and doing transactions.  A And wat was your enginneers response, ohh that could be a problem  because of the NUMA issue.  # How black and white do you want it.   G Of course thats before you consider that TPC-C and the Kingston Billing @ application both DBMS based also show that GS160/320's do behave< in the way your enginners suggested they did for actual DBMS apps.   G > Do you really think that's news?  Heck, numa isn't the only computing B > architecture that's suffered under *some* applications.  And theC > effect it causes isn't always a binary issue - either it's bad or E > good.  As in other things in life, the effects will vary along some F > sliding scale depending on the applications - from unnoticeable, allH > the way to unusable.  In the end, each business has to decide where on> > the scale it is for them.  That's one of the services that a) > consultant helps businesses understand.  >   9 Look DBMS OLTP applications arn't just some apps they are 7 a fairly major segment of apps that you would expect to  run on a GS system.   4 No one would care if this was a Webserver issue or a6 Java issue because the GS would be a pretty daft place to run this kind of app.  5 But high performance transaction systems was what the 5 GS320 was designed for or at least it was what Compaq  marketed it for.    @ > Again with this weak argument.  Do you really need to grasp soG > desperately at straws?  You can't possibly see that you've overstated F > the issues just a bit and back off?  Are you so completely incapable1 > of seeing the weaknesses in your own arguments?  >   G Sorry but again you are wrong, your marketing lead customers to believe D that not only would the GS160/320 be faster than the competition butC that it would also deliver much better throughput than the previous  generation of systems.  B The first claim was clearly untrue, as you must recollect you haveA never managed to justify the world leading claims for the GS320's ' in anything other than datacenter size.   A The second claim is also severely damaged by examples provided by > your own customers and your own benchmark results which showedB indifferent or 0 performance improvements moving from the older GS! platforms to the newer GS160/320.        > F >>>Your statements are more of a general "it's slower than gs140" as aH >>>way to warn folks away from it.  You state implicitly (and iirc, evenI >>>explicitly at times) that it's  That does qualify, imho, as an example I >>>of an extraordinary claim; therefore I ask for proof that this is true H >>>in a general sense.  And so far you've only proven that it's true forG >>>*some* applications.  If you can't support your claim that it's true > >>>in general, then withdraw it, or at least re-qualify it.    >>>  >>  F You marketing sweepingly stated that the GS160/320 is faster than the & GS140 this isn't true. Get used to it.  ; >>Hang on you are the parson who claims that you have never 9 >>seen an issue. I have simply pointed out that there are : >>customers and benchmark results that show that there are	 >>issues.  >  > > > Now you are telling a big lie.  You have not "simply stated"D > *anything*, but have consistently claimed - hell, even hyped - theH > fact that the numa effect is a huge problem.  You *never* claimed someE > applications and instead implied it is a general-case problem.  You H > have made this implication in many ways, for example just in the sheerF > number of notes where you discuss this problem, and also in the toneF > you use to harp on-and-on about it.  In this very note in fact.  SeeB > your first paragraph quoted at the top of this reply for a prime
 > example. >   > I have provide you with TPC-C, Kingston and your own customers= newsgroup posts which show that the NUMA problem is an issue.   0 What more do you need. To claim that I have lied> is as with most of your posts simply a distortion of the truth= and a clear example of where you need to do a little bit more  self aware.   A At the same time you have failed on numerous occasions to provide 8 examples of the performance tests that you claim in your5 normal hand wavey way to have had run on GS160/320's.   ? Up till now all you have provided is hand wavey hot air without > a since concrete example to support your case. Given you track< record concrete examples are going to be the only thing that is convincing.   Regards  Andrew Harrison    ------------------------------  $ Date: Wed, 5 Mar 2003 02:45:41 -0500* From: "Bill Todd" <billtodd@metrocast.net>1 Subject: OT:  National Moratorium to Stop the War 2 Message-ID: <Q5ednUHaAJy3NvijXTWc3A@metrocast.net>  L After last night's newscasts were full of descriptions of Dubya's increasingF rush to war (perhaps to start next week, according to some pundits), IL decided to take a look around to see what people were doing about it.  TurnsL out that today (March 5th) is the date of a planned 'moratorium' (letters toI legislators, work stoppages, teach-ins, civil disobedience - the gamut of L possible responses), and since the next major planned demonstration (a MarchK 15th march on Washington) may turn out to be after the fact those who would F like to have a chance to register their opposition might want to do soA today.  For further information, see www.internationalanswer.org.    - bill   ------------------------------  $ Date: Wed, 5 Mar 2003 12:27:02 -0500. From: "Leo Demers" <leo_dot_demers@HP_dot_com>% Subject: Re: Questions on SSH for VMS , Message-ID: <3e663568$1@usenet01.boi.hp.com>  D > It may please you to know that the "early adapter kit" for SSH was releasedK > today for HP internal use (and maybe for customers who would like to give  it aI > try, ask your HP account manager if like us you need it). It seems from  the docsK > that you do need TCP/IP V5.3. At the moment I'm reading the docs, just 33  pages. > 8 Actually that notice we a bit premature but it's up now.J We are please to announce the availability of the first hp TCP/IP ServicesD for OpenVMS SSHv2 Early Adopters Kit (EAK). The SSHv2 EAK kit is now& available for download. Please see the7 http://www.openvms.compaq.com/openvms/products/security   J  OpenVMS Security Web page, and follow the SSH link for additional details and download information.   K Notification of future updates to the SSH EAK will also be sent through the  ssh-openvms mailing list.    - Leo  --
 Leo Demers  OpenVMS Security Product Manager Leo_dot_Demers_at_HP_dot_COM   ------------------------------   Date: 5 Mar 2003 07:25:49 -0700 F From: lederman@star.enet.dec.DISABLE-JUNK-EMAIL.com (Bart Z. Lederman)5 Subject: Re: running autogen on alternate system disk . Message-ID: <3e6608ed$1_3@hpb10302.boi.hp.com>  A I sometimes have a system booted up from one disk, and would like > to set up another disk to boot from.  If I understand what the> original poster wanted, the idea would be to "pre-autogen" theB disk you are going to boot from before shutting down the currently running system.   A I can see how this might save a bit of time, because you wouldn't @ have to boot the new disk, wait for it to come up, autogen, then shut down and reboot.   = I don't think there is any supported way to do this.  I would A capture feedback on the running system (which you can do any time = with AUTOGEN), copy the feedback file to the 'new' disk, then = shutdown, reboot and autogen.  It might take a little longer,  but it would be safer.   --  (  B. Z. Lederman   Personal Opinions Only  8  Posting to a News group does NOT give anyone permission8  to send me advertising by E-mail or put me on a mailing  list of any kind.  5  Please remove the "DISABLE-JUNK-EMAIL" if you have a 5  legitimate reason to E-mail a response to this post.    ------------------------------  * Date: Wed, 5 Mar 2003 10:56:05 +0000 (UTC)5 From: "John Wallace" <johnwallace4@yahoo.dotco.dotuk>  Subject: Re: SCS/DECnet over IP / Message-ID: <b44l45$sok$1@helle.btinternet.com>   H "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message- news:KdxoU4hH+rOP@eisner.encompasserve.org... E > In article <3e62e175$0$9537$afc38c87@news.optusnet.com.au>, "Antony < Wardle" <antony.wardle@noooo_spammm_optusnet.com.au> writes: > >  > > SCS/DECnet over IP > >  > >  > > can we do this yet?  > & >    DECnet, yes, have been for years. > = >    SCS, LAT, MOM/MOP, and other non-routable protocols, no.  >   F Look, it's not a problem, just pop down to PC World /Staples and get aK couple of $100 WiFi boxes and some cocoa tin aerials, plug them in and just I bridge these non-routable protocols, then you can keep the network police  happy.  J Well, could it be any worse than attempting it with Translans? It probablyK could, actually, and one day someone will probably unintentionally find out C just how bad it can be... don't want to be there when that happens.    cheers john   ------------------------------   Date: 5 Mar 2003 11:31:50 -0600 @ From: brooks_r@encompasserve.org (Rob Brooks -- VMS Engineering)< Subject: Re: SCSI cluster disk thrashing between connections3 Message-ID: <KNCmmF$csn4p@eisner.encompasserve.org>   . lewis@PROBE.mitre.org (Keith A. Lewis) writes:  I > The problem is the two hosts are fighting over the disk.  Every time it J > gets accessed from the other machine, I get the following OPCOM message: > 7 >     Mount verification has completed for device DSA0:  > 2 > After a few dozen of these, I get something like > M >     8:46:06.74 Multipath access to device $5$DKA100: has been auto switched 4 >     from path PKC0.1 (MAZDA) to path MSCP (LUMINA) > M > And then soon after (maybe the next time the load eases), it switches back.  > L > Is there a way I can tell Mazda (and all other nodes) to go through Lumina; > if possible, using Mazda's connection only as a failover?   H Yes.  For the nodes that have both local and served paths to this deviceC (as MAZDA does), you can manually switch the path to use the served 0 path with $SET DEVICE/PATH=MSCP/SWITCH <devnam>.  K What apparently is happening is that the local paths on MAZDA are disturbed K such that I/O is failing.  Multipath searches all local paths for a working L path.  If none is found, we'll fail over to the remote path.  Once the localH path is found to be working again, we automatically fail back to a localH path, on the presumption that you'll want to use the local path wheneverI possible, since it's much more efficient.  We will not fail back a device 1 that has been manually switched to the MSCP path.    						--Rob  ----J Rob Brooks   VMS Engineering -- I/O Exec Group   brooks!cuebid.zko.dec.com   ------------------------------  * Date: Wed, 5 Mar 2003 10:59:35 +0000 (UTC)5 From: "John Wallace" <johnwallace4@yahoo.dotco.dotuk> = Subject: Re: Sending OPCOM or Security Audit events via email / Message-ID: <b44lam$svg$1@helle.btinternet.com>   8 "Jeff Cameron" <JCam90502@jcameron.com> wrote in message, news:BA89E604.57A0%JCam90502@jcameron.com...D > Has anyone have a simple solution for sending OPCOM messages for a specificI > class, and/or Security Audit events via email? It would be desirable to J > execute a DCL script so that I may filter and condense certain messages. >  > Thank you  > Jeff Cameron >   K I used to have something trivial which I used to email users a message when L they had been observed attempting to access files they shouldn't. SimplisticF rate limiting was built in. Would that be of help? I'd have to find it first, unfortunately...    regards  john   ------------------------------  % Date: Wed, 05 Mar 2003 08:50:53 -0500 ' From: Chris Olive <nospam@raytheon.com> = Subject: Re: Sending OPCOM or Security Audit events via email ; Message-ID: <7in9a.83$35.599@dfw-service2.ext.raytheon.com>    Jeff Cameron wrote: N > On 3/4/03 8:16 AM, in article _k49a.78$35.482@dfw-service2.ext.raytheon.com,, > "Chris Olive" <nospam@raytheon.com> wrote: >  >  >>Jeff Cameron wrote:  >>N >>>Has anyone have a simple solution for sending OPCOM messages for a specificJ >>>class, and/or Security Audit events via email? It would be desirable toK >>>execute a DCL script so that I may filter and condense certain messages.  >>>  >>@ >>This isn't another "contest" with your Unix friend, is it? 8-) >>2 >>(Couldn't resist asking this tongue-in-cheek...) >> >>Chris  >>----- 
 >>Chris Olive  >>Systems Consultant) >>Raytheon Technical Services Corporation  >>Indianapolis, IN >>, >>email: olivec(AT)indy(DOT)raytheon(DOT)com >> > @ > No, he's sleeping right now, waiting for his system to reboot. > L > I need this because my server has TELNET enabled on the web because I needN > to access my system from work, and even though I've got full security turnedK > with tough intrusion detection turned on. I notice that I've been getting H > several login attempts. I have the intrusion monitoring for certain ip& > addresses hang around for ten years. > M > I just want audit security alarms to have their text cleaned up to be breif  > and sent to my text pager. >  > Jeff >   I Well, I normally don't post anything purporting to be an answer unless I  F look into it to verify the facts.  I can't do that right now, so take 7 this as a vague memory and maybe entirely false, but...   F I thought I remember something about being able to receive audits and F opcom alarms to a mailbox.  If that's the case, then your solution, I D would think -- even if "home-grown", wouldn't be too hard.  Write a H listener at the other end of such a mailbox to clean up your alarms and E send 'em to your pager via an SMTP gateway (or paging gateway if you  ' have the software or a modem attached.)   D Maybe someone can verify or otherwise set me straight on this vague 	 memory...    Chris  -----  Chris Olive  Systems Consultant' Raytheon Technical Services Corporation  Indianapolis, IN  * email: olivec(AT)indy(DOT)raytheon(DOT)com   ------------------------------   Date: 5 Mar 03 18:47:49 +0100 ) From: p_sture@elias.decus.ch (Paul Sture) = Subject: Re: Sending OPCOM or Security Audit events via email ) Message-ID: <fthbt2NPD8$X@elias.decus.ch>   U In article <j0muXSOsujPr@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes: d > In article <3E6520D7.8650AAD8@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes: >> Paul Sture wrote:O >>> And that base processing should be robust in the case of the Audit Listener I >>> Mailbox. If your listening program terminates unexpectedly, the Audit I >>> Mailbox fills up, and anything wanting to write to it will then hang. K >>> LOGINOUT.EXE is a good example of that, as you can find yourself unable   >>> to login to fix the problem. >>  N >> I thought the only supported way of capturing OPCOM messages is to create aB >> pseudo terminal (PDT$ routines if I recall) and then issueing aO >> REPLY/ENABLE=xxx message to that pseudo terminal and then reading the output L >> which will be text formatted opcome messages just as they appear on OPA0: >>   > 6 > I said in "the case of the Audit Listener Mailbox". $ > I was referring to OPCOM messages.   ^^^^^^^^^^^^^^^^^^  B That of course should read "I was NOT referring to OPCOM messages"   >  >> Is there another way ?  >  > Dunno. >  > --   > Paul Sture --     --  
 Paul Sture   ------------------------------  + Date: Wed, 05 Mar 2003 10:46:35 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> @ Subject: Re: strange behaviour of SET ENVIRONMENT/NODE in SYSMAN; Message-ID: <01KT5V5TPLW29FQH4P@sysdev.deutsche-boerse.com>   & > > The question is "bug or feature?". > > J > > The HELP says to enclose a list of nodes in parentheses, and indeed ifJ > > one doesn't it doesn't work.  On the other hand, if one uses a logicalH > > name (in SYSMAN$NODE_TABLE) for the node list, then one can omit the > > parentheses. >  > DJAS01::DDACHTERA$ mc sysman > SYSMAN> doodleH > %CLI-W-IVVERB, unrecognized command verb - check validity and spelling >  \DOODLE\ 
 > SYSMAN>  > F > Note that the message indicates the CLI as the facility, not SYSMAN. > A > Since SYSMAN uses DCL (CLI$) to parse the command line, this is   > expected, documented behavior. > 7 > > With, say, SET PROTECTION, one cannot leave out the H > > parentheses---neither explicitly nor in the definition of a logical. > J > Yes, because any list as a qualifier value must be enclosed in paren.'s. > Entirely consistent.  I Let me rephrase that.  Why can one use a logical such as NODE1,NODE2 (no  B parentheses in the logical name) but if written out one needs the  parentheses?  H Is it because SET ENV/NOD=<logical name> passes the logical name itself H to SYSMAN, so it looks like just one node and not a list as long as the  CLI is handling it?    ------------------------------  % Date: Wed, 05 Mar 2003 07:24:41 -0500 2 From: rdeininger@mindspring.com (Robert Deininger)@ Subject: Re: strange behaviour of SET ENVIRONMENT/NODE in SYSMANL Message-ID: <rdeininger-0503030724410001@user-uinj4a8.dialup.mindspring.com>  J In article <01KT5V5TPLW29FQH4P@sysdev.deutsche-boerse.com>, Phillip Helbig+ <HELBPHI@sysdev.deutsche-boerse.com> wrote:   J >Let me rephrase that.  Why can one use a logical such as NODE1,NODE2 (no C >parentheses in the logical name) but if written out one needs the  
 >parentheses?  > I >Is it because SET ENV/NOD=<logical name> passes the logical name itself  I >to SYSMAN, so it looks like just one node and not a list as long as the   >CLI is handling it?  H Correct.  The parser sees a single item (the logical name), not a list. J So no parentheses are needed.  SYSMAN translates the logical name and gets7 a list of nodes, but that happens after CLI processing.   D If you SET ENV/NOD=<list of logical names>, you'll need parentheses.   ------------------------------  # Date: Wed, 05 Mar 2003 12:30:08 GMT - From: bdc@world.std.com (Brian 'Jarai' Chase)  Subject: Re: unix ) Message-ID: <b55hj-50s.ln1@pez.jarai.com>   G In article <b3sod7$rp0$1@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote: + > In article <en88j-q3l.ln1@pez.jarai.com>, 3 >    bdc@world.std.com (Brian 'Jarai' Chase) wrote:   G > > Yes, and I'm not trying to say that Unix is technically better than 2 > > TOPS-10 or any of its other contemporary OSes. > C > I wasn't talking about strengths and weaknesses of code execution @ > either.  I'm talking about the pros and cons of making sources= > available.  See my other post about the dangers of shipping 
 > sources.  C Yes, I saw your other post.  The one regarding the complications of H providing source because the customers start requesting that you includeI their contributions into your code tree; then they expect you to maintain I those contributions.  Also the complications of their code changes having J negative interactions with /other/ customers wanting to provide changes of
 their own.  G My opinion is that those are problems which are most painful for closed D development environments.  Unix, or more specifically the opensourceH incarnations of it, sprung up in the face of those challenges.  In fact,E they're entirely built up from user contributions!  So, the very same H processes which, for a closed and centrally managed OS, are problematic,G are the ones that opensource operating systems would not exist without.   E That's not to say that those sames processes don't cause problems and ? conflicts of interest, but the models and methods of opensource H programming have evolved to receive more benefits than harm from them.  # (Two steps forward, one step back.)   ) There are a number of reasons it works.     I * There is no one organization that eats the maintenance costs.  Instead  G   of there being dozens of even hundreds of paid developers, there are  G   many thousands of volunteer developers.  For the most part, each are  E   responsible for his or her own code.  If you fail to maintain your  F   contribution, and there's not enough value in that contribution for H   someone else to feel it worthy of taking ownership of it, it falls by    the wayside.  H * There are no promises to "customers" from the core of the development H   effort.  You're only entitled to what you pay for.  However, what you E   get for free is certainly some of the most remarkable stuff in the  G   world of software.  /If/ you're a customer, and you really want some  I   kind of guaranteed support, you can pay for it.  There are a number of  F   companies willing and able to do this, and at relatively reasonable I   prices because of the competition offered by everyone having access to  F   the sources.  This group of support organizations for hire will growA   proportionately to the demand for paid support and development.   I * The system is designed to be highly modular to facilitate distributed,  I   loosely connected (in time and space) development efforts.  The kernel  I   is separate from the utilities.  Anybody can easily extend the OS from  K   the command line tools perspective.  The only dependencies on the kernel  K   there are through the system call API.  Even within the kernel, all the   J   major subsystems tend to be very clearly delineated; device drivers are I   also well encapsulated, interfacing to the kernel through well defined  J   hooks.  Thousands of people contribute device drivers.  Filesystems are "   similarly "pluggable" creatures.  H * The coding style leans heavily towards straightforward clarity in its J   organization and style.  The practice of writing code that creates some K   desireable side-effect is notably absent, except perhaps in the few bits  5   that are coded in the assembly language primitives.   J * The coding is done in a portable language so that nearly all the effort I   put into development can be preserved as the world of hardware evolves  
   beneath it.     > > ...  My point is that it hasL > > the ability to survive and perpetuate itself better than any other OS.  N > > That's all; I'm not singing the technical praises of Unix above and beyondL > > all other OSes.  I'm saying it's hard to kill, because, really, no /one/K > > controls it anymore.  Between the companies who've already licensed and M > > have been developing their own Unix versions for years, and the untainted ? > > Unix-like OSes, there's no one place to go to eliminate it.  > C > On the contrary, this may be its Achilles' heel.  There is no one B > entity to babysit the bits, enforce standards, and have the lastB > say on Bad Ideas.  Do not misunderstand me.  Given a choice, I'dC > always vote for the side that distributes the knowledge; however, ? > there's a complete set of other problems that comes with that 	 > choice.   H The standards enforcer of the opensource world is a very effective one--I and it's not any organization--it's what comes about from the intertia of H the applications which run on Unix.  If any branch of Unix were to strayC off, trying to implement non-standard APIs without also maintaining J significant compatibility with some "least common denominator" of Unix, itG would very quickly die for the lack of application support.  If not, it H would at best be consigned to some little niche and mostly ignored--but E only if it had a few worthwhile tricks in its bag that no other Unix   offered.  E As for the Bad Ideas... If they're truly /Bad Ideas/ then, again, the F process of evolution comes into play.  Either the bad ideas are culledG from the system, or the system that retains them dies off because those F bad ideas cripple it.  I also think that the introduction of Bad IdeasH into any of the major opensource Unix OSes is not a very common thing.  H In my experience with NetBSD, there are some outright bloody peer reviewI sessions on the development lists.  I've not even seen Mediocre Ideas get J past the scathing criticism of them.  My understanding is that FreeBSD andI OpenBSD are equally critical, and that core kernel features of Linux must ? pass the review of Linus and/or other key competent developers.   G What happens is that all of the efforts compete to have the Best Ideas. K The *BSDs, owing to their similarities, tend to borrow code and ideas from  F eachother all the time.  In one release, they may all attack some new E terrority using different ideas--tackling it with each of their best  J guesses; by their next releases, they may all sync up on whichever of the ) approaches worked out most successfully.    5 > > .. The one exception being VMS, which still isn't K > > terribly portable--it seems to require herculean efforts of coding each N > > time it's moved to a new processor architecture--but it is still kicking,  > > more or less.  > = > Those kicks are called death throes.  Of course it requires ? > effort to support a new architecture.  You can either do this @ > effort in the OS or you can do it in the compiler that is used@ > to make the OS executable.  THERE IS NO SUCH THING AS PORTABLE > WITHOUT WORK.   I This is true.  I think with VMS it's quite a lot more work because of how E VMS was implemented, historically.  They've got to port compilers for G BLISS, MACRO-32, C, ADA, and (others?) because those languages have all G been used to implement various bits of the OS. In following comp.os.vms G recently, my understanding is that they're actually compiling VAX MACRO K code into their target platforms of Alpha of IA64.  I bet that's got to be  J a real pain, especially when trying to handle VAX assembly that relies on % various side effects of instructions.   D However, the opensource C compilers are fairly portable in their ownB rights.  (i.e. GCC, PCC, and LCC).  /That's/ probably an even moreD important part of the success of opensource Unix.  Sticking with theC evolutionary metaphor, you can say there's a symbiotic relationship 9 involved between the compilers and the operating systems.   G This is all the fault of Ken and Dennis, and their dangerous idea which  has taken on a life of its own.    -brian.    ------------------------------  $ Date: Wed, 5 Mar 2003 13:01:44 -0000D From: "Rupert Pigott" <roo@dark-try-removing-this-boong.demon.co.uk> Subject: Re: unix 2 Message-ID: <1046869304.495878@saucer.planet.gong>  : "Brian 'Jarai' Chase" <bdc@world.std.com> wrote in message# news:b55hj-50s.ln1@pez.jarai.com...  [SNIP]  K > This is true.  I think with VMS it's quite a lot more work because of how G > VMS was implemented, historically.  They've got to port compilers for I > BLISS, MACRO-32, C, ADA, and (others?) because those languages have all I > been used to implement various bits of the OS. In following comp.os.vms I > recently, my understanding is that they're actually compiling VAX MACRO L > code into their target platforms of Alpha of IA64.  I bet that's got to be  A I very much doubt that VAX MACRO poses a huge challenge. You have C a fairly well known original target (the VAX architecture), whereas ? if you're porting stuff like C across you're fighting more than ? one layer... ie : The ISA, the compiler, and the OS. Although I 6 would bet that the VMS sources are fairly tidy by now.   Cheers,  Rupert   ------------------------------   Date: 5 Mar 2003 18:20:06 GMT + From: Patrick Scheible <kkt@itchy.serv.net>  Subject: Re: unix * Message-ID: <b45f4m$6he$0@205.153.154.199>   jmfbahciv@aol.com writes:      > [about Unix] > C > On the contrary, this may be its Achilles' heel.  There is no one B > entity to babysit the bits, enforce standards, and have the lastB > say on Bad Ideas.  Do not misunderstand me.  Given a choice, I'dC > always vote for the side that distributes the knowledge; however, ? > there's a complete set of other problems that comes with that  > choice  E There's not one entity for all of Unix, but each of the BSDs and each < of the Linux distributions has a group babysitting its bits.  F Open source gives some protection to users against arbitrary decisionsD by companies.  Multiple open source versions of Unix gives even moreD protection.  Good ideas can get copied, while organizational failure? by one group still leaves other versions of Unix around.  While > converting wouldn't be painless, it would be easier than, say, converting from TOPS-10 to VMS.   
 -- Patrick   ------------------------------  $ Date: Wed, 5 Mar 2003 04:46:53 -0500  From: John Santos <JOHN@egh.com>8 Subject: Re: Upgrade to OPEN VMS 7.3-1 - Patches needed?6 Message-ID: <1030305044208.26948B-100000@Ives.egh.com>  ' On Tue, 4 Mar 2003, Jeff Goodwin wrote:    > N > Most of the released ECOs would be pertinent in my configurations, yours may	 > differ.  > E > Some of the V7.3-1 problems we ran into required new images not yet L > available in released patches.  The specific issues we received new images > for were:  > G > - HSJ tape drives go into Mount Verify state at dismount (every time) = > - Callable mail programs failing with RMS-F-BUG (very rare) E > - Non-fatal INCONSTATE bugchecks in error log (caused by SYS 2 ECO) 4 > - SMTP mail hanging in queue (caused by SYS 2 ECO)K > - Broken clusterwide mail/print broadcasts (caused by fix to previous two  > items)  A SYS 3 ECO was released yesterday (at least, I got mail about it.)   E The mail says it fixes the last 3 above (the third by backing out the L fix, so there is still a potential problem if you delete a process that has J pending asynch broadcasts.)  There is also a fix for something that soundsI a lot like item 1, tapes in mount verify on dismount.  I'm not sure about 
 the 2nd item.   ) > Last unfixed issue we have (so far) is:  > L > - RMS-F-DME error when searching logfiles (happens rarely, doesn't seem to* > be a PIOPAGES issue, but we're not sure) >  > -Jeff  >  > K > "Johan Nordebrink" <johan.nordebrink@se.flextronics.com> wrote in message 9 > news:d443f1b8.0303040258.70f9e72d@posting.google.com... @ > > I am planning to do an upgrade from OPEN VMS 7.2-1 to 7.3-1. > > G > > Anyone know of some patches(VMS, TCPIP, DECnet IV) that I must also 
 > install? > >  > > Thank you in advance!  > > 
 > > /Johan   --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  $ Date: Wed, 5 Mar 2003 13:43:51 -0500; From: "Brian Tillman" <Brian.Tillman@notnoone.notnohow.com> 8 Subject: Re: Upgrade to OPEN VMS 7.3-1 - Patches needed?$ Message-ID: <3e664557$1@news.si.com>  0 >http://ftp.digital.com.au/pub/ecoinfo/a731.html   That page says to go to G http://ftp.support.compaq.com.au/pub/ecoinfo/a731.html, but trying that " leads to a "page not found" error. --  I Brian Tillman         Internet: Brian.Tillman at smiths-aerospace dot com 5 Smiths Aerospace  Addresses modified to prevent SPAM. @ 3290 Patterson Ave. SE, MS Replace "at" with "@", "dot" with "." Grand Rapids, MI 49512-1991 8        This opinion doesn't represent that of my company   ------------------------------  * Date: Wed, 5 Mar 2003 11:17:04 +0000 (UTC)5 From: "John Wallace" <johnwallace4@yahoo.dotco.dotuk> E Subject: Re: VAXELN vs Posix Threads (Was: DECthreads problem on VAX) 1 Message-ID: <b44mbf$3rk$1@knossos.btinternet.com>   : "Dave Weatherall" <djweath@attglobal.net> wrote in message/ news:DTiotGxQ0bj6-pn2-nyzyGJ51wCpb@localhost... > > On Sun, 2 Mar 2003 18:09:54 UTC, cwhii_google_spam@yahoo.com > (C.W.Holeman II) wrote:  > < > > djweath@attglobal.net (Dave Weatherall) wrote in message1 news:<DTiotGxQ0bj6-pn2-6pHkIZkuND00@localhost>...  > > E > > > One thing I'm sure of, Posix based threading is nowhere near as J > > > elegant as VAX/ELN's JOB/Process model and the synchronisation tools > > > it provides. > > * > > Do you care to expound/expand on this? > F > I don't have time at the minute to writwe a long diatribe as I'm offF > to the UK for the rest of the week. Suffice to say for now that withC > DecThreads (aka Posix) one is obliged to manipulate a lot of data B > items that ELN does for you. e..g. (from memory I'm at home). toD > create an ELN process (thread of execution) within an ELN JOB (VMSF > process), I use CREATE_PROCESS with name, argument. In DecThreads, iF > have to create attributes, assign atttributes, and then I can create/ > the Thread. Deleting it all again is similar.  > D > To synchronise with one or more objects, in ELN I call WAIT_ANY orC > WAIT_ALL with a list of things that should be waited for. The ELN F > kernel does the work. In Threads I have to manipulate more attributeD > objects and then handle Mutexes and Condition variables myself. ItF > works but you can see the Unix as opposed to the RSX/VMS heritage of2 > ELN. I'm a VMS bigot so what can one expect :-). > F > Oh yes and then there's the little matter of error/status reporting.F > ELN's got it, Threads expects you to do it with exceptions. Great ifA > you're writing C, there a macros to do it. Not so easy in other ! > languages. My stuff is Fortran.  > ! > Enough, I've got to get to wok.  >  > -- > Cheers - Dave.  H UNIX heritage of ELN? Am I misunderstanding you here? Yes ELN is lighterI weight than VMS (no demand paging, no quotas, no privileges, no security, L etc) but there's no real UNIX heritage that I could see - e.g. ELN's got theK nice synchronisation you mention (which UNIX didn't) and (iirc) ELN doesn't # have that basic UNIX concept, fork.    'Twas nice software though.    cheers john   ------------------------------  $ Date: Wed, 5 Mar 2003 08:00:22 -0500  From: John Santos <JOHN@egh.com>E Subject: Re: VAXELN vs Posix Threads (Was: DECthreads problem on VAX) 6 Message-ID: <1030305075722.26948A-100000@Ives.egh.com>  ' On Wed, 5 Mar 2003, John Wallace wrote:    > < > "Dave Weatherall" <djweath@attglobal.net> wrote in message1 > news:DTiotGxQ0bj6-pn2-nyzyGJ51wCpb@localhost... @ > > On Sun, 2 Mar 2003 18:09:54 UTC, cwhii_google_spam@yahoo.com > > (C.W.Holeman II) wrote:  > > > > > > djweath@attglobal.net (Dave Weatherall) wrote in message3 > news:<DTiotGxQ0bj6-pn2-6pHkIZkuND00@localhost>...  > > > G > > > > One thing I'm sure of, Posix based threading is nowhere near as L > > > > elegant as VAX/ELN's JOB/Process model and the synchronisation tools > > > > it provides. > > > , > > > Do you care to expound/expand on this? > > H > > I don't have time at the minute to writwe a long diatribe as I'm offH > > to the UK for the rest of the week. Suffice to say for now that withE > > DecThreads (aka Posix) one is obliged to manipulate a lot of data D > > items that ELN does for you. e..g. (from memory I'm at home). toF > > create an ELN process (thread of execution) within an ELN JOB (VMSH > > process), I use CREATE_PROCESS with name, argument. In DecThreads, iH > > have to create attributes, assign atttributes, and then I can create1 > > the Thread. Deleting it all again is similar.  > > F > > To synchronise with one or more objects, in ELN I call WAIT_ANY orE > > WAIT_ALL with a list of things that should be waited for. The ELN H > > kernel does the work. In Threads I have to manipulate more attributeF > > objects and then handle Mutexes and Condition variables myself. ItH > > works but you can see the Unix as opposed to the RSX/VMS heritage of4 > > ELN. I'm a VMS bigot so what can one expect :-). > > H > > Oh yes and then there's the little matter of error/status reporting.H > > ELN's got it, Threads expects you to do it with exceptions. Great ifC > > you're writing C, there a macros to do it. Not so easy in other # > > languages. My stuff is Fortran.  > > # > > Enough, I've got to get to wok.  > >  > > -- > > Cheers - Dave. > J > UNIX heritage of ELN? Am I misunderstanding you here? Yes ELN is lighterK > weight than VMS (no demand paging, no quotas, no privileges, no security, N > etc) but there's no real UNIX heritage that I could see - e.g. ELN's got theM > nice synchronisation you mention (which UNIX didn't) and (iirc) ELN doesn't % > have that basic UNIX concept, fork.  >  > 'Twas nice software though.   E I parsed Dave's sentence as "Unix [heritage of DECthreads] as opposed  to RSX/VMS heritage of ELN."   >  > cheers > john   --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Wed, 05 Mar 2003 12:53:02 +0000 ' From: Andrew Harrison SUNUK Consultancy ! Subject: Re: VMS Backup solutions . Message-ID: <3E65F32E.1070906@nospamn.sun.com>   Bob Koehler wrote:Z > In article <3E6393A6.2070102@nospamn.sun.com>, Andrew Harrison SUNUK Consultancy writes: > E >>>   How many successful recoveries have you had despite a bad tape?  >>>  >>4 >>Well every server thats running is an example of a& >>sucessfull recovery. Thats over 120. >  > C >    That's not what I asked.  How many times have you been able to F >    recover data from a tape despite the fact that that tape was bad? > I >    Not from aonther tape, not from a similar system, from the bad tape.  >     < Well these arn't open real tapes so recovering data once the& tape has been damaged isn't that easy.  > So the answer is never, we just restore against the other copy/ why would you bother if you had a second copy ?   > BTW having OpenVMS or UNIX as the platform isn't going to make9 any difference to your ability to recover a damaged tape.    regards  Andrew Harrison    ------------------------------  % Date: Wed, 05 Mar 2003 09:13:34 -0500 & From: jlsue <jlsuexxxz@screaminet.com>! Subject: Re: VMS Backup solutions 8 Message-ID: <841c6v04k6e31li53a76gu8p7snbgis1r7@4ax.com>  . On Tue, 04 Mar 2003 01:18:52 GMT, Jeff Cameron <JCam90502@jcameron.com> wrote:   J >On 3/3/03 7:09 AM, in article jbr66vctcq43drnl16d6knscl8kv7idipu@4ax.com,* >"jlsue" <jlsuexxxz@screaminet.com> wrote: > > >> Actually, I've always wondered if the alias entries for theH >> VMS$COMMON.DIR-->[SYSx]SYSCOMMON.DIR are unnecessary, since the SYS$*E >> rooted logicals are searchlists anyway.  If they are necessary, it I >> could due to the the low-level facilities that are needed during early H >> boot processes.  But I'm not involved in the product development area0 >> so I don't get to work on investigating that. > K >When backing up the system disk, it is very important to get the directory L >alias structure backed up too, so that it is restored the same way. This is
 >[snip...]  B Uh, yeah.  You misunderstood my musings above.  I fully understandF (well, relatively fully anyway) the directory structures and the aliasA entries.  I think I explained these earlier in the post about the % problems with 3rd party backup tools.   C The above wandering about the need for the alias speaks more in the E general sense of "why?" (my would tell you I do ask that overly much)   A But, just to get back on track, re: 3rd-party VMS backup tools... E It's possible that some 3rd products do handle the aliases correctly. 1 Legato's newest version seem to say that they do.   B However, just because a 3rd party product doesn't handle them veryE well doesn't mean that the product won't work for you.  Sometimes you @ may need an extra disk drive available to which you create a VMSE BACKUP saveset of the system disk.  This could then be saved with the @ 3rd-party backup tool.  No, it's not a slick as a direct-to-tapeC backup, but it's relatively cheap to work-around and is well within  most IT groups' budgets.  D And, in response to Larry's note about other OpenVMS disk with alias> entries.... OUCH!  I've never needed to implement them for anyC application I've managed, but I can empathize with the problem that B will present.  The new Legato product could help out there, if you* need to get away from straight VMS BACKUP.   ------------------------------   Date: 5 Mar 2003 08:49:12 -0600 - From: Kilgallen@SpamCop.net (Larry Kilgallen) ! Subject: Re: VMS Backup solutions 3 Message-ID: <pP20T2VhKPwP@eisner.encompasserve.org>   a In article <841c6v04k6e31li53a76gu8p7snbgis1r7@4ax.com>, jlsue <jlsuexxxz@screaminet.com> writes:   F > And, in response to Larry's note about other OpenVMS disk with alias@ > entries.... OUCH!  I've never needed to implement them for anyE > application I've managed, but I can empathize with the problem that  > will present.   H But have you ever managed a system and implemented a method to _prevent_ users from creating aliases ?    ------------------------------   Date: 5 Mar 2003 09:48:19 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ! Subject: Re: VMS Backup solutions 3 Message-ID: <ptjmUkDp0mg2@eisner.encompasserve.org>   X In article <3E65F32E.1070906@nospamn.sun.com>, Andrew Harrison SUNUK Consultancy writes:  > > Well these arn't open real tapes so recovering data once the( > tape has been damaged isn't that easy. > @ > So the answer is never, we just restore against the other copy1 > why would you bother if you had a second copy ?   D    And if your disk dies while you're making that second tape?  Yes,@    I has some 3rd party ddisks that had a habbit of dying during    backups.   @ > BTW having OpenVMS or UNIX as the platform isn't going to make; > any difference to your ability to recover a damaged tape.     @    Guess again.  I HAVE restored VMS backups from damaged tapes,    no data loss.   ------------------------------   Date: 5 Mar 2003 10:25:41 -0600 + From: young_r@encompasserve.org (Rob Young) ! Subject: Re: VMS Backup solutions 3 Message-ID: <bfRNjItD9tX3@eisner.encompasserve.org>   q In article <ptjmUkDp0mg2@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: Z > In article <3E65F32E.1070906@nospamn.sun.com>, Andrew Harrison SUNUK Consultancy writes: > ? >> Well these arn't open real tapes so recovering data once the ) >> tape has been damaged isn't that easy.  >>  A >> So the answer is never, we just restore against the other copy 2 >> why would you bother if you had a second copy ? > F >    And if your disk dies while you're making that second tape?  Yes,B >    I has some 3rd party ddisks that had a habbit of dying during
 >    backups.  >   < 	It would be a poor system that requires two backups to makeH 	a second tape.  The way TSM works, is it does a daily export for DR so G 	the data is in two places.  Now *maybe* you just happened to write to  D 	a bad tape from your just completed backups.  If that is the case, G 	your backups failed (or actually a subset of your backups failed) you   	need to rerun them.  B 	And if a disk dies during backup?  Isn't that the point of having> 	backups, for recovery?  You may have to go to a prior backup.    A >> BTW having OpenVMS or UNIX as the platform isn't going to make < >> any difference to your ability to recover a damaged tape. >  > B >    Guess again.  I HAVE restored VMS backups from damaged tapes, >    no data loss. >   @ 	They never went through a fire OR have an operator "accidently"? 	degauss them, right?   In other words, damage is a loose term. ? 	They weren't all that damaged, they were somewhat "dinged" up. = 	And yes, VMS backup is better at having tapes survive dings.    				Rob    ------------------------------  % Date: Wed, 05 Mar 2003 16:29:09 +0000 ' From: Andrew Harrison SUNUK Consultancy ! Subject: Re: VMS Backup solutions . Message-ID: <3E6625D5.4040107@nospamn.sun.com>   Bob Koehler wrote:Z > In article <3E65F32E.1070906@nospamn.sun.com>, Andrew Harrison SUNUK Consultancy writes: >  > > >>Well these arn't open real tapes so recovering data once the( >>tape has been damaged isn't that easy. >>@ >>So the answer is never, we just restore against the other copy1 >>why would you bother if you had a second copy ?  >  > F >    And if your disk dies while you're making that second tape?  Yes,B >    I has some 3rd party ddisks that had a habbit of dying during
 >    backups.  > 3 We havn't had any data loss due to storage failure.   ; But we did then we could simply copy the first tape offline  in the media server.     > @ >>BTW having OpenVMS or UNIX as the platform isn't going to make; >>any difference to your ability to recover a damaged tape.  >  >  > B >    Guess again.  I HAVE restored VMS backups from damaged tapes, >    no data loss. >   6 Well of course, but that depends on whats on the tape.  4 The tape in question was one of 8 holding a database4 table. Since the tape itself was damaged we couldn't2 have recovered the database without the other copy of the tape.  - So in this case OpenVMS wouldn't have helped.   4 I have had data recovered from a tape where the tape3 is damaged but where the files we wanted to recover , were not on the damaged section of the tape.   ------------------------------  # Date: Wed, 05 Mar 2003 14:12:20 GMT , From: "James Gessling" <jgessling@yahoo.com>* Subject: [OT] ex-Etrade CEO house for saleA Message-ID: <8Bn9a.2844$BY7.127186025@newssvr15.news.prodigy.com>   B Since we did discuss Etrade and their VMS plans, I thought you allC might appreciate this.  One of the arguments for getting rid of VMS ? was that it is expensive. Not as expensive as this bozo's house  I bet.  See:  L http://www.mlslistings.com/common/properties/propertyDetail.asp?open=0&page=' 10&mls_number=310282&type=property&name    Regards,  J    ------------------------------  # Date: Wed, 05 Mar 2003 13:22:10 GMT # From: "John Smith" <a@nonymous.com>  Subject: Re: [OT}:to SueI Message-ID: <6Sm9a.153467$UXa.69096@news02.bloor.is.net.cable.rogers.com>   A "Kesav Tadimeti" <Kesav_Tadimeti@KeaneIndia.com> wrote in message C news:8EA11405E59BD611BA7100104B93C260FDDBAF@exdel01.del.mgsl.com... 	 > Hi Sue, D > I remember your saying in a recent post that HP is not advertising	 its OSes. F > I refer to ORACLE magazine's Jan-Feb 2003 edition. I happened to see a : > special ad section devoted to HP-UX 11i ( & ORacle 11i). >  > Perhaps I am wrong.     * 1) HP doesn't advertise what doesn't sell.1 2) Without advertising, VMS doesn't sell so well. 	 3) Goto 1   	 Catch-22.    ------------------------------   End of INFO-VAX 2003.126 ************************