0 INFO-VAX	Fri, 16 Feb 2001	Volume 2001 : Issue 94      Contents: Re: # of context switches ?  %LINK-W-MULDEF errors. Re: Another missed opportunity Re: Another missed opportunity Re: Another missed opportunity Re: Another missed opportunity	 Authorize 
 Re: Authorize 
 Re: Authorize ! Re: Availability Manager vs. AMDS  Bank of America is off4 Re: Can you remove LAT & DECNET from a Decserver 90?1 Delete, links etc was: Another missed opportunity + Re: Detecting modern hacker attempts on VMS  Re: ES40 upgrade?  Re: ES40 upgrade? & Re: Future outlook for OpenVMS admins?& Re: Future outlook for OpenVMS admins?& RE: Future outlook for OpenVMS admins?" Getting to the console on an Alpha Re: Google buys Deja Re: History of VAX releases  Interesting feature... Re: Itanium IS the Itanic  Re: Itanium IS the Itanic . Re: Looking for KA630 CPU Module User's Guide.. Re: Looking for KA630 CPU Module User's Guide. Re: Most
 Re: newbie' Re: no mail path to %smtp"fred@foo.bar" ' Re: no mail path to %smtp"fred@foo.bar"  OpenVMS and Supercomputing Re: OpenVMS Promotional Items  Re: Oracle RDB SQLSRV problems- Re: Oracle8i on OpenVMS/ES40-Platform anybody * Powerpoint (or other) demonstration of CMS0 Re: Problems with DECnet IV and Gigabit Ethernet Replaceing LP37 with LG15 Plus Re: TFT > 18.1" with 1280x1024
 track STOP/ID  Re: track STOP/ID  Re: track STOP/ID  Re: UNIX write vs VMS ?  Version numbers  Re: Version numbers  Re: Version numbers  Re: version numbers. Re: version numbers.% VMSINSTAL to POLYCENTER INSTALL tool?  RE: www.openvms.compaq.com RE: www.openvms.compaq.com Re: www.openvms.compaq.com RE: www.openvms.compaq.com  F ----------------------------------------------------------------------  % Date: Fri, 16 Feb 2001 08:35:08 +0000  From: Roy Omond <Roy@Omond.net> $ Subject: Re: # of context switches ?) Message-ID: <3A8CE63B.1AE0315F@Omond.net>    "Doug W." wrote:   > Peter Ljungberg wrote... > G > << Does anybody know if there is a accurate way to tell the number of I > context switches an (smp) OpenVMS (7.2-1H1) system is currently doing ?  >  >>  > N > Sadly, I do not know of a way to get this valuable information from VMS.  IfO > the scheduler bumped a counter in the PCB when a process went current, things " > would be a lot more interesting. > A > Glad to see someone is interested in counting context switches.   E Hmmm... going back to the days with the VAX 6000-400 and the VAX 9000 N with vector processors (I remember them well :-), there *were* indeed countersJ maintained for fast and slow vector processor context switches.  The hooksK are still in VMS (see HELP LEX F$GETJPI FAST_VP_SWITCH and SLOW_VP_SWITCH). P A fast context switch was one where the process was scheduled on the same vector  H processor, thus not requiring the restoration of quite a large amount of context.  SlowA was the opposite :-)   The difference could be quite substantial.   F Perhaps it's not that difficult to add to VMS in some future release ?  	 Roy Omond  Blue Bubble Ld.    ------------------------------  % Date: Fri, 16 Feb 2001 11:22:58 -0500 , From: "McNutt, Larry E." <Mcnutt@timken.com> Subject: %LINK-W-MULDEF errors. R Message-ID: <DA88FF10ADA8D211ADE700805F6507AF05BB8611@ctnhemail01.corp.timken.com>   Hello,  ; Recently I was awarded the privilege of managing 2 more VAX : systems that were transferred from another facility to HQ.= Thankfully, the previous admin did a good job and things have E been running smoothly. Now, we're going to see if we can fix that. ;)    On a VAX 4000-200 VMS 6.2-1H3 = I am trying to build some cobol applications and I am getting = several warnings from the linker. The output is listed below. A This didn't happen on an Alpha system and I tried some variations D of adding sys$library:vaxcrtl to the link options file, but no luck. TIA for your help.   Larry McNutt       McNutt@Timken.com    >@tkncom acollate cbl main  I $! **********************************************************************   I $! *                                                                    *   I $! * Procedure Name: TKNCOM.COM                                         *   I $! * Syntax: @TKNCOM program-name program-extension program-type        *   I $! *                 link/nolink-option                                 *   I $! *                                                                    *   I $! * This command procedure is used to compile C, COBOL, and FORTRAN    *   I $! * programs and, if a main program, link it to the REUSE shareable    *   I $! * image.                                                             *   I $! *                                                                    *   I $! * It can also be used to assemble Macro programs and create the      *   I $! * object file.                                                       *   I $! *                                                                    *   I $! **********************************************************************    $!   $ BC001_START:   $     WRT := WRITE SYS$OUTPUT   % $     IF P2 .EQS. ""  THEN P2 = "CBL"   % $     IF P3 .EQS. ""  THEN P3 = "SUB"   & $     IF P4 .EQS. ""  THEN P4 = "LINK"   $     _SOURCE_NAME = P1+"."+P2  < $     _COMP_MSG = "               ***** Running 'TKNCOM' for "+_SOURCE_NAME+" **  ***"   $     WRT _COMP_MSG   <                ***** Running 'TKNCOM' for ACOLLATE.CBL *****  9 $     _COMPILE_PARMS = "/CROSS_REFERENCE/LIST="+P1+".LST"   ; $     IF P2 .EQS. "C"        THEN      GOTO BC200_C_PROGRAM   ? $     IF P2 .EQS. "CBL"      THEN      GOTO BC300_COBOL_PROGRAM    $ BC300_COBOL_PROGRAM:  H $!   Create the environment control file for 805YE so that it alters the   $!   appropriate statements   " $     COPY I805YEA.CBL I805YEA.DAT  ; $     IF P3 .EQS. "MAIN"  THEN COPY I805YEB.VAX I805YEB.DAT   G $     IF P3 .EQS. "SUB"   THEN COPY I805YEB.VAX,I805YEB.SUB I805YEB.DAT    $!  ( $!   Run the source preprocessor program  3 $     WRT "Running the Source Preprocessor (805YE)"   ' Running the Source Preprocessor (805YE)    $     _HOLD_NAME   = P1+".HLD"    $     _DELETE_NAME = P1+".CBL.*"   $     ASSIGN "I805YEA.DAT"  DDA    $     ASSIGN "I805YEB.DAT"  DDB    $     ASSIGN ACOLLATE.CBL DDC    $     ASSIGN "D805YED.DAT"  DDD    $     RUN P805YE   $     DELETE I805YEA.DAT;*   $     DELETE I805YEB.DAT;*   $     DEASSIGN DDA   $     DEASSIGN DDB   $     DEASSIGN DDC   $     DEASSIGN DDD  & $     RENAME ACOLLATE.CBL ACOLLATE.HLD  % $     RENAME D805YED.DAT ACOLLATE.CBL    $!   $     WRT "Compiling"   	 Compiling   4 $     _COMPILE_PARMS = "/ANSI_FORMAT"+_COMPILE_PARMS  % $!    _COMPILE_PARMS = "/ANSI_FORMAT"   G $     COBOL /ANSI_FORMAT/CROSS_REFERENCE/LIST=ACOLLATE.LST ACOLLATE.CBL    $     DELETE ACOLLATE.CBL.*   & $     RENAME ACOLLATE.HLD ACOLLATE.CBL   $     GOTO BC500_LINK   
 $ BC500_LINK:   6 $     IF P3 .EQS. "SUB"  THEN          GOTO BC600_EXIT  6 $     IF P4 .EQS. "NOLINK"  THEN       GOTO BC600_EXIT   $     WRT "Linking"    Linking   5 $     LINK/MAP/CROSS_REFERENCE ACOLLATE,REUSEMAIN/OPT    REUSE/SHAREABLE   $ %LINK-W-WRNERS, compilation warnings  8         in module REUSE file $1$DIA1:[MISDSS]REUSE.VAX;2  , %LINK-W-MULDEF, symbol KILL multiply defined  ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7   . %LINK-W-MULDEF, symbol GETUID multiply defined  ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7   / %LINK-W-MULDEF, symbol GETEUID multiply defined   ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7   . %LINK-W-MULDEF, symbol GETCWD multiply defined  ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7   - %LINK-W-MULDEF, symbol _EXIT multiply defined   ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7   , %LINK-W-MULDEF, symbol EXIT multiply defined  ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7   . %LINK-W-MULDEF, symbol RENAME multiply defined  ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7   , %LINK-W-MULDEF, symbol OPEN multiply defined  ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7   - %LINK-W-MULDEF, symbol FOPEN multiply defined   ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7   - %LINK-W-MULDEF, symbol CHDIR multiply defined   ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7   . %LINK-W-MULDEF, symbol REMOVE multiply defined  ?         in module VAXCRTL file SYS$COMMON:[SYSLIB]VAXCRTL.EXE;7    $!  
 $ BC600_EXIT:    $     WRT "Finished!!!"    Finished!!!    >    ------------------------------    Date: 16 Feb 2001 08:50:48 -05004 From: koehler@eisner.encompasserve.org (Bob Koehler)' Subject: Re: Another missed opportunity 3 Message-ID: <Re17UWEfCX42@eisner.encompasserve.org>   e In article <3A8C83A1.A69382@earthlink.net>, "David J. Dachtera" <djesys.nospam@earthlink.net> writes:  > J > Does not "rm" "remove" the file and all it's links? Do we really want to  > change the behavior of DELETE?  H No.  UNIX rm removes one link.  When all hard links are gone the file is really deleted.   H The current VMS alias scheme addresses the quota issue.  If it's my fileH and I delete it it is gone.  The only ways I can think of to work quotasF fairly on the UNIX approach is to charge everyone that has a hard linkF or to charge the oldest hard link.  IMHO the latter would be a pain to
 implement.  F Somebody out there must know how UFS handles this.  I never turn quota7 on on any of my UNIX systems so I can't readily try it.   F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation = NASA GSFC Flight Software       | Federal Sector, Civil Group E                                 | please remove ".aspm" when replying    ------------------------------    Date: 16 Feb 2001 08:55:28 -05004 From: koehler@eisner.encompasserve.org (Bob Koehler)' Subject: Re: Another missed opportunity 3 Message-ID: <1YJuTZAbCoRL@eisner.encompasserve.org>   ` In article <96h5qk$1c9l$1@info.cs.uofs.edu>, bill@triangle.cs.uofs.edu (Bill Gunshannon) writes: > C > Who owns it??  Under Unix, no matter where the links point to the C > file still belongs to only one user.  If I own a file, and create A > a link to that file that happens to be in your directory (with  B > adequate permissions to do so, of course) the file still belongsE > to me.  If I delete the link in my directory, the file still exists B > because of the link in your directory, but it also still belongsG > to me and my quota would be charged for it.  How do VMS Quotas work??  > By owner or location?? >   A On UNIX, the owner is recorded in the inode.  Creating a new link H creates a new inode.  IIRC there's no difference between the first inode( and any other if they're all hard links.  F So who owns it?  Each inode could record a separate owner (OK, I gottaF go to a UNIX system and try that).  If I rm the original hard link why' should I still be considered the owner?   F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation = NASA GSFC Flight Software       | Federal Sector, Civil Group E                                 | please remove ".aspm" when replying    ------------------------------    Date: 15 Feb 2001 21:37:05 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> ' Subject: Re: Another missed opportunity H Message-ID: <y4ae7nvg26.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  M In VMS, continuing to charge to the original owner is certainly reasonable if N the file isn't actually deleted - in contrast to Unix, the directory structureN isn't required to access files, it serves as sugar for us poor humans with our? inexact and associative memories. The main point is that such a G reference-counting scheme actually stops somebody from deleting her own  file - not really a good idea.   	Jan   ------------------------------    Date: 16 Feb 2001 10:35:12 -05004 From: koehler@eisner.encompasserve.org (Bob Koehler)' Subject: Re: Another missed opportunity 3 Message-ID: <mOgJpeaGjk3d@eisner.encompasserve.org>   j In article <Re17UWEfCX42@eisner.encompasserve.org>, koehler@eisner.encompasserve.org (Bob Koehler) writes: > H > Somebody out there must know how UFS handles this.  I never turn quota9 > on on any of my UNIX systems so I can't readily try it.   F   OK, I tried it.  Looks like the new inode gets the same owner as theE   old inode, so the original creator controls it no matter where it's 	   linked.    ------------------------------  # Date: Fri, 16 Feb 2001 13:15:35 GMT 2 From: "William Hymen" <t18_pilot@hotmail.spam.com> Subject: Authorize= Message-ID: <XJ9j6.43186$gb1.1610032@news4.aus1.giganews.com>   ; Recently, I ran authorize on a test ID on a test VMS system = ( this system was not under security ) and reset the Password 8 lifetime.  This got one of the sys admins nose all bent.  7 So how did he know I ran authorize on the test user ID? " ps- I have bypass on the test box.   Bill   ------------------------------  % Date: Fri, 16 Feb 2001 09:20:43 -0500 0 From: David Beatty <David.Beatty.NOSPAM@sas.com> Subject: Re: Authorize2 Message-ID: <+DaNOl1lOLn8dMCr1APYnybcworp@4ax.com>  : VMS auditing would have reported that you changed password# lifetime and did it from AUTHORIZE.   1 On Fri, 16 Feb 2001 13:15:35 GMT, "William Hymen" # <t18_pilot@hotmail.spam.com> wrote:   < >Recently, I ran authorize on a test ID on a test VMS system> >( this system was not under security ) and reset the Password9 >lifetime.  This got one of the sys admins nose all bent.  > 8 >So how did he know I ran authorize on the test user ID?# >ps- I have bypass on the test box.  >  >Bill  >    ------------------------------    Date: 16 Feb 2001 22:45:41 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: Authorize- Message-ID: <87r90ysn3e.fsf@prep.synonet.com>   4 "William Hymen" <t18_pilot@hotmail.spam.com> writes:  = > Recently, I ran authorize on a test ID on a test VMS system ? > ( this system was not under security ) and reset the Password : > lifetime.  This got one of the sys admins nose all bent. > 9 > So how did he know I ran authorize on the test user ID? $ > ps- I have bypass on the test box.  " All changes to sysuaf are audited.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  # Date: Fri, 16 Feb 2001 15:06:54 GMT 1 From: "Mark D. Jilson" <jilly@clarityconnect.com> * Subject: Re: Availability Manager vs. AMDS2 Message-ID: <3A8D42CB.BC430129@clarityconnect.com>  F Because I expect that in the future the developers for the driver codeE will assume that everyone is using AvailMan, like say 2-3 years after G AMDS console has had it's last version shipped.  Then I expect that the H driver code will send some data that will blow up any old AMDS console. ! Maybe this won't happen but ....     Larry Kilgallen wrote: > h > In article <3A8BE1D2.CDB57ED8@clarityconnect.com>, "Mark D. Jilson" <jilly@clarityconnect.com> writes:K > > Yes the future will be Availability Manager.  Remember that there are 2 I > > pieces here.  The driver that supplies the data to interested parties F > > and the console that uses this data for it's nice pretty displays.K > > Under VMS the driver piece will still be developed as it has to keep up I > > with changes to VMS and and new pieces of performance/usage data that K > > can be collected.  The AMDS console is/will be retired and Availability L > > Manager will be the only console.  If you are implementing only now thenK > > work at implementing Availability Manager as this will create less work  > > for you in the future. > C > Since the recommendation is to put the console on a non-clustered C > system, why couldn't one just avoid upgrading the system that has  > the AMDS console ?   --  D Jilly	- Working from Home in the Chemung River Valley - Lockwood, NY0 	- jilly@clarityconnect.com			- Brett Bodine fan. 	- Mark.Jilson@Compaq.com			- since 1975 or so, 	- http://www.jilly.baka.com               -   ------------------------------  % Date: Fri, 16 Feb 2001 07:52:17 -0300e) From: fabio_compaq@ep-bc.petrobras.com.bre Subject: Bank of America is offuL Message-ID: <OFA6CE3C4C.9E249F24-ON032569F5.003B9957@ep-bc.petrobras.com.br>   What do they use ?  ; http://news.cnet.com/news/0-1007-200-4839299.html?tag=mn_hdM  + I know they are using Solaris . . .  :-))))a   Andrew are you there ?   Regardse   FC   ------------------------------    Date: 16 Feb 2001 04:12:08 +0800, From: Paul Repacholi <prep@prep.synonet.com>= Subject: Re: Can you remove LAT & DECNET from a Decserver 90??- Message-ID: <87vgqb4sfb.fsf@prep.synonet.com>u  % "Didier Morandi" <dmo@ims.ch> writes:   9 > I'd love to know why a Customer wants to stop DECnet... L > Unfortunately, I actually know the answer. He wants to stop DECnet routingL > on his network. When will people understand that DECnet and TCP/IP are TWO6 > DIFFERENT PRODUCTS !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!  > Normally a manager who would not know a net work from his, um, phone holder... ;)  F The trick is, add up all the suff you need. Be generous. Then when theE 'get all the not-IP off the net work, you say "Sure, I'll need 4 GigaeC switches, 6 new FO circuits to Timbucktoo, 120 DE500s, 6 master PCInF boxes for the Alphas. I'll get on to it as soon as YOU get the stuff."  < If he tells you you are being unresonable, stop SMB and IPX.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.u@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------    Date: 16 Feb 2001 21:57:27 +0800, From: Paul Repacholi <prep@prep.synonet.com>: Subject: Delete, links etc was: Another missed opportunity0 Message-ID: <87hf1uu3w8.fsf_-_@prep.synonet.com>  I Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> writes:e  + > "Bill Todd" <billtodd@foo.mv.com> writes:n  ; > > That would (in VMS) imply that this 'preferred' link be D > > identifiable at file creation time, or when the created file was0 > > subsequently entered in its first directory.  C > That already is the case. Some time after SET FILE/ENTER was madeiF > available (the facility was always there, but subject to a SMOP), itB > was realized that DELETE's behaviour could be improved by makingA > this distinction. It took two goes to get this right - the case4A > mentioned in another reply, when the alias entry is in the same:C > directory as the primary entry, was initially handled incorrectly 
 > (all IIRC).x   cre/dir [.a] cre/dir [.b] cre [.a]foo.fooE$ set file/ent=[.a]bar.bar [.a]foo.foo$ set file/ent=[.b]bar.bar [.a]foo.foo del [.a]foo.foo;1r  D dumpimng the headers of the bars, reveales a back link FID of 0,0,0, and a file name of foo.foo;1.e  E But, anal/disk flags the incorect backlink for [.b]foo.foo;1, but note the [.a] one...l  4 %ANALDISK-W-BACKLINK, directory (783,3,1) [A]B.DIR;1/         incorrect back link for entry FOO.FOO;1d       -- -< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.c@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------   Date: 16 Feb 2001 08:27:20 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)t4 Subject: Re: Detecting modern hacker attempts on VMS0 Message-ID: <96io98$g1f$3@n.ruf.uni-freiburg.de>  \ In article <3A8CBF28.A17BFDCB@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:O >In the past, hacker attacks were directed at login attempts, so it was easy to 6 >SHOW INTRUSION, ANA/AUDIT and looking at OPERATOR.LOG >sL >But now that attacks come in a much greater variety of flavours through theI >internet to a variety of ports, are there tools to detect/analyze hackerT >attempts ?  >kI >I am thinking of attempts to connect to various inactive ports, and then*O >checking the various log files for the applications that are on various ports.  > K >Does TCPIP services provide tools to monitor or alert when there are usage ) >patterns that warrant taking a look at ?o >mN >Also, how would one go about scanning web server logs to detect patterns that' >are consistent with hacking attempts ?   G I configured our router to reject inappropriate packets and send reject L messages to Opcom (via syslog (the service on VMS is provided by Multinet)).I So I am able to extract all sorts of statistics, not only VMS related bute for the whole network.   Regards,    Christoph Gartmann   H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, FRG                                               |H +--------- http://www.immunbio.mpg.de/home/english/menue.html ---------+   ------------------------------  % Date: Fri, 16 Feb 2001 08:35:42 -0600 1 From: "Dave Gudewicz" <david.gudewicz@abbott.com>E Subject: Re: ES40 upgrade?8 Message-ID: <96jdsf$6oq$1@fizban.fizban.pprd.abbott.com>  J I liked VAXen too.  Still have a few.  But they are yesterday and Alpha is today.   Dave...   @ "Steeples, Oliver" <Oliver.Steeples@compaq.com> wrote in messageH news:F498D199EDB12D468CD2C66680D308018B11E4@reoexc04.emea.cpqcorp.net...L > "Compaq's Alpha EV68 running at 833MHz will be used to upgrade The Big Q'sK > AlphaServer ES40 in the very near future, according to old Reg chum Terryi' > Shannon over at Shannon Knows Compaq.  >eF > And the 1GHz version of the chip, of which the company has more than enough, L > will appear in Compaq's AlphaServer GS line - aka Wildfire - sometime next > May. >HK > Indeed, the EV68 seems to have a future. According to SKC, IBM - which ispL > fabbing the EV68 - has clocked the chip up to 1.1GHz, thanks to its copperK > interconnect technology. That's higher than the debut speed of the EV68's ? > successor which was scheduled to come in at just over 1GHz. "  > F > I heard that Compaq are bringing out an ES45, basically an ES40 with 4*1Ghz5 > processors and upto 32Gb mem.  Me, I like my vax's.E >I > Oliver >  > -----Original Message-----8 > From: Dave Gudewicz [mailto:david.gudewicz@abbott.com]+ > Sent: Thursday, February 15, 2001 9:22 PMA > To: Info-VAX@Mvb.Saic.Com  > Subject: Re: ES40 upgrade? >B >  > Evil sounds good!  ;-) > 	 > Dave...n > C > "antony wardle" <antony.wardle@nospam.met.co.nz> wrote in messageI+ > news:byWi6.302$v4.16689@ozemail.com.au...S, > > our ds20e 's came with the 667 processor > >.- > > but when they boot up, they say they haveE > >A! > > 666 processors. Evil or what?O > >5
 > > cheers > >0
 > > antony > >  > >  > >  > >  > > > E > > > I'm havin' the same issue with DS20Es going from 500 to 667Mhz.C > > >T
 > > > Dave...L > > >  > > > <n > >  > >  >K   ------------------------------    Date: 16 Feb 2001 22:32:26 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: ES40 upgrade?- Message-ID: <87zofmsnph.fsf@prep.synonet.com>K    Dirk Munk <munk@home.nl> writes:  > > Don't know about the EV7 though, and I don't know if the newJ > architecture of the EV7 would be of any use in the ES40. As I understandG > it the EV7 is basically a EV6x with a 10 GB/sec network interface fore > inter-cpu communication.  > The ES40 is about one thousand pins per CPU short! Hum, 750 of em power and ground?   -- d< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------    Date: 16 Feb 2001 10:30:38 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>C/ Subject: Re: Future outlook for OpenVMS admins?NH Message-ID: <y43ddfasap.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  ! "rocko" <rocko1@home.com> writes:   N > We are waiting on a couple Compaq AlphaServer ES40's to arrive.  How do they > rate?:  L If they're the 833 MHz variety, they're the fastest processor on the planet.   	Jan   ------------------------------  % Date: Fri, 16 Feb 2001 07:53:57 -0300b) From: fabio_compaq@ep-bc.petrobras.com.brd/ Subject: Re: Future outlook for OpenVMS admins? L Message-ID: <OF9A9F3B86.4520AD45-ON032569F5.003BBEA1@ep-bc.petrobras.com.br>  . Or what will be the future of OpenVMS admins ?  B What are you planning to do if suddenly CPQ decides to stop OVMS !  < Forget the COE agreement..... if today we were in 2020 ! ! !   Regardsr   FC        D Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> em 16/02/2001 07:30:38r             Info-VAX@Mvb.Saic.Comt      / Assunto: Re: Future outlook for OpenVMS admins?y    ! "rocko" <rocko1@home.com> writes:T  I > We are waiting on a couple Compaq AlphaServer ES40's to arrive.  How dos they > rate?d  D If they're the 833 MHz variety, they're the fastest processor on the planet.n        Jan   ------------------------------  % Date: Fri, 16 Feb 2001 12:40:05 -0600N+ From: "Main, Kerry" <Kerry.Main@compaq.com>h/ Subject: RE: Future outlook for OpenVMS admins?ON Message-ID: <910612C07BCAD1119AF40000F86AF0D805284E00@kaoexc3.kao.cpqcorp.net>  K >>>> What are you planning to do if suddenly CPQ decides to stop OVMS ! <<<e  H What will happen if MS decides to stop support of Windows NT or Wiondows 2000?-  ; What will happen if Sun decides to stop support of Solaris?r  8 What will happen if HP decides to stop support of HP-UX?  7 What will happen if IBM decides to stop support of AIX?   % Stay tuned for tomorrow's episode ...-   :-)-  
 Kerry Main Senior Consultant: Compaq Canada Inc. Professional Servicese Voice: 613-592-4660u Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----) From: fabio_compaq@ep-bc.petrobras.com.brc, [mailto:fabio_compaq@ep-bc.petrobras.com.br] Sent: February 16, 2001 5:54 AMs To: Info-VAX@Mvb.Saic.Comr/ Subject: Re: Future outlook for OpenVMS admins?n    . Or what will be the future of OpenVMS admins ?  B What are you planning to do if suddenly CPQ decides to stop OVMS !  < Forget the COE agreement..... if today we were in 2020 ! ! !   Regards    FC        D Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> em 16/02/2001 07:30:38t             Info-VAX@Mvb.Saic.Com       / Assunto: Re: Future outlook for OpenVMS admins?r    ! "rocko" <rocko1@home.com> writes:n  I > We are waiting on a couple Compaq AlphaServer ES40's to arrive.  How doe they > rate?t  D If they're the 833 MHz variety, they're the fastest processor on the planet.i        Jan   ------------------------------  % Date: Fri, 16 Feb 2001 17:43:43 +0800 ' From: Tim Sneddon <tsneddon@olc.com.au>r+ Subject: Getting to the console on an Alphad< Message-ID: <2FCE1FC4E068D411877B00D0B7477F4D0E19A3@onlpc26>  D I am trying to get to the console on an AlphaStation 255 without anyE success. Break on the console seems to have been disabled, and I haverJ noticed that there is no BREAK key on the front of the box, CTRL/P doesn't work either.  " Any help is much appreciated. Tim.  D --------------------------------------------------------------------C "You can download an atmosphere and dial up a groove, but there's atC certain magic when three musicians and a dyslexic get together and h play in a room."
    -- BonoD --------------------------------------------------------------------   ------------------------------    Date: 16 Feb 2001 04:20:44 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: Google buys Dejar- Message-ID: <87r90z4s0z.fsf@prep.synonet.com>i  B For a minor giggle, ping localhost.deja.com. The perfect PHB test!     -- t< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.5@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------    Date: 16 Feb 2001 22:21:22 +0800, From: Paul Repacholi <prep@prep.synonet.com>$ Subject: Re: History of VAX releases- Message-ID: <874rxuu2sd.fsf@prep.synonet.com>t  / "Nick Ogbourne" <okeburnit@bigpond.com> writes:s  D > Can anyone point me to information that lists the release date andH > withdrawal dates of VAX and Alpha machines in the period 1995 to 2000?  G Ah, get the VMS FAQ, and go to the hobiest licence page. have a look at- the mongo list of alphas...1   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.4@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Fri, 16 Feb 2001 11:41:48 -0700n% From: Dean Woodward <deanw@rdrop.com>I Subject: Interesting feature...D) Message-ID: <3A8D746C.FF54A45A@rdrop.com>i  C One of the guys from the other side of the building (Internet Svcs) A just got a new toy from that company named after our nearest starRE (Not mentioning names lest I awaken AH).  He showed me a feature that  I found most interesting.d  C This is a little 1u rackmount box.  It's console runs even when theFC system itself is powered down- in fact, one of the commands you canaA issue is 'poweron', and suddenly all the fans and drives wake up.   B Forgiving my ignorance of high-end Alphas, but do any of them knowA how to do this, and if so, is it a feature likely to migrate down B to low end boxes that might compete with this? (Say, the rackmount
 DS10 series)?h  E (Of course, and OTOH, it's a feature for when your box isn't running,6/ so maybe not of much interest to VMS users. ;-)r   ------------------------------  % Date: Fri, 16 Feb 2001 08:36:49 -0500r4 From: "Bochnik, William J" <BochnikWJ@bernstein.com>" Subject: Re: Itanium IS the Itanic> Message-ID: <2B37459189B0D211BE710000F8EF9D8508908C65@NTS0147>  J This message is in MIME format. Since your mail reader does not understand< this format, some or all of this message may not be legible.  ' ------_=_NextPart_001_01C0981D.8533E77E  Content-Type: text/plain;. 	charset="iso-8859-1"a  L Funny you should say this - I've been saying for YEARS that Compaq (actuallyK Digital at the time) should have got together with someone like Nintendo or L Sony, and get the Alpha out there as a kick a** system.  All I ever got back% from such comments were blank stares.-   Oh well.   -----Original Message----- From: JF Mezei Sent: February 15, 2001 5:46 PMt To: Info-VAX@mvb.saic.com:" Subject: Re: Itanium IS the Itanic    * fabio_compaq@ep-bc.petrobras.com.br wrote:? > Itanium is "commited to servers", and people dont buy serverse/ > at home - except one crazy friend of mine :-)a  E I disagree. Home games and the apetite for teenagers to have a fasterp computerL than their neighbour will drive the IA64 to the consumer market if it offers& enough performance and made available.  G Alpha could have taken the consumer market if it had been available andlI marketed. Quake etc would be running on Alpha, and in fact, it would have  make- Alpha based games kill off the consumer 8086.r  F The problem with intel is that competition is forever driving the 8086 speedsK up. Until the 8086 hits a solid wall, IA64 probably won't get a significants" enough edge to make it worthwhile.  K I have a feeling that when the IA64 project was started, they didn't expectnE that they would be able to push the 8086 as far as they have done so.e  H So again, the whole question remains: how much farther can they push the 8086L before it hits a wall ? Is Pentium 4 an indication that incremental advancesJ will be smaller and smaller, allowing other chips to speed past the 8086 ?   William J. Bochnik   Systems Programmer Operations, CMIS WPO-17-187, Ext. 2467h Nextel 1130 - 917 416 7103        ' ------_=_NextPart_001_01C0981D.8533E77E? Content-Type: text/html; 	charset="iso-8859-1"g+ Content-Transfer-Encoding: quoted-printabley  1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">a <HTML> <HEAD>9 <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =  charset=3Diso-8859-1">@ <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
 5.5.2651.65">r) <TITLE> Re: Itanium IS the Itanic</TITLE>u </HEAD>- <BODY>  H <P><FONT SIZE=3D2 FACE=3D"Arial">Funny you should say this - I've been =I saying for YEARS that Compaq (actually Digital at the time) should have = H got together with someone like Nintendo or Sony, and get the Alpha out =A there as a kick a** system.&nbsp; All I ever got back from such = & comments were blank stares.</FONT></P>  0 <P><FONT SIZE=3D2 FACE=3D"Arial">Oh well.</FONT> </P>  B <P><FONT SIZE=3D2 FACE=3D"Arial">-----Original Message-----</FONT>7 <BR><FONT SIZE=3D2 FACE=3D"Arial">From: JF Mezei</FONT>o@ <BR><FONT SIZE=3D2 FACE=3D"Arial">Sent: February 15, 2001 5:46 =	 PM</FONT>lB <BR><FONT SIZE=3D2 FACE=3D"Arial">To: Info-VAX@mvb.saic.com</FONT>? <BR><FONT SIZE=3D2 FACE=3D"Arial">Subject: Re: Itanium IS the =I
 Itanic</FONT>S </P> <BR>  F <P><FONT SIZE=3D2 FACE=3D"Arial">fabio_compaq@ep-bc.petrobras.com.br =
 wrote:</FONT> E <BR><FONT SIZE=3D2 FACE=3D"Arial">&gt; Itanium is &quot;commited to =n1 servers&quot;, and people dont buy servers</FONT>eC <BR><FONT SIZE=3D2 FACE=3D"Arial">&gt; at home - except one crazy =  friend of mine :-)</FONT>6 </P>  I <P><FONT SIZE=3D2 FACE=3D"Arial">I disagree. Home games and the apetite =". for teenagers to have a faster computer</FONT>G <BR><FONT SIZE=3D2 FACE=3D"Arial">than their neighbour will drive the =C/ IA64 to the consumer market if it offers</FONT> ? <BR><FONT SIZE=3D2 FACE=3D"Arial">enough performance and made =  available.</FONT>6 </P>  F <P><FONT SIZE=3D2 FACE=3D"Arial">Alpha could have taken the consumer =* market if it had been available and</FONT>H <BR><FONT SIZE=3D2 FACE=3D"Arial">marketed. Quake etc would be running =0 on Alpha, and in fact, it would have make</FONT>B <BR><FONT SIZE=3D2 FACE=3D"Arial">Alpha based games kill off the = consumer 8086.</FONT>. </P>  A <P><FONT SIZE=3D2 FACE=3D"Arial">The problem with intel is that =u5 competition is forever driving the 8086 speeds</FONT>dI <BR><FONT SIZE=3D2 FACE=3D"Arial">up. Until the 8086 hits a solid wall, =e, IA64 probably won't get a significant</FONT>: <BR><FONT SIZE=3D2 FACE=3D"Arial">enough edge to make it = worthwhile.</FONT> </P>  F <P><FONT SIZE=3D2 FACE=3D"Arial">I have a feeling that when the IA64 =. project was started, they didn't expect</FONT>G <BR><FONT SIZE=3D2 FACE=3D"Arial">that they would be able to push the =e( 8086 as far as they have done so.</FONT> </P>  H <P><FONT SIZE=3D2 FACE=3D"Arial">So again, the whole question remains: =. how much farther can they push the 8086</FONT>H <BR><FONT SIZE=3D2 FACE=3D"Arial">before it hits a wall ? Is Pentium 4 =. an indication that incremental advances</FONT>I <BR><FONT SIZE=3D2 FACE=3D"Arial">will be smaller and smaller, allowing =e+ other chips to speed past the 8086 ?</FONT>t </P>  A <P><B><FONT SIZE=3D2 FACE=3D"Arial">William J. Bochnik</FONT></B>D </P>  A <P><B><FONT SIZE=3D2 FACE=3D"Arial">Systems Programmer</FONT></B> @ <BR><B><FONT SIZE=3D2 FACE=3D"Arial">Operations, CMIS</FONT></B>E <BR><B><FONT SIZE=3D2 FACE=3D"Arial">WPO-17-187, Ext. 2467</FONT></B> < <BR><B><FONT SIZE=3D2 FACE=3D"Arial">Nextel 1130 - 917 416 = 7103</FONT></B>6 </P> <BR> <BR>   </BODY>t </HTML>l) ------_=_NextPart_001_01C0981D.8533E77E--    ------------------------------    Date: 16 Feb 2001 22:59:00 +0800, From: Paul Repacholi <prep@prep.synonet.com>" Subject: Re: Itanium IS the Itanic- Message-ID: <87n1bmsmh7.fsf@prep.synonet.com>t  / JF Mezei <jfmezei.spamnot@videotron.ca> writes:m  O > So again, the whole question remains: how much farther can they push the 8086oN > before it hits a wall ? Is Pentium 4 an indication that incremental advancesL > will be smaller and smaller, allowing other chips to speed past the 8086 ?  E Just saw on the Reg that Intel have had to push the core voltage *up*aC on the P4, from 1.70 to 1.75V. So it seems they are struggling withc either yield or speed, or both.t   -- c< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.t@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  ! Date: Fri, 16 Feb 01 11:02:09 GMT  From: jmfbahciv@aol.com 7 Subject: Re: Looking for KA630 CPU Module User's Guide. + Message-ID: <96j5qj$97t$1@bob.news.rcn.net>o  2 In article <rW0j6.482$c5.397137@news1.iquest.net>,!    "JD" <dyson@jdyson.com> wrote:a >t8 >"Timothy Stark" <sword7@speakeasy.org> wrote in message; >news:EH0j6.12640$EW3.484591@e3500-chi1.usenetserver.com...- <snip>  D >> Also, I still am not finished my emulator yet.  However, I now am' >> considering to release my unfinished J >> emulator sources and let developers to finish my emulator because I now8 >> realized that finishing my emulator is taking forever >>H >The last parts of a project are often the most difficult, frustrating,  time consuming >or any of the above!!!m  ? Yes.  There is also the problem of declaring a project finished-> even though you know of stuff that has to be fixed, developed,< etc.  JMF's method of teaching this to guys who did know how> to stop was to just edict.  Tim doesn't have that luxury.  TheB middle of a development cycle is the most fun.  It's the beginning; (writing specs and dealing with marketroids) and the endingpA (documentation, field test, mundane problems such as installation0> fumble finger prevention) that is the pits.  I don't know many= people who preferred the beginning and endings to the middle.r     >0I >I am really interested in getting my hands on your code (to play with), a but I really doo+ >understand how hard you must be working!!!p  : And there's quite a number of things that need to be done.; After verifying that the system can crash, take a dump, andt6 continue, the next thing I think that needs to be done= is to  1.  verify that a multi-pack structure can be emulatedo< (Tim's way out of disk space for any interesting work on the7 -10 side)  2.  a non-RP06 structure can be accessed viaE) MOUNT/DISMOUNT--IOW private structures.  c  ; I think those are the most important.  Now, I understand ifn: Tim doesn't want to do this part because it really doesn't7 have much to do with the emulation work.  I'd also liker? to see that system swap a lot, IOW, a real test of timesharing.:   And I got more ;-).o   /BAH  ' Subtract a hundred and four for e-mail.    ------------------------------   Date: 16 Feb 2001 13:57:01 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)n7 Subject: Re: Looking for KA630 CPU Module User's Guide.g, Message-ID: <96jbjd$1dmb$1@info.cs.uofs.edu>  > Well, if everyone here is moving to emulators, perhaps someone? has a few Console cabkits they don't need.  I would like to putv@ together another couple of MicroVAXen but lack the little widget= from the back panel.  Anybody feel like sending me a couple??n  > Emulators are nice, and fun, but they just lack something when compared to the real hardware.   bill   -- -J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   n   ------------------------------  % Date: Fri, 16 Feb 2001 09:26:28 +00004 From: Roy Omond <Roy@Omond.net>i Subject: Re: Mostw) Message-ID: <3A8CF244.57DA51BB@Omond.net>n   Malcolm Smeaton wrote:  G > Where can I find the latest version of MOST file browser for VMS. Thed* > present one I am using came with CSWING.  2 Don't know about MOST, but take a look at LOOK :-)  2 LOOK is *very nice*;  as always, you'll find it at ftp://ftp.wku.edu/vms/fileserv  ' Mmm... just checked, MOST is there too.e  	 Roy Omondl Blue Bubble Ltd.   ------------------------------  % Date: Fri, 16 Feb 2001 09:32:08 +0000b From: Roy Omond <Roy@Omond.net>a Subject: Re: newbiee( Message-ID: <3A8CF398.B20250E@Omond.net>   Andy wrote:s  G > I just started a position where i will be using open VMS extensively,cJ > and am wondering if this NG has any recommendations on books that covers/ > a broad range,...intro, commands, admin, etc.  >tI > Any ideas appreciated,....local bookstores were not much help,..nothing 
 > in stock   (No criticism intended :-)  F This approach (i.e. going to yer local bookstore) seems to have becomeC endemic, very probably due to the inadequacy of the "documentation"DA that comes with, ahem, other systems.  VMS is not like that.  The > standard VMS documentation is truly excellent, covering pretty: much everything you'll ever need (certainly for a newbie).  E So, rather than wasting your time at the local bookstore, just settle < down in a nice comfortable sofa with the VMS documentation !  	 Roy Omonde Blue Bubble Ltd.   ------------------------------  % Date: Fri, 16 Feb 2001 08:47:41 +0100l, From: Didier Morandi <Didier.Morandi@gmx.ch>0 Subject: Re: no mail path to %smtp"fred@foo.bar"& Message-ID: <3A8CDB1D.52FA53A1@gmx.ch>  > "MAIL$PROTOCOL_SMTP" = "TCPIP$SMTP_MAILSHR" (LNM$SYSTEM_TABLE)  F I have the same route, and the same database as the other system whereH it works. If I use the Kepner-Tregoe method that I learned when I workedF in the TSC in 1983, the answer to the question "what has changed sinceE the last time it worked" is: I upgraded OpenVMS from 7.1 to 7.2-1 andp UCX V4 to TCP/IP V5.0A.r   D.   John Vottero wrote:n > : > What's the value of the logical name MAIL$PROTOCOL_SMTP?   ------------------------------    Date: 16 Feb 2001 10:35:54 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>P0 Subject: Re: no mail path to %smtp"fred@foo.bar"H Message-ID: <y4wvar9dhh.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  A The M in SNMP is "management", absolutely no relation to SMTP and A mail. Whether you want the SNMP listener is quite another matter.-   	Jan   ------------------------------  % Date: Fri, 16 Feb 2001 15:48:09 -0300p) From: fabio_compaq@ep-bc.petrobras.com.bro# Subject: OpenVMS and SupercomputingeL Message-ID: <OF3F8CE602.C8A54D1B-ON032569F5.006720F9@ep-bc.petrobras.com.br>  , Is there any supercomputer running OpenVMS ?  = I searched www.top500.org and I didint discover any referenced5 about it  . . .  I dont know if  it is possible . . .e   Regardsf   FC   ------------------------------    Date: 16 Feb 2001 22:18:13 +0800, From: Paul Repacholi <prep@prep.synonet.com>& Subject: Re: OpenVMS Promotional Items- Message-ID: <878zn6u2xm.fsf@prep.synonet.com>   4 "Sue Skonetski" <susan.skonetski@compaq.com> writes:  : > Well that is certainly an idea for one to one marketing. > sues > < > "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message( > news:3A8AD241.51B713A6@videotron.ca...: > > You want promotional items that people will remember ? > > L > > Some high quality condoms with the "VMS" logo imprinted on the condom as > well > > as the caption:   > > "always up when you need it"  C Perhaps, from others coments, it could be a special CA promotion :)o   -- n< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.t@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.   ------------------------------  % Date: Fri, 16 Feb 2001 10:49:42 +0000u. From: Peter Jackson <peter.jackson@oracle.com>' Subject: Re: Oracle RDB SQLSRV problemse* Message-ID: <3A8D05C6.3947024B@oracle.com>   "McNutt, Larry E." wrote:o  ; > The default location for SQLSRV log files is sys$manager.-3 > They are named SQS_nodename_DIS*.LOG - dispatcherF% > and SQS_nodename_MON*.log - monitorc  H There are also the executor logs which go in the login directory for the service owner.  
 Peter Jackson  Rdb Support 	 Oracle UKg   ------------------------------  # Date: Fri, 16 Feb 2001 11:16:48 GMT  From: Dirk Munk <munk@home.nl>6 Subject: Re: Oracle8i on OpenVMS/ES40-Platform anybody' Message-ID: <3A8D0C16.7EB5D8D3@home.nl>   
 Hi Joseph,  H We are in the process of converting Oracle databases form VMS 6.2/OracleF 7.1 to VMS 7.2/Oracle 8.x. Because of an architecture change in OracleG (this already happened in Oracle 7.3) we noticed a initial drop (!!) insC performance of 30%. We were able to compensate this drop by severaloG means. One of them was to increase the maximum size of the mailboxes toaF to the maximum 32 kB. Oracle uses a lot of buffered IO in the shape ofB mailboxes, and increasing the size of the mailboxes reduces system	 overhead.c  = We also saw some poor programming techniques. Apparently some F applications fetched their data row by row, and changing that by usingG an array, things ran a lot quicker. (I'm not a database expert, but I'm $ sure you can figure out what I mean)  F I don't know if the disk cabinet is already present, but we decided toF configure our own disk cabinet. We use the 4300 shelves with 15000 rpmG SCSI disks behind a double HSG80. These disks are so fast, that a HSG806G is not even capable of using their full potential. I got 35 MB/sec readdG speed on a internal shelf in the ES40, and 18 MB/sec on the HSG80 usingrF a internal HSG80 performance utility. So when the HSG90 comes were are
 ready :-))  D Because of this very high speed on the internal shelves, we will useH internal disks in the ES40 for scratch areas, including table spaces forE sorting etc. Only 'real data' will be stored on the SAN. You will notnE find the 15000 rpm disks in the ES40 specs, but they are available !!t  H I have been told by Oracle experts that Oracle advises to use big stripeD sets (up to 256 GB) for maximum performance. I suppose this is true,E because this way all disk in the stripe set are statistically used atwG the same IO rate. Assuming most IO is random IO, rather than sequential F IO, this will get you maximum performance avoiding hot spots on single disks.  F The only save way to do this is using striped mirror sets,  = Raid 10,H =Raid 0+1. The reads to a mirror set are distributed over all members ofD the set, so you can again increase the IO capacity by increasing the1 number of disks in a mirror set to more than two.s  G And at last you can experiment with the block size as you will know. WenG haven't done any experimenting, but I tend to believe that larger block F size will yield higher performance with today's hardware. The overheadG of transporting more data per IO is less important than the overhead ofh
 more IO's.           joseph altea wrote:l   > Hi,o. > I am currently tasked with setting up a VLDB( > for a compaqES40 Oracle8i. Has anybody. > done this type of thing out there. I am sure2 > they have. I would like to discuss a few points. >h7 > In particular what kinds of strategies have you found . > useful for setting up the RAID on the RA8000 > using Oracle, etc....m > 3 > Any response from parties involved with Oracle ono$ > OpenVMS would be most appreciated. >  > jaltea@dsw.net   ------------------------------  % Date: Fri, 16 Feb 2001 14:21:01 -0000i' From: "Paul Arman" <home@paularman.net>i3 Subject: Powerpoint (or other) demonstration of CMSi1 Message-ID: <96jd28$jrh$1@lyonesse.netcom.net.uk>o  K The company I am working for at the moment has asked me about using CMS foriJ their code management. I have used CMS extensively before, and I am tryingG to get some information that is suitable for a demo to the managers andA# other members of the IT department.   D Is anyone aware of where I could get either a document (or ideally aJ PowerPoint demo) which gives a good overview of what CMS does, how you canI use generations, classes groups etc, as well as some more techie info fori the techies in the department.   Many many thanks.i  
 Paul Arman   ------------------------------  + Date: Fri, 16 Feb 2001 08:26:49 +0100 (MET)e& From: Rudolf Wingert <win@fom.fgan.de>9 Subject: Re: Problems with DECnet IV and Gigabit Ethernett6 Message-ID: <200102160722.IAA23871@sinet1.fom.fgan.de>   Hello,  H the problem was, that I did not reboot the system after configuration ofD the LINE and CIRCUITE. After reboot I could use the Gigabit EthernetD adapter for the DECnet protocoll. But I will have one more question:G if you will execute the DEFINE LINE EWA-1 command, the NCP utility will8E ask you for the line parameters. One of them is the number of RECEIVE=D BUFFERs. The displayed range is 1...1024. But you can't specify 1024D RECEIVE BUFFERs for DECnet via Gigabit Ethernet. OK DECnet IV is notB the state of the art. But Alphas with oneGB memory are also older.   Best regards Rudolf Wingerte   ------------------------------  % Date: Fri, 16 Feb 2001 10:19:15 -0600a From: bramblet@fuller.com ' Subject: Replaceing LP37 with LG15 Plusr* Message-ID: <3A8D5303.A29EDC0C@fuller.com>   Hello Managers,O. I have an AlphaServer 400 running Open VMS 6.2G We have a "musuem piece" of a printer that Compaq is tired of servicingiD every week. It is about a 2 hour drive from their office to ours. WeG purchased a LG 15 plus on their recommendation. They said that it would H be a "easy" replacement. Just shut down queue on old LP37 and add new LG 15 and restart queue.t  ( It turns out that it is not that simple.B This printer is connected to a  LPV11 q-bus printer controller. NoB matter what the tech does it doesn't work. This is a data products connection.r  D Does anyone have experience on how to set this up?? We don't want toF have to connect serially but would like to just drop and add newone. IF think that it is a configuration problem. I called the other person in? town who has the same server that we do and theirs is connectedF	 serially.2  ( Any help would be appreciated very much.     --
 Ron Bramblett  Systems Adminstrator Fuller Brush Company   ------------------------------   Date: 16 Feb 2001 08:01:54 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)A' Subject: Re: TFT > 18.1" with 1280x1024k0 Message-ID: <96impi$g1f$1@n.ruf.uni-freiburg.de>  X In article <3A8C65B7.D9232823@infopuls.com>, Christof Brass <brass@infopuls.com> writes:@ >I dare to ask this question because the screen will be attached >to a VMS system.e: >The dealer offered a Compaq 18.1" TFT display with native6 >resolution of 1280x1024. I expect the pixels to be to= >narrow/small to use my favourite font sizes. The dealer saysFA >there are no bigge TFT displays available with native resolutionl9 >of 1280x1024. As I have read that with TFTs it is highlyD9 >recommended to use their native resolution as they would ? >otherwise scale the picture by calculating "intermediate" dots @ >or only displaying the smaller resultion in a smaller area than? >the screen's size I'm seeking for a combination of screen size0= >and resolution which displays at the same size as my 20" CRTD	 >monitor.r >"- >Question1: is the pixel size of TFT's fixed?u   Yes, it is.t  > >Question2: is the quality reduced if the native resolution is4 >1600x1200 (19.6" TFT) but I'm using only 1280x1024?  I It lies in the principle of a TFT that the pixels are located at specific B spots. Thus, either you provide the data for every single pixel orM a) you provide less data (e.g. 800 x 400) and therefore the picture displayed H    is considerably smaller or the TFT uses four pixels for every single :    pixel data => the picture is large but you see the dotsJ b) you provide more data => the screen has to somehow "reduce" the data in0    order to assign it to the less pixels it has.  I I am writing this on an Eizo 18.1" TFT which I higly appreciate (viewable-F image size 45.9 cm or 18.07 inches). There is now an Eizo 19.6" with a resolution of 1600 x 1200 .0   Regards,    Christoph Gartmannp  H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, FRG                                               |H +--------- http://www.immunbio.mpg.de/home/english/menue.html ---------+   ------------------------------  % Date: Fri, 16 Feb 2001 08:52:41 +0100 , From: Didier Morandi <Didier.Morandi@gmx.ch> Subject: track STOP/ID& Message-ID: <3A8CDC48.84E3DA0C@gmx.ch>  5 Is there a way to know which process did a $ stop/id?KC Is the PCB of the killed process updated with such info in any way?i( Is there another solution to my problem?   Thanks,.   D. --+ http://didier.morandi.free.fr/index_us.html    ------------------------------  # Date: Fri, 16 Feb 2001 11:55:28 GMT0= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)r Subject: Re: track STOP/ID0 Message-ID: <009F7B6D.E90FB226@SendSpamHere.ORG>  U In article <3A8CDC48.84E3DA0C@gmx.ch>, Didier Morandi <Didier.Morandi@gmx.ch> writes::6 >Is there a way to know which process did a $ stop/id?D >Is the PCB of the killed process updated with such info in any way?) >Is there another solution to my problem?n >e >Thanks, >o >D.r >--n, >http://didier.morandi.free.fr/index_us.html  G On the Freeware V3.0 and V4.0 CDs is a [TMESIS] directory with containslG a simple Alpha system service intercept.  The example intercept "hooks"cH both $CREPRC and $DELPRC to display the caller's PID on the console.  ItH should be very easy to modify for your needs.  I'd suspect you'd want toF know the PID of the process deleted as well as the PID of the deleting process.  G With a little effort, you could combine this code with a process to log F these $DELPRC calls to a file.  Since the intercept is in kernel mode,H you cannot write to a file with normal file system calls but you can putG the intercept information into a mailbox.  Simply write a program which = loops reading the mailbox contents and writes them to a file.g  H The system service intercept template was published as a way to simplifyG hooking services.  It is not the most efficient mechanism by any means.aG It would be fine for the occasional $DELPRC call but I would not use iteG to intercept any service which is used extensively throughout VMS.  FornG example, if you decide to use it to intercept $ENQ and $DEQ, don't comer7 crying to me about the performance lag on your system. w   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMr            sO city, n., 1. a place where trees are cut down and streets are named after them.t   ------------------------------    Date: 16 Feb 2001 07:29:46 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)n Subject: Re: track STOP/ID3 Message-ID: <6C++ki4noS+L@eisner.encompasserve.org>a  U In article <3A8CDC48.84E3DA0C@gmx.ch>, Didier Morandi <Didier.Morandi@gmx.ch> writes:-7 > Is there a way to know which process did a $ stop/id?G  E VMS auditing will record a "use of privilege" audit for the caller ofiH SYS$DELPRC when GROUP or WORLD privilege is used (and auditing enabled).? Thus no audit will be performed in the following circumstances:a  3 	Target process has the same UIC as Subject process : 	Target process is in the same Job Tree as Subject processF 	Subject process privileges are due to image installed with privilege, 		and thus part of the TCB  D Obviously the third condition is not applicable to the STOP/ID case,> unless you have a _very_ non-standard system configuration :-)  N ==============================================================================N Great Inventors of our time: Al Gore -> Internet; Sun Microsystems -> ClustersN ==============================================================================   ------------------------------    Date: 16 Feb 2001 10:24:07 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>   Subject: Re: UNIX write vs VMS ?H Message-ID: <y466ibaslk.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  M In Unix, write() is a syscall. In VMS, it is done by the C RTL, I presume. In N any case, VMS doesn't do such stupid things, so you don't have to worry. OTOH,N you can keep the loop if properly coded - it will execute once, and that's it.   	Jan   ------------------------------  % Date: Fri, 16 Feb 2001 02:49:32 -05000- From: JF Mezei <jfmezei.spamnot@videotron.ca>R Subject: Version numbers+ Message-ID: <3A8CDB8A.51C0B77@videotron.ca>   J There was discussion about how to handle the 32k limit of version numbers.  4 My suggestion for this is to add a switch to PURGE :  N PURGE/RESET would rename files so that the lowest numbered version would be 1.  N PURGE/RESET/KEEP=0 would not delete any file, but just do the renumbering down to 0.s  I This way, when applications are started, they can use a simple command toh< reset their log files to numbers that are always manageable.  ! Right now, the easiest way is to:a  	 PURGE *.*s RENAME *.*  ;1  0 However you lose multiple versions of the files.   ------------------------------  % Date: Fri, 16 Feb 2001 04:50:35 -0800 3 From: Jeff Coffield <Jeffrey@DigitalSynergyInc.com>t Subject: Re: Version numbers5 Message-ID: <3A8D221B.55CB679B@DigitalSynergyInc.com>m   JF Mezei wrote:m  L > There was discussion about how to handle the 32k limit of version numbers. >06 > My suggestion for this is to add a switch to PURGE : >>P > PURGE/RESET would rename files so that the lowest numbered version would be 1. >xP > PURGE/RESET/KEEP=0 would not delete any file, but just do the renumbering down > to 0.o >,K > This way, when applications are started, they can use a simple command toe> > reset their log files to numbers that are always manageable. >h# > Right now, the easiest way is to:a >d > PURGE *.*a > RENAME *.*  ;1 >:2 > However you lose multiple versions of the files.   How about the following:   $       ! RENAME.COM $       ! P1 = File name $iJ $       HIGH_VER        = F$EXTRACT(1,5,F$PARSE(F$SEARCH(P1),,,"VERSION"))P $       LOW_VER         = F$EXTRACT(1,5,F$PARSE(F$SEARCH(P1+";-0"),,,"VERSION")), $       INDEX           = F$INTEGER(LOW_VER)- $       STOP            = F$INTEGER(HIGH_VER)@ $       COUNT           = 1- $ LOOP:m4 $       ORIG            = P1 + ";" + F$STRING(INDEX)4 $       NEW             = P1 + ";" + F$STRING(COUNT) $       RENAME 'ORIG' 'NEW'/LOG-# $       INDEX           = INDEX + 1-# $       COUNT           = COUNT + 10) $       IF INDEX .LE. STOP THEN GOTO LOOPs   ------------------------------    Date: 16 Feb 2001 15:58:37 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>p Subject: Re: Version numbersH Message-ID: <y44rxuy8rm.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  . Before renaming, check the file really exists.   	Jan   ------------------------------  % Date: Fri, 16 Feb 2001 09:11:37 -05000# From: Jim Agnew <agnew@hsc.vcu.edu>o Subject: Re: version numbers.P+ Message-ID: <3A8D3519.3F1182D8@hsc.vcu.edu>@   There was posted a looong time ago a com file that actually looked for files that had version numbers greater than whatever was your$ limit... Maybe the archives have it?  h by the way, WHERE are the archives, i used to know, but it got stored on a bad sector in my brane..  ;-)   J.   John Macallister wrote:  > N > One facet of version numbers which may not be obvious, or perhaps overlookedL > until it rears its ugly head, is the passage of time. We have several dataM > and program directories which have been in continuous existence since early N > 1981. 32767 versions may appear vast but it represents only 4-5 versions perJ > day. A regular job running every six hours, producing a log file or someL > other new file each time, would be hitting the 32767 limit now. I've foundJ > that I've had to manually rename a few multiple version files in various* > places to avoid version number problems. > M > It takes even less time for FTP and DECnet log files to reach the limit and,I > there's nothing in the standard procedures to prevent a version  numbernI > problem arising. Simple purging isn't enough to avoid the version limiteK > problem: it's also necessary to rename the files to lower version numberslN > e.g. to 1,2,...,n if keeping n versions. I've modified our local server .com@ > FTP and DECnet files to rename as well as purge the log files. > N > Unless you're able to globally rename all files to, say, version 1 from timeK > to time it's a good idea to review the distribution of version numbers onaL > your disks e.g. by simply using the dir command and tabulating the versionL > numbers. Doing this will help to warn you where version numbers are rising > to problem values. >  > John > D > Name: John B. Macallister  E-mail: j.macallister1@physics.ox.ac.ukJ > Post: Nuclear and Astrophysics Laboratory, Keble Road, Oxford OX1 3RH,UKC > Phone: +44-1865-273388 (direct)  273333 (reception)  273418 (Fax)e   ------------------------------  % Date: Fri, 16 Feb 2001 10:29:12 -0500e- From: "Peter Weaver" <peter.weaver@stelco.ca>o Subject: Re: version numbers.e4 Message-ID: <uHbj6.131708$Z2.1721364@nnrp1.uunet.ca>  0 "Jim Agnew" <agnew@hsc.vcu.edu> wrote in message% news:3A8D3519.3F1182D8@hsc.vcu.edu...tD > There was posted a looong time ago a com file that actually lookedA for files that had version numbers greater than whatever was your & > limit... Maybe the archives have it? >i  C Here is a procedure that I run from time to time, it uses DFU V2.7.,   $!0 $!      CHECK_DISK_PROBLEMS.COM  PRW 19-DEC-2000 $!C $!      A procedure that uses DFU to check for common disk problemsa that may $!      be devloping.  $! $!      Call it with: $!         @PRW$UTL:CHECK_DISK_PROBLEMS [diskname [max_ver
 [large_ver]]] 
 $!      where 9 $!         diskname is a disk drive, defaults to SYS$DISK C $!         max_ver is an integer, we look for directories that have 	 more than ; $!              max_ver versions of one file, defaults to 5dA $!         large_ver is an integer, we look for files that have ao versionc@ $!              number greater than large_ver, defaults to 20000 $!* $ if "''p1'" .eqs. "" then p1 := sys$disk:& $ if 'f$integer(p2) .le. 0 then p2 = 5* $ if 'f$integer(p3) .le. 0 then p3 = 20000 $!D $ write sys$output "Checking ''p1' for directories over 128 blocks." $!% $ assign/nolog/user_mode no dfu$nosmgg& $ dfu sear /file=*.dir/siz=min=128 'p1 $! $ write sys$output -D     "Checking ''p1' for directories with more than ''p2' versions of	 one file"s $!% $ assign/nolog/user_mode no dfu$nosmgo $ dfu dir /ver='p2    'p1  $!D $ write sys$output "Checking ''p1' for files that have reached ''p3'
 versions." $!% $ assign/nolog/user_mode no dfu$nosmgi $ dfu sear /ver=min='p3 'p1e $!   ------------------------------  % Date: Fri, 16 Feb 2001 09:46:03 +0100 , From: Didier Morandi <Didier.Morandi@gmx.ch>. Subject: VMSINSTAL to POLYCENTER INSTALL tool?% Message-ID: <3A8CE8CB.86854C9@gmx.ch>h  G I ask my question for the last time before I start writing the program.s  > Does anyone know about such tool or related documentation, the! Polycenter doc annexe A excepted?h   D. a -- e+ http://didier.morandi.free.fr/index_us.htmlp   ------------------------------   Date: 16 Feb 2001 00:52 CSTn' From: carl@gerg.tamu.edu (Carl Perkins)h# Subject: RE: www.openvms.compaq.com - Message-ID: <16FEB200100525280@gerg.tamu.edu>c  5 "Farrell, Michael" <MFarrell@voltdelta.com> writes...aJ }As an enhancement to File-11, how about treating the version number as anM }unsigned integer and then we could have 65535 versions which wouldd put thise- }kind of a problem off for a lot longer time.r } 
 }Mike Farrell.  B This is not a good plan. It would wipe out the concept of relativeF version numbers. You know - things like foo.bar;-1 to get the previous3 version, foo.bar;-0 to get the oldest version, etc.i   --- Carl   ------------------------------  % Date: Fri, 16 Feb 2001 08:45:09 -0500a4 From: "Bochnik, William J" <BochnikWJ@bernstein.com># Subject: RE: www.openvms.compaq.comv> Message-ID: <2B37459189B0D211BE710000F8EF9D8508908C66@NTS0147>  J This message is in MIME format. Since your mail reader does not understand< this format, some or all of this message may not be legible.  ' ------_=_NextPart_001_01C0981E.AF6BEAE0  Content-Type: text/plain;  	charset="iso-8859-1"i  L Um, no.  The "shorthand" specifications of ;-1, ;-2 etc do NOT correspond toJ the signed integers - they are just that - shorthand for relative numbers.E Increasing the size of the version should not affect their operation.aH Saying ;-1 says "give me one back from the highest number".  It does notB return a negative version (in the sense that -32767 to -1 could beI represented as 32768 - 65535 (or vice versus at a binary level)).  Makingh" the version unsigned should be ok.   -----Original Message----- From: carl@gergl1.gerg.tamu.edu  Sent: February 16, 2001 1:52 AMV To: Info-VAX@mvb.saic.comg# Subject: RE: www.openvms.compaq.comh    5 "Farrell, Michael" <MFarrell@voltdelta.com> writes...lJ }As an enhancement to File-11, how about treating the version number as anH }unsigned integer and then we could have 65535 versions which wouldd put this- }kind of a problem off for a lot longer time.0 } 
 }Mike Farrello  B This is not a good plan. It would wipe out the concept of relativeF version numbers. You know - things like foo.bar;-1 to get the previous3 version, foo.bar;-0 to get the oldest version, etc.    --- Carl   William J. Bochnik   Systems Programmer Operations, CMIS WPO-17-187, Ext. 2467a Nextel 1130 - 917 416 7103        ' ------_=_NextPart_001_01C0981E.AF6BEAE0r Content-Type: text/html; 	charset="iso-8859-1"t+ Content-Transfer-Encoding: quoted-printable1  1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">d <HTML> <HEAD>9 <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =a charset=3Diso-8859-1">@ <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
 5.5.2651.65">r* <TITLE> RE: www.openvms.compaq.com</TITLE> </HEAD>n <BODY>  4 <P><FONT SIZE=3D2 FACE=3D"Arial">Um, no.&nbsp; The =H &quot;shorthand&quot; specifications of ;-1, ;-2 etc do NOT correspond =F to the signed integers - they are just that - shorthand for relative =E numbers.&nbsp; Increasing the size of the version should not affect =0H their operation.&nbsp; Saying ;-1 says &quot;give me one back from the =G highest number&quot;.&nbsp; It does not return a negative version (in =-G the sense that -32767 to -1 could be represented as 32768 - 65535 (or =sD vice versus at a binary level)).&nbsp; Making the version unsigned = should be ok.</FONT></P>  B <P><FONT SIZE=3D2 FACE=3D"Arial">-----Original Message-----</FONT>) <BR><FONT SIZE=3D2 FACE=3D"Arial">From: =e  carl@gergl1.gerg.tamu.edu</FONT>@ <BR><FONT SIZE=3D2 FACE=3D"Arial">Sent: February 16, 2001 1:52 =	 AM</FONT> B <BR><FONT SIZE=3D2 FACE=3D"Arial">To: Info-VAX@mvb.saic.com</FONT>0 <BR><FONT SIZE=3D2 FACE=3D"Arial">Subject: RE: = www.openvms.compaq.com</FONT>s </P> <BR>  ? <P><FONT SIZE=3D2 FACE=3D"Arial">&quot;Farrell, Michael&quot; = / &lt;MFarrell@voltdelta.com&gt; writes...</FONT>aF <BR><FONT SIZE=3D2 FACE=3D"Arial">}As an enhancement to File-11, how =. about treating the version number as an</FONT>G <BR><FONT SIZE=3D2 FACE=3D"Arial">}unsigned integer and then we could =f0 have 65535 versions which wouldd put this</FONT>D <BR><FONT SIZE=3D2 FACE=3D"Arial">}kind of a problem off for a lot = longer time.</FONT>2+ <BR><FONT SIZE=3D2 FACE=3D"Arial">} </FONT> 6 <BR><FONT SIZE=3D2 FACE=3D"Arial">}Mike Farrell</FONT> </P>  I <P><FONT SIZE=3D2 FACE=3D"Arial">This is not a good plan. It would wipe =." out the concept of relative</FONT>F <BR><FONT SIZE=3D2 FACE=3D"Arial">version numbers. You know - things =* like foo.bar;-1 to get the previous</FONT>I <BR><FONT SIZE=3D2 FACE=3D"Arial">version, foo.bar;-0 to get the oldest =t version, etc.</FONT> </P>  0 <P><FONT SIZE=3D2 FACE=3D"Arial">--- Carl</FONT> </P>  A <P><B><FONT SIZE=3D2 FACE=3D"Arial">William J. Bochnik</FONT></B>l </P>  A <P><B><FONT SIZE=3D2 FACE=3D"Arial">Systems Programmer</FONT></B> @ <BR><B><FONT SIZE=3D2 FACE=3D"Arial">Operations, CMIS</FONT></B>E <BR><B><FONT SIZE=3D2 FACE=3D"Arial">WPO-17-187, Ext. 2467</FONT></B>-< <BR><B><FONT SIZE=3D2 FACE=3D"Arial">Nextel 1130 - 917 416 = 7103</FONT></B>  </P> <BR> <BR>   </BODY>  </HTML> ) ------_=_NextPart_001_01C0981E.AF6BEAE0--1   ------------------------------  % Date: Fri, 16 Feb 2001 11:26:13 -0700e% From: Dean Woodward <deanw@rdrop.com> # Subject: Re: www.openvms.compaq.com/) Message-ID: <3A8D70C5.728B039D@rdrop.com>    BochnikWJ@bernstein.com wrote:  N > Um, no.  The "shorthand" specifications of ;-1, ;-2 etc do NOT correspond toL > the signed integers - they are just that - shorthand for relative numbers.G > Increasing the size of the version should not affect their operation.oJ > Saying ;-1 says "give me one back from the highest number".  It does notD > return a negative version (in the sense that -32767 to -1 could beK > represented as 32768 - 65535 (or vice versus at a binary level)).  Making0$ > the version unsigned should be ok. >  > -----Original Message-----! > From: carl@gergl1.gerg.tamu.eduD > 7 > "Farrell, Michael" <MFarrell@voltdelta.com> writes...cL > }As an enhancement to File-11, how about treating the version number as anJ > }unsigned integer and then we could have 65535 versions which wouldd put > this/ > }kind of a problem off for a lot longer time.h > }  > }Mike Farrell  > D > This is not a good plan. It would wipe out the concept of relativeH > version numbers. You know - things like foo.bar;-1 to get the previous5 > version, foo.bar;-0 to get the oldest version, etc.p  E Changing from signed to unsigned only doubles the number of availablehH versions, and still does nothing to solve the issue of what happens whenF you run out- neither does changing the version field to 32 bits, it'llC just take longer to get there (I shudder to think that if they were H available, someone would want to use them all... I can't think of a goodE reason to have 32K versions of a file around online, let alone 64K or  heaven forbid 4G versions)  H The issue at hand is, whatever the limit is, what should happen when youC bump into it?  Should the filesystem renumber them all downward andhB continue, delete all but the top copy and renumber it ;1, or fail?  @ I'll offer the opinion that it should fail, as it did here.  AnyF software that generates that much file data (either generations of theG same file name or different file names) without some built-in mechanismtA to deal with it probably ought to be looked and re-thought a bit.   H That said, a mechanism to renumber all existing versions starting at oneG has been put forward, which, if run in batch mode, would have prevented0H this particular error.  (Although it needs a little more error checking-@ what if versions 201, 202, and 204 exist but not 203?  Should itA renumber them 1,2,3, or 1,2,4?  Perhaps a [NO]CONTIGUOUS option?)i   ------------------------------  % Date: Fri, 16 Feb 2001 10:38:59 -0800/! From: Shane.F.Smith@Healthnet.comr# Subject: RE: www.openvms.compaq.comAD Message-ID: <OFDC9ACDB0.F539CC59-ON882569F5.006606AE@foundation.com>  H The question is, how far down into DCL is the translation from "-1" to = "theF number one less than the top number" done? Because it's so universallyH accepted, and because the idea just plain makes sense, I'd say it's way=  H down at the lower levels. That would mean there's a lot of signed longs=  inCH various pieces of code carrying it down there. If so, that's a lot of c= odesH to change and test. That's before you start looking at customer code th= at" manipulates versions the same way.   Shane           H "Bochnik, William J" <BochnikWJ@bernstein.com> on 02/16/2001 05:45:09 A= M    To:   Info-VAX@Mvb.Saic.Com= cc:=  $ Subject:  RE: www.openvms.compaq.com        H Um, no.=A0 The "shorthand" specifications of ;-1, ;-2 etc do NOT corres= pondD to the signed integers - they are just that - shorthand for relativeF numbers.=A0 Increasing the size of the version should not affect theirH operation.=A0 Saying ;-1 says "give me one back from the highest number= ".=A0 ItH does not return a negative version (in the sense that -32767 to -1 coul= d beH represented as 32768 - 65535 (or vice versus at a binary level)).=A0 Ma= king" the version unsigned should be ok.   -----Original Message----- From: carl@gergl1.gerg.tamu.eduo Sent: February 16, 2001 1:52 AM  To: Info-VAX@mvb.saic.coml# Subject: RE: www.openvms.compaq.comh  5 "Farrell, Michael" <MFarrell@voltdelta.com> writes...eH }As an enhancement to File-11, how about treating the version number as=  anaH }unsigned integer and then we could have 65535 versions which wouldd pu= t  this- }kind of a problem off for a lot longer time.t } 
 }Mike Farrell.  B This is not a good plan. It would wipe out the concept of relativeF version numbers. You know - things like foo.bar;-1 to get the previous3 version, foo.bar;-0 to get the oldest version, etc.t   --- Carl   William J. Bochnik   Systems Programmer Operations, CMIS WPO-17-187, Ext. 2467e Nextel 1130 - 917 416 7103           =m   ------------------------------   End of INFO-VAX 2001.094 ************************