1 INFO-VAX	Thu, 13 Jul 2000	Volume 2000 : Issue 389       Contents:+ Re: %DCL-W-USGOTO, target of GOTO not found " * Return address problem with MAIL4 Re: ANNOUNCING: FreeCiv 1.11.0 port for OpenVMS Axp.! Apache for Open VMS, performance? 1 Re: C-Kermit on VMS v7.2's latest TCP/IP release?  Re: DECnet login w/ ext. auth.. Re: Default protection for logfiles (VMS 7.1).. Re: Default protection for logfiles (VMS 7.1).+ Re: Direct OPCOM messages to e-mail account + RE: Direct OPCOM messages to e-mail account  Re: disable spawn in VMSMail?  Re: disable spawn in VMSMail?  RE: disable spawn in VMSMail?  Re: disable spawn in VMSMail?  Re: disable spawn in VMSMail?  Re: disable spawn in VMSMail?  Re: disable spawn in VMSMail?  Re: duplicate file names Re: error parsing 'snapshot' Re: FTP to/from Alpha boxes 8 Re: How can I dismount a drive with files still opened ? Re: ibm 18 gig drives  Lexical F$GETJPI Re: Lexical F$GETJPI Re: Lexical F$GETJPI Re: Lexical F$GETJPI Re: Lexical F$GETJPI Re: Lexical F$GETJPI Re: Lexical F$GETJPI Re: Lexical F$GETJPI4 Re: Looking for VMS source code (any version really)' Re: Mac-Decnet via Webster Multiport LT ' Re: Mac-Decnet via Webster Multiport LT ) Re: Minimum Page/Swap File Size in Blocks ) Re: Minimum Page/Swap File Size in Blocks ) Re: Minimum Page/Swap File Size in Blocks ) Re: Minimum Page/Swap File Size in Blocks ) RE: Minimum Page/Swap File Size in Blocks " MOre then one common SYSDUMP file?& Re: MOre then one common SYSDUMP file?& Re: More then one common SYSDUMP file?& Re: MOre then one common SYSDUMP file? new problem with VMS 5.5 New software available for VMS Odp: VMS Pascal question1 Re: OpenVMS loses big, was: RE: Compaq advertises * RE: OVMS V7.2-1: Passwords Case-Sensitive?' Porting assistance for GNU code (GnuGP)  Re: ppp dialup on VAX again  Recall: subscribe & SQL+ODBC options, inexpensive or free? UCX sockets and aux server UCX sockets and aux server UCX sockets and aux server Re: UCX sockets and aux server! Re: VAX to DEC Cobol differences. ! Re: VAX to DEC Cobol differences.  VMS Pascal question  Re: VMS Pascal question  Re: VMS Pascal question  Re: VMS Pascal question  VMS Pascal vs Ansi vs Borland  Re: VMS tuning wsdefault Re: VMS tuning wsdefault Re: What is VMS   F ----------------------------------------------------------------------  % Date: Thu, 13 Jul 2000 09:40:30 +0200 ' From: "Dr. B. Weikl" <weikl@ilt.fhg.de> 4 Subject: Re: %DCL-W-USGOTO, target of GOTO not found2 Message-ID: <8kjrpe$5dp$1@nets3.rz.RWTH-Aachen.DE>  1 Lorin Ricker <Lorin.Ricker@t-netix.com> wrote in: > 418E68E524A8D311ACCE00508B78866A7680A8@exchange.t-netix.com...  I > I've often wished that the -W-USGOTO condition was actually a -E- level E > conditions, which would make trapping it with an ON ERROR condition F > handler possible (I've never had much use for ON WARNING, since thatI > generates too many spurious conditions to worry about).  OTOH, it might G > be nice to extend DCL's ON-handlers to be multi-level, so one *could*  > establish things like: > * >   $ ON WARNING THEN GOTO Warning_Handler& >   $ ON ERROR THEN GOTO Error_Handler& >   $ ON FATAL THEN GOTO Fatal_Handler > C > (or am I missing something obvious, equivalent or better, handler  > style-wise?).   ' I always use an error handler like this % $ on warning then gosub error_handler  ...  $error_handler:  $ status=$status ...   G where error_handler decides whether to abort the procedure or continue. F IMHO you always have to use an error handler with on warning. Consider this little example:  3 $! Procedure to move files from one disk to another " $ on error then goto error_handler $ xopy 'p1 'p2 $ delete/noconfirm 'p1 ...   I Alas, I've mistyped copy. And what does the procedure? It just deletes my  file, " because xopy gives only a warning:F %DCL-W-IVVERB, unrecognized command verb - check validity and spelling> Only with an on warning error handler you will catch this one!F But I have to admit, it didn't help me in the case of a missing endif.   -- Bruno Fraunhofer-ILT mailto:weikl@ilt.fhg.de    ------------------------------  + Date: Thu, 13 Jul 2000 10:57:52 -0500 (CDT) # From: Ron <RROCKWELL@cctr.umkc.edu> + Subject: * Return address problem with MAIL . Message-ID: <01JRPS9BGX9UREE4LM@CCTR.UMKC.EDU>  F   I was under the impression that the system logical mail$system_flagsF was reponsible for handling the return address of local mail among our: cluster. Can anyone point out what might be going on here?  C   Some of our nodes are using our cluster alias, others the cluster D full-name alias, and others *nothing at all*... they give the return address of ::USERNAME.   SYSMAN> set env/clu + %SYSMAN-I-ENV, current command environment: $         Clusterwide on local cluster<         Username RROCKWELL    will be used on nonlocal nodes  # SYSMAN> do sh log mail$system_flags 2 %SYSMAN-I-OUTPUT, command execution on node NODE0A/    "MAIL$SYSTEM_FLAGS" = "7" (LNM$SYSTEM_TABLE) 2 %SYSMAN-I-OUTPUT, command execution on node NODE01/    "MAIL$SYSTEM_FLAGS" = "7" (LNM$SYSTEM_TABLE) 2 %SYSMAN-I-OUTPUT, command execution on node NODE14/    "MAIL$SYSTEM_FLAGS" = "7" (LNM$SYSTEM_TABLE) 2 %SYSMAN-I-OUTPUT, command execution on node NODE03/    "MAIL$SYSTEM_FLAGS" = "7" (LNM$SYSTEM_TABLE) 2 %SYSMAN-I-OUTPUT, command execution on node NODE07/    "MAIL$SYSTEM_FLAGS" = "7" (LNM$SYSTEM_TABLE) 2 %SYSMAN-I-OUTPUT, command execution on node NODE0B/    "MAIL$SYSTEM_FLAGS" = "7" (LNM$SYSTEM_TABLE) 2 %SYSMAN-I-OUTPUT, command execution on node NODE04/    "MAIL$SYSTEM_FLAGS" = "7" (LNM$SYSTEM_TABLE) 2 %SYSMAN-I-OUTPUT, command execution on node NODE05/    "MAIL$SYSTEM_FLAGS" = "7" (LNM$SYSTEM_TABLE) 2 %SYSMAN-I-OUTPUT, command execution on node NODE02/    "MAIL$SYSTEM_FLAGS" = "7" (LNM$SYSTEM_TABLE)    SYSMAN> do mail nl: rrockwell 2 %SYSMAN-I-OUTPUT, command execution on node NODE0A2 %SYSMAN-I-OUTPUT, command execution on node NODE012 %SYSMAN-I-OUTPUT, command execution on node NODE142 %SYSMAN-I-OUTPUT, command execution on node NODE032 %SYSMAN-I-OUTPUT, command execution on node NODE072 %SYSMAN-I-OUTPUT, command execution on node NODE0B2 %SYSMAN-I-OUTPUT, command execution on node NODE042 %SYSMAN-I-OUTPUT, command execution on node NODE052 %SYSMAN-I-OUTPUT, command execution on node NODE02  /     # From                 Date         Subject   &     1 CCTR::RROCKWELL      13-JUL-2000&     2 CCTR::RROCKWELL      13-JUL-2000&     3 CCTR::RROCKWELL      13-JUL-2000&     4 LOCAL.CCTR::RROCKWEL 13-JUL-2000&     5 LOCAL.CCTR::RROCKWEL 13-JUL-2000&     6 LOCAL.CCTR::RROCKWEL 13-JUL-2000&     7                      13-JUL-2000&     8 CCTR::RROCKWELL      13-JUL-2000&     9 CCTR::RROCKWELL      13-JUL-2000&    10                      13-JUL-2000&    11 CCTR::RROCKWELL      13-JUL-2000&    12 LOCAL.CCTR::RROCKWEL 13-JUL-2000&    13 LOCAL.CCTR::RROCKWEL 13-JUL-2000&    14 LOCAL.CCTR::RROCKWEL 13-JUL-2000&    15                      13-JUL-2000&    16 CCTR::RROCKWELL      13-JUL-2000&    17 CCTR::RROCKWELL      13-JUL-2000&    18 CCTR::RROCKWELL      13-JUL-2000       Ron Rockwell   Systems Administrator )   Information Services, Central Computing %   University of Missouri, Kansas City    ------------------------------  # Date: Thu, 13 Jul 2000 08:32:02 GMT % From: "P.Lj" <plj@byron.ext.telia.se> = Subject: Re: ANNOUNCING: FreeCiv 1.11.0 port for OpenVMS Axp. 2 Message-ID: <396D7E40.B4E8F84A@byron.ext.telia.se>  
 Ah! Great!  	 >>> ^P.Lj        Roger Tucker wrote:   1 > Announcing FreeCiv 1.11.0 port for OpenVMS AXP.  > A > (Freeciv is a multiplayer strategy game, released under the GNU ; >  General Public License.  It is generally comparable with , >  Civilization II, published by Microprose. >  See http://www.freeciv.org/)  > 9 > My updates to FreeCiv for the port to OpenVMS have been < > added to the freeciv CVS latest development release and is' > available in the latest CVS snapshot.  > < > Also the latest stable release 1.11.0 (although it doesn't= > include my changes) does compile and run correctly with the ! > changes in the latest snapshot.  > > > You will need to get down the latest CVS snapshot to get the< > three files needed ([.vms]readme.txt, [.vms]vms_build.com, > and [.server]sernet.c) > < > Thanks go to the freeciv team for such a good game, and to; > Colin Blake for the port of Mozilla.  If anyone want's to 1 > take up where I left off, they are welcome too.  > 9 > The following information is from the [.vms]readme.txt:  >  > ------------------------- I > This document includes instructions on how to build Freeciv for OpenVMS = > Alpha.  I used Compaq C V6.2-003 on OpenVMS Alpha V7.2, but ; > other versions of the C compiler and OpenVMS should work.  > : > 1.  Make sure TCP/IP is installed correctly and running. > . > 2.  Download and install the latest Mozilla.! >     I used the M16 release from I >     http://www.openvms.compaq.com/openvms/products/ips/get_mozilla.html 2 >     Follow the release notes there to install it. >     Run it, and make sure it at least works. > $ > 3.  Get the latest version of GLIB7 >     I used 1.9.7 from ftp://ftp.gtk.org/pub/gtk/v1.2/  >     UnBzip2 it, untar it% >     rename the directory to [.glib] ; >     Copy in glibconfig.h_vms from the bottom of this file   >         to [.glib]glibconfig.h9 >     This directory is only needed for the header files. 2 >     (I hope to get glibconfig.h_vms added to the >      next release of glib) > % > 4.  Get the lastest version of GTK+ 7 >     I used 1.2.8 from ftp://ftp.gtk.org/pub/gtk/v1.2/  >     UnBzip2 it, untar it$ >     rename the directory to [.gtk]/ >     This is only needed for the header files.  > % > 5.  Get the latest version of IMLIB 9 >     ftp://ftp.gnome.org/pub/GNOME/stable/sources/imlib/  >         I used 1.9.8 >     UnBzip2 it, untar it& >     rename the directory to [.imlib]? >     copy the imlib_vms_build.com from the bottom of this file " >         to [.imlib]vms_build.com >     set def [.imlib]7 >     Execute the vms_build and make sure imlib builds. > >     (There will be several compiler warnings but I choose to7 >      leave them so maybe someday they will be fixed.) : >     run [.utils]IMLIB_CONFIG.EXE and make sure it works.? >     If you have problems, make sure the logical etc exists or C >     hard-code the correct directory in your sys$login:.imrc file. = >     (I hope to get theses changes added to the next version  >      of IMLIB) > 4 > 6.  Get the latest stable version of Freeciv down.6 >     UnBzip2 it, untar it to the directory [.freeciv]' >         $ bzip2 -d freeciv-1_11_0.bz2 % >         $ tar -xvf FREECIV-1_11_0.; @ >     make sure Freeciv.h is in [.FREECIV-1_11_0.CLIENT.GUI-GTK]6 >         (You can rebuild this with GNV SED and rc2c)% >     $ set def [.FREECIV-1_11_0.vms]  >     $ @vms_build.com> >         You will get a few warnings but theses should be OK.? >     Read the vms_build.com file or the unix install.; file if  >     you have problems  >  > 7.  Todo list: > 0 >     Create a standard VMS binary distribution.A >     Get changes needed into GLIB and IMLIB source distributions D >     Get AUTOCONFIG to work on VMS so that VMS can use the standard >         UNIX make-files.1 >     Get RedHat Linux .RPM files to work on VMS. & >         (Now I'm really dreaming...) >  > ----- A > Most of the problems I've found seem to be in all versions, but B > let me know if you run this, have fixes for the OpenVMS port, or1 > can contribute to make the installation easier.  >  > Roger    ------------------------------   Date: 13 Jul 2000 16:02:46 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)* Subject: Apache for Open VMS, performance?, Message-ID: <8kkp76$1nl@gap.cco.caltech.edu>  B Can somebody at Compaq provide benchmark or other performance testC information for Apache on OpenVMS vs. Apache on Linux (or Tru64) on J identical hardware?   And while you're at it, how about a test against theJ OSU and WASD servers - which is what most OpenVMS customers are using now.  E The bits which most interest me, and presumably anybody serious about  "e-business" are:   !  1.  "raw" html page serving rate T  2.  scripted html page serving rate (speed for pages produced by Perl/java scripts)F  3.  transaction processing rates (max rate of form based input, to a )      database back end and/or flat files) $  4.  all of the above, through https  @ Apache does a lot of IO, but since it is essentially a read onlyI application, there is at least in this instance a significant chance that K the results will be comparable.   If I had to guess though, I'd expect that I the currently feeble file caching that OpenVMS offers is going to cause a J major performance hit when the tests hit a large number of pages in randomJ order, unless this version of Apache implements its own cache, as does the OSU server.    Thanks,    David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech     ------------------------------   Date: 13 Jul 2000 16:37:57 GMT0 From: fdc@watsun.cc.columbia.edu (Frank da Cruz): Subject: Re: C-Kermit on VMS v7.2's latest TCP/IP release?5 Message-ID: <8kkr95$8cs$1@newsmaster.cc.columbia.edu>   I In article <418E68E524A8D311ACCE00508B78866A7680A9@exchange.t-netix.com>, / Lorin Ricker  <Lorin.Ricker@t-netix.com> wrote: > : Does anyone have current experience with the latest C-Kermit< : release (which I presume to be ckv196-axp-vms72-ucx50.exe)? : regarding the *very* latest TCP/IP v5.0(A) under v7.2?  I may ) : just be confused on terminology here...  : B : VMS 7.2 inflicted, for better (mostly) or worse, that new TCP/IP@ : package (the one where all the filenames changed prefixes fromA : "UCX$..." to "TCPIP$..." --- boy, that! broke a few com-files!) A : on us.  I've been calling this "the TCP" package (as opposed to < : "the UCX" package).  However, as noted above, the C-Kermit9 : distribution kit is named with "-ucx50" ... same thing?  : . Yes.  Except it was built under 5.0, not 5.0A.  A : Any compatibility or operational experiences, good or bad, with * : C-Kermit 196/197 and this latest TCP/IP? : E You'll have to ask somebody who has this setup.  The "-ucx50" version @ was sent in by such a person, but I haven't been exposed to this, version myself (I have 4.2 here) (and 2.0!).  E : NB to the Kermit team:  If by "UCX 5.0" you really mean "the latest C : TCP/IP release for VMS", some fine-tuning of nomenclature on your . : web-site would be helpful too.  ;-)  Thanks. : A It's hard to keep up with the constant renaming / repositioning / B buying and selling of the many and varied VMS TCP/IP products (andD VMS itself for that matter) (not to mention ex-DEC's Unix product)).D For sanity, we try to pick a name and stick to it -- at least in theC file naming conventions.  As far as I know, this particular product 1 hasn't been officially called UCX for many years.   E What's the actual name of the product?  The last time I looked it was & "DEC TCP/IP Services for OpenVMS(tm)".   - Frank    ------------------------------  % Date: Thu, 13 Jul 2000 09:11:18 -0400 - From: "Jerry Alan Braga" <jabraga@golden.net> ' Subject: Re: DECnet login w/ ext. auth. , Message-ID: <8kkf61$e99$1@cougar.golden.net>  K how where you able to get external auth to work.  I tried it once but could  not    What were the steps involved,    1) logicals  2) settings  3) command procedures  etc     C "Michael T. Davis" <DAVISM@er6.eng.ohio-state.edu> wrote in message 4 news:8kfl3c$mth$1@charm.magnus.acs.ohio-state.edu... > F > We are running Advanced Server 7.2a under OpenVMS 7.2-1 on an Alpha.J > Most of our users' accounts are configured with external authentication. I E > would like to use a DECnet (Phase IV) object running under a user's 
 account toG > perform various tasks on the account, but authenticating the user via  DECnetF > seems to be a problem.  Say I have a DECnet object defined such that	 access to / > it will require a username and password, i.e.  > ; >                        NODE"username password"::"MYTASK="  > L > Using this syntax with an account that is externally authenticated results in@ > the following error (using a DCL OPEN statement, for example): > J > %DCL-E-OPENIN, error opening NODE"username password"::"MYTASK=" as input$ > -RMS-E-ACC, ACP file access failed> > -SYSTEM-F-INVLOGIN, login information invalid at remote node > J > Is there something we can do to fix this, or is this due to a deficiency in the/ > current implementation of VMS authentication?  > 	 > Thanks,  >  Mikes > --E >              Michael T. Davis              |    Systems Specialist:i ChE,MSEg; >   E-mail: davism@er6.eng.ohio-state.edu    | Departmental1 Networking/ComputingL >            -or- DAVISM+@osu.edu            |     The Ohio State UniversityL > http://www.er6.eng.ohio-state.edu/~davism/ |     197 Watts, (614) 292-6928   ------------------------------    Date: 13 Jul 2000 12:26:20 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>e7 Subject: Re: Default protection for logfiles (VMS 7.1). H Message-ID: <y4u2duqqb7.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>   I presume RMS_DEFPROT applies.  N Patch LOGINOUT.EXE to explicitly change the protection on the log file to what	 you want.a  L Any reason the batch job itself cannot modify the protection of its log file: while running? What good is the G:W supposed to do anyway?   	Jan   ------------------------------  % Date: Thu, 13 Jul 2000 13:19:00 +0100i- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk> 7 Subject: Re: Default protection for logfiles (VMS 7.1).e) Message-ID: <396DB3B4.93155112@bbc.co.uk>e  3 Thanks, I'm a bit embrassed to have missed that buto at least I know now.   Cheers     >? > >  >e > SYSGEN>  HELP PARAM RMS_FILE >  > Parameters >m >   RMS_FILEPROT > G >        RMS_FILEPROT determines the default file protection for systemeI >        processes such as those that create the error log, operator log,EG >        and job controller. It also determines default file protection I >        for processes created by the job controller (all interactive ands >        batch processes). > [...]    --6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.ukb  A I speak for myself only and my views in no way represent those of  MedAS or the BBC.    ------------------------------  % Date: Thu, 13 Jul 2000 10:07:01 -0400w# From: Jim Agnew <agnew@hsc.vcu.edu>e4 Subject: Re: Direct OPCOM messages to e-mail account+ Message-ID: <396DCD05.4F0438B2@hsc.vcu.edu>h  E Hey, it would if you put it into a batch job, write it out to a file, = and .... no.... aw forgetit...  my Ford truck broke down thisDA morning.....  they had a VACUUM hose on the radiator instead of ak  heatproof HEATER hose... bleah..       arturo saavedra wrote: > L > Actually.. scratch that.. that's not gonna work.. :(  i think i've had too > much coffee today..  >  > ''''''''''''''''''* > how about reply/enable=cluster/user=dude > M > and have mail forwarded from dude to your internet mail account .. that waye0 > you get all notification for cluster changes.. >  > absh >  > -----Original Message-----8 > From: Boyle, Darren [mailto:boyledj@bankofbermuda.com]& > Sent: Tuesday, July 11, 2000 1:07 PM > To: Info-VAX@Mvb.Saic.ComE6 > Subject: RE: Direct OPCOM messages to e-mail account > G > > From:         system@SendSpamHere.ORG[SMTP:system@SendSpamHere.ORG]  > >eL > > If there's a cluster transition, how does a PeeCee know that the clusterK > > is transitioning as opposed to someone just tripped over the network or = > > a power cable to something in between PeeCee and cluster?o > >eK > The PC detects the "Blocking Activity" of a quorum hang or the "ProposinghJ > reconfiguration" of a state transition messages displayed on the console  > through the serial connection. > L > > In my experience, and the testament of many posting here, the cluster isL > > likely to significantly over-shadow uptime (by several magnitide) of the > > NT box.L > >VL > Oh Yes, I couldn't agree more, the Heroix services or OMS stations work inL > the same way as the old BRS systems in that if one goes away the secondaryN > one takes the console connection to monitor for such events.  So even thoughH > a stable PC only reboots 20 times every minute 8-) the output is still > caught by the secondary PC.o > 
 > - Darren > H > **********************************************************************E > This message and any files transmitted with it are confidential andyL > may be privileged and/or subject to the provisions of privacy legislation.J > They are intended solely for the use of the individual or entity to whom > theyM > are addressed. If the reader of this message is not the intended recipient,nD > please notify the sender immediately and then delete this message.K > You are notified that reliance on, disclosure of, distribution or copyingr  > of this message is prohibited. >  > Bank of BermudasH > **********************************************************************   ------------------------------  % Date: Thu, 13 Jul 2000 11:44:22 -0400A0 From: arturo saavedra <arturo.saavedra@wcom.com>4 Subject: RE: Direct OPCOM messages to e-mail accountC Message-ID: <NDBBJFNBJJHKADILJHIJMEJFEKAA.arturo.saavedra@wcom.com>x  J hahaha.. put it into a batch job and have another process monitor that log file .. and .. and...  ;)t     -----Original Message-----* From: Jim Agnew [mailto:agnew@hsc.vcu.edu]& Sent: Thursday, July 13, 2000 10:07 AM To: Info-VAX@Mvb.Saic.Com"4 Subject: Re: Direct OPCOM messages to e-mail account    E Hey, it would if you put it into a batch job, write it out to a file, = and .... no.... aw forgetit...  my Ford truck broke down thisSA morning.....  they had a VACUUM hose on the radiator instead of ac  heatproof HEATER hose... bleah..       arturo saavedra wrote: >EL > Actually.. scratch that.. that's not gonna work.. :(  i think i've had too > much coffee today..E >  > ''''''''''''''''''* > how about reply/enable=cluster/user=dude > I > and have mail forwarded from dude to your internet mail account .. that- wayP0 > you get all notification for cluster changes.. >T > absn >x > -----Original Message-----8 > From: Boyle, Darren [mailto:boyledj@bankofbermuda.com]& > Sent: Tuesday, July 11, 2000 1:07 PM > To: Info-VAX@Mvb.Saic.Comc6 > Subject: RE: Direct OPCOM messages to e-mail account >nG > > From:         system@SendSpamHere.ORG[SMTP:system@SendSpamHere.ORG]  > > L > > If there's a cluster transition, how does a PeeCee know that the clusterK > > is transitioning as opposed to someone just tripped over the network or0= > > a power cable to something in between PeeCee and cluster?R > >LK > The PC detects the "Blocking Activity" of a quorum hang or the "Proposing J > reconfiguration" of a state transition messages displayed on the console  > through the serial connection. > L > > In my experience, and the testament of many posting here, the cluster isL > > likely to significantly over-shadow uptime (by several magnitide) of the > > NT box.1 > >-L > Oh Yes, I couldn't agree more, the Heroix services or OMS stations work inL > the same way as the old BRS systems in that if one goes away the secondaryG > one takes the console connection to monitor for such events.  So evenv thoughH > a stable PC only reboots 20 times every minute 8-) the output is still > caught by the secondary PC.1 >u
 > - Darren > H > **********************************************************************E > This message and any files transmitted with it are confidential andnL > may be privileged and/or subject to the provisions of privacy legislation.J > They are intended solely for the use of the individual or entity to whom > theyB > are addressed. If the reader of this message is not the intended
 recipient,D > please notify the sender immediately and then delete this message.K > You are notified that reliance on, disclosure of, distribution or copyingM  > of this message is prohibited. >C > Bank of BermudanH > **********************************************************************   ------------------------------  % Date: Thu, 13 Jul 2000 10:12:25 -0400o# From: Jim Agnew <agnew@hsc.vcu.edu> & Subject: Re: disable spawn in VMSMail?+ Message-ID: <396DCE49.F9BA8473@hsc.vcu.edu>s  < hhmm..  you really need to diddle with the UAF to do that..   @ if your application does NOT spawn subprocesses, you can disable8 subprocessed in the uaf w/o hurting things..  good luck.  " all my solutions involved the UAF.  + also, one could patch the mail image... ;-)l   j.   Lonny Balderston wrote:n > 1 > (previously posted on vmsnet.alpha 22-JUN-2000)- > 5 > Is there an easy way to disable the "spawn" commandn8 > that can be found in in VMSMail? [new: without messing8 > with UAF flags. I have no control over the application7 > we are running, probably cannot change UAF parametersa8 > without causing problems. Just want to disable "spawn"/ > in VMS Mail. Alpha 4100 running VMS 7.1-1H1.]  >  > Thank you, Lonny Balderston   > email: lbalders at gte dot net   ------------------------------  % Date: Thu, 13 Jul 2000 10:56:26 -0400t- From: "Peter Weaver" <peter.weaver@stelco.ca> & Subject: Re: disable spawn in VMSMail?. Message-ID: <smrm4q64nd633@corp.supernews.com>  C If the users can only enter MAIL from a command procedure then makedC sure the command procedure disables TMPMBX. The user can still sendoA mail but can not spawn without TMPMBX. A nice feature for INSTALLyF would be to allow us to say REPLACE SYS$SYSTEM:MAIL.EXE /PRIV=NOTMPMBX< and then invoking MAIL would automatically remove the TMPMBX
 privilege.  	 $DO_MAIL:G! $ Old_priv = f$setprv("NOTMPMBX")/ $ MAIL! $ Privs = f$setprv("''old_priv'")t    : BTW: I just noticed that both F$SETPRV and F$SETPRIV work.   -- Peter Weaver  < Lonny Balderston wrote in message <396CD23A.1A9B@gte.net>...0 >(previously posted on vmsnet.alpha 22-JUN-2000) > 4 >Is there an easy way to disable the "spawn" command7 >that can be found in in VMSMail? [new: without messingl7 >with UAF flags. I have no control over the application 6 >we are running, probably cannot change UAF parameters7 >without causing problems. Just want to disable "spawn" . >in VMS Mail. Alpha 4100 running VMS 7.1-1H1.] >s >Thank you, Lonny Balderston >email: lbalders at gte dot net    ------------------------------  % Date: Thu, 13 Jul 2000 17:06:42 +0100v- From: "POWERS, John" <John.POWERS@sema.co.uk> & Subject: RE: disable spawn in VMSMail?; Message-ID: <D30A62ABC710D211AEE100A0C9D615EE316410@REAES2>h  + There is, unfortunately, a snag with this..I   > -----Original Message-----4 > From: Peter Weaver [mailto:peter.weaver@stelco.ca] > Sent: 13 July 2000 15:56 > To: Info-VAX@Mvb.Saic.Comc( > Subject: Re: disable spawn in VMSMail? >  > E > If the users can only enter MAIL from a command procedure then make E > sure the command procedure disables TMPMBX. The user can still send C > mail but can not spawn without TMPMBX. A nice feature for INSTALLtH > would be to allow us to say REPLACE SYS$SYSTEM:MAIL.EXE /PRIV=NOTMPMBX> > and then invoking MAIL would automatically remove the TMPMBX > privilege. >  > $DO_MAIL:a# > $ Old_priv = f$setprv("NOTMPMBX"). > $ MAIL# > $ Privs = f$setprv("''old_priv'")s >  > < > BTW: I just noticed that both F$SETPRV and F$SETPRIV work. >  > -- > Peter Weaver > > > Lonny Balderston wrote in message <396CD23A.1A9B@gte.net>...2 > >(previously posted on vmsnet.alpha 22-JUN-2000) > >-6 > >Is there an easy way to disable the "spawn" command9 > >that can be found in in VMSMail? [new: without messingo9 > >with UAF flags. I have no control over the applicationl8 > >we are running, probably cannot change UAF parameters9 > >without causing problems. Just want to disable "spawn".0 > >in VMS Mail. Alpha 4100 running VMS 7.1-1H1.] > >n > >Thank you, Lonny Balderston! > >email: lbalders at gte dot net  >  >   G This will achieve the stated aim of disabling spawn from mail, but will < also cripple mail in a rather unfortunate way. See example..   $ mail foo.bar hecate::ins $ set proc /priv=notmpmbxa $ mail foo.bar hecate::ins; %MAIL-E-LOGLINK, error creating network link to node HECATEhG -SYSTEM-F-NOPRIV, insufficient privilege or object protection violationb $n  E Ouch! You cannot send any mail across the network. Mail only works on F your local node. As far as the original querant is concerned, the sad H news is that AFAIK, there is *no* way of doing this without getting yourG hands dirty and fiddling with the UAF (either changing captive flag, orp0 the prclm) - which Lonny said was not an option.  F Sorry, I know this is not very helpful. All I can think of to help is G that you will have to get the assistance of your system manager to messh with SYSUAF.    -- Cheers, John  F  - Note  This message represents my opinions and nothing else, not theI   opinion of SEMA, my family, or the cricket club - though my dog Meg didsE   nod in agreement whilst I was typing. If you have any problems thenvD   please complain to her (or me, but not SEMA, my family or the CC).  ,          K ___________________________________________________________________________-B This email is confidential and intended solely for the use of the H individual to whom it is addressed. Any views or opinions presented are E solely those of the author and do not necessarily represent those of c Sema Group. M If you are not the intended recipient, be advised that you have received thiseI email in error and that any use, dissemination, forwarding, printing, or s- copying of this email is strictly prohibited.v  E If you have received this email in error please notify the Sema Groupd. Helpdesk by telephone on +44 (0) 121 627 5600.K ___________________________________________________________________________    ------------------------------  % Date: Thu, 13 Jul 2000 08:31:01 -0700 * From: "George Shouse" <USENET@Shouses.com>& Subject: Re: disable spawn in VMSMail?, Message-ID: <8kknbk029q6@enews4.newsguy.com>  8 "Peter Weaver" <peter.weaver@stelco.ca> wrote in message( news:smrm4q64nd633@corp.supernews.com...E > If the users can only enter MAIL from a command procedure then make0E > sure the command procedure disables TMPMBX. The user can still send C > mail but can not spawn without TMPMBX. A nice feature for INSTALLsH > would be to allow us to say REPLACE SYS$SYSTEM:MAIL.EXE /PRIV=NOTMPMBX> > and then invoking MAIL would automatically remove the TMPMBX > privilege. >  > $DO_MAIL:B# > $ Old_priv = f$setprv("NOTMPMBX")  > $ MAIL# > $ Privs = f$setprv("''old_priv'")f >V >,< > BTW: I just noticed that both F$SETPRV and F$SETPRIV work.  L F$SETPOPPAONFIRE will work, too.  The significant portion on all lexicals is@ the first 4 characters after F$ (F$xxxx) except for FAO and PID.   --) George Shouse      http://www.shouses.comr) ----------------------------------------- ) Always A Fan of the World Champion Lakers " Retire #99 and Hang the Banners!!!   ------------------------------  % Date: Thu, 13 Jul 2000 13:01:12 -0400i- From: "Peter Weaver" <peter.weaver@stelco.ca>e& Subject: Re: disable spawn in VMSMail?- Message-ID: <smrteqlnd635@corp.supernews.com>O  ! POWERS, John wrote in message ...-, >There is, unfortunately, a snag with this.. >: > ...0 >$ mail foo.bar hecate::ins" >$ set proc /priv=notmpmbx >$ mail foo.bar hecate::ins < >%MAIL-E-LOGLINK, error creating network link to node HECATE> >-SYSTEM-F-NOPRIV, insufficient privilege or object protection	 violation  > ...t  D Right, the only place I ever had the need to disable spawn from MAIL+ we had a cluster so this was not a problem.M   ------------------------------  % Date: Thu, 13 Jul 2000 13:02:48 -0400a- From: "Peter Weaver" <peter.weaver@stelco.ca>n& Subject: Re: disable spawn in VMSMail?. Message-ID: <smrthm67nd674@corp.supernews.com>  B George Shouse wrote in message <8kknbk029q6@enews4.newsguy.com>...9 >"Peter Weaver" <peter.weaver@stelco.ca> wrote in messagee > ...a >>= >> BTW: I just noticed that both F$SETPRV and F$SETPRIV work.t > A >F$SETPOPPAONFIRE will work, too.  The significant portion on allt lexicals is A >the first 4 characters after F$ (F$xxxx) except for FAO and PID.  > ...   3 Right, I didn't have my cup of coffee this morning.s   ------------------------------  # Date: Thu, 13 Jul 2000 17:41:22 GMTh From: byatesiii@my-deja.comL& Subject: Re: disable spawn in VMSMail?) Message-ID: <8kkuvu$oic$1@nnrp1.deja.com>s  - In article <smrteqlnd635@corp.supernews.com>,e0   "Peter Weaver" <peter.weaver@stelco.ca> wrote:# > POWERS, John wrote in message ...s. > >There is, unfortunately, a snag with this.. > >e > > ...i > >$ mail foo.bar hecate::insm > >$ set proc /priv=notmpmbx > >$ mail foo.bar hecate::ins > > >%MAIL-E-LOGLINK, error creating network link to node HECATE@ > >-SYSTEM-F-NOPRIV, insufficient privilege or object protection > violationa > > ...r >gF > Right, the only place I ever had the need to disable spawn from MAIL- > we had a cluster so this was not a problem.t >d >dC In any event, this would work for us. There would be no instance in-F which our users would send mail across the network to another cluster. So, thanks for the tip!uG Now, if someone could explain prclm to me, and why prclm=1 allows about & 5 processes, 2 allows about 10, etc...    & Sent via Deja.com http://www.deja.com/ Before you buy.l   ------------------------------  % Date: Thu, 13 Jul 2000 14:44:17 +0100i  From: steven.reece@quintiles.com! Subject: Re: duplicate file namesf> Message-ID: <8025691B.004B9E24.00@qedilc01.qedi.quintiles.com>   John,pM I have seen this _exact_ behaviour on a pair of AlphaServers with shared SCSIuI disks hanging off of a dual redundant pair of HSZ70s.  This was under VMSj 7.1-1H2.  P IIRC the file in question was a system log, although I can't remember from which	 facility.s   Steve.    9 John Nixon (jorlnixon at worldnet dot att dot net) wrote:hK >>>When I looked at the original disk directory, I saw several files of the  EXACT name,RH right down to the version number. They had different file-ids, different cre/mod dates,G they were different sizes and had different data.  These were log files  being kept aroundeD for statistical purposes.  (Thank god they are not Production data)!  C If I do a $DIR   filename.dat;1    I only see one file, but if I dooG $DIR filename.dat;* it lists both (or all three or four) files with theP
 same name.  @ In order to look at the older one, I have to COPY FILENAME.DAT;*E to   TMP.DAT;0,   which forces it to copy both files to TMP.DAT;1 and  TMP.DAT;2 etc...  I I can't imagine RMS allowing this to happen, but I can't think of any way- any of= our applications could be creating files while bypassing RMS.v  L Any Ideas?  Has anyone ever saw this before?   If I am not the only one thatA has seen this I think I would have a better chance of getting VMSb engineeringt involved.<<<   ------------------------------  % Date: Thu, 13 Jul 2000 15:25:40 +0100n* From: "Peter Allingham" <Peter@nospam.org>% Subject: Re: error parsing 'snapshot's) Message-ID: <8kkjgl$cr3$1@soap.pipex.net>   F We had this problem a few months ago. Our solution was manually to addF BACKUP.CLD from the ALPBACK04_071 kit to DCLTABLES and then reinstall.  4 Looks like a 'feature' in the Kitinstal.com routine.   Peter Allinghamu GAP Ltdt
 www.gap.co.uk.          J "Wayne Sewell" <wayne@tachysoft.xxx.560753.killspam.00bb> wrote in message( news:LxknSygwplyE@tachxxsoftxxconsult...D > In article <009ECE9F.4FE10774.6@sternwarte.uni-erlangen.de>, Horst2 Drechsel <ai05@sternwarte.uni-erlangen.de> writes:) > >>Date: Mon, 10 Jul 2000 10:20:03 -0300 5 > >>From: "Boyle, Darren" <boyledj@bankofbermuda.com>@) > >>Subject: RE: error parsing 'snapshot'c > >>2 > >>      Maybe it's just my ignorance but what is* > >>  /SYS$BACK:[000000]LU12_00-07-07.LU -F > >> Not a VMS qualifier to my knowledge, this may explain why you are	 getting ae2 > >> CLI error.  As for batch jobs not running ??.
 > >> - Darren  > >>>h0 > >>>    BACKUP/IMAGE/NOASSIST/BLOCK=65000/LOG -- > >>>    /SYS$BACK:[000000]LU12_00-07-07.LU -e > >>>     DISK1:? TEMP:LU12_00-07-07.BCK/SAVE/LABEL=LUCKY/REWIND/IGNORE=INTERLOCKm > >>>t- > >>> %CLI-F-SYNTAX, error parsing 'SNAPSHOT'c@ > >>> -CLI-E-ENTNF, specified entity not found in command tables > >>>V > >h >s >  >lB > This looks like some sort of version mismatch between backup and
 dcltables.K > The snapshot qualifier was introduced in 7.1-2 and is also present in 7.2o and L > later.  Some patches to 7.1 backup also include it.  It appears as if yourI > backup image is trying to find the snapshot qualifier, but the originalr 7.1oI > version of dcltables is in effect and original 7.1 does not define this*J > qualifier.   The error message is telling you *exactly* what the problem is:tC > the snapshot qualifier is not present in the version of dcltablesv currently in > effect for this process. >f >t >u >o > > -------------------------- > >    Dear Darren,r > >.K > >    thanks for your reply. I'm sorry, I omitted something when includingf the H > > above error message in my InfoVAX posting. The actual command reads: > >n/ > >>    BACKUP/IMAGE/NOASSIST/BLOCK=65000/LOG - 0 > >>    /LIS=SYS$BACK:[000000]LU12_00-07-07.LU - > >>     DISK1:*? TEMP:LU12_00-07-07.BCK/SAVE/LABEL=LUCKY/REWIND/IGNORE=INTERLOCK- > >> > >-L > >    i.e. it's just the /LIS= qualifier which was missing (SYS$BACKUP is aK > > logical defined as abbreviation of the path pointing to our backup listaJ > > file directory). That's ok and has always been running fine, and is by no" > > means reason of the CLI error. > >  > >>L > >>    This problem occurs since we applied ECO  ALPSYS20_071 (well, that'sK > >> my guess, because since  a b o u t  that time these errors occur, I am,E > >> not 100% sure that our problem is related to this system patch).e > >> > I > If it includes backup.exe and/or backupshr.exe, it probably is.  If theoJ > snapshot qualifier was introduced in the patch, you need to do something toJ > pick up the new dcltables, such as logging in again before trying to useH > backup.  The patch probably suggested a reboot after installation.  If this had1 > been done, you would never have seen the error.o >  >h > >eK > >    The actual reason is still not clear. I could meanwhile recover fromuD > > this behaviour, but don't know which of these steps was actually > > responsible: > >o > >    1) run AUTOGENfF > >    2) installed ECO ALPF11X04_071 (recommended in combination with	 ALPSYS20)DB > >    3) deactivated JAVA JDK 1.1.8-5 kit (i.e. @JAVA$STARTUP and @JAVA$SETUP 0 > >       no more invoked during system startup) > >    4) booted serverl > >u > K > If it works now, I imagine the reboot brought dcltables in synch with the J > backup images.  It is probable that logging off and back on before using backup > would have done it too.o >  > >mE > >    We run into this situation, because we wanted to install Adobei Acrobat G > > Viewer for VMS, which needs the JAVA development kit, which in turn  couldhE > > only be installed after a couple of system ECOs, one of which wase	 ALPSYS20.  > >e> > >    I personally suspect that JAVA was causing the problem. > >  > 
 > I doubt it.f >n >o > -- >eL ============================================================================ ===i: > Wayne Sewell, Tachyon Software Consulting  (281)812-0738 wayne@tachysoft.xxxc: > http://www.tachysoft.xxx/www/tachyon.html and wayne.htmlH > change .xxx to .com in addresses above, assuming you are not a spambot :-)n > L ============================================================================ ===rL > Otter, on dining with Bluto:"It's perfectly safe if you keep your arms and legs > away from his mouth."*   ------------------------------  # Date: Thu, 13 Jul 2000 17:09:57 GMTf+ From: "Andy Proctor" <aproctor@hotmail.com>n$ Subject: Re: FTP to/from Alpha boxes4 Message-ID: <963508267.148540@talita.eclipse.net.uk>   Jesper,e  G Many Many thanks for the Help so far. I have obtained the data - pasted.L below. It would seem that they are not running the file UCX$FTPD_STARTUP.COM( as no FTPD log files are created..right?  G From my small knowledge, would the ACL entries on the UCX$FTP directoryi cause this?M  A As always, much appreciation to all those who have helped so far,*J particularly Jesper. The good thing is I'm learning loads, so I won't have to ask these questions again!h   Best regards   Andy Proctor  7 Jesper Naur <jesper.naur@post.tele.dk> wrote in messages& news:8kiqsm$6ca$1@news.inet.tele.dk...# > We shall need from your customer:  >l > Output from: >i > $ ucx sh service ftp/fullf   Service: FTP-                            State:     Enabled G Port:               21     Protocol:  TCP             Address:  0.0.0.0oH Inactivity:          5     User_name: UCX$FTP         Process:  UCX$FTPDC Limit:              10     Active:      0             Peak:       0a  9 File:         SYS$SYSDEVICE:[UCX$FTP]UCX$FTPD_STARTUP.COMa Flags:        None   Socket Opts:  Rcheck Schecke0  Receive:            0     Send:               0  K Log Opts:     Acpt Actv Dactv Conn Error Exit Logi Logo Mdfy Rjct TimO Addrt9  File:        SYS$SYSDEVICE:[UCX$FTP]UCX$FTPD_STARTUP.LOGw   Security  Reject msg:  not definedM  Accept host: 0.0.0.0n  Accept netw: 0.0.0.0   $ > $ dir/full sys$system:ucx$ftpd.exe   Directory SYS$COMMON:[SYSEXE]a  2 UCX$FTPD.EXE;1                File ID:  (7659,2,0)0 Size:          220/222        Owner:    [SYSTEM]" Created:   18-APR-1995 13:59:01.63( Revised:   17-MAY-2000 11:02:29.71 (100) Expires:   <None specified>o Backup:    <No backup recorded>  Effective: <None specified>s Recording: <None specified>i File organization:  Sequential Shelved state:      OnlineJ File attributes:    Allocation: 222, Extend: 0, Global buffer count: 0, No" version limit, Contiguous best try1 Record format:      Fixed length 512 byte records  Record attributes:  None RMS attributes:     None Journaling enabled: NoneA File protection:    System:RWED, Owner:RWED, Group:RWED, World:RET Access Cntrl List:  None    Total of 1 file, 220/222 blocks.    8 > $ dir/full sys$sysdevice:[ucx$ftp]ucx$ftpd_startup.com  ! Directory SYS$SYSDEVICE:[UCX$FTP]n  > UCX$FTPD_STARTUP.COM;1                    File ID:  (6242,8,0)9 Size:            2/3          Owner:    [UCX$AUX,UCX$FTP]o" Created:   17-MAY-2000 11:07:26.46& Revised:   17-MAY-2000 11:07:26.68 (3) Expires:   <None specified>p Backup:    <No backup recorded>a Effective: <None specified>0 Recording: <None specified>T File organization:  Sequential Shelved state:      OnlineH File attributes:    Allocation: 3, Extend: 0, Global buffer count: 0, No
 version limitRF Record format:      Variable length, maximum 0 bytes, longest 66 bytes4 Record attributes:  Carriage return carriage control RMS attributes:     None Journaling enabled: None? File protection:    System:RWED, Owner:RWED, Group:RE, World:REl Access Cntrl List:  None   Total of 1 file, 2/3 blocks.  . > $ dir/full sys$sysdevice:[000000]ucx$ftp.dir      Directory SYS$SYSDEVICE:[000000]  3 UCX$FTP.DIR;1                 File ID:  (6241,10,0)m9 Size:            3/6          Owner:    [UCX$AUX,UCX$FTP]r" Created:   17-MAY-2000 11:07:26.23& Revised:   17-MAY-2000 11:07:26.33 (1) Expires:   <None specified>o Backup:    <No backup recorded>I Effective: <None specified>h Recording: <None specified>t File organization:  Sequential Shelved state:      OnlineH File attributes:    Allocation: 6, Extend: 0, Global buffer count: 0, No4 default version limit, Contiguous, MoveFile disabled"                     Directory fileI Record format:      Variable length, maximum 512 bytes, longest 512 bytesn4 Record attributes:  No carriage control, Non-spanned RMS attributes:     None Journaling enabled: None< File protection:    System:RWE, Owner:RWE, Group:RE, World:E Access Cntrl List:L (UNKNOWN=%X82,SIZE=%D168,FLAGS=%X0400,ACCESS=%X06900000,DATA=%X00000003,%X33 9D62DE,%X000002B7,%X00010000,s  L %X00000000,%X00000000,%X339D62DE,%X00000175,%X00020000,%X00000000,%X00000005" ,%X339D6DB9,%X0000030D,%X00010000,  L %X00000000,%X00000006,%X339D6E90,%X00000232,%X00010000,%X00000000,%X00000004" ,%X339D62DE,%X000002F9,%X00010000,  L %X00000000,%X00000001,%X339D62DE,%X000001C3,%X00020001,%X00000000,%X00000002" ,%X339D62DE,%X00000210,%X00020002,  B %X00000000,%X00000000,%X00000000,%X00000000,%X00040000,%X00000000)   Total of 1 file, 3/6 blocks. >w! > Listing of the UCX$FTP account:- >R > $ set def sys$system > $ mc authorize > UAF> SH UCX$FTPn  # username: UCX$FTP    Owner: UCX$FTPs4 Account: UCX$FTP    UIC: [375,1] ([UCX$AUX,UCX$FTP]) CLI: DCL     Tables: DCLTABLES  Default: SYS$SYSDEVICE:[UCX$FTP]
 LGICMD: LOGINs Flags: Restrictedt! Primary days: Mon Tue Wed Thu Frit Secondary days:    Sat Sun) Primary: "0000..."   Secondary: "0000..."l+ Day hours: "0123..."   Secondary: "0123..."o7 Network: ##### FULL ACCES #####  ##### FULL ACCES #####t$ Batch:   No access         No access$ Local:   No access         No access$ Dialup:  No access         No access$ Remote:  No access         No access2 Expiration: (None)  Pwd minimum: 6  Login fails: 0/ Pwdlifetime: 90 00:00 Pwd change: (pre-expired)mE Last login: (None) (interactive), 12-JUL-2000 14:15 (non-interactive) ' Max jobs: 0 Fillm:   50  Bytlm:  108000(( Max acct jobs: 0 Shrfillm: 0  Pbytlm:  0) Max de tach: 0 BIOlm:   18  JTquota: 4096_! Prclm:  8 DIOlm:  18  Wsdef:  350_! Prio:  8 ASTlm:  100  Wsquo:  512n% Queprio: 8 TQElm:  15  Wssextent: 512 ( CPU:  (None) Englm:  100  Pollquo: 10240 Authorized Privilegesr
 NETMBX TMPMBXo Default Privileges
 NETMBX TMPMBX    >s. > If possible, instruct the customer to insert >e >  $ SET VERIFY  >iB > at the beginning of sys$sysdevice:[ucx$ftp]ucx$ftpd_startup.com,A > and provide the contents of ucx$ftpd_startup.log resulting fromt# > an FTP attempt after this change.   G After "SET VERIFY" at the beginning of the file "ftpd_startup.com", ftppI attempt from PC or self attempt from Alpha gives no UCX$FTPD.LOG file but_. only UCX$FTPD_STARTUP.LOG file. This contains:  H  %SYSTEM-F-NOPRIV, insufficient privilege or object protection violation8   UCX$FTP      job terminated at 13-JUL-2000 14:20:30.82     Accounting information:gK   Buffered I/O count:              73         Peak working set size:   2672.K   Direct I/O count:                 7         Peak page file size:    38928sK   Page faults:                    459         Mounted volumes:            0pA   Charged CPU time:           0 00:00:00.82   Elapsed time:     0n 00:00:00.95o     >BC > Also, see if any ucx$ftpd.log files are created, and include them 
 > as well. >.  & No new ucx$ftpd.log files are created.   ------------------------------  # Date: Thu, 13 Jul 2000 12:40:55 GMT  From: hanghien@my-deja.comA Subject: Re: How can I dismount a drive with files still opened ? ) Message-ID: <8kkdcj$a4u$1@nnrp1.deja.com>F   Hi guys,  4 Thanks all of you who have responded to my question.F What I found was a guy on our system who used the LD utility to create@ a container file, and renamed it to .EXE file to do his test andG connected and mounted it to the lda1: logical disk. I dismounted lda1:,lB disconnected the file, and was able to dismount dka200: afterward.   Regards, John    0 > If this is a known file (INSTALLed image, then. > $INSTALL REMOVE USWWVM$DKA200:[000000]MMINSTA > may do it, unless some application is actually using that file,r> > in which case the application will have to close it or exit.= > You may want to discover how it was INSTALLed and with whata> > qualifiers first, so you can fix the application when you're( > finished doing whatever with the disk.  $ MMINST.EXE isn't an installed image.     >> Process name PID File name   >> 00000000 [000000]INDEXF.SYS;1  >> 00000000 [000000]MMINST.EXE;1  G >The first method we used was to do a sh dev/fi before dismounting, andl? stop proc on all id's accessing the >device, before dismounting1  = I stopped the proc with "stop/id=000000" that logged me off .u    & Sent via Deja.com http://www.deja.com/ Before you buy.6   ------------------------------  / Date: Thu, 13 Jul 2000 09:03:14 +0200 (MET DST)c& From: Rudolf Wingert <win@fom.fgan.de> Subject: Re: ibm 18 gig drives6 Message-ID: <200007130703.JAA01240@sinet1.fom.fgan.de>   Hello,  J sorry, but I do have expirience with the DNES disk only with OpenVMS priorJ V7.1-2 (with patch VMS712_UPDATE_V300). Under this versions, the DNES does> not work with this versions. Under OpenVMS V7.1-2 (with patch H VMS712_UPDATE_V300) I have tested the following IBM disk drives with theF KZPCM controller: DRHS, DMVS and the newest one DDYS. All this disk doH work with DS20 and AlphaServer 1000A. With AlphaStation 200 I don't knowG at this moment. Which firmware version does the AlphaStation have (best  would be V5.5)?o   Regards Rudolf Wingert   ------------------------------  # Date: Thu, 13 Jul 2000 10:50:11 GMTa From: guyab@my-deja.comp Subject: Lexical F$GETJPIi) Message-ID: <8kk6sv$5ie$1@nnrp1.deja.com>   8 I seem to be having an unlikely situation on my cluster.  / Hardware - 2 x Alpha 2100 with dual 4/233 896MB $ O/S      - VMS 6.2-1h3 with full PVS Problemc When I issue the command* $ write sys$output f$getjpi("","username")< I get my username displayed as expected as I do for UIC etc. When I issue the command' $ write sys$output f$getjpi("","owner")e I only get a blank line.E I have got all of the UAF records set up with meaningful names in thex< owner field and I would like to extract it a single command. Anybody any ideas.   -------------- Guy Barrettt TNT Post Group Note# These are my personal opinions only-    & Sent via Deja.com http://www.deja.com/ Before you buy.t   ------------------------------  % Date: Thu, 13 Jul 2000 13:26:56 +02008% From: "Fred Zwarts" <F.Zwarts@KVI.nl>i Subject: Re: Lexical F$GETJPI . Message-ID: <8kk921$m5k$1@info.service.rug.nl>  H <guyab@my-deja.com> wrote in message news:8kk6sv$5ie$1@nnrp1.deja.com...: > I seem to be having an unlikely situation on my cluster. >=201 > Hardware - 2 x Alpha 2100 with dual 4/233 896MBi& > O/S      - VMS 6.2-1h3 with full PVS	 > Problemd > When I issue the command, > $ write sys$output f$getjpi("","username")> > I get my username displayed as expected as I do for UIC etc. > When I issue the command) > $ write sys$output f$getjpi("","owner"), > I only get a blank line.G > I have got all of the UAF records set up with meaningful names in thet> > owner field and I would like to extract it a single command. > Anybody any ideas.  & I don't know what you expected to see.F According to the documentation "owner" returns the PID of the parent = process.= A normal interactive process has no parent. A subprocess has. 
 E.g., try:   $ SPAWNP' $ write sys$output f$getjpi("","owner")l $ logout   ------------------------------  % Date: Thu, 13 Jul 2000 13:29:26 +0200,% From: "Fred Zwarts" <F.Zwarts@KVI.nl>  Subject: Re: Lexical F$GETJPIi. Message-ID: <8kk96n$m96$1@info.service.rug.nl>  H <guyab@my-deja.com> wrote in message news:8kk6sv$5ie$1@nnrp1.deja.com...: > I seem to be having an unlikely situation on my cluster. >=201 > Hardware - 2 x Alpha 2100 with dual 4/233 896MB & > O/S      - VMS 6.2-1h3 with full PVS	 > Problemt > When I issue the command, > $ write sys$output f$getjpi("","username")> > I get my username displayed as expected as I do for UIC etc. > When I issue the command) > $ write sys$output f$getjpi("","owner")h > I only get a blank line.G > I have got all of the UAF records set up with meaningful names in theo> > owner field and I would like to extract it a single command. > Anybody any ideas.  1 Maybe you want to use the system service $GETUAI. B Unfortunately, there is no DCL lexical function equivalent for it.I You could write a small program to extract the information and store it =r in a DCL symbol.   ------------------------------  # Date: Thu, 13 Jul 2000 12:56:50 GMTb% From: Berty <Guy.Barrett@tntpost.com>: Subject: Re: Lexical F$GETJPI ) Message-ID: <8kkeaa$atj$1@nnrp1.deja.com>e  . In article <8kk921$m5k$1@info.service.rug.nl>,(   "Fred Zwarts" <F.Zwarts@KVI.nl> wrote:& > <guyab@my-deja.com> wrote in message# news:8kk6sv$5ie$1@nnrp1.deja.com...c< > > I seem to be having an unlikely situation on my cluster. > >=203 > > Hardware - 2 x Alpha 2100 with dual 4/233 896MB0( > > O/S      - VMS 6.2-1h3 with full PVS > > Problem- > > When I issue the command. > > $ write sys$output f$getjpi("","username")@ > > I get my username displayed as expected as I do for UIC etc. > > When I issue the command+ > > $ write sys$output f$getjpi("","owner")e > > I only get a blank line.E > > I have got all of the UAF records set up with meaningful names in7 thea@ > > owner field and I would like to extract it a single command. > > Anybody any ideas. >p( > I don't know what you expected to see.H > According to the documentation "owner" returns the PID of the parent =
 > process.? > A normal interactive process has no parent. A subprocess has.e > E.g., try: > 	 > $ SPAWN ) > $ write sys$output f$getjpi("","owner")p
 > $ logout >i >o   --8 What I am trying to get is the owner string from my UAF. The UAF shows -   > Username: BILLB                            Owner:  BILL BLOGGS? Account:  AHE                              UIC:    [100,100242]t
 ([LGG,BILLB])d< CLI:      DCL                              Tables: DCLTABLES   So I was thinking that if % f$getjpi("","username") returns BILLB0 and ' f$getjpi("","UIC") returns [100,100242]- and." f$getjpi("","account") returns AHE  A then sureley I should be able to extract the owner field as well. D Sounds like I need the gospel according to Hoff to explain this one.) Yours slightly confused and a bit miffed.t0 But I refuse to RTFM. I'm a techie and we don't.2 -------------------------------------------------- Guy Barretto Infrastructure Team Leader TNT Post Group2 Note. This post only contains my personal opinion.    & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  % Date: Thu, 13 Jul 2000 15:20:37 +0100 - From: Tim Llewellyn <tim.llewellyn@bbc.co.uk>> Subject: Re: Lexical F$GETJPIx) Message-ID: <396DD035.AD89D69B@bbc.co.uk>    Berty wrote:   >i: > What I am trying to get is the owner string from my UAF. > The UAF shows -e >c@ > Username: BILLB                            Owner:  BILL BLOGGSA > Account:  AHE                              UIC:    [100,100242]s > ([LGG,BILLB]),> > CLI:      DCL                              Tables: DCLTABLES >t > So I was thinking that if.' > f$getjpi("","username") returns BILLBs > ando) > f$getjpi("","UIC") returns [100,100242]  > andy$ > f$getjpi("","account") returns AHE >UC > then sureley I should be able to extract the owner field as well.t  L Well, I was doing this just yesterday. Hacked up a real nasty report parsing  8 the output of UAF LIS/FULL with F$ELEMENT, F$EXTRACT andL friends. Could't find another way quickly if there is one. Only a report for ae) bean counter's, so I'm not that bothered.   H You can read the UAF a record at a time I guess, but that is unsupported like what I did is.e   > F > Sounds like I need the gospel according to Hoff to explain this one.+ > Yours slightly confused and a bit miffed.V2 > But I refuse to RTFM. I'm a techie and we don't.  = How can you call youself a techie if you DON'T RTFM (at least,< occasionally). How do you handle system upgrades etc. Do you> have that telepathic link to Compaq support upgrade installed?    --o6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.ukx  A I speak for myself only and my views in no way represent those ofn MedAS or the BBC.a   ------------------------------  % Date: Thu, 13 Jul 2000 07:03:37 -0700=1 From: Dave C. <DCantorNODCSPAM@shore.net.invalid>= Subject: Re: Lexical F$GETJPIg9 Message-ID: <0a2d0c2c.26ba17f7@usw-ex0108-062.remarq.com>   / In article <8kkeaa$atj$1@nnrp1.deja.com>, Berty-  <Guy.Barrett@tntpost.com> wrote: .. > So I was thinking that ifr' > f$getjpi("","username") returns BILLBR > andt) > f$getjpi("","UIC") returns [100,100242]1 > anda$ > f$getjpi("","account") returns AHE4 > then sureley I should be able to extract the owner > field as well.  6 Yes, you should, and you will.  But OWNER of a process; is not a UIC; it is the PID of its parent process, if there  is one.  Try this:   $ SPAWNd( %DCL-S-SPAWNED, process CANTOR_1 spawned: %DCL-S-ATTACHED, terminal now attached to process CANTOR_1' $ WRITE SYS$OUTPUT F$GETJPI("","OWNER")s 0002320E $ LOGOUT8   Process CANTOR_1 logged out at 13-JUL-2000 10:17:18.828 %DCL-S-RETURNED, control returned to process Dave Cantor   Dave C.a    4 > Sounds like I need the gospel according to Hoff to > explain this one.e+ > Yours slightly confused and a bit miffed.t2 > But I refuse to RTFM. I'm a techie and we don't.4 > --------------------------------------------------
 > Guy Barrett  > Infrastructure Team Leader > TNT Post Group4 > Note. This post only contains my personal opinion.( > Sent via Deja.com http://www.deja.com/ > Before you buy.v       * Sent from AltaVista http://www.altavista.com Where you can also find related Web Pages, Images, Audios, Videos, News, and Shopping.  Smart is Beautiful    ------------------------------  # Date: Thu, 13 Jul 2000 15:43:21 GMTo% From: Berty <Guy.Barrett@tntpost.com>A Subject: Re: Lexical F$GETJPIC) Message-ID: <8kko2k$inb$1@nnrp1.deja.com>y  ) In article <396DD035.AD89D69B@bbc.co.uk>,.    tim.llewellyn@bbc.co.uk wrote: >  >/ > Berty wrote: >$ > >e< > > What I am trying to get is the owner string from my UAF. > > The UAF shows -5 > >)B > > Username: BILLB                            Owner:  BILL BLOGGSC > > Account:  AHE                              UIC:    [100,100242]: > > ([LGG,BILLB])o@ > > CLI:      DCL                              Tables: DCLTABLES > >d > > So I was thinking that ifg) > > f$getjpi("","username") returns BILLB  > > andi+ > > f$getjpi("","UIC") returns [100,100242]  > > andf& > > f$getjpi("","account") returns AHE > >eE > > then sureley I should be able to extract the owner field as well.i >eF > Well, I was doing this just yesterday. Hacked up a real nasty report parsing  >t: > the output of UAF LIS/FULL with F$ELEMENT, F$EXTRACT andC > friends. Could't find another way quickly if there is one. Only as
 report for > ac+ > bean counter's, so I'm not that bothered.: >X> > You can read the UAF a record at a time I guess, but that is unsupportede > like what I did is.  >w > > H > > Sounds like I need the gospel according to Hoff to explain this one.- > > Yours slightly confused and a bit miffed.i4 > > But I refuse to RTFM. I'm a techie and we don't. >s? > How can you call youself a techie if you DON'T RTFM (at leastu> > occasionally). How do you handle system upgrades etc. Do you@ > have that telepathic link to Compaq support upgrade installed? >f >  -- 8 > Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project2 > MedAS at the BBC, Whiteladies Road, Bristol, UK.C > Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uk  >yC > I speak for myself only and my views in no way represent those of  > MedAS or the BBC.3 >o >.   --2 I have managed to produce a work round as follows.   $ uaf list/brief [10,*]0, $ convert/fdl=suppuaf sysuaf.lis suppuaf.dat  H This produces an small indexed file of the uaf's that I'm interested in.D In the command procedure WELCOME.COM I have then added the following lines.  * $       username = f$getjpi("","USERNAME")* $       name = f$edit(username,"collapse") $               !e: $       open/read/share inf disk$quorum:[admin]suppuaf.dat# $       read inf/key='name realnamei $       close infl+ $       realname = f$extract(0,12,realname)i  F This has the effect of extracting the owner field that I am interested in.rG What I was trying to do however, was to get this information from a oneaE line command. Producing indexed files and 4 lines of code seem a long C winded way when we all know there is usually an easier way of doingG things.W2 -------------------------------------------------- Guy Barrett, Infrastructure Team Leader TNT Post Group2 Note. This post only contains my personal opinion.8 PS I do read occasionally, but don't tell my users that.    & Sent via Deja.com http://www.deja.com/ Before you buy.3   ------------------------------  % Date: Thu, 13 Jul 2000 13:20:29 -0400%- From: "Peter Weaver" <peter.weaver@stelco.ca>% Subject: Re: Lexical F$GETJPI,. Message-ID: <smruisiend654@corp.supernews.com>  7 Berty wrote in message <8kko2k$inb$1@nnrp1.deja.com>...0 > ...) > + >$       username = f$getjpi("","USERNAME")f+ >$       name = f$edit(username,"collapse")y >$               !; >$       open/read/share inf disk$quorum:[admin]suppuaf.dats$ >$       read inf/key='name realname >$       close inf, >$       realname = f$extract(0,12,realname) >I< >This has the effect of extracting the owner field that I am
 interested >in.D >What I was trying to do however, was to get this information from a onelF >line command. Producing indexed files and 4 lines of code seem a longD >winded way when we all know there is usually an easier way of doing >things.   > ...s    D Why not install GETUAI (off the freeware CD I think, but I have been2 wrong at least twice today)and do it in two lines;  5 $ getuai /owner=realname "''f$getjpi("","USERNAME")'"p $ sh sym realnamel   REALNAME = "Peter Weaver"2   ------------------------------    Date: 13 Jul 2000 11:49:32 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> = Subject: Re: Looking for VMS source code (any version really)8H Message-ID: <y4zonmqs0j.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  ? wayne@tachysoft.xxx.560753.killspam.00bb (Wayne Sewell) writes:T  M > What you think makes sense for a license is not relevant, especially if youtL > are basic on a different licensing model.  What is relevant are the actualM > terms of the license.  If it says "no third party transfer", that's exactlylB > what it means, whether you are talking about a sale or a loan.    K Well yes, that's suppposed to be the state of the art. OTOH, a German courtPM recently decided that once you (Microsoft, in this case) have sold a piece ofRD software, you cannot invoke copyright to restrict resale. It will be* interesting to see what comes out of this.   	Jan   ------------------------------  % Date: Thu, 13 Jul 2000 17:53:05 +0930uA From: "Geoff Roberts" <geoffrobx@stmarksx.ppx.catholicx.edux.aux>20 Subject: Re: Mac-Decnet via Webster Multiport LT3 Message-ID: <W%eb5.34718$N4.1349649@ozemail.com.au>   - "Hans Vlems" <hvlems@iae.nl> wrote in message " news:8khp9h$6l4$1@news.IAEhv.nl...@ > You need DECnet for the MAC. There used to be a product in the	 Pathworksa > familyB > that supported the MAC. I believe it is no longer available, but perhaps. > some kind  > soul can give you the kit.  C Hmm, don't know what version of pathworks/when it was around by any- chance?   E > There may be third party vendors that still sell a DECnet stack forg the MAC.  ( Possibly.  I'll see what I can dig up...  H > Personally I prefer DECnet, but then again the rest of the world seems to2 > turn  to IP, which may be the way to go for you.  ; Yeah, I can do it with IP if necessary.  Just a thought....c   Cheers  
 Geoff Robertsd Computer Systems Manager Saint Mark's College Port Pirie,, South Australiad6 geoffrob at stmarks dot pp dot catholic dot edu dot au ICQ: 1970476   ------------------------------  % Date: Thu, 13 Jul 2000 17:53:49 +0930TA From: "Geoff Roberts" <geoffrobx@stmarksx.ppx.catholicx.edux.aux>r0 Subject: Re: Mac-Decnet via Webster Multiport LT3 Message-ID: <C0fb5.34720$N4.1349662@ozemail.com.au>   = "Zane H. Healy" <healyzh@shell1.aracnet.com> wrote in messagea0 news:v69b5.453$sp6.146054@typhoon.aracnet.com...  G > There *used* to be DECnet for the Mac available from the company that  makeseF > DAVE (SMB for the Mac).  A portion of that product is available with the H > eXodus X-Windows software for the Mac, but I think all you can do with it iso# > use it as an X-Windows transport.b  8 Hmm, could be interesting.  I'll look into it, thanks...   Cheers  
 Geoff Roberts  Computer Systems Manager Saint Mark's College Port Pirie,  South Australia-6 geoffrob at stmarks dot pp dot catholic dot edu dot au ICQ: 1970476   ------------------------------   Date: 13 Jul 2000 03:53 CSTs' From: carl@gerg.tamu.edu (Carl Perkins)t2 Subject: Re: Minimum Page/Swap File Size in Blocks- Message-ID: <13JUL200003530671@gerg.tamu.edu>i  5 "Boyle, Darren" <boyledj@bankofbermuda.com> writes...rE }	What's the absolute minimum file size in blocks of the Primary page G }& swap files for OpenVMS 7.1 (VAX) and 7.1-2 (Alpha).  The systems areiM }configures with 448Mb & 4096Mb respectively.  Obviously this should still besH }big enough to accommodate the start-up process  until SYPAGESWPFILES is' }called to install the secondary files.  }TIA,e }Darreno   The absolute minimum is zero.-  K VMS will run just fine with no page or swap files (right up until you starto running out of memory, anyway).c  H It pretty much has to be able to, considering that you can boot VMS from@ a CD and you obvioulsy can't have any writeable files on the CD.   --- Carl   ------------------------------    Date: 13 Jul 2000 12:23:46 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>h2 Subject: Re: Minimum Page/Swap File Size in BlocksH Message-ID: <y4wviqqqfh.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  M 0. With that amount of memory, do you expect any paging or swapping to occur?@   	Jan   ------------------------------  % Date: Thu, 13 Jul 2000 11:33:52 +0100t- From: Tim Llewellyn <tim.llewellyn@bbc.co.uk> 2 Subject: Re: Minimum Page/Swap File Size in Blocks) Message-ID: <396D9B10.3E16ECD8@bbc.co.uk>    Carl Perkins wrote:   7 > "Boyle, Darren" <boyledj@bankofbermuda.com> writes... M > }       What's the absolute minimum file size in blocks of the Primary pagenI > }& swap files for OpenVMS 7.1 (VAX) and 7.1-2 (Alpha).  The systems arejO > }configures with 448Mb & 4096Mb respectively.  Obviously this should still beAJ > }big enough to accommodate the start-up process  until SYPAGESWPFILES is) > }called to install the secondary files.t > }TIA,c	 > }Darrena >a > The absolute minimum is zero.u  C Yup, I remember a satellite MicroVAX II who's paging disk had died,f5 but it still booted in the cluster, just VERY slowly.t  -- 6 Tim Llewellyn, OpenVMS Infrastructure, Remarcs Project0 MedAS at the BBC, Whiteladies Road, Bristol, UK.A Email tim.llewellyn@bbc.co.uk. Home tim.llewellyn@cableinet.co.uko  A I speak for myself only and my views in no way represent those of  MedAS or the BBC.n   ------------------------------   Date: 13 Jul 2000 07:42:51 CDT= From: wayne@tachysoft.xxx.293778.killspam.0223 (Wayne Sewell)T2 Subject: Re: Minimum Page/Swap File Size in Blocks. Message-ID: <6l6pAA4VBikw@tachxxsoftxxconsult>  X In article <396CBC87.2E6DE4E4@ohio.edu>, "Richard D. Piccard" <piccard@ohio.edu> writes:H > One item that I have not seen mentioned in this thread is dump files. J > In most situations it would be good to have at least a minimal dump fileH > on the system disk, to help diagnose problems early in the boot.  WhenI > tight on disk space, I have usually added the lines in MODPARAMS.DAT to J > use the primary paging file for dumping, and had no swap or dump files. E > This let me have a usefully large primary page file with less spacee! > sitting idle most of the time. e > J > As Hoff and many others here well know, setting this up requires care toG > avoid having disk blocks stomped on:  rename the swap and dump files,  > reboot, THEN delete them!  >     N Actually, disk corruption is no longer a problem with recent versions of vms. K The system now holds the dump file open for the life of the boot so nothing , else can allocate the blocks and reuse them.  O That said, the procedure above is still required, because it's the only way youa can truly delete the dump file.    Wayne    [stuff deleted]m     --  O ===============================================================================-M Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxxT: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-) O ===============================================================================-O Otter, on dining with Bluto:"It's perfectly safe if you keep your arms and legse 			away from his mouth."   ------------------------------  % Date: Thu, 13 Jul 2000 12:14:38 -0300 1 From: "Boyle, Darren" <boyledj@bankofbermuda.com>o2 Subject: RE: Minimum Page/Swap File Size in BlocksK Message-ID: <9F664D538536D411BD3200508B6FF01A33369B@bdant027.bda.bobda.com>!   > From: 	Jan@ > Vorbrueggen[SMTP:jan@fsnif.neuroinformatik.ruhr-uni-bochum.de] > H > 0. With that amount of memory, do you expect any paging or swapping to > occur? >  > 	Jan > K No and we tune to avoid it having small page files and no swap files, there L was a different motive to my questioning in that I'm redesigning the startup procedures.d - Darren    F **********************************************************************C This message and any files transmitted with it are confidential andEJ may be privileged and/or subject to the provisions of privacy legislation.M They are intended solely for the use of the individual or entity to whom they"L are addressed. If the reader of this message is not the intended recipient, B please notify the sender immediately and then delete this message.I You are notified that reliance on, disclosure of, distribution or copyingf of this message is prohibited.   Bank of BermudaDF **********************************************************************   ------------------------------  / Date: Thu, 13 Jul 2000 08:13:42 +0200 (MET DST)s& From: Rudolf Wingert <win@fom.fgan.de>+ Subject: MOre then one common SYSDUMP file?c6 Message-ID: <200007130613.IAA00709@sinet1.fom.fgan.de>   Hello,  H is it possible to have more then one common SYSDUMP file? If yes, how toJ must I name them? I would like to have one common SYSDUMP file for a group; of nodes, but not for all (e.g. eight groups of five node).e   TIA and regards Rudolf Wingert   ------------------------------  # Date: Thu, 13 Jul 2000 10:05:53 GMTy From: guyab@my-deja.com / Subject: Re: MOre then one common SYSDUMP file?t) Message-ID: <8kk49p$3uq$1@nnrp1.deja.com>-  6 In article <200007130613.IAA00709@sinet1.fom.fgan.de>,)   Rudolf Wingert <win@fom.fgan.de> wrote:. > Hello, >uG > is it possible to have more then one common SYSDUMP file? If yes, howN toF > must I name them? I would like to have one common SYSDUMP file for a group8= > of nodes, but not for all (e.g. eight groups of five node).e >u  > TIA and regards Rudolf Wingert >a > < The following is an exert from an answer I receiced from DEC    #                          COMMONDUMPo#                          ----------tF If you have several nodes booting off a common system disk, disk spaceG constraints may make it impossible to maintain a separate dump file forhG each node.  In order to conserve disk space a single "Common" dump filee; can be created for use by all nodes booting from that disk.u  D One drawback to using a shared, or common dump file, is in the eventC of multiple nodes crashing in rapid succession.  If this occurs theoB later crashes will write over earlier ones.  In this case the dumpC file will contain the dump from the last node, or may be corrupted.:  D Use the following steps to create a "Common" dump file in a cluster:  > 1. Examine each node sharing the common dump file to determine=    required file size.  The size of largest dump file is usedT    to create the common file.3  9 2. Create the dump file in the common directory using thec    following SYSGEN command:  A      SYSGEN> CREATE SYS$COMMON:[SYSEXE]SYSDUMP-COMMON.DMP/SIZE=<>e  G 3. Create an alias file name for the file, in the system specific root,o8    of each node sharing the dump, using the DCL command:  7      $ SET FILE SYS$COMMON:[SYSEXE]SYSDUMP-COMMON.DMP -a8      _$    /ENTER=SYS$SYSDEVICE:[SYS1.SYSEXE]SYSDUMP.DMP  F    You can ensure that all the alias file names point to the same areaG    By executing a DIRECTORY/FULL on the alias and checking the FILE ID.t  B 4. Add the entry, DUMPFILE=0, to each node's MODPARAMS.DAT file to5    disable AUTOGEN from modifying the dump file size.-  C 5. Add the following commands to the site specific startup files ofr#    each node sharing the dump file:p  '      $ ANALYZE/CRASH <synonym filename>0      COPY <output filespec>B	      EXITn    #                        SEPARATEDUMP #                        ------------z* How to return to using separate dumpfiles:  F 1. Use SET FILE/REMOVE to remove the file aliases on all nodes sharing    the common dump file:  =      $ SET FILE/REMOVE SYS$SYSDEVICE:[SYSn.SYSEXE]SYSDUMP.DMPo  E 2. Rename the old dumpfile in SYS$COMMON to the SYS$SPECIFIC:[SYSEXE]hF    directory for the node it is sized for. (Rename it to SYSDUMP.DMP.)  < 3. Create the new dumpfiles in SYS$SPECIFIC all other nodes.   4. Reboot all cluster nodes.    )                         DUMPOFFSYSTEMDISKl)                         -----------------.C Technological advances now allow extremely large and diverse memory.A subsystems to be configured on individual nodes and in VMSClusterSA environments.  The dumpfile size required to capture the contents D of memory has become a significant consideration in planning for newD and existing configurations.  Often times a dumpfile will be omittedC or its size restricted, as a trade off to disk space consumption on  the system disk.  E In order to address the need to capture valid dump information in thecF case of a system crash, a new feature called "Dumpfiles Off the SystemA Disk" (DOSD) has been introduced in OpenVMS VAX, version 6.2, and  OpenVMSD Alpha, version 7.1.j  H This feature still allows you to maintain a dumpfile on the system disk;E however, if you choose to relocate the dumpfile, or, in the case of acF VMSCluster environment, enable certain nodes to use a DOSD device, the2 following conditions and restrictions must be met:     OpenVMS VAX and Alpha    ---------------------l4   1. The dump device cannot be part of a volume set..   2. The DOSD device must be ODS-2 structured.>   3. The DOSD device must contain a directory identical to the@      root directory on the boot disk, i.e., [SYSn.SYSEXE], where3      "SYSn" is the same as the current SYS$SYSROOT.n<   4. The dumpfile must resides in the above directory on the      DOSD device.b,   5. The dumpfile must be named SYSDUMP.DMP.@   6. Compaq strongly recommends that the dump device also not be      part of a shadow set.A   7. Although not a requirement, Compaq recommends that you mount +      the dump device during system startup.r@   8. For the Crash Log Utility Extractor (CLUE) to support DOSD,?      you must define the logical name CLUE$DOSD_DEVICE to point :      to the dump file to be analyzed after a system crash.     OpenVMS VAX (only)   ------------------<   1. The system must be connected directly to, and must boot      from, CI controllers.<   2. The dump device must physically connect to the same two+      HSx CI controllers as the boot device.n?   3. The dump device cannot be MSCP unit zero.  Only units 1 tom/      4095 decimal (1 to FFF hex) are supported.h?   4. Version 4.3 or later of the VAX 7000 console firmware mustb@      be installed in order to use DUMPSTYLE for DOSD on VAX 7000      type systems.,   5. The system cannot be a VAX 9000 System.?   6. A stub SYSDUMP.DMP must reside in the SYS$SYSTEM directoryt@      in order for ERRORLOG buffers to be saved.  The size of the      minimal dump is:w  >        MIN_SYSDUMP.DMP = (ERRORLOGBUFFERS * ERLBUFFERPAGES) +1  ;   8. The DOSD device volume label must include "DOSD_DUMP",a&      i.e., DOSD_DUMP_12, 12_DOSD_DUMP.          Note:;          The volume label will be revalidated if the system :          crashes.  The label must match the label that was<          read when the system booted.  If the label does not(          match, no dump will be written.  A   ***Caution***Caution***Caution***Caution***Caution***Caution***sA   THE DUMPFILE MUST NOT BE DELETED ONCE THE SYSTEM HAS MAPPED IT.iA   If deleted after mapping, a subsequent system crash could causegB   file corruption because the dump will be written to the location,   where the deleted dumpfile used to reside.      PROCEDURE FOR IMPLEMENTING DOSD:  G   1. Ensure that all conditions listed in the subject section have beent.      examined and apply to your configuration.  @   2. If the system currently has a dumpfile mapped in the system
 directory,(      rename the dumpfile to SYSDUMP.SAV.  G   3. Use SYSGEN, or AUTOGEN, to create the dumpfile on the DOSD device.   @        SYSGEN> CREATE <dmpdev>:[SYSn.SYSEXE]SYSDUMP.DMP/SIZE=nnn  D      Including a line in the system's MODPARAMS.DAT file and runningF      AUTOGEN through the GENFILES phase will create a dump file on the      DOSD device.t  %        DUMPFILE_DEVICE = $nnn$ddcuuuui          Note:B          According to the V7.2 AUTOGEN.LIS and help file, the disk:          that DUMPFILE_DEVICE points to must be in quotes:    >          $!   Create directories for DOSD. Put reminder in the>          $!   report file to set the DUMP_DEV console env var.>          $!   Add help text that DUMPFILE_DEVICE value must be!          $!   enclosed in quotes.H              Note:E            Character strings (for parameters such as SCSNODE and file @            specifications for page, swap and dump files) must be,            enclosed in quotation marks, e.g. DUMPFILE_DEVICE="DISK$DOSD:"          Note:E          On Alpha systems a list of possible devices may be included.l;          VAX systems do not incorporate this functionality.e  E       Other database articles discuss the proper sizing of the systemn       dumpfile.   ;   4. Ensure the SYSGEN parameter DUMPSTYLE is set properly.-  D      DUMPSTYLE specifies the method of writing system dumps.  It's aG      32-bit mask, whose value is the sum of the values of the bits that.G      have been set (see examples below).  Remaining or undefined values-      are reserved to Compaq.  B      The following bits are defined for the DUMPSTYLE parameter in      OpenVMS V6.2 through V7.2:lE      +------------+-------------------------------------------------+ E      |     BIT    |               DESCRIPTION                       | E      | (set value)|                                                 |BE      +------------+---+---------------------------------------------+0E      |     0      | 0 | Full memory dump.  The entire contents of   |TE      |            |   | physical memory will be written to the dump |nE      |            |   | file.                                       | E      |            +---+---------------------------------------------+oE      | (00000001) | 1 | Selective dump (SYSGEN default).  The       |hE      |            |   | contents of memory will be written to       |oE      |            |   | the dump file selectively to maximize       |tE      |            |   | the usefulness of the dump file while       | E      |            |   | conserving disk space.                      | E      +------------+---+---------------------------------------------+kE      |     1      | 0 | Minimal console output.                     |gE      |            +---+---------------------------------------------+aE      | (00000002) | 1 | Full console output (includes stack dump,   | E      |            |   | register contents, and so on).              |aE      +------------+---+---------------------------------------------+iE      |     2      | 0 | Dump to system disk. (VAX only)             | E      |            +---+---------------------------------------------+.E      | (00000004) | 1 | Dump to alternate disk. (See note 1 below.) |oE      +------------+---+---------------------------------------------+-E      |     3      | 0 | Do not compress. (Alpha only)               |rE      |            +---+---------------------------------------------+DE      | (00000008) | 1 | Enable dump file compression.               |aE      +------------+---+---------------------------------------------+oE      |    4-14    |     (Reserved to Compaq.)                       |nE      +------------+---+---------------------------------------------+aE      |    15      | 0 | Disable use of bits 16 - 27. (VAX only)     |nE      +------------+---+---------------------------------------------+pE      | (00008000) | 1 | Enable use of bits 16 - 27.                 |sE      |            |   | (See note 3 below.)                         | E      +------------+---+---------------------------------------------+$E      |   16-27    |     Range of DOSD unit numbers. (VAX only)      |gE      | (0FFF0000) |     (1-4096 decimal, 1-FFF hex)                 |oE      +------------+-------------------------------------------------+cE      |   28-31    |     (Reserved to Compaq.)                       | E      +------------+-------------------------------------------------+b      Notes:m>        [1]Alpha systems do not support dumping to an alternateA           disk prior to V7.1   [2]VAX systems do not support dumpp8           compression   [3]Specific to VAX 7000 systems.        Examples:=        - On VAX systems, the value of 4 directs the system to->          send a full dump, with minimal console output, to the          alternate disk.?        - On Alpha systems, the value of 9 directs the system toj#          compress a selective dump.i        DUMP COMPRESSION:H      On Alpha systems, system managers can save space on the system diskA      and, in the event of a crash, save time recording the system  memory,-E      by using the OpenVMS Alpha dump compression feature.  Unless the E      system manager overrides the default AUTOGEN calculations, i.e.,gG      setting DUMPSTYLE in MODPARAMS.DAT, uses the following algorithms: G        Note: To enable DOSD on the ALPHA with compressed selective dumpt(              the dumpstyle should be 13.  D      a. On a system with less than 128 MB of memory, the system setsE         DUMPSTYLE to 1 (a raw selective dump) and sizes the dump file8         appropriately.E      b. On a system with 128 MB of memory or greater, the system setsaE         DUMPSTYLE to 9 (a compressed selective dump), and creates theaH         dump file at two thirds the value of the corresponding raw dump.  B   5. Shut down and reboot the system using the following format to      designate the DOSD device:r        OpenVMS VAX      -----------@      - On VAX 7000 configurations, use bits 16 through 27 of the"        DUMPSTYLE system parameter.  A        Note: On VAX 7000 systems bit 15 must be set to enable theaB              DOSD unit in bits {27:16}.  Bits sixteen (16) throughA              twenty-seven (27) will be checked by SYSBOOT for the @              unit number of the dump device, if bit fifteen (15)              is also set.n  ?        Example: To use unit thirty (30) as the dump device, andxA                 have a full (as versus a subset) dump take place:                    $MC SYSGEN1                 SYSGEN>  SET DUMPSTYLE %X001E8006 &                 SYSGEN>  WRITE CURRENT  &                    Reboot the VAX 7000  ( %X0    0    1    E    8    0     0     6) 0000 0000 0001 1110 1000 0000  0000  0110b8           \       / \  /             \  /           Bit#D            \     /    |               |         /  (0) = 0 Full DumpG              \ /      |               |        /   (1) = 1 Full Consoler OutputE               |       |               |       /          at Bug CheckHD               |       |               +------>     (2) = 1 Use  Dump OffiD               |       |                       \          System Disk deviceE               |       |                        \   (3) = . <Reserved>o               |       | 8               |       |            / (12) = . <Reserved>8               |       |           /  (13) = . <Reserved>8               |       +--------->    (14) = . <Reserved>G               |                   \  (15) = 1 Enable Flag for bits 27 -e 16@               |                    \        for DOSD on VAX 7000               |=B               +------------>         (27->16) DOSD Unit Number ...<                                               Unit 30 in hex  =      - On other configurations, use bits 16 through 27 of R3.o  0          31 28 27         16 15 12 11          01         +-----+-------------+-----+-------------+=1         | 0's | Dump Device | 0's | Boot Device |i1         +-----+-------------+-----+-------------+            Example:/           R3 = 00440000,  Unit  0 = Boot Device-/                   ^       Unit 68 = Dump Devicer!                   |             ^b?                   |             +- 68 is the decimal equivalent 5                   +--------------- of 44 hexadecimal.n        OpenVMS Alpha      -------------F      - Use the following format to specify the dump device environment/        variable DUMP_DEV at the console prompt:n  H          >>> SHOW DEV (a list of devices, including disks, is displayed)$                Example: DUA1.4.0.2.3  E              Then if the new disk to contain the dump is DUA1 use thet              following:   *          >>> SET DUMP_DEV device-name[...]5                Example: >>> SET DUMP_DEV DUA1.4.0.2.3d  <          Note: 4 paths to the device can be set-up for DOSD.  G   6. Once the system is rebooted the renamed dump file (see step 2) mayi      be deleted.  F Going on from what this information is telling us, we shaould have theE ability to create multiple dumpfiles for different groups of nodes inh one common area.  D Sorry that the answer is a bit long winded, but better too much than not enough.i -------------------------- Guy Barretts Infrastructire Team Leader TNT Post Group   Note.*F The contents of this reply are my own opinions and have no relation to( the organisation for which I am working.    & Sent via Deja.com http://www.deja.com/ Before you buy.>   ------------------------------  % Date: Thu, 13 Jul 2000 10:22:37 -0400A  From: norm.raphael@jamesbury.com/ Subject: Re: More then one common SYSDUMP file? 4 Message-ID: <C225691B.004E15CE.00@jklh21.valmet.com>  C It seems to me you can create them in the common area and name themoD anything that helps you keep them straight in your head, then do the appropriate-  G SET FILE/ENTER=sys$specific:[sysexe]sysdump.dmp correct_common_filenamea   for each root in your group(s).h   -Normm        ) win@fom.fgan.de on 07/13/2000 02:13:42 AM.   To:   Info-VAX@mvb.saic.com7 cc:3, Subject:  MOre then one common SYSDUMP file?         Hello,  H is it possible to have more then one common SYSDUMP file? If yes, how toJ must I name them? I would like to have one common SYSDUMP file for a group; of nodes, but not for all (e.g. eight groups of five node).    TIA and regards Rudolf Wingert   ------------------------------  # Date: Thu, 13 Jul 2000 15:49:57 GMT 1 From: "Mark D. Jilson" <jilly@clarityconnect.com> / Subject: Re: MOre then one common SYSDUMP file? 2 Message-ID: <396DE50F.2B5823E4@clarityconnect.com>  ? Certainly possible.  You just point your SYSDUMP.DMP alias to aeD different file.  We like to make the 'actual' file name be something" useful but you could have the fileF SYS$SYSDEVICE:[MUMBLE.MUMBLE.MUMBLE]MUMBLE.MUMBLE be the 'actual' fileH that you create the SYSDUMP.DMP alias to look to.  Of course if I was toG support your systems after this I might have to physically harm you ;*)r   Rudolf Wingert wrote:l >  > Hello, > J > is it possible to have more then one common SYSDUMP file? If yes, how toL > must I name them? I would like to have one common SYSDUMP file for a group= > of nodes, but not for all (e.g. eight groups of five node).r >   > TIA and regards Rudolf Wingert   -- gD 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: 13 Jul 2000 16:48:43 GMT0 From: FAZEKAS Mihaly <michael@goliat.eik.bme.hu>! Subject: new problem with VMS 5.5L, Message-ID: <8kkrtb$5tg$1@goliat.eik.bme.hu>  8 My other (new :) ) machine (MicroVAX II, and 2 of RA81).? Probe booting from DUA0:, but cannot boot (DUA0 is system disk,r
 DUA1 is not).t  H The old "system user" of this machine say: "possibly cluster member,..."7 Ok, but i not have VAX cluster (but have many VAX :) ).   3 How can i boot this machine (example: without net). $ How can i get a new SYSTEM password?, (the old "system user" is loose :) password)   -- t  mailto:michael@goliat.eik.bme.hu Phone: 463-1966e   ------------------------------  % Date: Thu, 13 Jul 2000 03:10:06 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)' Subject: New software available for VMSeL Message-ID: <rdeininger-1307000310070001@user-2iveafc.dialup.mindspring.com>   Greetings VMS folks,  C Here's an opportunity for an interesting diversion for you and your.F alpha, and maybe some VMS evangelism as well.  Join the Great InternetC Mersenne Prime Search (GIMPS), now available for an Alpha near you!   = If your alpha has grown weary of looking for extraterrestrial F space-aliens (seti@home), maybe you should give it a new hobby.  Teach= it to hunt for huge Mersenne prime numbers in its spare time.   E Here's the idea in a nutshell.  See the web pages mentioned below forc all the gory details.   A 1. Pick a prime number P in the neighborhood of 10 million or so.o  E 2. Consider the Mersenne number M(P) = 2^P - 1.  This is a REALLY bigi(    number - around 10 million bits long.    .E 3. If M(P) has any "small" factors, throw it away and pick another P.tE    Mersenne composites are a dime a dozen, and nobody wants any more.     sF 4. If M(P) has no small factors, do a "Lucas-Lehmer" test to determine     whether it is a prime number.    fE 5. If the LL test proves that M(P) is prime, hurray!  You are famous!6H    Only 38 Mersenne primes are known.  You just discovered the 39th one!D    It's likely the biggest known prime number of ANY kind.  How longH    will your record stand?  And you discovered a new perfect number as a	    bonus.n    t/ 6. If M(P) is not prime, go back to step one...d    E Ernst Mayer's Lucas-Lehmer program, Mlucas, tests these huge Mersenne-E numbers to see if they are prime.  Ready-to-link VMS object files areSD now available on Ernst's server, thanks to my humble efforts portingF the program from Tru64 Unix to VMS.  ("Porting" consisted of compilingF the Fortran 90 source with /FLOAT=IEEE_FLOAT, so don't thank me, thankH the Compaq Fortran folks.)  I've also provided the VMS readme file and a* few possibly useful .COM files for Mlucas.  E Since Fortran 90 isn't available for Vaxes, there's no Vax version of  Mlucas.  Sorry.   C If you decide to join the hunt, please drop me an e-mail and let meaF know.  When there's enough interest, we can form a VMS team to competeH against the PeeCee herds, and hopefully discover the next record-setting@ prime number on a VMS system.  (That's the VMS evangelism part.)  G I'll shut up now and direct you to some web pages and README files that- tell the rest of the story.t     Useful URLse ===========m   The GIMPS home page:  %     http://www.mersenne.org/prime.htmh  C This is the home page for the Great Internet Mersenne Prime Search.eE Everything (almost?) else is linked from here.  GIMPS is dominated bymD Wintel and lintelux folks, and a lot of the information here doesn'tA apply to VMS users.  Neverthless, this is the home of the projectn- and everyone who participates should read it.      The PrimeNet home page:s  !     http://mersenne.org/primenet/.  I PCs connect to the PrimeNet server automatically to get work assignments oI and report results.  The VMS software doesn't have the automatic connect  H feature (yet) so that part of the server doesn't do us any good. There'sB still a lot worth reading at this site.  A few clicks away via theH Source Code link, you will find the Mlucas program, with the source codeG and binaries for various platforms, including VMS Alpha.  I'll save you " some hunting and include the URLs:  !     ftp://hogranch.com/pub/mayer/n,     ftp://hogranch.com/pub/mayer/README.html  4     ftp://hogranch.com/pub/mayer/bin/ALPHA_VMS/*.ZIP=     ftp://hogranch.com/pub/mayer/bin/ALPHA_VMS/README_VMS.TXTM    D Start with the two README files, which will tell you how to download everything else.    F You'll need UNZIP to expand the VMS binary files. If you don't have it/ already, you can get a working version via ftp:   )     ftp://ftp.wku.edu/vms/unzip.alpha_exen  F Download this file in binary mode and use it to unzip the VMS binaries
 for GIMPS.   -- o Robert Deininger rdeininger@mindspring.comf   ------------------------------  # Date: Thu, 13 Jul 2000 15:25:06 GMT.+ From: "Michal Pakula" <mpakula@gask.com.pl>e! Subject: Odp: VMS Pascal question 0 Message-ID: <mblb5.1029$wy2.24072@news.tpnet.pl>  C > When looking at "HELP RTL SMG$ SMG$whatever", or any of the other , > documentation, you'll see stuff like this: [...]hL > You should ignore the "VMS usage". All that matters for data type purposesJ > is the "type". You may also need to check the "mechanism" part. The "[]"K > around an argument in the "Format" section indicates that the argument isn > optional.  > J > All you need is to know how declare and pass those data types in Pascal. >t > But even better...J > You should have (I do) a file called SYS$LIBRARY:PASCAL$SMG_ROUTINES.PASE > on your system. In this file is the exact definition of all the SMG 	 routines.t [...]n Thanks!aI I suppose, that the second sugestion is better. I think so because when IpH look at the description of SMG$CREATE_MENU, at "choices" argument, I see something like this:  .         VMS usage: static array of char_string         type: character string         access: read onlys          mechanism: by descriptor  E and I think that I _have_to_ use array of strings, not just charactertJ string, because as stated later it is: "Static array in which each element= corresponds  to  an  item  to  be displayed  in  the  menu.".iK Anyway - I shall search the PASCAL$SMG_ROUTINES.PAS file and hope I'll findu what I'm looking for.    Thanks       Michal Pakulab   ------------------------------    Date: 13 Jul 2000 11:13:08 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> : Subject: Re: OpenVMS loses big, was: RE: Compaq advertisesH Message-ID: <y43dles89n.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  + yyyc186.illegaltospam_@flashcom.net writes:(  @ > The nice part is that Oracle has been relegated to little moreI > than an off-line data storage mechanism.  In the world of real-time and C > reliability this client learned the hard way...Oracle can't play.|  F So what _do_ your clients use as a database system? Inquiring minds...   	Jan   ------------------------------  % Date: Thu, 13 Jul 2000 12:29:18 -0300-1 From: "Boyle, Darren" <boyledj@bankofbermuda.com> 3 Subject: RE: OVMS V7.2-1: Passwords Case-Sensitive?tK Message-ID: <9F664D538536D411BD3200508B6FF01A33369C@bdant027.bda.bobda.com>t  K Aren't 7.2 passwords now available in both cases, i.e.  upper & lower.  I'm L sure I read somewhere that you can now specify a lowercase or case-sensitiveI password by including it in quotes, an unquoted password is automatically+ uppercased.  i.e.-  ( "ThisPassword" is passed as ThisPassword but & ThisPassword is passed as THISPASSWORD  ) Check it out in the new features in help.  - Darren   > ----------< > From: 	David J. Dachtera[SMTP:djesys.nospam@earthlink.net]* > Sent: 	Wednesday, July 12, 2000 10:58 PM > To: 	Info-VAX@Mvb.Saic.Com2 > Subject: 	OVMS V7.2-1: Passwords Case-Sensitive? >  > Hi, Folks, > J > Ran into an interesting problem on a client site today. My boss upgradedI > their Pathworks server from V7.1 of OVMS to V7.2-1 and V6.-something of-$ > Pathworks to Advanced Server V7.2. > F > Now, they have an interesting problem. When attempting to log in viaF > PowerTerm, they can't log in unless they have the shift-lock on when > typing their password. >  > Anyone have any ideas? >  > --   > David J. Dachtera- > dba DJE Systems- > http://www.djesys.com/ > < > Unofficial Affordable OpenVMS Home Page and Message Board:! > http://www.djesys.com/vms/soho/- >     F **********************************************************************C This message and any files transmitted with it are confidential and-J may be privileged and/or subject to the provisions of privacy legislation.M They are intended solely for the use of the individual or entity to whom they-L are addressed. If the reader of this message is not the intended recipient, B please notify the sender immediately and then delete this message.I You are notified that reliance on, disclosure of, distribution or copying- of this message is prohibited.   Bank of BermudatF **********************************************************************   ------------------------------  # Date: Thu, 13 Jul 2000 17:33:38 GMT-, From: alphaman-nix-spam@hsv.sungardtrust.com0 Subject: Porting assistance for GNU code (GnuGP)) Message-ID: <8kkuhh$o46$1@nnrp1.deja.com>-   Hi,-  C I am interested in porting GnuGP (think of it as the GNU version ofpA PGP) to OpenVMS.  I am green when it comes to porting U*X code to  OpenVMS, however.o  < To date, I've got the porting library off the Q's site, haveB successfully installed it, and am wading through some of the finalA tidbits of the installation and getting the gnugp files ready foruE OpenVMS.  But wading through the makefile is tedious (it's been yearsvD since I've looked at one at this level of detail), and I'm sure thatA I'm going to run into many, many more issues in my efforts (e.g.,cD another I can see is that the makefile.in is made from automake, butE we've been asked NOT to put Perl on our system.  If my efforts are tohC make their way back into the gnu source, then there must be someoneeD willing to help on this project with Perl installed on their system,' otherwise this will be a one-off port.)h  B So, what I would like is some mentoring from someone familiar withF porting gnu type C programs, especially if that individual is familiar, with the workings of the new porting libraryC (http://www.openvms.digital.com/openvms/products/ips/porting.html).pD The level of effort may be as simple as just answering an occasionalF question to actually participating in the port and helping to maintain	 the code.m  D Any takers out there?  Anyone else interested in seeing GnuGP ported who has porting experience?.   TIA, Aaronr    & Sent via Deja.com http://www.deja.com/ Before you buy.e   ------------------------------  % Date: Wed, 12 Jul 2000 23:36:18 -04000* From: David A Froble <davef@tsoft-inc.com>$ Subject: Re: ppp dialup on VAX again- Message-ID: <396D3932.FB0AEC61@tsoft-inc.com>t  & "Brian Schenkenberger, VAXman-" wrote: > F > In article <396c993c@amhnt2.amherst.edu>, jlodom@amherst.edu writes: > >wM > >And to add to his question: Is it possible to use PPP on a Vaxstation 3100, > >running Multinet?G > >(And where would I find I good modem to connect to said Vaxstation?)t > >m > F > FWIW, if you're willing to spend a few $$$s (~$250), there are boxesE > with will do this for you.  One such unit is the 3com OfficeConnectSE > 56K LAN modem.  This box automatically logs into the ISP and estab- E > lishes a PPP connection.  The VAX, or Alpha, or any other box which E > can speak the TCP/IP lingo, connects to the LAN modem via 10bT RJ45 G > connections on the LAN modem.  You can connect up to 4 devices to the G > 3com LAN modem and it will provide DHCP service, NAT, and *on demand*/E > dialup to the ISP.  There are similar offerings from other mfg's, I 2 > believe intel has one I've seen in catalogs too. > H > With such boxes you get a modem (V.90), hub, minimal routing features,G > NAT, DHCP server, and a slew of other little features like the on de- < > mand connection with very little hassle and configuration. > G > BTW, I'm no 3com fan.  I've just had one of these units to play with.  >  > --Q > VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM   J I've a similar story.  Bought a rather cheap device from OnSale, called anO internet sharing device (Hawking Technology).  I did need to add a modem, got a P MultiTech 56K analogue modem.  Thing does NAT, on-demand dial-up, and some otherN stuff I'm not using.  Connects to the network and unlimited nodes can use it. N Using Gated something or other (hey, I'm not into TCP/IP) some of my VMS nodesP can telnet and ftp and such directly over the internet.  Also supports dedicatedO lines, DSL, etc.  Can also allow limited (1 or 2) local computers to be visable-O to the internet.  Cost was 90-100 dollars, before modem.  Check around, there'st* some neat (for the price) stuff available.   Dave   --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  % Date: Thu, 13 Jul 2000 12:38:20 -0500 % From: "Crosson, Mary" <mary@UMKC.EDU>E Subject: Recall: subscribeP Message-ID: <95A711A70065D111B58C00609451555C048A1C12@umkc-mail02.wins.umkc.edu>  < Crosson, Mary would like to recall the message, "subscribe".   ------------------------------   Date: 13 Jul 2000 15:40:42 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)/ Subject: SQL+ODBC options, inexpensive or free?-, Message-ID: <8kkntq$1nl@gap.cco.caltech.edu>  H What are the options, if any, for a free (or very inexpensive) SQL+ODBC E program for OpenVMS Alpha?   One of my users has a project which may cK require this, performance shouldn't be that big an issue, but cost will be.   I I've seen posts here of people working on, or thinking about working on, nF the various open source database projects.  Have any of these resulted in a working OpenVMS port?  G Am I the only one who finds it odd that the OpenVMS guys are working so K hard on Apache without, apparently, having provided in any way for a serverjJ database smaller than RDB or Oracle?  I wonder what the intended market isG for that work anyway.  The only possible market I see is for a swarm ofMG DS10s running Apache feeding into a GS series RDB/Oracle back end.  Ie,hE there would already have to be a huge investment in the "core" of the G cluster so that the very high price of the web servers/cluster licensestF doesn't _seem_ so very high by comparison.  Because a cluster of DS10sG without such an expensive core is going to seem just as expensive as itmF really is - possible customers are going to blink twice at the clusterK license prices, and drop dead when they see the RDB/Oracle prices, and willlK probably "make do" with Linux/mysql on the DS10s instead.  Assuming they gooK with DS10s at all, since they could just as well do Intel based machines if  they don't need OpenVMS. e   Regards,   David Mathog mathog@seqaxp.bio.caltech.eduW? Manager, sequence analysis facility, biology division, Caltech 0   ------------------------------  % Date: Thu, 13 Jul 2000 10:58:22 +0200:6 From: Graham Van Der Vaart <Graham.vd.Vaart@ast.co.za># Subject: UCX sockets and aux server P Message-ID: <47DBB722BF1BD21199E00080C8372C1B31A5CE@ntsnexc1.new.iscorltd.co.za>   Hi  L I have a problem developing a socket dispatcher using UCX Auxiliary service. What I want is....K Windows client always initiates. Auxiliary service on OpenVMS listens as in  UNIX inetd on port 7650.  I set up the service as follows:   ucx set service SSDS2 -  /port=7650 - /prot=TCP -r /flag=listen - /user=tcpip$inetd - ) /file=dra1:[users.x.taap300]taap300.com -m /process=taap300 -1 /log=(file=dra1:[users.x.taap300]taap300.log,all)   - On the client side ......... 	gethostbyname()g 				get servbyname() 				htons()e 				socket()
 				connect())
 				send()  % On the server side.........		select()l
 				recv()  7 But my select times out waiting for data. ie returns 0.lI Here is my select ... nFound = select(32, &fdReadFds, 0 , 0 ,&tTimeOut );i  ; Just too mention that this all works on UNIX successfully.   Any help would be valuable.-  - If I telnet <hostname> 7650 from another host D I  can see the Socket buffer bytes when sho dev <device_socket> /con   RegardsD Graham       Graham Van der vaart Senior Systems Analyst AST-Abraxas Kwazulu-Natal    Tel:	034 3148152 Cell:	083 327 0801 Fax:	034 31 8844   ------------------------------  % Date: Thu, 13 Jul 2000 11:30:33 +0200:6 From: Graham Van Der Vaart <Graham.vd.Vaart@ast.co.za># Subject: UCX sockets and aux serveroP Message-ID: <47DBB722BF1BD21199E00080C8372C1B31A5CF@ntsnexc1.new.iscorltd.co.za>   > Hi > E > I have a problem developing a socket dispatcher using UCX Auxiliary.
 > service. > What I want is....J > Windows client always initiates. Auxiliary service on OpenVMS listens as > in > UNIX inetd on port 7650." > I set up the service as follows: >  > ucx set service SSDS2 -  > /port=7650 -
 > /prot=TCP -e > /flag=listen - > /user=tcpip$inetd -i+ > /file=dra1:[users.x.taap300]taap300.com -n > /process=taap300 -3 > /log=(file=dra1:[users.x.taap300]taap300.log,all)i > / > On the client side ......... 	gethostbyname()S > 				get servbyname()
 > 				htons()h > 				socket() > 				connect()  > 				send() > ' > On the server side.........		select()( > 				recv() > 9 > But my select times out waiting for data. ie returns 0. K > Here is my select ... nFound = select(32, &fdReadFds, 0 , 0 ,&tTimeOut );  > = > Just too mention that this all works on UNIX successfully. . > Any help would be valuable.e > / > If I telnet <hostname> 7650 from another hosteF > I  can see the Socket buffer bytes when sho dev <device_socket> /con > 	 > Regardss > Graham >  >  >  > Graham Van der vaart > Senior Systems Analyst > AST-Abraxas Kwazulu-Natalp >  > Tel:	034 3148152 > Cell:	083 327 0801 > Fax:	034 31 8844 >  >    ------------------------------  % Date: Thu, 13 Jul 2000 11:58:45 +0200s6 From: Graham Van Der Vaart <Graham.vd.Vaart@ast.co.za># Subject: UCX sockets and aux server6P Message-ID: <47DBB722BF1BD21199E00080C8372C1B31A5D0@ntsnexc1.new.iscorltd.co.za>   > Hi > E > I have a problem developing a socket dispatcher using UCX Auxiliary 
 > service. > What I want is....J > Windows client always initiates. Auxiliary service on OpenVMS listens as > in > UNIX inetd on port 7650." > I set up the service as follows: >  > ucx set service SSDS2 -e > /port=7650 -
 > /prot=TCP -i > /flag=listen - > /user=tcpip$inetd -n+ > /file=dra1:[users.x.taap300]taap300.com -p > /process=taap300 -3 > /log=(file=dra1:[users.x.taap300]taap300.log,all)i > / > On the client side ......... 	gethostbyname()d > 				get servbyname()
 > 				htons()n > 				socket() > 				connect()V > 				send() > ' > On the server side.........		select()u > 				recv() > 9 > But my select times out waiting for data. ie returns 0. K > Here is my select ... nFound = select(32, &fdReadFds, 0 , 0 ,&tTimeOut );l > = > Just too mention that this all works on UNIX successfully. S > Any help would be valuable.e > / > If I telnet <hostname> 7650 from another hosttF > I  can see the Socket buffer bytes when sho dev <device_socket> /con > 	 > Regardsn > Graham >  >  >  > Graham Van der vaart > Senior Systems Analyst > AST-Abraxas Kwazulu-NatalU >  > Tel:	034 3148152 > Cell:	083 327 0801 > Fax:	034 31 8844 >  >    ------------------------------  % Date: Thu, 13 Jul 2000 09:03:20 -0400e= From: "Stephane Paquin" <stephane.paquin@nospam.ispatnet.com>s' Subject: Re: UCX sockets and aux server,3 Message-ID: <s6jb5.19521$227.372731@nnrp1.uunet.ca>V  L     What you are doing is tell inetd to wait on port TCP 7650 and when thereI is a connection request to satrt a new process and execute the procedure.n  D     Since inetd will create a process on the VMS machine and run theC taap*.com procedure for you, all you need to do in the process is a L sys$assign to SYS$NET to get the channel and then exchange messages. I do itL with QIOs directly instead of send,recv. Should you also specify a number of connections allowed ?C       Best luck.     Stphane Paquinr     Ispat Sidbec  A Graham Van Der Vaart <Graham.vd.Vaart@ast.co.za> wrote in messagelJ news:47DBB722BF1BD21199E00080C8372C1B31A5CE@ntsnexc1.new.iscorltd.co.za... > Hi > E > I have a problem developing a socket dispatcher using UCX Auxiliaryp service. > What I want is....J > Windows client always initiates. Auxiliary service on OpenVMS listens as in > UNIX inetd on port 7650." > I set up the service as follows: >e > ucx set service SSDS2 -  > /port=7650 -
 > /prot=TCP -  > /flag=listen - > /user=tcpip$inetd -t+ > /file=dra1:[users.x.taap300]taap300.com -. > /process=taap300 -3 > /log=(file=dra1:[users.x.taap300]taap300.log,all)t > . > On the client side ......... gethostbyname() > get servbyname()	 > htons()r
 > socket() > connect()- > send() >-& > On the server side......... select() > recv() >c9 > But my select times out waiting for data. ie returns 0.lK > Here is my select ... nFound = select(32, &fdReadFds, 0 , 0 ,&tTimeOut );g > < > Just too mention that this all works on UNIX successfully. > Any help would be valuable.i > / > If I telnet <hostname> 7650 from another hosttF > I  can see the Socket buffer bytes when sho dev <device_socket> /con >h	 > Regardsc > Graham >  >  >" > Graham Van der vaart > Senior Systems Analyst > AST-Abraxas Kwazulu-Natal  >p > Tel: 034 3148152 > Cell: 083 327 0801 > Fax: 034 31 8844 >e >t >y   ------------------------------  % Date: Thu, 13 Jul 2000 03:37:09 -0700a5 From: Richard  <maher_rjNOmaSPAM@hotmail.com.invalid>t* Subject: Re: VAX to DEC Cobol differences.9 Message-ID: <00e8e5d8.617b3398@usw-ex0110-076.remarq.com>    Hi,o  7 A place I was at a few months ago was getting a similar ; problem using lib$cvt_dx_dx to convert strings to numerics.s8 It seems the compiler is generating different descriptor; classes between VAX and Alpha. I think the problem could bee9 traced back to the fact that DEC COBOL (unlike VAX COBOL)h7 supports UNsigned integers. You can finally stick 65535c into an unsigned word integer!  # A couple of other random thoughts:-h  ; Using the address of the RMS_STS of a file to jump into its  FAB and RAB no longer works.  ; EXTERNAL variables now produce NOshr PSECTS by default. AndI4 the size in bytes of these PSECTS is rounded up to a multiple of 8.  8 DEC COBOL has a screen section that allows you to create9 SMG$ forms in your working-storage. Has anyone used it? Ie haven't had a chance.S   Regards Richard Maher.     * Sent from AltaVista http://www.altavista.com Where you can also find related Web Pages, Images, Audios, Videos, News, and Shopping.  Smart is Beautifula   ------------------------------    Date: 13 Jul 2000 08:56:07 -0500 From: briggs@eisner.decus.orgo* Subject: Re: VAX to DEC Cobol differences.+ Message-ID: <VAE$c89RvWqz@eisner.decus.org>    In article <1137A4A23A51D311B2D600105A1D5213019AEEFB@seantexch.unitedad.com>, Terry Marosites <TMarosites@unitedad.com> writes:u >  > J > I am migrating to an Alpha (dec Cobol) from a Vax .I just copied my codeN > over and recompiled. A 100% of the programs compile without a problem  99.9%< > of the programs ran Ok , But I have a few that went amuck. >  > One was very strange.    > M >              The program on the VAX accepted a variable with spaces as zeror > i.e. >  >              If string = zero % >              Display "Bla bla bla".u > I >              On the VAX the display was performed on the Alpha was not.u  / You should look at the compile time qualifiers:    	/CONVERT=LEADING_BLANKS andp 	/CHECK=DECIMAL   6 This compiles in code to handle the troublesome cases.  * 	John Briggs			briggs@alpha.tst.tracor.com   ------------------------------  # Date: Thu, 13 Jul 2000 13:05:52 GMTw+ From: "Michal Pakula" <mpakula@gask.com.pl>  Subject: VMS Pascal question. Message-ID: <Q8jb5.479$wy2.8628@news.tpnet.pl>   Hello,H I have to write a Pascal program for VMS, similar to Midnight Commander,K using SMG$ routines. I found a small example on how to use SMG$ routines intK Pascal but it covers only few functions and I can't find good documentationhG on those routines. By "good documentation" I mean text where I can find K _exact_ description on using SMG$ functions in _Pascal_, not a general one./C I've already found general documentation, but it uses VMS datatypes I descriptions and it's not so easy to guess what Pascal types shoud I use.* I use VMS V. 5.4-3.* Thanks in advances       Michal Pakulas   ------------------------------   Date: 13 Jul 2000 09:20 CST/' From: carl@gerg.tamu.edu (Carl Perkins)e  Subject: Re: VMS Pascal question- Message-ID: <13JUL200009203315@gerg.tamu.edu>h  / "Michal Pakula" <mpakula@gask.com.pl> writes...a }Hello, I }I have to write a Pascal program for VMS, similar to Midnight Commander,eL }using SMG$ routines. I found a small example on how to use SMG$ routines inL }Pascal but it covers only few functions and I can't find good documentationH }on those routines. By "good documentation" I mean text where I can findL }_exact_ description on using SMG$ functions in _Pascal_, not a general one.D }I've already found general documentation, but it uses VMS datatypesJ }descriptions and it's not so easy to guess what Pascal types shoud I use. }I use VMS V. 5.4-3. }Thanks in advance }  }    Michal Pakula  A When looking at "HELP RTL SMG$ SMG$whatever", or any of the other * documentation, you'll see stuff like this:       SMG$PUT_LINE  H          The Write Line to Virtual Display routine writes a line of textF          to a virtual display, beginning at the current virtual cursor          position.            Formatl  J            SMG$PUT_LINE  display-id ,text [,line-advance] [,rendition-set]  J                          [,rendition-complement] [,flags] [,character-set]  %                          [,direction]t           Argumentsl           display-id  !          OpenVMS usage identifiera*          type          longword (unsigned)           access        read only#          mechanism     by referencenH          Specifies the virtual display affected. The display-id argumentI          is the address of an unsigned longword that contains the displayt          identifier.  >          Display-id is returned by SMG$CREATE_VIRTUAL_DISPLAY.  
       text  "          OpenVMS usage char_string'          type          character stringo           access        read only$          mechanism     by descriptorF          The characters to be written to the virtual display. The textF          argument is the address of a descriptor pointing to the text. [... and so on...]  J You should ignore the "VMS usage". All that matters for data type purposesH is the "type". You may also need to check the "mechanism" part. The "[]"I around an argument in the "Format" section indicates that the argument is 	 optional.s  H All you need is to know how declare and pass those data types in Pascal.   But even better...H You should have (I do) a file called SYS$LIBRARY:PASCAL$SMG_ROUTINES.PASM on your system. In this file is the exact definition of all the SMG routines..6 It contains about 160KB of stuff that looks like this:  N (*    SMG$PUT_LINE                                                          *)N (*                                                                          *)N (*    Write Line to Virtual Display                                         *)N (*                                                                          *)N (*    The Write Line to Virtual Display routine writes a line of text to a  *)N (*    virtual display.                                                      *)N (*                                                                          *)  & [ASYNCHRONOUS] FUNCTION smg$put_line (         display_id : UNSIGNED;A         text : [CLASS_S] PACKED ARRAY [$l2..$u2:INTEGER] OF CHAR;n+         line_advance : INTEGER := %IMMED 0;a-         rendition_set : UNSIGNED := %IMMED 0;u4         rendition_complement : UNSIGNED := %IMMED 0;%         flags : UNSIGNED := %IMMED 0;m-         character_set : UNSIGNED := %IMMED 0; >         direction : UNSIGNED := %IMMED 0) : INTEGER; EXTERNAL;  J By comparing what is in the help, and/or in this file, with what is in theF example you have you should be able to get what you need to know about@ how to interpret the type and mechanism information into Pascal.  = For example, the example file PASCAL$EXAMPLES:SMG_EXAMPLE.PAS ! has a call to the above function:   /     Status := SMG$Put_Line(Display_Id_2, Text);y  D Comparing things, including the variable declarations in the example= program, shows you that what is meant when the help specifiesp  '          type          character string-$          mechanism     by descriptor  L is for the variable to be a PACKED ARRAY [x] OF CHAR and passed with nothingJ special done. Every time the type and mechanism match this for any routine" you can do exactly the same thing.  M I expect that the SMG routines only use a few different data types - integers K both signed and unsigned, text strings, and perhaps arrays of these things.e   --- Carl   ------------------------------   Date: 13 Jul 2000 15:45:19 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)u  Subject: Re: VMS Pascal question0 Message-ID: <8kko6f$gjr$1@n.ruf.uni-freiburg.de>  \ In article <Q8jb5.479$wy2.8628@news.tpnet.pl>, "Michal Pakula" <mpakula@gask.com.pl> writes:I >I have to write a Pascal program for VMS, similar to Midnight Commander,SL >using SMG$ routines. I found a small example on how to use SMG$ routines inL >Pascal but it covers only few functions and I can't find good documentationH >on those routines. By "good documentation" I mean text where I can findL >_exact_ description on using SMG$ functions in _Pascal_, not a general one.D >I've already found general documentation, but it uses VMS datatypesJ >descriptions and it's not so easy to guess what Pascal types shoud I use. >I use VMS V. 5.4-3.  H It is well documented in the VMS-documenation and in the Pascal manuals.M SMG$-routines use the same calling-standard as all the other system services.nK Thus, if you already wrote programs using LIB$-routines it is just the samen with SMG$-routines.n   Regards,    Christoph Gartmannt  H -----------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, FRG                                               |H +------------ http://www.immunbio.mpg.de/english/menue.html -----------+   ------------------------------  % Date: Thu, 13 Jul 2000 10:33:30 -0600i- From: Lorin Ricker <Lorin.Ricker@t-netix.com>   Subject: Re: VMS Pascal questionI Message-ID: <418E68E524A8D311ACCE00508B78866A7680B2@exchange.t-netix.com>:  J Carl's advice is right on target, Michael.  We've also written extensivelyH in DEC Pascal, using many of the RTL's, including SMG$.  If you run intoG specific problems, please post and I'll be happy to cross-check against < something similar we've done to help and advise if possible.  K You should find a group of Pascal source files in SYS$LIBRARY, all named ofeJ the form PASCAL$<rtl>_ROUTINES.PAS (e.g., substitute SMG for <rtl> in thatL form)... These are provided for many (but not all) of the RTLs during PascalL installation.  I've actually concocted a few more for those that don't "comeI out of the box", like TPU, CLI and MAIL routines; if you're interested, I  could post those that I've got.a  L I also recall a very good book, published by Digital Press (back in the goodG old days), which dealt extensively and exclusively with programming VMS E applications using system services and RTLs *in DEC Pascal* ... we've-K recently moved home and offices from Portland to Denver, so many things arecH still in boxes, and that book's one of them, so I don't recall author or8 title... but maybe someone else can help with a posting?  K Oh, hey, wait!  I just tried searching Amazon.com for "DEC Pascal" (books),uJ and got the hit!  (Is this a great planet or what!)  It's "Writing VAX/VMSF Applications Using Pascal", by Theo de Klerk... go get it at Amazon or5 Powell's or wherever... it's still readily available.I  E Also, be sure to check out the os-doc-set for the manual-let "OpenVMSaI Programming Interfaces: Calling a System Routine".  One of VMS's originalaL strengths (which the Q marketers have fully forgotten, if they ever knew it)G was the idea of "Common Programming Language Environment", meaning that K applications can be "rolled" by programming in several languages, as neededrL and chosen for their strengths and features... a truly modular approach, andI one which supports powerful reusability traits as well.  This means that,tK for instance, you can take some old, yet-useful FORTRAN routines, add a fewgL new C routines, and meld the whole together with a "main program" written inL Pascal, to concoct an application... presuming you attend to standard issuesI of modular interfaces/parameter lists and data structures, etc.  Good VMSrI application designer/programmers can be (should be) fairly agnostic about I "language war" issues, instead focusing on choosing the best/optimum toold for a given task.i  I <rant>  But I'll never forgive Digital for failing to include VAX SCAN ingL the ported-to-Alpha language repertory!  (Yes, I'm aware that it's availableK via DECUS, but it *should'a been* ported, supported and further developed!)  </rant>i  G IMHO, DEC Pascal is an excellent language for writing VMS applications,uG either by itself or in combination with other languages (as above), andiK interfaces very well to system data structures, specifically because PascaleG "understands data" and supports strong data typing so well.  Again, the L above PASCAL$*_ROUTINES and the book give you great guidance... and once youL get the hang of setting up parameter lists and underlying data TYPEs, it all7 becomes routine, yet effective, efficient and reusable.h  H And, although DEC Pascal adheres to the language and academic standards,F this is *not* your professor's weenie little teaching language!  DEC'sE implementation is a full, industrial strength compiler, with numeroustC extensions (great for VMS, but having the usual negative effects onaL "portability", if that's important).  It also shares the use of the same GEMH code-generator back-end that DEC C and FORTRAN (I think) use, so you getK high quality binaries/executables out of it.  All in all, DEC Pascal is onenL of old Digital's best kept secrets, and I doubt that many Q folks understand' what an asset it is in the VMS arsenal.s   regards,  @ Lorin Ricker                            Lorin.Ricker@T-NETIX.com
 Director, JMS 5 T-NETIX, Inc.                           (303)705-5575- 67 Inverness Drive East 8 Englewood, Colorado 80112      http://www.LockTrack.com/   ------------------------------  % Date: Thu, 13 Jul 2000 22:47:02 +05003 From: Dan <dan@vrx.net>m& Subject: VMS Pascal vs Ansi vs Borland' Message-ID: <396E0096.1C2DBF22@vrx.net>m  F I'm an old Borland Pascal (and turbo pascal programmer), although manyA moons ago I did programming on VMS (Vax11/750 vms 3.x and 4.x) soi nothing that applies now.   E I'm having a hard time adjusting to the new VMS pascals. I think I'veyD gotten too used to having things like "blockread, blockwrite" and soC forth. and the string manipulation fun which is the core of all theA programs I write.a   Dan.  $ ---There are always possibilities...   ------------------------------  % Date: Wed, 12 Jul 2000 23:48:24 -0400p* From: David A Froble <davef@tsoft-inc.com>! Subject: Re: VMS tuning wsdefaultn- Message-ID: <396D3C08.AC802CAD@tsoft-inc.com>s   geraldine wrote: > D >  I have a tuning question for OpenVMS 7.1 running on an alpha box.< > We are experiencing what we think is slow performance whenF > installing/linking Oracle RDBMS software on the machine.  When we do' > this we are the only user on the box.u > I > When monitoring the working set usage for the process that is doing theeH > install, it fluctuates wildly and furthermore goes below the wsdefaultI > to almost to values of 200 to 300.  Why is this?  I thought a process's : > working set should NEVER go below the wsdefault setting. > A > Quotas have been jacked up really high to see if that makes anyh8 > difference and doesn't seem to.  Quotas are currently: >  > WSDEFAULT:  500000 > WSQUOTA:    900000 > WSEXTENT:   900000  N First, put your working set default back to some reasonable number, like 500. P If your installation does multiple commands, between each one it must go and getN the 500000 pages, whether you are going to use them or not.  No offense meant,O but this was a really bad move.  Jacking up numbers on a system without knowingDJ what you are doing rarely has the desired result, and usually creates moreM problems.  Of the three, WSEXTENT is the only one you need to set high if you L want a process to have access to large amounts of physical memory. In such aK case, the system is capable of determining what it can use.  Of course, the2P SYSGEN parameters mentioned by others can limit what a specific process can use,& regardless of the working set numbers.  N A bigger hammer isn't always the solution.  It definitely is NOT when it's too big to pick up.n   Dave   -- M4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  # Date: Thu, 13 Jul 2000 15:41:25 GMT)1 From: "Mark D. Jilson" <jilly@clarityconnect.com> ! Subject: Re: VMS tuning wsdefaultt2 Message-ID: <396DE30F.8F9D7C76@clarityconnect.com>  H This is incorrect.  Working set quotas are just gates in the working setA creation & expansion code.  These quotas DO NOT make VMS give theeE process any memory to use.  Again watch the physical memory usage viaeH the PPGCNT & GPGCNT items and the actual working set size via WSSIZE and? you can give us a better idea as to what is actually happening.p   David A Froble wrote: O > First, put your working set default back to some reasonable number, like 500.pR > If your installation does multiple commands, between each one it must go and get? > the 500000 pages, whether you are going to use them or not.      -- lD 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: Thu, 13 Jul 2000 10:10:14 -0400g# From: Jim Agnew <agnew@hsc.vcu.edu>f Subject: Re: What is VMS+ Message-ID: <396DCDC6.2C3F0380@hsc.vcu.edu>v  G You're close, VMS took the best features of Unix and several other os',u8 and you can look up the FAQ that comes up in this list..  C Get an account on that machine, and the HELP command should get you1, started if you really want to play with it..   I *THINK* the faq is ata   http://www.openvms.compaq.com   ! guys, correct me if i'm wronge...d   j.   lasermon wrote:g > L > There is a VAX machine where I work.  I was wondering what VAX is.  I haveK > been told that it is VMS.  What is VMS?  From what I understand, they are M > UNIX like OS's.  What is their relation, are they the same thing, where canc$ > I get more information on the two?   ------------------------------   End of INFO-VAX 2000.389 ************************ developing a socket dispatcher using UCX Auxiliary service. What I want is....K Windows client always initiates. Auxiliary service on OpenVMS listens as in  UNIX inetd on port 7650.  I set up the service as follows:   ucx set service SSDS2 -  /port=7650 - /prot=TCP -r /flag=listen - /user=tcpip$inetd - ) /file=dra1:[users.x.C-=Xܝѱu"wMŁUeiӮTDa#k+e[,>-˺:7Ĳ+ [7/+㨍_&ж*v{QK9SYB[(E`T=ޤgu
zf'|I\̤Da&ŶE<_' dI!Bta^;<mVHh!K~(AXӂwѹ8qذ)	$c<M2DeO;	$eW{I_<&?<{p	O	PJ.H1+[$[`tXu-5'R ɼrꚤT'D*I]R^@"oPF%
p}2viXaS-[5>aԈ&1KY=
<2I6OC2cفVl)@-̑eVBs=bQq\ơˢ+902У[MD4'F=ir@Y*MxM]SHo)6XH>:GYV(
k73#D,vuy:qg5>j<	Kv)^!xKjj+57F90U]wh8[
/-м	*O&k]=?]dPkz3j̺~ VWFq[-_g!qJ^|x+f
nNɜCCD,ޒTFj<GS;!g?e!%Nqaee)-~A0˺T33\Շ䨺o'JzO:&LKIyaK:bmX#}%4zt?mn8o|CZ~c"\lL0~0p͒}ɞ'܏<m'L]A{N"m1em=heh(v,c~se`Ѿ7
8hxֻ:7QRd>?$,hɺ@cܤ2tlJe{yvCKݱ?][*b:kD9.ifFYGAU	_)y?S@1C@KdJSs(84p}{J[Buj'X0
AS{EFm|S" |By9Xe2E5m^MOg`@ O"m!.X
S9q.UZTɑ"?)\J֬vI`B?spN{
2ݗ8c)sT+iG*ǠF9*tdQMYFiN͢C3[2O|W\  8*yWTiz}.B@a`47i==TK;W0E2Ư
z"
M55gl3