1 INFO-VAX	Mon, 08 Sep 2003	Volume 2003 : Issue 497       Contents:3 ANN: some new freeware and freeware updates for VMS 7 Re: ANN: some new freeware and freeware updates for VMS 7 Re: ANN: some new freeware and freeware updates for VMS $ Re: Backup/Copy tree with alias file$ Re: Backup/Copy tree with alias file$ Re: Backup/Copy tree with alias file BCC08 % Re: Choosing between ASTs and Threads * copy bootable CD to disk and boot from it?. Re: copy bootable CD to disk and boot from it? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger? Re: Do we need a DCL debugger?, Re: Documentation for $setenv system service$ Limit Processor utilisation per user( Re: Limit Processor utilisation per user( Re: Limit Processor utilisation per user( Re: Limit Processor utilisation per user: Re: Limit Processor utilisation per user (Class Scheduler): RE: Limit Processor utilisation per user (Class Scheduler), Looking for JAVA-based  "terrminal emulator"9 Re: OpenVMS Itanium system access for developers via DSPP 9 Re: pgp 2.6.3ia multi05 finally running on OpenVMS V7.3-1 C using a FALCO 5000 as a console terminal on an ALPHAstation 255/233 G Re: using a FALCO 5000 as a console terminal on an ALPHAstation 255/233   F ----------------------------------------------------------------------  % Date: Sun, 07 Sep 2003 11:47:20 -0700 + From: Mark Berryman <mark@theberrymans.com> < Subject: ANN: some new freeware and freeware updates for VMS% Message-ID: <3f5b1ac9@cpns1.saic.com>   H The following packages have recently been placed on my server.  You can 
 find them at:   ) http://mvb.saic.com/freeware/submissions/  or ftp://mvb.saic.com/submissions/   H (so named because that is where I told the Freeware Coordinator that is  where he could find them).   Ghostscript V8.11 F packaged as a PCSI install kit with a few fixes so that it will build H correctly on either a VAX or an Alpha.  I also added supporting code so B that you can use the new -dAlignToPixels=0 setting on versions of G DECWindows earlier than V1.3 (which makes for some really nice-looking  G font rendering).  This version will also detect whether or not you are  B running a version of DECWindows that still has Display Postscript 6 support and build the ghostscript fontmap accordingly.   Mozplugger V1.3.0 G This is a port of a program, written for Linux, that allows standalone  G programs to be used as a plugin for Mozilla.  For example, the default  F config uses XPDF to display PDF files.  Mozplugger will activate XPDF C and then place it in the browser window.  MPG123 and Timidity also  E provide great support for viewing those web pages with sound on them.    MPG123 v0.59s-mh4 > I fetched a copy of MPG123 from the author's web site (which, I apparently, has not been touched since Oct 2000) and fixed the code that  I allows one to specify a URL as the MP3 filename so that it works on VMS.  H   I also rewrote the sound support code for VMS since the original code F was constantly allocating and deallocating buffers and eventually ran D into memory allocation errors if you played a long enough playlist. D With the addition of URL support, MPG123 can now be used to support $ streaming MP3 files with Mozplugger.  
 XPDF v2.02pl1 > The fixes here were mainly in the build procedure (there is a E configure.com that performs the VMS equivalent of GNU configure that  B needed several updates) as well as some fixes to T1lib so that it H correctly finds all of the font files XPDF tries to pass to it.  If you G have ghostscript, edit the XPDFRC file to use the font files that come  < with ghostscript and set the two variables t1libControl and I freetypeControl to high (this is documented in the file).  Doing so will  E cause XPDF to render anti-aliased fonts that look as nice as acrobat   reader does on the PC.  I All of the packages build without error on any supported version of VMS.  H   I do not have the resources to test on earlier versions.  However, if F you try to build them on earlier versions and they don't work, let me 4 know what errors you get and I will try to fix them.  
 Mark Berryman    ------------------------------  % Date: Sun, 07 Sep 2003 22:15:25 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> @ Subject: Re: ANN: some new freeware and freeware updates for VMS' Message-ID: <3F5BF44D.D28389B2@fsi.net>    Mark Berryman wrote: > I > The following packages have recently been placed on my server.  You can  > find them at:  > + > http://mvb.saic.com/freeware/submissions/  > or! > ftp://mvb.saic.com/submissions/  > I > (so named because that is where I told the Freeware Coordinator that is  > where he could find them). >  > Ghostscript V8.11 G > packaged as a PCSI install kit with a few fixes so that it will build I > correctly on either a VAX or an Alpha.  I also added supporting code so C > that you can use the new -dAlignToPixels=0 setting on versions of H > DECWindows earlier than V1.3 (which makes for some really nice-lookingH > font rendering).  This version will also detect whether or not you areC > running a version of DECWindows that still has Display Postscript 8 > support and build the ghostscript fontmap accordingly. >  > Mozplugger V1.3.0 H > This is a port of a program, written for Linux, that allows standaloneH > programs to be used as a plugin for Mozilla.  For example, the defaultG > config uses XPDF to display PDF files.  Mozplugger will activate XPDF D > and then place it in the browser window.  MPG123 and Timidity alsoG > provide great support for viewing those web pages with sound on them.  >  > MPG123 v0.59s-mh4 ? > I fetched a copy of MPG123 from the author's web site (which, J > apparently, has not been touched since Oct 2000) and fixed the code thatJ > allows one to specify a URL as the MP3 filename so that it works on VMS.I >   I also rewrote the sound support code for VMS since the original code G > was constantly allocating and deallocating buffers and eventually ran E > into memory allocation errors if you played a long enough playlist. E > With the addition of URL support, MPG123 can now be used to support & > streaming MP3 files with Mozplugger. >  > XPDF v2.02pl1 ? > The fixes here were mainly in the build procedure (there is a F > configure.com that performs the VMS equivalent of GNU configure thatC > needed several updates) as well as some fixes to T1lib so that it I > correctly finds all of the font files XPDF tries to pass to it.  If you H > have ghostscript, edit the XPDFRC file to use the font files that come= > with ghostscript and set the two variables t1libControl and J > freetypeControl to high (this is documented in the file).  Doing so willF > cause XPDF to render anti-aliased fonts that look as nice as acrobat > reader does on the PC. > J > All of the packages build without error on any supported version of VMS.I >   I do not have the resources to test on earlier versions.  However, if G > you try to build them on earlier versions and they don't work, let me 6 > know what errors you get and I will try to fix them.  D Any chance you can put up object code distributions like Hunter does with some packages?    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sun, 07 Sep 2003 22:43:38 -0700 + From: Mark Berryman <mark@theberrymans.com> @ Subject: Re: ANN: some new freeware and freeware updates for VMS% Message-ID: <3f5bb49b@cpns1.saic.com>    David J. Dachtera wrote: > Mark Berryman wrote: >  [package descriptions deleted] > F > Any chance you can put up object code distributions like Hunter does > with some packages?   I Actually, I left the .OBJ, .OLB, and .EXE files I generated when I built  E each package on my system in the directory before I zipped it up.  I  I could definitely reorganize the package but I was running into deadlines  < for the Freeware CD and wanted to get something working out.  I So, if you fetch the package and unzip it, you will find executables and  9 object files (or object libraries) somewhere in each kit.   
 Mark Berryman    ------------------------------  % Date: Mon, 08 Sep 2003 02:33:59 +0800 , From: Paul Repacholi <prep@prep.synonet.com>- Subject: Re: Backup/Copy tree with alias file - Message-ID: <87llt0mors.fsf@prep.synonet.com>   ( David M Smith <dsmit115@csc.com> writes:  : > On Thu, 04 Sep 2003 14:43:54 +0200 (MET), Phillip Helbig- > <HELBPHI@sysdev.deutsche-boerse.com> wrote:   0 >>Well, DISK:[SYSn]SYSCOMMON.DIR is an alias forD >>DISK:[000000]VMS$COMMON.DIR.  (Actually, since both of these occurE >>in logicals defined as search lists, I've never understood why this A >>is necessary; perhaps it goes back to the very old days of VMS, / >>before search-list logicals were implemented?   : > This is a question I've often wondered about, too -- why@ > (technically) are the SYSCOMMON.DIR aliases setup on a cluster= > common system disk? As you point out, with the logical name B > SYS$SYSROOT defined as a search list, I don't see any reason why > this is/was done.   E > I've never found anyone who can make even a plausible response. I'd , > love to find out from one of the HP folks.  E This dates back to V3/V4 days and the initial cluster code that would > use [SYSn.foo] and then [SYSn.SYSCOMMON.foo] to find stuff. NoE logicals need apply. In fact, a niggling memory is trying to say that F at the time the code did this, logicals had not been initialised. ThisE is really really early in the boot process. Search lists existed, but C did not have anything in them yet. Search lists happened in 3.[456] * I think. Before 3.7, aka the cluster of 1.  8 And no, I'm not going to break out the fiche to check :)   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------   Date: 7 Sep 2003 14:19:00 -0700 . From: spamsink2001@yahoo.com (Alan E. Feldman)- Subject: Re: Backup/Copy tree with alias file = Message-ID: <b096a4ee.0309071319.3cf17ad4@posting.google.com>   a Paul Repacholi <prep@prep.synonet.com> wrote in message news:<87llt0mors.fsf@prep.synonet.com>... * > David M Smith <dsmit115@csc.com> writes: > < > > On Thu, 04 Sep 2003 14:43:54 +0200 (MET), Phillip Helbig/ > > <HELBPHI@sysdev.deutsche-boerse.com> wrote:  >   2 > >>Well, DISK:[SYSn]SYSCOMMON.DIR is an alias forF > >>DISK:[000000]VMS$COMMON.DIR.  (Actually, since both of these occurG > >>in logicals defined as search lists, I've never understood why this C > >>is necessary; perhaps it goes back to the very old days of VMS, 1 > >>before search-list logicals were implemented?  >   < > > This is a question I've often wondered about, too -- whyB > > (technically) are the SYSCOMMON.DIR aliases setup on a cluster? > > common system disk? As you point out, with the logical name D > > SYS$SYSROOT defined as a search list, I don't see any reason why > > this is/was done.  >   G > > I've never found anyone who can make even a plausible response. I'd . > > love to find out from one of the HP folks. > G > This dates back to V3/V4 days and the initial cluster code that would @ > use [SYSn.foo] and then [SYSn.SYSCOMMON.foo] to find stuff. NoG > logicals need apply. In fact, a niggling memory is trying to say that H > at the time the code did this, logicals had not been initialised. ThisG > is really really early in the boot process. Search lists existed, but E > did not have anything in them yet. Search lists happened in 3.[456] , > I think. Before 3.7, aka the cluster of 1. > : > And no, I'm not going to break out the fiche to check :)    @ So why not have the initial cluster code use [SYSn.foo] and thenE [VMS$COMMON.foo] ? (Substitute VMS$COMMON with whatever it was called  back then.)     F The question remains: Why is [VMS$COMMON] aliased as [SYSn.SYSCOMMON]?     Disclaimer: JMHO Alan E. Feldman    ------------------------------   Date: 7 Sep 2003 18:14:27 -0500 - From: Kilgallen@SpamCop.net (Larry Kilgallen) - Subject: Re: Backup/Copy tree with alias file 3 Message-ID: <UzoV7QQVVvYJ@eisner.encompasserve.org>   n In article <b096a4ee.0309071319.3cf17ad4@posting.google.com>, spamsink2001@yahoo.com (Alan E. Feldman) writes:  B > So why not have the initial cluster code use [SYSn.foo] and thenG > [VMS$COMMON.foo] ? (Substitute VMS$COMMON with whatever it was called 
 > back then.)   A That would prevent having two different versions of the operating ? system on the same disk.  (That is not presently supported, but F this piece of VMS does not prevent it, and to the best of my knowledgeC the only piece of VMS that prevents it is upgrade and installation.   H > The question remains: Why is [VMS$COMMON] aliased as [SYSn.SYSCOMMON]?  . I thought I explained that in my earlier post.   ------------------------------  * Date: Sun, 7 Sep 2003 21:30:19 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: BCC08$ Message-ID: <bjg81b$skt$2@online.de>  I I have a BCC08 cable.  It doesn't seem to work as a serial cable between  E a terminal and a VMS machine.  Am I correct that this is a "special"  F cable since pins 8 and 9 are shorted together and thus normally won't 2 work (but will on a VAXstation 2000 or something)?  ( Is there any other use for such a cable?   ------------------------------  $ Date: Sun, 7 Sep 2003 23:43:52 -0400* From: "Bill Todd" <billtodd@metrocast.net>. Subject: Re: Choosing between ASTs and Threads2 Message-ID: <4cOcnZYx9JFgZ8aiXTWJkA@metrocast.net>  9 "Vince" <vince@the[NOSPAM]berrymans.com> wrote in message 7 news:IoB6b.10527$lY2.9532@newssvr29.news.prodigy.com... F > I am new to VMS and am currently trying to learn some of the ways inD > which programming is unique on VMS.  One question I am looking forF > clarification on is the difference between using ASTs and threads on! > systems with more than one CPU.  > G > If I read it right, a program designed around ASTs cannot make use of J > more than one CPU at a time since only one AST can execute at a time and= > the main program can't execute while an AST is in progress.  > G > With threads, however, it appears I can have as many sections of code  > executing as there are CPUs. >  > Is my understanding correct?  H While I haven't used VMS since long before kernel threads were supportedG there, on other OSs which schedule process threads in the kernel (or in J kernel code itself) the performance-optimal solution tends to be something like the following:   J 1.  Create as many threads in your process as the number of processors youD expect to be available to it (this may be less than the total numberG available, if other processes or system tasks will need one or two; you H don't want many *more* threads than available processors, since that can? increase contention among them - i.e., effectively increase the I multi-programming level - beyond what is necessary to keep the processors  busy).  K 2.  Create a central work queue from which the threads can obtain things to  do.   I 3.  Within each thread, write the same AST-oriented code you'd write in a G single-threaded implementation (modulo any inter-thread synchronization I requirements), using asynchronous I/O and picking a new item off the work G queue each time the thread would otherwise suspend for synchronous I/O. I This allows each thread (= each processor) to remain busy all the time as 6 long as there's enough work to keep them all occupied.  K Whether the AST routines continue the operation at AST level or simply dump J the appropriate continuation record onto your work queue is an interestingG choice:  if the former, one thread (processor) can become overloaded by H backed-up AST activities (assuming that ASTs are always delivered to theJ thread that caused them) while the other threads (processors) sit idle; ifG the latter, you may lose the benefit of processor (and processor cache) L affinity as continuation activities move to other processors from the one onI which the operation previously executed (though if other AST activity has H already made use of the cache this may not be all that disadvantageous).  K If your statement is correct that only a single AST can execute at one time K in a process no matter how many threads it has, then the choice is obvious: L use the AST as briefly as possible to dump the continuation information onto your work queue.   - bill   ------------------------------  * Date: Sun, 7 Sep 2003 22:33:20 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)3 Subject: copy bootable CD to disk and boot from it? $ Message-ID: <bjgbng$vq5$1@online.de>  G I'm trying to figure out the best way to upgrade VAX systems with no CD D drive on them.  (OK, in the long run I'll probably have every systemE disk a shadow set and just do one upgrade on a system with a CD drive D then upgrade the other systems by using the new disk as a new shadowG master, but that's in the future for now.)  Can I do an image backup of H the distribution CD to a disk then boot from this disk as if it were the> CD (i.e. get into the menu etc)?  (Even though it might not beF supported, I know I can copy the .% savesets to disk and upgrade from I that, but I'd like to avoid that hassle.)  Would this also work on ALPHA  ; in the event I don't have a (working) CD drive on a system?    ------------------------------   Date: 7 Sep 2003 18:17:24 -0500 - From: Kilgallen@SpamCop.net (Larry Kilgallen) 7 Subject: Re: copy bootable CD to disk and boot from it? 3 Message-ID: <oss9XW0PenAI@eisner.encompasserve.org>   w In article <bjgbng$vq5$1@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: I > I'm trying to figure out the best way to upgrade VAX systems with no CD  > drive on them.   Use an Infoserver.  I > master, but that's in the future for now.)  Can I do an image backup of J > the distribution CD to a disk then boot from this disk as if it were the" > CD (i.e. get into the menu etc)?   Yes.    > Would this also work on ALPHA = > in the event I don't have a (working) CD drive on a system?    Yes.   ------------------------------  % Date: Mon, 08 Sep 2003 02:14:07 +0800 , From: Paul Repacholi <prep@prep.synonet.com>' Subject: Re: Do we need a DCL debugger? - Message-ID: <87u17ompow.fsf@prep.synonet.com>   . lewis@mazda.mitre.org (Keith A. Lewis) writes:   > Mark Berryman <Mark.Berryman@Mvb.Saic.Com> writes in article <3f562eb9$1@cpns1.saic.com> dated Wed, 03 Sep 2003 18:11:05 -0700: H >>   o If line-editing across a line-wrap can't be fixed in the terminal% >>     driver, work around it in DCL.  > A > I second that!  Unix and Linux "bash" shells have it already.     / TECO has been able to do it for about 20 years.    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Mon, 08 Sep 2003 02:25:04 +0800 , From: Paul Repacholi <prep@prep.synonet.com>' Subject: Re: Do we need a DCL debugger? - Message-ID: <87pticmp6n.fsf@prep.synonet.com>   / Kilgallen@SpamCop.net (Larry Kilgallen) writes:   D > Presuming you mean "qualifiers", why not use a compiled language ?   >> And just like Unix.  C > I would guess _not_ "just like Unix".  Unix expansion of filespec B > wildcards is handled quite differently from VMS.  Instead of "/"B > Unix uses "-", and I don't know enough about Unix to say how, or0 > whether, it handles syntax for something like:  2 > 	/SELECT=(SIZE=(MINIMUM=500,MAXIMUM=1000),NOACL)  E "Just like unix" means every utility rolls its own. See the man pages F for dd for the egarious example. Or write a unix program that uses VMS' syntax. It is all a SMO(non standard)P.    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Sun, 07 Sep 2003 20:43:14 GMT * From: Michael Lemke <lemkemch@t-online.de>' Subject: Re: Do we need a DCL debugger? . Message-ID: <1103_1062967394@news.t-online.de>  V On Sat, 06 Sep 2003 20:43:39 -0500, "David J. Dachtera" <djesys.nospam@fsi.net> wrote: > Michael Lemke wrote: > > Z > > On Wed, 03 Sep 2003 21:12:44 -0500, "David J. Dachtera" <djesys.nospam@fsi.net> wrote: > > > Mark Berryman wrote: > > > >  > > > > Mark Berryman wrote: > > > > > Guy Peleg wrote:	 > > > > > N > > > > >> The DCL engineering team is currently evaluating new features to be8 > > > > >> added in future version of VMS (post V7.3-2).
 > > > > >>R > > > > >> We were wondering if folks think that developing a DCL debugger will be/ > > > > >> useful. It won't be as sophisticated R > > > > >> as the VMS debugger, I was thinking about implementing step,examine and > > > > >> set break. 	 > > > > > 	 > > > > > 	 > > > > > O > > > > >   o I'd like a command (SET PROCESS/PARSE=PIPE_ASSUMED ?) that causes N > > > > >     DCL parsing to always assume the command was prefixed with PIPE. > > >  > > > That WOULD be cool, yes. > > G > > Absolutely.  I am astonished it isn't there yet.  Seems so obvious.  >  > A later thought: how 'bout:  >  > $ SET [NO]PIPE >  > ....rather like SET [NO]ON?  > 
 > Example: >  > $ SET PIPE$ > $ SHOW SYSTEM | SEARCH SYS$PIPE RW6 > $ IF (&$STATUS .AND. %XFF) .NE. 1 THEN EXIT &$STATUS > $ WRITE SYS$OUTPUT "% Oops!"! > $ SHOW SYSTEM > SHOW_SYSTEM.LIS  > $ EXIT >   < Yes, looks better.  But whatever the syntax point is to get , get some way to make that PIPE verb default.  ! > > > > I thought of another one:  > > > > H > > > > A P0 parameter that contains the entire command that invoked theI > > > > procedure, exactly as typed.  (If not P0, then a lexical that can  > > > > retrieve it.)  > > >  > > ; > > We need more: the equivalent of Unix' "$@"  Currently I ; > > can't write a wrapper procedure around an image because ? > > I can't pass the parameters properly.  I consider this much ' > > more important than a DCL debugger.  > E > Don't have access to a running UN*X machine just now. What's $@ do?   9 Well, it is a little complicated to explain which is why  < I didn't elaborate and assumed you know.  Let me try in DCL:8 Let's call the $@ parameter P@.  Then I can do in a .COM procedure PROC.COM:    $ @ANOTHERPROC "''P@'"  0 and ANOTHERPROC would see as P1 exactly what was3 passed as P1 to PROC.COM, including all spaces that 7 possibly are in the expansion of P1.  P2 would be P2 of 9 the caller and so on.  In other words I can transparently ; pass on parameters.  Currently I can't.  "''P1'" will break : if P1 = "bla""bla" or somesuch, i.e., if there is a clever4 combination of double quiotes and spaces.  That will2 confuse DCL royally.  Sorry I can't come up with a7 real example, has been too long (years) that I ran into 5 that and I gave up after I realized it can't be done.   9 The P@ would also allow to pass qualifiers via procedures & to images.  E.g. a SUBMIT.COM could to   $ DEFINE SYS$PRINT NL: $ SUBMIT "''P@'"  1 and the thing could be used exactly like the real 7 submit command with all qualifiers supported by SUBMIT.    > > > > And while I am at it: why not access to all CLI$ routines.: > > Then I could write a procedure with switches and stuff, > > just like an image.  And just like Unix. > 7 > What's missing among the CLI$ routines that you need?   4 I want to use them in a command procedure.  Not just? P1 to P8.  So I can do @PROC /TEFF=9700 /MIX=(OVER=1,LENGTH=17) 4 So there is no difference between a procedure and a 7 complied program (except from the @).  That is actually < what I meant with `just like Unix'.  There it doesn't matterB whether I am calling a script or a program.  I can't tell from the syntax.    > C > FYI: I don't recall how just now, but you can ask one of the CLI$ H > routines to pass you the verb from the command line. This is (I shouldF > think, dunno fer sure) how SUBMIT.EXE knows whether to behave like aJ > SUBMIT command or a PRINT command, aside from the obvious differences inH > the .CLD code. Both verbs are served by the same image. Likewise, COPY' > and APPEND, DELETE and PURGE, others.  > F > I believe one of the CLI$ calls will return the entire command line.A > Maybe just LIB$GET_FOREIGN will do the same. Not sure just now.   = All yes, but access is from a compiled language.  I want DCL.    Michael    ------------------------------  # Date: Sun, 07 Sep 2003 20:50:43 GMT * From: Michael Lemke <lemkemch@t-online.de>' Subject: Re: Do we need a DCL debugger? . Message-ID: <1104_1062967843@news.t-online.de>  V On Sat, 06 Sep 2003 21:06:12 -0500, "David J. Dachtera" <djesys.nospam@fsi.net> wrote: > Marc Van Dyck wrote: > >  > > Another few suggestions :  > > 2 > > - Automatic command completion, like in Bash ; > % > Why bother when you can abbreviate?     I can't abbreviate file names.    B The completion would also list all possible qualifiers.  In fact, 9 DCL could do this much easier than Unix because DCLTABLES ? has all the information.  Unix doesn't know the options so bash B has to be taught first.  Actually, not sure bash can do this.  ButC tcsh does (my standard shell, I configured it to do command editing C like EDT. Rather ironic Unix does this better than the real thing.)    Michael    ------------------------------  # Date: Sun, 07 Sep 2003 22:28:14 GMT > From: Michael Austin <maustin@no-more-spam.firstdbasource.com>' Subject: Re: Do we need a DCL debugger? 9 Message-ID: <2iO6b.495$vV.228@newssvr23.news.prodigy.com>    JF Mezei wrote:  > Tom Linden wrote:  > , >>Why not port bc from unix and pipe to it?  >  > K > If you really want VMS to take off big time, port Microsoft Office to VMS P > (cough, shudder). Disable/remove all the dangerous features and all those thatO > don't abide by standards and then make it a robust trustable system. You'd be C > selling VMS workstations by the tens of thousands to enterprises.  > G > VMS would have a working Office while Linux would be left out to dry.    Jeff,   C Have you tried using OpenOffice from OpenOffice.Org?  it reads and  G writes MSOffice files without the bugs in MSOffice... Runs on Windows,  C Linux and MAC OS X.  I have used it for a while and it works great.      Michael Austin.  reply-to address munged.   ------------------------------  $ Date: Sun, 7 Sep 2003 19:12:27 -0400% From: "John Vottero" <John@mvpsi.com> ' Subject: Re: Do we need a DCL debugger? / Message-ID: <vlneqs4tl8oh25@news.supernews.com>   < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3F5A8977.4558C24A@fsi.net... ! > david20@alpha1.mdx.ac.uk wrote:  > > ? > > In article <3F5806A4.A555F66F@fsi.net>, "David J. Dachtera"  <djesys.nospam@fsi.net> writes:  > > >Martin Hunt wrote: 
 > > >> [snip] L > > >> Ability to easily change the default SUBMIT behaviour to /NOPRINT - IL > > >> know there are workarounds, but it is frustrating to wonder where theL > > >> log file has gone, then realise that it has been printed and deleted.< > > >> Does anyone actually want their log files doing this? > > > G > > >One not-so-obvious work-around for that is to not have a SYS$PRINT  > > >queue.  > > > J > > >You can also use the VERB utility to extract the definition of SUBMITL > > >and change the default value of the PRINT qualifier to be the name of a > > >non-existant queue. > > > * > > >More than one way to skin that cat... > > >  > > I > > How about SUBMIT/REMOTE to a queue other than sys$batch on the remote  nodeK > > together with a /NOLOG or /NOPRINT to stop getting log files printed on  the  > > remote node. >  > I might request that as: > C > $ SUBMIT/REMOTE=(NODE=node_name,QUEUE=queue_name,USER=username) - . > local_filespec/NOPRINT/PARA=(p1,p2,p3...,p8) >   H Go to http://www.mvpsi.com and download JAMS.  As long as you're runningJ less than 25 jobs per day, the free license is all you need.  Then you can do things like:   F $ JAMS SUBMIT REMNODE::REMJOB/QUE=blah/USER=blah/PARAM=(blah, blah...)  G It's not exactly what you're looking for but it is quite a step up from  SUBMIT/REMOTE.  L While you can use most of the SUBMIT qualifiers with the JAMS SUBMIT commandK (remote or local), you probably won't need to.  A JAMS Job definition knows H the values you want for the qualifiers so you'll probably wind up doing:   $ JAMS SUBMIT REMNODE::REMJOB    John Vottero JAMS Developer   ------------------------------  % Date: Sun, 07 Sep 2003 19:27:32 +0200 2 From: martin@radiogaga.harz.de (Martin Vorlaender)' Subject: Re: Do we need a DCL debugger? ; Message-ID: <3f5b6a84.524144494f47414741@radiogaga.harz.de>   0 David J. Dachtera (djesys.nospam@fsi.net) wrote: > Marc Van Dyck wrote: > > Another few suggestions :  > > 2 > > - Automatic command completion, like in Bash ; > % > Why bother when you can abbreviate?   E You've got a point there. But there's more to it: bash also completes 8 path and file names. Which could prove useful sometimes.   cu,    Martin --  D                        |  Martin Vorlaender  |  VMS & WNT programmer1   OpenVMS: When you    |  work: mv@pdv-systeme.de E   KNOW where you want  |     http://www.pdv-systeme.de/users/martinv/ 8   to go today.         |  home: martin@radiogaga.harz.de   ------------------------------   Date: 7 Sep 2003 21:34:04 -0500 - From: Kilgallen@SpamCop.net (Larry Kilgallen) ' Subject: Re: Do we need a DCL debugger? 3 Message-ID: <0vTxTiuhylNN@eisner.encompasserve.org>   p In article <3f5b6a84.524144494f47414741@radiogaga.harz.de>, martin@radiogaga.harz.de (Martin Vorlaender) writes:2 > David J. Dachtera (djesys.nospam@fsi.net) wrote: >> Marc Van Dyck wrote:  >> > Another few suggestions : >> >  3 >> > - Automatic command completion, like in Bash ;  >>& >> Why bother when you can abbreviate? > G > You've got a point there. But there's more to it: bash also completes : > path and file names. Which could prove useful sometimes.  C I prefer that my typographic errors be detected rather than hidden.    ------------------------------  % Date: Sun, 07 Sep 2003 22:00:56 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> ' Subject: Re: Do we need a DCL debugger? ' Message-ID: <3F5BF0E8.4D64EF9C@fsi.net>    Larry Kilgallen wrote: > ] > In article <1103_1062894093@news.t-online.de>, Michael Lemke <lemkemch@t-online.de> writes: 
 > > [snip] > > And just like Unix.  > C > I would guess _not_ "just like Unix".  Unix expansion of filespec B > wildcards is handled quite differently from VMS.  Instead of "/"B > Unix uses "-", and I don't know enough about Unix to say how, or0 > whether, it handles syntax for something like: > 9 >         /SELECT=(SIZE=(MINIMUM=500,MAXIMUM=1000),NOACL)   F In UN*X-land, the "shell" (a.k.a. "CLI") doesn't really care about "-"C or anything else other than its own metacharacters. It's up to each D individual command program to get the command line and interpret it.  F UN*X doesn't really have a "CLI" as we know it in the form of DCL. The< "layer" between the kernel and the user is called a "shell".   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sun, 07 Sep 2003 22:07:51 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> ' Subject: Re: Do we need a DCL debugger? ' Message-ID: <3F5BF287.BDAE2D8C@fsi.net>    Martin Vorlaender wrote: > 2 > David J. Dachtera (djesys.nospam@fsi.net) wrote: > > Marc Van Dyck wrote: > > > Another few suggestions :  > > > 4 > > > - Automatic command completion, like in Bash ; > > ' > > Why bother when you can abbreviate?u > G > You've got a point there. But there's more to it: bash also completesa: > path and file names. Which could prove useful sometimes.   How are ambiguities resolved?e  < Example: Selecting between elements of a search list such asC SYS$SYSROOT. I want to edit files in both SYS$SPECIFIC:[SYSMGR] andw8 SYS$COMMON:[SYSMGR]; however, both of those are found asG SYS$SYSROOT:[SYSMGR] or SYS$MAMAGER:. How would the CLI know which file ? I want? I still have to type the entire filespec to resolve the 
 ambiguity.  B Could be useful to some, but I might find little or no use for it.   ...IMHO.   -- a David J. Dachtera  dba DJE Systemsd http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/e   ------------------------------  % Date: Sun, 07 Sep 2003 22:11:45 -0500w1 From: "David J. Dachtera" <djesys.nospam@fsi.net>n' Subject: Re: Do we need a DCL debugger?t' Message-ID: <3F5BF371.E34E2331@fsi.net>t   Michael Lemke wrote: > X > On Sat, 06 Sep 2003 21:06:12 -0500, "David J. Dachtera" <djesys.nospam@fsi.net> wrote: > > Marc Van Dyck wrote: > > >u > > > Another few suggestions :t > > > 4 > > > - Automatic command completion, like in Bash ; > >e' > > Why bother when you can abbreviate?s >   > I can't abbreviate file names. > C > The completion would also list all possible qualifiers.  In fact, ; > DCL could do this much easier than Unix because DCLTABLESC > has all the information.    D Not always. You may find some layered products (DCSC is one example)G where the .CLD simply gives P1 as $rest_of_line, and the command stringrG itself is parsed internally. In that specific case, extracting the .CLDrD from the image does yield useful info., but you can't count on that.  ' > Unix doesn't know the options so basheD > has to be taught first.  Actually, not sure bash can do this.  ButE > tcsh does (my standard shell, I configured it to do command editingaE > like EDT. Rather ironic Unix does this better than the real thing.)r  C As I mentioned in another post, I'd likely find too many situationsl@ where ambiguous filespecs would result in incorrect completions.   -- r David J. Dachteraa dba DJE Systemso http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/1   ------------------------------  % Date: Mon, 08 Sep 2003 06:24:57 +0200i2 From: martin@radiogaga.harz.de (Martin Vorlaender)' Subject: Re: Do we need a DCL debugger?D; Message-ID: <3f5c0499.524144494f47414741@radiogaga.harz.de>   + Michael Lemke (lemkemch@t-online.de) wrote:o4 > "David J. Dachtera" <djesys.nospam@fsi.net> wrote: > > Marc Van Dyck wrote: > > >  > > > Another few suggestions :t > > > 4 > > > - Automatic command completion, like in Bash ; > > ' > > Why bother when you can abbreviate?e >o" > I can't abbreviate file names.   > D > The completion would also list all possible qualifiers.  In fact, ; > DCL could do this much easier than Unix because DCLTABLES  > has all the information.  C Not so fast. There's DCLTABLES (but with the undocumented "cliflagst? (foreign)" CLD statement), then there are foreign commands, andr there is DCL$PATH.  ? I don't think qualifier completion would be (largely) feasible.    cu,s   Martin -- nG  Your mouse has moved.     | Martin Vorlaender  |  VMS & WNT programmerr4  Windows must be restarted | work: mv@pdv-systeme.deH  for the change to take    |    http://www.pdv-systeme.de/users/martinv/;  effect. Reboot now? [OK]  | home: martin@radiogaga.harz.de    ------------------------------   Date: 8 Sep 2003 06:40:10 GMTo/ From: "Dave Weatherall" <djweath@attglobal.net>e5 Subject: Re: Documentation for $setenv system service 5 Message-ID: <DTiotGxQ0bj6-pn2-neyj6s1sd7Cv@localhost>i  C On Sun, 7 Sep 2003 00:25:23 UTC, Z  <zarlenga@conan.ids.net> wrote:s  * > Howard S Shubs <howard@shubs.net> wrote:L > : From what I've been able to determine from looking at the docs (which I I > : did for this purpose about a year ago), the $SETENV service can only eL > : modify a subset of console environment variables.  That subset does NOT  >  > Where is $SETENV documented? > / > There's no mention of it in the Master Index.r  F Sadly, I suspect the Master Index is not what it once was. I remember > being disappointed earlier this year when I  went looking for 
 something.   -- r Cheers - Dave.   ------------------------------  $ Date: Sun, 7 Sep 2003 23:52:09 +02002 From: "AWN_VSIS" <anty_SPAM_email_adres@email.com>- Subject: Limit Processor utilisation per userf6 Message-ID: <3f5ba886$0$49108$e4fe514c@news.xs4all.nl>  B Is it possible to limit a user tot a maximum of for example 30% of8 processing time on an Alpha800 running OPEN VMS 7.1.2 ??  6 Please respond to awakeren@xs4all.nl or in this group.  	 A newbee,r Arjan:   ------------------------------  $ Date: Sun, 7 Sep 2003 23:54:29 +02002 From: "AWN_VSIS" <anty_SPAM_email_adres@email.com>1 Subject: Re: Limit Processor utilisation per usera6 Message-ID: <3f5ba902$0$49112$e4fe514c@news.xs4all.nl>  L Forgot to tell, the user is connecting with a TELNET session and has its own useraccount.    ? "AWN_VSIS" <anty_SPAM_email_adres@email.com> schreef in berichts0 news:3f5ba886$0$49108$e4fe514c@news.xs4all.nl...D > Is it possible to limit a user tot a maximum of for example 30% of: > processing time on an Alpha800 running OPEN VMS 7.1.2 ?? >s8 > Please respond to awakeren@xs4all.nl or in this group. >t > A newbee,l > Arjan  >p >g >V >.   ------------------------------  # Date: Sun, 07 Sep 2003 22:37:50 GMTo> From: Michael Austin <maustin@no-more-spam.firstdbasource.com>1 Subject: Re: Limit Processor utilisation per usere: Message-ID: <2rO6b.498$mk1.249@newssvr23.news.prodigy.com>   AWN_VSIS wrote:   N > Forgot to tell, the user is connecting with a TELNET session and has its own > useraccount. >  > A > "AWN_VSIS" <anty_SPAM_email_adres@email.com> schreef in berichtw2 > news:3f5ba886$0$49108$e4fe514c@news.xs4all.nl... > D >>Is it possible to limit a user tot a maximum of for example 30% of: >>processing time on an Alpha800 running OPEN VMS 7.1.2 ?? >>8 >>Please respond to awakeren@xs4all.nl or in this group. >> >>A newbee,l >>Arjann >> >> >> >> >  >  >    In AUTHORIZE see:o UAF> HELP MOD /CPUTIME MODIFY      /CPUTIMEe            /CPUTIME=time  E       Specifies the maximum process CPU time for the CPU field of thenG       UAF record. The maximum process CPU time is the maximum amount of F       CPU time a user's process can take per session. You must specifyF       a delta time value. For a discussion of delta time values, referD       to the OpenVMS User's Manual. The default is 0, which means an       infinite amount of time.   ------------------------------  % Date: Sun, 07 Sep 2003 22:19:29 -0500l1 From: "David J. Dachtera" <djesys.nospam@fsi.net>r1 Subject: Re: Limit Processor utilisation per user ' Message-ID: <3F5BF541.4FA42ADD@fsi.net>s   Michael Austin wrote:e >  > AWN_VSIS wrote:  > P > > Forgot to tell, the user is connecting with a TELNET session and has its own > > useraccount. > >s > >tC > > "AWN_VSIS" <anty_SPAM_email_adres@email.com> schreef in berichtr4 > > news:3f5ba886$0$49108$e4fe514c@news.xs4all.nl... > >rF > >>Is it possible to limit a user tot a maximum of for example 30% of< > >>processing time on an Alpha800 running OPEN VMS 7.1.2 ?? > >>: > >>Please respond to awakeren@xs4all.nl or in this group. > >>
 > >>A newbee, 	 > >>Arjan  > >> > >> > >> > >> > >n > >i > >f >  > In AUTHORIZE see:: > UAF> HELP MOD /CPUTIME  F Not quite. He wants to limit an active process to no more than 30% CPUE Util at any given time, not a limit on CPU time total for the processA	 lifetime.w  G To answer the OP, not as you might think. There are parameters that canmB be manipulated, process priority being the most obvious. This willD control how often a process gets scheduled into the CPU, but not how% much it will use during those slices.m   -- g David J. Dachtera  dba DJE SystemsL http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/p   ------------------------------   Date: 7 Sep 2003 18:15:42 -0500n- From: Kilgallen@SpamCop.net (Larry Kilgallen)aC Subject: Re: Limit Processor utilisation per user (Class Scheduler)d3 Message-ID: <IWDpaL07wtcy@eisner.encompasserve.org>l  k In article <3f5ba886$0$49108$e4fe514c@news.xs4all.nl>, "AWN_VSIS" <anty_SPAM_email_adres@email.com> writes:o  D > Is it possible to limit a user tot a maximum of for example 30% of: > processing time on an Alpha800 running OPEN VMS 7.1.2 ??  G You need to look for "Class Scheduler" in the documentation.  There are>E two different ways to do it, both too complex to avoid looking at the : documentation.  The Master Index is (usually) your friend.   ------------------------------  $ Date: Sun, 7 Sep 2003 22:48:40 -0400' From: "Main, Kerry" <kerry.main@hp.com>lC Subject: RE: Limit Processor utilisation per user (Class Scheduler)sR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB0C7BA3@tayexc19.americas.cpqcorp.net>  F Re: VMS class scheduler in V7.1-2 .. Not sure what version it came in, but -o   SYSMAN> help class=20    If error then exit   Regardsu  
 Kerry Main Senior Consultanti Hewlett-Packard (Canada) Co.! Consulting & Integration Servicese Voice: 613-592-4660  Fax   : 613-591-4477 Email: kerryDOTmain@hpDOTcom-     (remove the DOT's and replace with "."'s)u OpenVMS DCL - the original .COMi =20t   >=20 > -----Original Message-----9 > From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net]=20l! > Sent: September 7, 2003 7:16 PMY > To: Info-VAX@Mvb.Saic.Como >=20; > In article <3f5ba886$0$49108$e4fe514c@news.xs4all.nl>,=20t6 > "AWN_VSIS" <anty_SPAM_email_adres@email.com> writes: >=20I > > Is it possible to limit a user tot a maximum of for example 30% of=20i< > > processing time on an Alpha800 running OPEN VMS 7.1.2 ?? >=20B > You need to look for "Class Scheduler" in the documentation. =20? > There are two different ways to do it, both too complex to=20p= > avoid looking at the documentation.  The Master Index is=209 > (usually) your friend. >=20   ------------------------------  # Date: Sun, 07 Sep 2003 23:06:18 GMTi> From: Michael Austin <maustin@no-more-spam.firstdbasource.com>5 Subject: Looking for JAVA-based  "terrminal emulator"u: Message-ID: <KRO6b.501$bR1.138@newssvr23.news.prodigy.com>  H I am looking for a terminal emulator - preferrably free whereby I could F   have a web page that a user can select a pre-configured command and F have a JAVA-based terminal emulator pop up, execute the command, wait L n-seconds and then exit.  like - execute QUOTAWATCH program from a web page.  G I see several on the commercial side (PowerTerm, Reflections), but was  , wondering of there was a freeware version...   Michael Austin.H   ------------------------------  $ Date: Sun, 7 Sep 2003 22:33:59 -0500( From: Wayne Sewell <wayne@tachysoft.com>B Subject: Re: OpenVMS Itanium system access for developers via DSPP/ Message-ID: <00A25918.983FDA75.7@tachysoft.com>o   >yE >I'm sorry that the website isn't clear about internet access to the VI >systems.  When we receive a request, we set up an Alpha system with the  J >cross tools and an Itanium system.  The partner gets exclusive access to F >them, privs., and you may use FTP and TELNET after sending us the IP C >address that you will be coming from.  Both the AMTCs in Reading, cI >England and Littleton, Mass. support Internet access as well as on-site n >visits.    L This is great and we will probably make use of it, but is it possible to getN the cross-compilers for installation on my own systems as a preliminary step? J Seems to me that it would be more efficient for everybody if I could get aD clean compile and link of the software *before* tying up one of your Alpha/Itanium system pairs.      WayneiO ===============================================================================pN Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html   uO ===============================================================================oN Butler:"Gentlemen!"  Curly(as he and other Stooges look around):"Who came in?"   ------------------------------  * Date: Mon, 8 Sep 2003 00:36:58 -0500 (CDT) From: sms@antinode.orgB Subject: Re: pgp 2.6.3ia multi05 finally running on OpenVMS V7.3-1) Message-ID: <03090800365800@antinode.org>-  4 From: Doc.Cypher <Use-Author-Address-Header@[127.1]>  N > I'll happily host a more friendly build if someone has one they've worked onL > and are happy to share.  It'll go nicely with something I rescued with the; > WayBack machine - http://vmsbox.cjb.net/HowPGPWorks.html.(  E    You might look at "http://www.antinode.org/dec/sw/pgp.html".  It'seD been a while since I did anything with it, but there are DESCRIP.MMS files.  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode.orgM    Saint Paul  MN  55105-2547h   ------------------------------  * Date: Sun, 7 Sep 2003 20:17:50 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)L Subject: using a FALCO 5000 as a console terminal on an ALPHAstation 255/233$ Message-ID: <bjg3pe$osm$1@online.de>   Subject says it all.  E Actually, it probably doesn't matter if it's the console or not, and e? probably doesn't matter what type of system it is connected to.a  A I posted a while back on this question, and after I couldn't get  7 anything to work, concluded that it is a cable problem.s  H I realised there are at least 50 types of serial cable: each end can be F DEC 423 or, if not, then large or small and male or female.  That's 5 G possibilities times two ends is 25.  Multiply by 2 since the cable can a be crossed or straight-through.0  H So, I tried out my cables.  I have about 6 different kinds (discounting 1 whether or not they are straight-through or not).k  F The only cable I have which seems to work has a big DB25 at each end, I one male and one female.  The Falco (in contrast to, say, a VT320) has a tE female connector.  So, I connected the male end to the Falco and the gF female end to a DB25-to-DB9 adapter.  (The ALPHAstation has male DB9'sF for the serial ports; obviously, the adapter is female on both sides.)   This now seems to work.   I The remaining problem is that the cursor isn't well behaved.  Sometimes, eE some of the output appears twice---where it should be, and somewhere  = else.  Things like MONITOR have some of the output misplaced.   H There is nothing obviously wrong going by the output of SHOW TERMINAL.  A However, there seem to be things in the Falco setup which aren't   indicated in SHOW TERMINAL.n  E Does anyone have a FALCO connected to a VMS machine and have it work c with no problems?a  D Years ago, using the same terminal with the same machine, I seem to F recall that it DID work fine, so I am thinking that my FALCO setup is  not entirely correct.s  D It is a rather long cable, and in contrast to the situation several G years ago, there are more "live" cables nearby.  Could it be a problem a with line noise?   ------------------------------  * Date: Sun, 7 Sep 2003 21:28:59 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)P Subject: Re: using a FALCO 5000 as a console terminal on an ALPHAstation 255/233$ Message-ID: <bjg7uq$skt$1@online.de>  D In article <bjg3pe$osm$1@online.de>, helbig@astro.multiCLOTHESvax.de3 (Phillip Helbig---remove CLOTHES to reply) writes: e  H > The only cable I have which seems to work has a big DB25 at each end, K > one male and one female.  The Falco (in contrast to, say, a VT320) has a hG > female connector.  So, I connected the male end to the Falco and the wH > female end to a DB25-to-DB9 adapter.  (The ALPHAstation has male DB9'sH > for the serial ports; obviously, the adapter is female on both sides.)  G Should have said "adapter is small on the female side and large on the n
 male side".		n   ------------------------------   End of INFO-VAX 2003.497 ************************ecome overloaded by H backed-up AST activities (assuming that ASTs are