PB5840@UTKVM1.BITNET.UUCP (05/27/87)
Date: 26 May 1987, 12:24:12 LCL
From: Mike Newman x6755 PB5840 at UTKVM1
To: INFO-VAX at SRI-KL.ARPA
Does anyone have experience with Cincom's relational database products
on the VAX? We have IBM's IMS and Cincom's SUPRA/MANTIS installed on
an MVS/SP IBM 3081D, and we apparently will be implementing a Student
Academic Management System on a VAX host. I would be particulary interested
in talking with anyone who has worried about 'connectivity issues' with
such a set up.
Mike Newman, Associate Director for Administrative Computing Services,
The University of Tennessee at Knoxville, BITNET: PB5840@UTKMV1BRUCE@UC780.BITNET.UUCP (06/20/87)
>> Does anyone out there know if it is possible to use RUNOFF to get a file >> suitable for insertion into a help library, given a file designed to be >> printed documentation? In particular, can the .HEADER LEVEL type of >> directives be instructed to produce single numbers, as required. Yes it is possible because the vax version of kermit uses runoff to produce the kermit help files, but I can't find it on my system.... bruce <bruce@uc780> on bitnet
BRUCE@UC780.BITNET.UUCP (06/20/87)
in a previous info-vax submission.................... 1) Will VMS schedule jobs at two priority levels "fairly" or will the lower priority jobs never get reasonable response? I know the answer depends heavily on what job mix we have, so I guess I want to know if anyone has set up their system to divide their users into two or more priority classes. (I'm running an 8500 with mostly student users, a few large SAS or MACSYMA users.) 2) Is their a way to set a process priority limit downward such that the user can't set it back? The documentation for SET PROCESS/PRIORITY and SYS$SETPRI seem to be misleading. It suggests that the user can't set the base priority higher than the existing base priority when in fact they can set it back to the former (uaf specified) value. It looks like I would have to poke around in a system data structure; does anyone know how to do it? Thanks. ...................................... on 1, interactive response will be just find on priority 3, iff (if and only if) you NEVER have a job that takes the full cpu for more then just a second or so. If one of you priority 4 users starts something up which is only CPU bound (1000! or such) then the priority 3 users will see the system halt. What VMS could use in this case is a class based scheduling system (class a gets 80 % minimum, class b gets 15 % minimum, class C gets 3% minimum, class D gets 2% minimum). Each class gets to run, but each only up to its % maximum unless there is free time. on 2 I can't help. bruce <bruce@uc780> on bitnet
DHASKIN@CLARKU.BITNET (Denis W. Haskin, Manager, Technical Services) (06/22/87)
> Date: 19 JUN 87 11:14-EDT > From: BRUCE%UC780.BITNET@WISCVM.WISC.EDU > > >> Does anyone out there know if it is possible to use RUNOFF to get a file > >> suitable for insertion into a help library, given a file designed to be > >> printed documentation? In particular, can the .HEADER LEVEL type of > >> directives be instructed to produce single numbers, as required. > > Yes it is possible because the vax version of kermit uses runoff to produce > the kermit help files, but I can't find it on my system.... > > bruce <bruce@uc780> on bitnet Aha... I was waiting to see whether there's another way to do it, but I guess not. You can do it, but you have to plan for it *before* writing your file, e.g. if you want to do it to some existing documentation you're in for a bit of editing. Bruce was right that the Kermit documentation uses it; at the bottom of this message is an excerpt from VMSMIT.RNH. It would be processed with RUNOFF/VARIANT=SYSTEM VMSMIT.RNH to produce a help file, and RUNOFF VMSMIT.RNH to produce a manual. Hope this helps... % Denis W. Haskin Manager, Technical Services % % ----------------------------------------------------------------------- % % DHASKIN@CLARKU.BITNET Office of Information Systems (617)793-7193 % % Clark University 950 Main Street Worcester MA 01610 % -------------------------- excerpt from VMSMIT.RNH ------------------------- VR SYSTEM P0 ap LM1.RM70 I-1 IF SYSTEM 1 Kermit ENDIF SYSTEM Kermit is a file transfer protocol. It allows the transfer of files over [... etc ...] a message indicating the default transfer terminal when it is started. I-1 IF SYSTEM 2 BYE ELSE SYSTEM 1 BYE ENDIF SYSTEM This command will cause Kermit-32 (when in local mode) to tell the other [... etc ...] the acknowledgement that this is being done, it will exit to VMS. b.i+10 Kermit-32>BYE b I-1 IF SYSTEM 2 CONNECT ELSE SYSTEM 1 CONNECT ENDIF SYSTEM The CONNECT command will allow you to connect in as a virtual terminal [... etc ...]
JOE@FHCRCVAX.BITNET (06/25/87)
Hello everyone,
I've been away from the nets for a while, but I just recently became
a member of the BITnet community, so I'm back in gear. To refresh your
memories, I authored VERB, as well as a few lesser known utilities,
such as FILE, UNMESSAGE, STATUS and so forth.
I just wanted to let everyone know my new permanent Email address,
(bitnet - joe@fhcrcvax). Feel free to contact me. If you aren't on
bitnet and you send me mail, let me know how to send mail back to you!
If you have a copy of VERB, you might want to see if it's the most recent.
After posting to the net and sending it to DECUS I did find one glaring
bug and one cosmetic one. The glaring bug was ignoring the cliflags parameter
(on aything except cliroutines.) You can test it with a command definition
such as:
DEFINE VERB SD
IMAGE EXE$DIR:SD, cliflags(foreign)
If you create that definition then use VERB to get it back you may not
get the "cliflags(foreign)" part, if you don't get it then you should
ask me for a newer version.
The second bugfix is cosmetic. If you start out with a verb which has
multiple DISALLOWS clauses, then use VERB on it, you may end up with one
(often VERY long) disallows clause (each one or'ed together). If you
have this problem then you also might want the newer version.
Also, if anyone out there is beta testing VMS 5.0, please let
me know if VERB still works! Of course I'd also love to hear about whats
new, whats changed, and so forth.....
Speaking of DECUS (I mentioned it above right?) I'd like to
contact Martin Minow in regards to his revision of the Unix compress
program which appeared on the 86 fall tapes - I made some some enhancements
to allow it to work with indexed files. Are you out there Martin?
If anyone on Bitnet has the latest release of the CMU PTY driver,
could you contact me about sending me a copy? Thanks in advance...
Well, it's nice to be back in business.
Cheers,
Joe Meadows Jr. bitnet - JOE@FHCRCVAX
voice - (206) 467-4970PB5840@UTKVM1.BITNET (07/18/87)
Date: 16 July 1987, 09:03:20 LCL
From: Mike Newman x6755 PB5840 at UTKVM1
SMC 200
To: INFO-VAX at SRI-KL.ARPA
cc: Faye Muly MULY at UTKCS1
>has anyone ever witnessed a halon discharge? what actually happens in the room?
>are people knocked down? are raised floor tiles lifted? are suspended ceiling
>tiles blown out? are temporary walls knocked out? is the planet moved from its
>orbit?
>
>no one here has actually witnessed a halon discharge. we only know that it
>costs $3000 (at least) to recharge our system.
The Center for Computer Integrated Engineering and Manufacturing at the
University of Tennesse at Knoxville installed a halon discharge system
in their computing facility. They required the installer to set the
thing off (how else do you know it really works?). The installer filled
the tank with a gas that would be visible, and they videotaped the
discharge.
If anyone is coming through Knoxville and would like to see the facility
(and the videotape), they can contact
Dean William T. Snyder
124 Perkins Hall
The University of Tennessee
Knoxville, Tennessee 37996
(615)-974-5321
Ms. Faye Muly is the System Manager for the CCIEM facility and can be
reached at MULY@UTKCS1.BITNET.TRAN@FRMEU51.BITNET (03/19/88)
Date: 18-MAR-1988 08:48:45.55
From: TRAN Van-Tue TRAN AT FRMEU51
To: EARN::"INFO-VAX@KL.SRI.COM"
Subj:
Hi , I would like to know if someone has to-day a VAXSTATION
VS3200 in his office . I appreciate greatly to have a feeling from
user on this station .
Many thanks in advance
TRAN@FRMEU51.BITNETGHC@NIHKLMB.BITNET (05/12/88)
Date: 12-MAY-1988 08:15:38.58 From: Gerson H Cohen GHC AT NIHKLMB To: GATEWAY::"info-vax@kl.sri.com",GHC Subj: RE: VMS Mailboxes Tom Mish askes for some examples of mailbox use in VMS for inter-process communication. Take a look in the "Guide to Programming on VAX/VMS", Order No. AI-Y503B-TE for some examples in Fortran (my mother tongue).
GHC@NIHKLMB.BITNET (06/14/88)
Date: 14-JUN-1988 09:07:47.72 From: Gerson H Cohen <GHC@NIHKLMB> To : <info-vax@kl.sri.COM> Original To: GATEWAY::"zwarts@hgrrug51",GATEWAY::"info-vax@kl.sri.com",GHC Subj: RE: Disk visibility in LAVC. >Date sent: Wed, 8 Jun 88 11:40:00 N >Received: by PSUVM (Mailer X1.25) id 0060; Tue, 14 Jun 88 01:52:45 EDT >Date: Wed, 8 Jun 88 11:40:00 N >Reply-To: INFO-VAX@KL.SRI.COM >Sender: INFO-VAX Discussion <INFO-VAX@VTVM2> >From: ZWARTS%HGRRUG51.BITNET@CUNYVM.CUNY.EDU >Subject: Disk visibility in LAVC. >Comments: To: info-vax@kl.sri.com >To: GERSON H COHEN <GHC@NIHKLMB> > >About half a year ago we started with a LAV cluster. A Vax 8300 is the boot >node. Up to last week we have had only diskless VAXstations 2000 as satellite >nodes. So, the only disks in the cluster were the disks of the Vax 8300. Those >disks were visible and accessible from all satellites. Last week we added a >VAXstation 2000 with its own disk. I modified MODPARAMS.DAT for this node to >include MSCP_LOAD=1, so that it would serve its disk for the cluster. Indeed, >all other satelite nodes see this disk and have access to this disk. However, >the Vax 8300 still only sees its own local disks. SHOW DEVICE on the boot >member does not show the disk of the satelite. Any attempt to access the disk >from the boot member fails with an error stating that there is no such device. >Is there a SYSGEN parameter to be set on the boot member to make it accept >disks served by satelites? > ------------------------------------------------------ When I first brought up my LAVC some 6 months ago I had a similar problem. Check to see if you have a process by the name CONFIGURE running on your 8300. We have an 8530 and the process was absent. It is the responsibility of this process to find new served disks when they come on line and the problem was that I had a "rich" SYCONFIG.COM that installed several drivers and then had the line "$ STARTUP$AUTOCONFIGURE_ALL == 0". This is OK with a non-cluster system, but as one of the system management manuals points out, if such language is used in a cluster environment, CONFIGURE does not start up and your satellite disks cannot be seen. BTW - if you reboot your 8300 without shutting down the VAXstation you should then see the missing disk(s). Below is part of my SYCONFIG.COM file: ---------------------------------------------------------------- $! $! Site-specific device configuration procedure. $! [---] $ STARTUP$AUTOCONFIGURE_ALL == 0 <- the culprit $ @SYS$SYSTEM:STARTUP CONFIGURE <- the fix
GHC@NIHKLMB.BITNET (06/16/88)
Date: 16-JUN-1988 09:24:45.28
From: Gerson H Cohen <GHC@NIHKLMB>
To : <info-vax@kl.sri.COM>
Original To: GATEWAY::"info-vax@kl.sri.com"
Subj: RE: Moving users to a new Vax
In reply to a query from SYSMAN@NMSUNM1.BITNET Re: Moving users to a new Vax
I have suggested the following. I am reposting it to the list because of the
usefull command procedure which I included in the response.
============================================================================
From: VGER::GHC "Gerson H Cohen" 16-JUN-1988 09:16
To: GATEWAY::"sysman@nmsuvm1",GHC
Subj: Re: Moving users to a new Vax
I have done almost exactly what you require, excepting that several of my
drives came along onto the new system. If you don't bring any drives,
you'll have to copy the system disk (image mode) onto tape and restore it
on the new machine. As for copying quotas, I have a procedure (originally
from a DECUS symposium tape which I am attaching. Execute it on each of
your current drives and then execute to resultant command file on the new
machine. BTW - I use it from time to time to record tha quotas on each on
my devices in case I loose a drive and have to recreate what was there on
a new pack.
GHC
---------------------------------------------------------------------------
$ verify = 'F$VERIFY(0)'
$ GOTO L1$
$!
$! DISKQUOTA.COM
$!
$! This command procedure will extract the currently set disk quotas for
$! a disk volume and prepare a command file to be used to restore those quotas.
$! The two parameters give the disk name with the volume whose quotas are to
$! be scanned and the file name for the command file to be used to restore the
$! quotas. This command procedure is used because the BACKUP utility will not
$! allow the QUOTA.SYS file to be saved online.
$!+ DISKQUOTA
$! The DISKQUOTA command procedure will create a command file to
$! restore the current disk quota settings (using the DISKQUOTA program).
$! Execute the command procedure by doing:
$!
$! @D$:DISKQUOTA device[:] [file_spec]
$!
$! The output file specification defaults are taken from
$! SYS$DISK:QUOTAS.COM.
$!-
$! Created: Frank J. Nagy 9-Oct-81 Fermilab Accelerator Controls
$!
$! Modifications:
$! 11-Apr-82 FJN Cleaned up use of VER_MODE
$! 30-Aug-82 FJN Use VMS V3 features
$! 03-Sep-82 FJN Remove ";" added by F$PARSE
$! 21-Jul-87 GHC Check for UIC by character string
$!=============================================================================
$L1$:
$!
$! If no P1 parameter specified, exit. If p1 is given, remove any trailing
$! colons.
$!
$ IF p1 .EQS. "" THEN $ EXIT F$VERIFY(verify).OR.1
$ p1 = p1 - ":"
$!
$! Set default for file specification parameter.
$!
$ p2 = F$PARSE(p2,"SYS$DISK:QUOTAS.COM") - ";"
$!
$! Create temporary command file to run DISKQUOTA
$!
$ OPEN/WRITE OUT: DISKQUOTA.TMP
$ WRITE OUT: "$ RUN SYS$SYSTEM:DISKQUOTA"
$ WRITE OUT: "USE ",p1
$ WRITE OUT: "SHOW [*,*]"
$ WRITE OUT: "EXIT"
$ CLOSE OUT:
$!
$! Run DISKQUOTA to get disk space allocated by UIC
$!
$@DISKQUOTA.TMP/OUTPUT=DISKSPACE.TMP
$ DELETE DISKQUOTA.TMP;*
$!
$! Sort into order by UIC
$!
$ SORT/KEY=(POS:2,SIZE:6,NUMBER:1)/KEY=(POS:9,SIZE:6,NUMBER:2) -
DISKSPACE.TMP DISKSPACE.TMP
$ PURGE DISKSPACE.TMP
$!
$! Now read the DISKQUOTA output file and setup the restoring command file.
$!
$ OPEN/READ/ERR=ERR_NO_DISK SPACE: DISKSPACE.TMP
$ OPEN/WRITE/ERR=ERR_NO_WRITE RESTORER: 'P2'
$ ON CONTROL_Y THEN $ GOTO EXIT_ALL
$ ON ERROR THEN $ GOTO EXIT_ALL
$!
$! Skip first record in disk space file
$!
$ READ SPACE: DUMMY
$!
$! First lines of restorer are to RUN DISKQUOTA and USE the disk and to
$! CREATE the QUOTA.SYS file.
$!
$ WRITE RESTORER: "$ RUN SYS$SYSTEM:DISKQUOTA"
$ WRITE RESTORER: "USE ",P1
$ WRITE RESTORER: "CREATE"
$!
$! Loop reading the DISKQUOTA output file and writing the restorer file
$!
$LOOP:
$ READ/END=EOF_SPACE SPACE: space_buf
$ uic_space = F$EXTRACT(0,16,space_buf)
$!
$! Check for [0,0]. If so then operation is "MOD", else operation is "ADD"
$!
$ op_type = "ADD" !Default operation type
$ IF F$EXTRACT(0,5,uic_space) .EQS. "[0,0]" THEN -
op_type = "MOD"
$!
$! Get quota values from input
$!
$ uic_perm = F$EXTRACT(31,10,space_buf)
$ uic_over = F$EXTRACT(49,10,space_buf)
$ WRITE RESTORER: -
op_type," ",uic_space,"/PERM=",uic_perm,"/OVER=",uic_over
$ GOTO LOOP
$!
$! All done with input
$!
$ EOF_SPACE:
$!
$ EXIT_ALL:
$ CLOSE SPACE:
$!
$! End the restorer command file.
$!
$ WRITE RESTORER: "REBUILD"
$ WRITE RESTORER: "EXIT"
$ CLOSE RESTORER:
$!
$! Cleanup
$!
$ DELETE DISKSPACE.TMP;*
$ PURGE 'P2'
$ EXIT F$VERIFY(verify).OR.1
$!
$! Error Messages
$!
$!
$ ERR_NO_DISK:
$ WRITE SYS$OUTPUT "%DISKUSAGE-F-NO_DISK, no disk space file found."
$ GOTO EXIT_ALL
$!
$ ERR_NO_WRITE:
$ WRITE SYS$OUTPUT "%DISKUSAGE-F-NOWRITE, restorer command file not opened."
$ GOTO EXIT_ALL
$!GHC@NIHKLMB.BITNET (06/17/88)
Date: 16-JUN-1988 13:40:23.42
From: Gerson H Cohen <GHC@NIHKLMB>
To : <info-vax@kl.sri.COM>
Original To: GATEWAY::"info-vax@kl.sri.com",GHC
Subj: Spawn from detached process
I am trying to spawn a subprocess in order to submit a batch job from a
detached process. The call to lib$spawn returns with LIB$_NOCLI. I have
tried the same sequence from an interactive process with success. I have
checked tha process quotas for the detached process and it has 2 subprocesses
permitted. The spawn command is on the next line
ok = lib$spawn( full_spawn_cmd(1:full_spawn_cmd_l) )
and has the form "submit file/param=qw".
Does anyone have any ideas about this problem?carl@CITHEX.CALTECH.EDU (Carl J Lydick) (06/24/88)
> I am trying to spawn a subprocess in order to submit a batch job from a > detached process. The call to lib$spawn returns with LIB$_NOCLI. I have > tried the same sequence from an interactive process with success. I have > checked tha process quotas for the detached process and it has 2 subprocesses > permitted. The spawn command is on the next line > > ok = lib$spawn( full_spawn_cmd(1:full_spawn_cmd_l) ) > > and has the form "submit file/param=qw". > > Does anyone have any ideas about this problem? Yes, the error message you get says it all: %LIB-F-NOCLI, no CLI present to perform function LIB$SPAWN requires a Command-Language Interpreter (CLI), but there is no CLI present in your detached process. I conclude from this that you're starting the detached process with something like: $ RUN/DETACHED myprogram This starts up you program with no CLI. A workaround is to create a procedure (myprocedure.com) with the line: $ RUN myprogram Then start the detached process with a command of the form: $ RUN/INPUT=myprocedure.com/detached SYS$SYSTEM:LOGINOUT This creates a process complete with DCL as the CLI which then runs your program.
GHC@NIHKLMB.BITNET (06/25/88)
Date: 24-JUN-1988 10:04:30.69 From: Gerson H Cohen <GHC@NIHKLMB> To : <info-vax@kl.sri.COM> Original To: GATEWAY::"info-vax@kl.sri.com",GHC Subj: Re: Spawn from detached process. Resolution In view of the numerous responses that I have received about my problem using LIB$SPAWN from a detached process, I feel that a posting of the resolution is required. Several different approaches were presented; all but one are correct. The problem derives from the fact that my detached process was scheduled without invoking SYS$SYSTEM:LOGINOUT.EXE. Hence there is no CLI available even to carry out the spawn operation. The erroneous suggestion is to simply include the CLI (DCL) as the last argument to LIB$SPAWN; with no CLI the call itself cannot be serviced. Proposed were: 1) Fire up the original detached process with RUN/UIC.../INPUT=...COM SYS$LOGIN:LOGINOUT.EXE instead of directly. 2) Use $CREPRC and create a subprocess using the command file I wanted in the BATCH as input. The log file would be specified as output. 3) Use $SNDJBC. (I acknowledge a pretty piece of VAXC code). 4) Have the parent process write a command file and close it with DISP=SUBMIT. My problem has several limitations with tend to force the selection of only two of the above solutions. My original effort was a bit cloddy and spawned a process to submit a process. Let's do it in only one step. The ultimate process may be run as a subprocess (it will for a start) or as a batch process (it will ultimately). Since I want to direct the log file to a particular place, option 4 is not best. Option 1 requires changing the philosophy of a third party product which I am modifying in this effort. I settled on option 2 as the easiest to code, creating a subprocess in which I can test the program which will ultimately become a part of the package. I do think that option 3 is the cleanest approach and will be the ultimate choice, but first I have to learn how to select the proper batch queue, pass the required parameter and specify the location of the output log. My sincere thanks to all who took the time to write.
TWWALLIS@TWSUVM.BITNET (08/01/88)
I am looking for a good, introductory level book on VAX/VMS. It will be
a resource that will supplement my user's manual. Does anyone out there
have any ideas about a book like this? It should be commercially available.
It would be nice if it has a chapter on editing files with EDIT/TPU.
Thanks in advance.
Thomas Wallis
BITNET: TWWALLIS@TWSUVMdavidli@umn-cs.cs.umn.edu (Dave Meile) (08/05/88)
In article <8808031622.AA00241@ucbvax.berkeley.edu> TWWALLIS@TWSUVM.BITNET writes: >I am looking for a good, introductory level book on VAX/VMS. It will be >a resource that will supplement my user's manual. Does anyone out there >have any ideas about a book like this? It should be commercially available. >It would be nice if it has a chapter on editing files with EDIT/TPU. >Thanks in advance. If you're getting the VMS 5.0 release, you already have an "introductory level" book which seems to me to fit what you want. It's the VMS GENERAL USER'S MANUAL, part of the Base Documentation Set. (DEC # AA-LA98A-TE) You might be able to order them separately, though I'd expect the cost to be fairly high from DEC. It has an adequate chapter on editing files using the EVE and EDT editors, as well as a comprehensive DCL dictionary, etc. If most of the users of the Simulation Resource where I work were not already technically oriented (they program Monte Carlo simulation in FORTRAN...), I would get a lot of copies of this book to hand out when a new person came to the VAX and VMS. -- Dave Meile, systems manager Simulation Resource for Stochastic Micropopulation Models
NED@YMIR.BITNET (Ned Freed) (08/05/88)
Thomas Wallis asks for a good introduction to VAX/VMS. I suggest using one
of the brand new manuals in the VMS 5.0 documentation set -- the VMS General
User's Manual (AA-LA98A-TE). I think this is a super book for people who
are just starting out with VMS and need something that will get them up
to speed in order to do real work.
This book is the smaller size of manual and it is softbound. Its kinda chunky
since its about 1.5 inches thick, but not too bad. It starts with logging in,
takes you a little ways into DCL, covers MAIL and PHONE, then more information
on files and processes, then logical names, then DCL and command procedures,
and finally a tutorial on EVE and RUNOFF. The rest of the book is a sort of
miniature reference manual. It covers the various DCL commands, EVE commands,
RUNOFF commands, and a few other miscellaneous things like SORT, EDT and
TFF.
There's nary a word about programming in anything but DCL in there, which (in
my opinion) is good because no single language would ever be appropriate to
every "general user".
This book is not casual in any sense. It is only for people who want to
actually accomplish something on VMS, not for dabblers. In this regard it
achieves something I've rarely seen an introductory manual do -- it is still
useful after you cease to be a novice user.
Ned Freed, PMDF Project
P.S. It also passes a personal test of mine... there's an ASCII code chart in
it. I'm ashamed to say that I've gotten out of the habit of putting those in
the documentation I write...