[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference bulova::decw_jan-89_to_nov-90

Title:DECWINDOWS 26-JAN-89 to 29-NOV-90
Notice:See 1639.0 for VMS V5.3 kit; 2043.0 for 5.4 IFT kit
Moderator:STAR::VATNE
Created:Mon Oct 30 1989
Last Modified:Mon Dec 31 1990
Last Successful Update:Fri Jun 06 1997
Number of topics:3726
Total number of notes:19516

452.0. "Error 12DB821C" by USMRM4::PKADOW (Crashed&burning on the learning curve) Wed Mar 22 1989 08:42

    Several times a day I lose the process in am working on.  The window
    and process just disappear into never-never land.  Sometimes I
    lose the controlling process also.  The error left behind is:
    
    Decterm Controller process stopped
    Message Number 12DB821C
    
    Any help will be greatly appreciated.
    Thanks in advance - Paul

T.RTitleUserPersonal
Name
DateLines
452.1*decterm.noteXDELTA::SIMMWed Mar 22 1989 13:128
�  <<< Note 452.0 by USMRM4::PKADOW "Crashed&burning on the learning curve" >>>
�                              -< Error 12DB821C >-

    Paul, this one's been discussed over in the DECterm notes file...
    (HANNAH::DECW$DISK:[PUBLIC]DECTERM, KP7, etc...)

    Steve

452.2DECTERM Notes file says to ask hereZEKE::KOZIKOWSKIHugs for HealthTue May 16 1989 10:5011
    Hi! 
    
    I'm having the same problem.  Before coming here I checked the
    DECTERM notes file, and they said, check BULOVA::DECWINDOWS,
    and wait for the next release.  ANY ideas?  
    
    I have only had this happen to me since I have attached an
    LN03R ScriptPrinter and installed the appropriate software.
    
    Derek

452.3This is a real problem for us7412::LINDSEYTue May 16 1989 11:1119
    
    A user I am trying to support here has the same problem.  She is
    running on a VS2000 with 6 meg of memory.  On a regular basis, her
    DECterm Controller Process aborts with the message number mentioned
    in note 1.  She is not processing any REGIS or Sixel files, often
    only has 2 windows open (one that is set host to a timesharing cluster)
    and is doing as mundane a task as editing a file.
    
    I have read ALL the notes already in this file and in the DECterm notes
    file and it doesn't appear that anyone is addressing the issue and from
    what I can see there are many people experiencing the same problem.
    
    I have tried increasing various process quotas as noted in the manual,
    but nothing seems to alleviate the problem.  Please, can someone help?
    
    Sue
    
    

452.4Maybe GBLmumble params are too low...18891::HARLEYPeople just don&#039;t freak out like they used to...Tue May 16 1989 17:5219
You may be running out of GBLPAGES, I've seen DECterm get upset when I had this
parameter set too low. On my GPX (VMS V5.2-42N), it's set to 25000; the entry in
VMSPARAMS.DAT says that DECwindows wants a minimum of 15K. This is a "cheap"
parameter anyways, so crank it up and see if it helps. Oh yeah, you may want to
bump up GBLSECTIONS and GBLPAGFIL as well, just to play it safe...

Here's what VMSPARAMS looks like:

DECWGB  GBLPAGES                                ADD       5000      GBA0
DECWGB  GBLPAGES                                MIN       15000     GBA0
DECWGB  GBLPAGFIL                               ADD       5000      GBA0
DECWGB  GBLPAGFIL                               MIN       6024      GBA0
DECWGB  GBLSECTIONS                             ADD       50        GBA0
DECWGB  GBLSECTIONS                             MIN       300       GBA0

The entries for GAA0 and GCA0 are identical.

/Harley

452.5Not below the minimum required7412::LINDSEYWed May 17 1989 13:5215
    
    Currently I have GBLPAGES set at 30700 (7200 free), GBLSECTIONS at 300
    (83 free), and GBLPAGFIL at 6024.  None are below the minimum...
    
    Someone suggested turning off the "saved scrolling" as that utilitizes
    alot of memory (in the customization menu) and I have tried that and
    am waiting to see if that has any effect.
    
    If turning off the scrolling doesn't help, I will try increasing the
    GBL params.  Its worth a shot.
    
    Thanks,
    
    Sue

452.6Still no resolutionHPSRAD::LINDSEYWed Jun 14 1989 14:519
    
    I have decreased the number of lines of saved scrolling and have
    increased all her GBL params and still the decterm controller
    stopped message is occuring...
    
    Any other ideas...this problem is getting real old.
    
    Sue

452.7LAVC problem?HANNAH::MESSENGERBob MessengerWed Jun 14 1989 18:148
Re: .6

Maybe it has something to do with the network connection (I'm thinking of the
other note that talked about improperly terminated Thinwire ports etc.).  Is
the VS2000 diskless?

				-- Bob

452.8I don't think termination is the problemHPSRAD::LINDSEYTue Jun 20 1989 13:2212
    
    No, the VS2000 has one disk on it and I believe there is no problem
    with the thinwire port - it has a t connector at the cpu and a single
    connector at the dec-connect box.  Any thing else ethernet related I 
    should look for...
    
    Thanks,
    
    Sue
    
    

452.9Message 12DB821C (AGAIN!)TAMARI::MCGOVERNSzechuan VanillaMon Jul 31 1989 12:5525
    
    
    This problem was mentioned ina previous note, but that note provided
    no resolution and has been dormant since June.  Here we go again:
    
    Periodically (with annoying regularity) my DECwindows terminal
    processes stop with no warning.  I am doing mundane tasks only,
    like reading mail or editing small files.  I get the following message:
                                                               
    DECTERM CONTROLLER PROCESS STOPPED:  MESSAGE NUMBER 12DB821C.
    
    I then get a message on my Session Manager that DECwindows is starting
    the terminal server again, but it fails to do so. 
    
    VPA shows no anomalies during these events.
    
    Please provide information to SOLVE this problem.
    
    I will provide whatever system information is necessary.  For starters:
    VS2000 with 14 MB running as satellite of a uVAX3500-based LAVC.
    I can provide sysgen, modparams, etc. information on request.   
                                              
    Michael McGovern
    

452.10And Another Thing...TAMARI::MCGOVERNSzechuan VanillaMon Jul 31 1989 12:562
    P.S.  I followed the advice in the previous note to no avail.

452.11It happens to me too!WHYNOW::NEWMANWhat, me worry? YOU BET!Wed Aug 02 1989 09:2013
    I too have just started to get this message.  It seems to have started
    ever since I upgraded to DECwindows V2 EFT1.  I am also running VMS
    T5.2-1 but it did not start until I upgraded to V2 of windows.
    
    If I get the message it comes up in the "operator" window when I
    attempt to QUIT out of the session manager.  When this occurs,
    sometimes DECwindow's gets blown away and I have to restart it.  Other
    times a CTRL-F2 restores the workstation.
    
    Any information would be greatly appreciated.  I too will supply
    whatever information I can.  My system is a 13mb VSII/GPX with 1 RD54,
    2 RD53's, and 2 RRD50's

452.12LESLIE::LESLIEHalfway to Paradise (-56)Wed Aug 02 1989 11:1510
    I, too, have seen this a lot since upgrading to DW V2 IFT, then EFT. I
    have billyuns and billyuns of global pages and sectiobs free.
    
    What's going on, folks? What info do you need to fix this that I can
    supply?
    
    Andy
    (LESLIE::  is a 14mb VS2000 Colour GPX with 2 RD54's, VMS T5.2-1 and DW
    V2 EFT1)

452.13STAR::ORGOVANVince OrgovanThu Aug 03 1989 19:2613
    I'm not sure I understand the problem, but in any case you shouldn't
    be seeing the 2db821c errors. In .0 it sounds like these were
    occurring from the terminal emulator controller, in .2 it sounded 
    like these were coming from the session manager?

    We're not seeing this at all on our systems. In fact, it's been 
    very nice to be rid of these messages since the transport buffering
    fixes were made. 

    It sure sounds like there a bug under here, but I don't even know
    what to tell you to check. Can you post a precise description of
    what clients are reporting this message? 

452.14Cannot bring up DecWindows at allSOLOWA::SOLOWAYThu Aug 17 1989 11:0435
    This was originally posted in a new note, before it was pointed out to
    me that notes already existed on the topic.  I would like to add a few
    comments while I'm reposting it, however:
    
       1) I only saw the 2db821c message once during the booting process;
    normally when I boot the node I see no decwindows-related error
    messages. 
    
       2) This is unlike all the other unsolved 2db821c-related bugs, in
    that I can't bring up DecWindows at all.
    
       3) I will be happy to provide any further information anyone wants
    to investigate this problem.  
    
    What follows is the note as originally posted:
    
    I have a problem that is preventing my workstation from coming up with
    DecWindows.  If this is not the proper place to get help for a bug (and
    I suspect it isn't), I would appreciate it if someone would point me to
    the proper place.
    
    My workstation (a vs 2000) was running decwindows more or less happily
    until recently.  The only thing I can think of that changed was that my
    disk utilization has grown steadily.  Anyway, I have 6mb of memory and
    45000 free disk blocks (out of the original 300,000).  Suddenly, my
    display froze -- I could move the mouse pointer, and I could hear the
    disk crunch when I moved it, but otherwise the system was dead as a
    doornail.  I have never been able to successfully boot the system since
    then; it comes up normally but does not start decwindows.  I can only
    log in over the LAT or in the operator window at the top of the
    display.  At least once I have seen the message "Message 2db821c"
    displayed during booting; I seem to remember seeing this same message a
    few times in the past when the DecWindows terminal controller crashed.


452.15STAR::ORGOVANVince OrgovanThu Aug 17 1989 21:324
    It sounds like the DECwindows server can't start for some reason.
    Then when loginout tries to connect to the server it gets Xlib I/O
    errors (2db821c). Check sys$manager:decw$server*.log;*

452.16me too, but different...DECWET::SIEBOLDI19G = I18N in GermanFri Aug 18 1989 13:1214
It looks this happens to me too, but a little different. Yesterday evening
I QUIT my session (GPX, 9Meg, RD54) and left the room without watching.
This morning the screen was a black and white copy of the last color screen
from yesterday, but not the login screen. In the upper left corner there
were two messages 'MESSAGE NUMBER 02DB821C'. Stopping the Decwindows
processes and restarting DW with DECW$STARTUP did not help, BUT hitting
CTRL/F2 (or F3, the OPA0 sequence) brought the login screen up!

This happens whenever I quit a session.

Thomas

PS: VMS V5.2 and DW FT1 V2

452.17One more system with the same problemHPSCAD::LARRIVEERuss LarriveeFri Aug 18 1989 16:2812
    I built VMS V5.2 from scratch and installed DECwindows at the same
    time. Everytime the DECwindows starts I get a "Message 02DB821C"
    from OPCOM. I use ctrl F2 to clear the screen and I proceed to get
    the same message once again. 
    
    I did see some X lib messages in decw$server_error_0.log once. 
    Unfortunately I haven't seen them since.
    
    Seems like this is more than coincidence.
    
    Russ

452.18STAR::ORGOVANVince OrgovanMon Aug 21 1989 18:399
    RE: .8 getting 02db821c every time DECwindows starts
    
    You certainly shouldn't be seeing this. Please check sys$manager:
    decw$startserver.com to verify that the following line is intact:
    
    	$ set terminal/permanent/nobroadcast OPA0:
    
    Vince

452.19SET TERMINAL is thereNITMOI::LARRIVEERuss LarriveeMon Aug 21 1989 21:2110
    I checked the file and did find the set terminal command referenced in .9
    
    I should mention that DECWindows never really starts up.  It gets as
    far as starting WSA0 and continues to display "MESSAGE 02DB821C" at the
    top of the screen in groups of two. Everytime I use ctrl-F2 to clear
    the screen WSA0 seems to start up again and reports the same hex
    number.
    
    Russ

452.20At least one problem has been fixedHPSCAD::LARRIVEERuss LarriveeTue Aug 29 1989 12:467
    In reference to .10, I  was able to fix the problem by running 
    ANALYZE/DISK_STRUCTURE/REPAIR on the system disk. There were some 
    bad files (some were decwindow files) which apparently were the 
    source of the problem.
    
    Russ

452.21You were right; my disk was corrupted.SOLOWA::SOLOWAYWed Aug 30 1989 17:4821
Well, I did like Russ said.  I ran analyze /repair twice to get
rid of all the errors (there were a bunch; unfortunately I didn't
keep a list).  Then I booted, and discovered I had 30000 more disk 
blocks than when I started.  Unfortunately, my problem didn't go 
away.

So I assume that something got corrupted somehow; at some point
I'm going to have to rebuild the disk.

Anyway, thanks to Russ for pointing me in the right direction.
            <<< Note 1204.11 by HPSCAD::LARRIVEE "Russ Larrivee" >>>
                    -< At least one problem has been fixed >-

    In reference to .10, I  was able to fix the problem by running 
    ANALYZE/DISK_STRUCTURE/REPAIR on the system disk. There were some 
    bad files (some were decwindow files) which apparently were the 
    source of the problem.
    
    Russ


452.22my terminal emulator isn't working right!HYDRA::SCHUMANNGates&#039;R&#039;UsFri Sep 08 1989 00:2423
Re .0

I have the same problem. I can reproduce the problem at will by doing the
following:

1) from VWSLAT> C ASGARD !(This is an 8800)
2) @WL.COM		! reproduced below
3) try to pop other windows
presto, the display hangs for a few seconds (except for live pointer), then
all terminal windows are vaporized and the 12db821c message appears in the
session manager window.

(The com file merely writes a steady stream of null lines to the window.)

Please help!! Is there a fix for this bug??


DISK$DESIGNDISK22:[ADESIGN.PCHIP.R0.GRA]WL.COM;1

$loop:
$write sys$output ""
$goto loop

452.23Another 2db821c problem, with a workaroundDECEAT::NOFSINGERThu Sep 14 1989 11:4427
  After a recent system crash, I have not been able to start a session in
decwindows due to the server displaying the message 2db821c. 

 I looked around in the system files and didn't see anything unusual 
(to my untrained eye).  I read this notes file and tried the suggested remedy
(analyze/disk/repair), no help.

  Out of habit I logged into the system account on my station and it worked! 
That is the session manager correctly started with no messages.  This lead me
to believe that there must be some file that the sm reads when its starting up
that keeps some status. And for my normal account the status was "sm already
started".

  On that belief, I got a directory of the modified files since the problem 
started. None of the files seemed conspicous (they were not connected with 
decwindows or my home directory).  Continuing with this line of thought I 
created a extra user account. When logging in with its own directory everything
is fine. When I set its login directory to be the same as my usual account, I 
got the message 2db821c.

  Thinking that perhaps my resource definition files were at fault I got rid of
them. This didn't help.

  Where is the session manager getting this bogus info? Or, more generally,
 what the heck is going on?

452.24LESLIE::LESLIEAndy ��� Leslie. Fat was then....Thu Sep 14 1989 18:075
    What do the SYSUAF quotas look like for both accounts? If one is
    substantially different to the other, adjust and try again...
    
    - ���

452.25STAR::ORGOVANVince OrgovanFri Sep 15 1989 20:086
    Try checking the file protections on:
    
    	sys$share:decw*.exe
    	decw$system_defaults:decw*.dat,decw$*.uid
    

452.26UpdateHYDRA::SOLOWAYTue Sep 19 1989 11:2121
    Regarding my previous notes --
    
    After rebuilding my disk, things went well for awhile.  But just now I
    had my terminal windows suddenly die with the famous error number
    displayed in the session manager window.  I ran analyze/disk_str, and 
    discovered a couple of errors:
    
$ analyze/disk_str
_Device: dua0:
%VERIFY-I-OPENQUOTA, error opening QUOTA.SYS
-SYSTEM-W-NOSUCHFILE, no such file
%VERIFY-I-BADHIGHWATER, file (2300,7,1) [SYS0.SYSCOMMON.SYSEXE]FTSVQUEUE.DAT;2
        inconsistent highwater mark and EFBLK
%VERIFY-W-DELHEADER, file (2656,2,1) [SYS0.SYSCOMMON.SYSLIB]DCLTABLES.EXE;12
        marked for delete
%VERIFY-W-LOSTHEADER, file (2656,2,1) DCLTABLES.EXE;12
        not found in a directory
    
    I'm not sure if these errors are connected with the problem; I believe
    that I get them whenever I do an install.  

452.27JAMMER::JACKMarty JackTue Sep 19 1989 11:442
    These errors are not connected with the problem.

452.28One cause and cure, and a question on LOGOUTFOOBAR::KABELdoryphoreThu Sep 21 1989 18:5523
    The infamous error number has appeared on my system recently as a
    result of some fidding I did to my DECW$LOGIN.COM file.  Unfiddling
    removed the problem.
    
    What I had done (wrong) was to insert a symbol definition as
    follows:
	$	lo*gout	== "@'disk':<'topdir'.com>decw_logout "
    into the file.  After logging out and logging in again, subsequent
    logouts and logins gave me two login screens (logo and box) and
    heaps o' trouble.  I quickly reverted to the previous version of my
    DECW$LOGIN.COM _and_ rebooted, and the problems disappeared.  My
    system is a VS3100 running Desktop VMS (5.1-1B).  I wouldn't presume
    that this is the only way to get the error message, but it certainly
    is one way.
    
    In a related vein, I used to have a logout procedure defined by
        $       lo*gout = "@'disk':<'topdir'.com>decw_logout "
    in the DECW$LOGIN.COM, and noticed that it wasn't being run since
    moving to this system (it worked on a VS2000 4MB VMS 5.0), from
    which I recently upgraded.  How can I specify a logout procedure?  I
    used it to reassign OPCOM messages to opa0 and more-or-less
    gracefully kill miscellaneous processes when I quit my session.

452.29Large terminal windows seem to cause the problem.HYDRA::SOLOWAYMon Sep 25 1989 14:175
    I've noticed that if I give myself a really large window and TYPE a
    large file, I consistently get the error.  "set term/page=63/wid=132"
    guarantees a crash, but it also happens a lot with my normal window
    size, 54X80.

452.30Problem seems to be fixed nowHPSRAD::SOLOWAYWed Oct 11 1989 10:422
    The new release of DecWindows seems to fix the problem.

452.31DECterm Controller stops # 12DB821CUJEST::SOPRANOLauri 264-7821Wed Nov 01 1989 16:2210
    I have a VAXstation 2000 running VMS 5.1-1, DECwindows and 6MB memory.
    Periodically, the DECterm controller process stops, killing all
    DECterm windows (usually 2 or 3).  The message number is always: 
    12DB821C.  Then the DECterm controller starts automatically again.
    
    This only happens when I am using WPS-PLUS/ALL-IN-1 in a DECterm
    window.  It happened more frequently when I was editing a "wide"
    document, although it does happen with a standard size document.
    
    Any ideas?
452.32Try hereAGBEAR::HORNERA.G.Bear, Old fashion teddy bearWed Nov 01 1989 17:238
Try these...

        Dave

            -< VMS V5.3 kit available with DW V2, see note 1639.0 >-
 Topic  Author               Date         Repl  Title
   452  USMRM4::PKADOW       22-MAR-1989     8  Error 12DB821C
  1204  TAMARI::MCGOVERN     31-JUL-1989    21  Message 12DB821C (AGAIN!)
452.34Moved by Moderator - AJLUJEST::SOPRANOLauri 264-7821Thu Nov 02 1989 14:017
    I have a VAXstation 2000 with 6MB running VMS V5.1-1 and DECwindows.
    Periodically while editing a WPS-PLUS/ALL-IN-1 file in a DECterm
    window, the DECterm controller process stops.  The error number is 12DB821C
    every time.  It kills all of my DECterm sessions (generally 2 to
    3) and immediately starts the DECterm controller again.
    
    Does anyone know what might be wrong?
452.35STAR::ORGOVANVince OrgovanFri Nov 03 1989 05:523
    x2DB821C is XLIB-E-IOERROR. Note 1532.0 in this conference gives a
    good description of the problem, and describes how it is fixed in
    VMS V5.3.