Conference csc32::consolemanager

473.0. "1.5-003 problems" by OPCO::TSG_SJM (Coming live to you from Rosebery) Fri Nov 11 1994 06:30

    
    John,
    
    We copied the new versions of the following images.  This was in
    response to a conversation the other night between yourself and Martin
    Ellis.
    
    CONSOLE$CONMON.EXE 1.5-003
    CONSOLE$ENS_DAEMON.EXE 1.5-003
    CONSOLE$DAEMON.EXE 1.5-003
    CONSOLE$EXTRACT.EXE 1.5-003
    
    Put these images into CONSOLE$IMAGE and restarted PCM.  PCM ran for a
    while OK, but then we lost connection to about 13 systems, and was
    unable to regain connection.
    
    When connecting to a system which is believed to be active, the session
    just hangs, and the console daemon and control processes disappear, but
    the notify process remains active.
    
    The following is the logfile for one of the control processes, which
    may shed some light.
    
    $ set noverify
    POLYCENTER Console Manager
    Console Controller Daemon Version V1.5-003
    Copyright (c) 1994 Digital Equipment Corporation. All Rights Reserved
    
    Console port open failure for STCGWY
    Open failure on LAT terminal (server: STC1AS  port: LC-1-5)
    %SYSTEM-F-ABORT, abort
    Console port open failure for STCV17
    Open failure on LAT terminal (server: STC1AS  port: LC-2-8)
    %SYSTEM-F-ABORT, abort
    %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual
    address=0000009B, PC=000086D4, PSL=03C00005
    
      Improperly handled condition, image exit forced.
    
            Signal arguments              Stack contents
    
            Number = 00000005                0037EA00
            Name   = 0000000C                0037EA01
                     00000000                00350500
                     0000009B                00366F50
                     000086D4                00000000
                     03C00005                201C0000
                                             7FC9DB04                                        
                                             7FC9DAE4
                                             00018C0D
                                             00008494
    
            Register dump
    
            R0 = FFFFFFF8  R1 = 83FF6050  R2 = 00000000  R3 = 0037EA00
            R4 = 00000000  R5 = 00366F50  R6 = 00019484  R7 = 00000000
            R8 = 00000000  R9 = 00000000  R10= 00000000  R11= 00000000
            AP = 7FC9DA60  FP = 7FC9DA20  SP = 7FC9DA9C  PC = 000086D4
            PSL= 03C00005
     
      SYSTEM       job terminated at 11-NOV-1994 11:17:42.77
    
      Accounting information:
      Buffered I/O count:             515         Peak working set size:    3772
      Direct I/O count:               197         Peak page file size:     15933
      Page faults:                   2182         Mounted volumes:             0
      Charged CPU time:           0 00:00:03.57   Elapsed time:     0 00:00:36.75
    
    
    I have since restored the daemon image back to 1.5-000, and all is now
    working fine. We will be replacing our DECserver 500 with a DECserver
    900 hub this weekend, so that should hopefully also fix a few
    unexplainables.
    
    Cheers
    Steve

Hmm,

  The only differences in the new and old image is the way the "child"
  controllers start up, they are run through loginout so that we can
  create seperate log files for each child to help with debugging.

  I have found from experience that "%SYSTEM-F-ABORT, abort" is typically
  caused by a problem connecting to the terminal server.

  As for the accvio, well, I will have to take a look. Would it be possible
  for you to mail me a copy of your database in export format?

  Throw those DECserver 500's away ASAP, we had no end of trouble with
  ours when we used them for testing, they are the worst servers digital
  produced IMHO.

Cheers,
Phil