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

Conference 7.286::postscript_printing

Title:Digital PostScript printers and their associated software
Moderator:REGENT::LASKOHER
Created:Wed Jan 24 1990
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:7230
Total number of notes:31971

7060.0. "%DCPS-F-INTERNAL_ERROR Invalid state change" by STKHLM::HEGNELIUS () Tue Jan 28 1997 03:35

DCPS symbiont crash suddenly.

A customer suffers from DCPS symbiont crashes now and then.
Earlier they had DCPS 1.2 and had a crash nearly every week.
I suggested them to upgrade to DCPS v1.3 but the crashes still
happens. They are running VMS v.6.2-1H3 and LPS$server is v 5.1-15.

The messages in opcom looks like this:

%%%%%%%%%%%  OPCOM  22-JAN-1997 15:02:28.11  %%%%%%%%%%%
Message from user SYSTEM on BORPRD
Process SYMBIONT_21: %DCPS-F-INTERNAL_ERROR, Internal error detected at line 427
of file DCPS$:[BUILD.V13]DP_STMGT.C;1, Invalid state change, cannot continue
-JBC-W-NOMOREQMGR, no more queue managers found

%%%%%%%%%%%  OPCOM  22-JAN-1997 15:02:28.15  %%%%%%%%%%%
Message from user QUEUE_MANAGE on BORPRD
%QMAN-E-SYMDEL, unexpected symbiont process termination

%%%%%%%%%%%  OPCOM  22-JAN-1997 15:02:28.16  %%%%%%%%%%%
Message from user QUEUE_MANAGE on BORPRD
-NONAME-F-NOMSG, Message number 064B82A4

The queue manager still works because all batchqueues are still running.

I havn't found any traces to what could cause this. He just restart 
DCPS$startup and everything works again.

Any suggesions?

Eva Hegnelius
Telephone support
Sweden
T.RTitleUserPersonal
Name		DateLines
7060.1May be due to insufficient process memory.HANNAH::KAMATHMon Feb 03 1997 10:3819
    Eva,
    
    	I suspect this error is due to insufficient memory
    	for the symbiont process.  Observe the results doing
    	one of these :
    
    		- Increase page file quota for the system.
    	          You need to stop and restart the job controller &
    	          queue manager to get the new value into effect.
    
    		- If DCPS$MAX_STREAMS logical is defined and value is
    		  greater than 1 then try decreasing the value.   You
                  should stop and restart all the DCPS queues subsequent
                  to this change.
    
       If you notice the same behaviour doing these changes please
       capture the DCPS trace for further analysis.
    
    -Harish Kamath.
7060.2probably not the page file quotaSTKHLM::HEGNELIUSTue Feb 04 1997 07:4538
    Unfortunatelly the customer have a page file quota of 800.000 pages 
    on this system.
    
    Here comes a output from the account file:
    
    Username:          SYSTEM            UIC:               [SYSTEM]
    Account:           <start>           Finish time:       22-JAN-1997
    15:02:28.14
    Process ID:        000118C2          Start time:         4-DEC-1996
    17:43:48.26
    Owner ID:                            Elapsed time:               48
    21:18:39.88
    Terminal name:                       Processor time:              0
    01:39:21.29
    Remote node addr:                    Priority:          4
    Remote node name:                    Privilege <31-00>: 00004001
    Remote ID:                           Privilege <63-32>: 00000020
    Remote full name:
    Queue entry:                         Final status code: 064B82A4
    Queue name:
    Job name:
    Final status text: <no text>
    
    
    Page faults:          1749560        Direct IO:             465659
    Page fault reads:      144359        Buffered IO:          1209018
    Peak working set:        9968        Volumes mounted:            0
    Peak page file:         64784        Images executed:            1
     
    So, it should not be a problem with page file.
    
    I have seen the IPMT cases on problems with this on VMS 7.0 but this
    customer is running VMS 6.2-1H3
    
    Any other suggestions?
    
    Regards
    Eva