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

Conference virke::mrmemo

Title:VAX MAILGATE for MEMO
Moderator:STKHLM::OLSSON
Created:Sat Feb 25 1989
Last Modified:Tue May 14 1996
Last Successful Update:Fri Jun 06 1997
Number of topics:216
Total number of notes:933

90.0. "MRMEMO and INSVIRMEM problems" by STKAI2::RNORDSTROM (Polaris is it !!) Tue Jul 23 1991 13:48

    Hi,
    
    	I have a customer that has had a problem with mrmemo that
        confuses me. After the customer's system had been up for 25 
        days he started to get fatal insvirmem (insufficient virtual
        memory) when anyone tried to use mrmemo. It seemed like the 
        mrmemo server process had "eaten" to much virtual memory and
        now it couldn't claim more. Restarting the server process
        solved the problem temporarily but we excpect to see the
        problem again.
        When the insvirmem messages started the server process had a 
        peak virtual size of 14700. After the process was restarted	
        peak virtual size was back to approx. 10500. Now to the problem,
        the customer claims that each day peak virtual size increments
        with approx. 200 pages. It seems like the server process never 
        releases any virtual pages just "eats" them.
        Virtualpagecount is 65000 and pql_dpgflquota is 16000. 
        Has anybody seen this behaviour before or perhaps it's even 
        expected ??? Is there some special recommendations for sysgen
        parameters when using mrmemo ???
        If some one out there could help me I would surely be glad.
    
    
                     Regards, Richard TSC Lule� Sweden
T.RTitleUserPersonal
Name		DateLines
90.1Same problemEEMELI::VESALAINENDECnet Phase V = OSI for the MassesThu Jul 25 1991 11:5435
    
    We just received a similar call from our customer as stated in .0.
    I have enclosed an extract from the MRMEMO log. This happened
    after the server has been running more than 20 days.
    
    Anyone seen this before?
    
    
Time: 1991-07-18 12:59:56.05; message from server MRMEMO2:

%LIB-F-INSVIRMEM, insufficient virtual memory
%MRMEMO-W-TRACE, traceback forced from the Server Handler
-MRMEMO-I-FSM, idle, connected to MR and MEMO after message indicated in MR mailbox
-MRMEMO-I-RING, ring: '52 19 ][ 4A 52 19 ][ 4A 52 19.', substates: 00000430
%TRACE-W-TRACEBACK, symbolic stack dump follows
module name     routine name                     line       rel PC    abs PC
SRVLOG          SRV$LOG_HANDLER                  3943      000001C9  001E379D
----- above condition handler called with exception 00158214:
%LIB-F-INSVIRMEM, insufficient virtual memory
----- end of exception message
SRVMRD          SRV$MRD_DISASS                  11129      00000092  001F2025
SRVACT          SRV$ACT_M                        4710      0000003D  001EB1A9
SRVDSP          SRV$DSP_FSM                      3310      00000115  001E3299
SRVMMO          SRV$MMO_ONE_LIFE                 4376      000001EF  001E03BF
SRVMMO          SRV$MMO_MAIN                     4208      00000017  001E01C3
                                                           00239BAE  00239BAE
KOTERM          KOTERM                            804      00000039  002375A2
                                                           00239B89  00239B89
KODOC           KODOC                            1768      00000097  002347E4
                                                           00239B89  00239B89
                                                           00264DB2  00264DB2
ADA$ELAB_DDS    ADA$ELAB_DDS                               0000000E  001C0C0E
                                                           00239B89  00239B89
90.2I'm working on it !STKAI2::RNORDSTROMPolaris is it !!Mon Jul 29 1991 13:2110
    .-1
    
    Hi !
    
    	My plan is to wait until Stefan Persson or Anders Olsson, two of
        the Swedish developers of MRMEMO is back from their vacation.
        Then I will confront them with the problem, we'll post whatever
    	we can come up with on this problem.
    
                         Regards Richard
90.3Workaround by restarting server regularlySTKOFF::SPERSSONPas de problemeThu Aug 08 1991 10:1513
    
    Hi,
    
    We have not tracked down the reason for this problem. Regrettably, we
    have very little resources to do so in near future. Therefore, I
    suggest you workaround the problem by restarting the server at regular
    intervals (once every night, week-end or whatever seems fit).
    
    Sorry for not being more helpful.
    
    regards,
    
    	Stefan
90.4Any logfile tracebacks?STKHLM::OLSSONAnders Olsson, SIP SwedenWed Aug 14 1991 21:1016
    Some problems with memory that was never deallocated were fixed in
    MRMEMO V2.1. But if you are already running 2.1, I don't know what the
    problem is.

    One thing...
    I think there might be bug in the SNA API that fails to release some
    memory when a connection attempt fails. So if the link between MRMEMO
    and MEMO is down for long periods, memory might be lost while MRMEMO
    tries to reconnect. I'm not sure about this, however - it requires some
    testing.

    Are there any error messages/tracebacks in the server logfile or does
    the memory grow under "perfect" conditions too (nothing in the logfile)?


    Anders
90.5SNA-API V2.3 does eat memory!STKHLM::OLSSONAnders Olsson, SIP SwedenThu Sep 12 1991 20:305
    The theory that the SNA API eats memory when connection attempts fail,
    has now been confirmed. The problem is fixed in S2.3-6, according to
    note 2999 in FORTY2::MAILBUS.

    Anders
90.6ahaKETJE::VANHOOSTEGuide to ShadowlandWed Nov 06 1991 17:204
That's why I got these messages when trying to connect to the VBO IBM system 
a couple of weeks ago !

					Marc VH.