[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

179.0. "Another MEMO server crash" by EEMELI::VESALAINEN (DECnet Phase V = OSI for the Masses) Wed Jun 02 1993 13:42

    The Finnish Telecom is using MRMEMO to connect MEMO users to X.400
    though MRX.
    
    They have reported a problem where MEMO Server is crashing,
    I have enclosed a log file. Any ideas what's going wrong.
    
Time: 1993-06-02 13:06:50.21; message from server MRMEMO5:de%MRMEMO-F-MESSED, i`
%MRMEMO-W-TRACE, traceback forced from the Server Handler
-MRMEMO-I-FSM, idle, connected to MR and MEMO after message indicated in MR mai`
-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 PCde
SRVLOG          SRV$LOG_HANDLER                  3943      000001C9  001E379D
----- above condition handler called with exception 0893803C:
%MRMEMO-F-MESSED, internal error in VAX MAILGATE for MEMO
----- end of exception message
ALLPAC          ALL$PAC_END_LLIDF                2621      000000BA  001E94E0
ALLPAC          ALL$PAC_DOC_UNIT                 4247      0000007F  001EA59F
ALLPAC          ALL$PAC_MSGITMLST_TO_GIUBUF      4394      000000BA  001EA6EF
SRVACT          SRV$ACT_M                        4773      000000B4  001EB220
SRVDSP          SRV$DSP_FSM                      3310      00000115  001E3299
SRVMMO          SRV$MMO_ONE_LIFE                 4376      000001EF  001E03BF
SRVMMO          SRV$MMO_MAIN                     4208      00000017  001E01C3
T.RTitleUserPersonal
Name
DateLines
179.1Unknown problem - need .NBS file or logSTKHLM::OLSSONAnders Olsson, SIP SwedenWed Jun 02 1993 20:4219
    When this happens, I presume there is a message in the MR mailbox that
    is "stuck" (i.e. repeatedly fetched by MRMEMO, which crashes again and
    again).

    Can this be reproduced at will by sending a certain message?
    When (and if) this happens again, do you think that you could have the
    customer save the .NBS file of the trouble message (i.e. the first
    message in MRMEMO's mailbox) so that I can have a look at it?

    Another thing that could be of help is if the logical name
    MRMEMO$LOGMASK is defined (/SYSTEM) to "20" when this happens. A formatted
    dump of the message will then be written to the MRMEMO log file.
    
    I don't know what's wrong but a guess is that it could have something
    to do with large messages, which MRMEMO sometimes doesn't handle very
    well.

    Anders
179.2Too big message causing the same problemEEMELI::ALADINMon Jun 07 1993 09:39170
	Hey, Anders !

	Another customer of ours, VTKK has obviously got the same
	problem which Ari mentioned about. I just got a log from
	them, which starts when MRMEMO receives a 470 blocks long
	message. It is obviously a message, that is too big ! 
	The thing which a little confuses me is that the
	first log is stamped at 21:30 and the stack dump log message
	is stamped first at 23:33. 
	Is it the long message that causes the stack dump later or what ??
	
	Here is the log from our customer:


Time: 1993-05-11 21:30:31.28; message from server MRMEMO1:
%MRMEMO-I-DISREJ, distribution rejected by MEMO system:
-MRMEMO-S-RCERRTXT_02,   Memo is too big.
Time: 1993-05-11 23:33:30.76; message from server MRMEMO1:
%MRMEMO-F-MESSED, internal error in VAX MAILGATE for MEMO
%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: '1A 1A 1A 1A 1A 1A 1A 1A 1A 19.', substates: 00000430
%TRACE-W-TRACEBACK, symbolic stack dump follows
module name     routine name                     line       rel PC    abs PC
SRVLOG          SRV$LOG_HANDLER                  3943      000001C9  001E379D
----- above condition handler called with exception 0893803C:
%MRMEMO-F-MESSED, internal error in VAX MAILGATE for MEMO
----- end of exception message
ALLPAC          ALL$PAC_END_LLIDF                2621      000000BA  001E94E0
ALLPAC          ALL$PAC_DOC_UNIT                 4247      0000007F  001EA59F
ALLPAC          ALL$PAC_MSGITMLST_TO_GIUBUF      4394      000000BA  001EA6EF
SRVACT          SRV$ACT_M                        4773      000000B4  001EB220
SRVDSP          SRV$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
                                                           0026537D  0026537D
ADA$ELAB_DDS    ADA$ELAB_DDS                               0000000E  001C0C0E
                                                           00239B89  00239B89
                                                           00239B89  00239B89
.
.
.
.    The log continues here until:
.
.
.
.
.
Time: 1993-05-11 23:59:32.11; message from server MRMEMO1:
%MRMEMO-F-MESSED, internal error in VAX MAILGATE for MEMO
%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$LOG_HANDLER                  3943      000001C9  001E379D
----- above condition handler called with exception 0893803C:
%MRMEMO-F-MESSED, internal error in VAX MAILGATE for MEMO
----- end of exception message
ALLPAC          ALL$PAC_END_LLIDF                2621      000000BA  001E94E0
ALLPAC          ALL$PAC_DOC_UNIT                 4247      0000007F  001EA59F
ALLPAC          ALL$PAC_MSGITMLST_TO_GIUBUF      4394      000000BA  001EA6EF
SRVACT          SRV$ACT_M                        4773      000000B4  001EB220
SRVDSP          SRV$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
                                                           0026537D  0026537D
ADA$ELAB_DDS    ADA$ELAB_DDS                               0000000E  001C0C0E
                                                           00239B89  00239B89

%SNA-I-UNBINDREC, UNBIND received from IBM application
Time: 1993-05-11 23:59:46.87; message from server MRMEMO1:
%MRMEMO-F-UNBIND, UNBIND request received from MEMO Gateway
%SNA-I-UNBINDREC, UNBIND received from IBM application
Time: 1993-05-12 00:00:55.62; message from server MRMEMO1:
%MRMEMO-F-UNBIND, UNBIND request received from MEMO Gateway
%SNA-I-UNBINDREC, UNBIND received from IBM application
Time: 1993-05-12 00:02:00.59; message from server MRMEMO1:
%MRMEMO-F-UNBIND, UNBIND request received from MEMO Gateway
.
.
.
.    Again the log continues..
.
.
.
.

Time: 1993-05-12 01:04:57.99; message from server MRMEMO1:
%MRMEMO-F-UNBIND, UNBIND request received from MEMO Gateway
%SNA-I-UNBINDREC, UNBIND received from IBM application
Time: 1993-05-12 01:06:01.20; message from server MRMEMO1:
%MRMEMO-F-UNBIND, UNBIND request received from MEMO Gateway
%SYSTEM-F-ACCVIO, access violation, reason mask=01, virtual address=E2D440F8, PC=80000010, PSL=03C00004

  Improperly handled condition, image exit forced.

	Signal arguments	      Stack contents

	Number = 00000005		 80264240
	Name   = 0000000C		 00000002
		 00000001		 7FF1A090
		 E2D440F8		 7FF1A078
		 80000010		 00000004
		 03C00004		 00030001
					 00000000
					 00010001
					 00000000
					 05000001

	Register dump

	R0 = 03C00000  R1 = E2D440F8  R2 = 00010001  R3 = 01455A7C
	R4 = 0145812A  R5 = 003171BC  R6 = 01458132  R7 = 7FFEE1A8
	R8 = 0030BFB4  R9 = 0030BFC0  R10= 0030BFBC  R11= 003171BC
	AP = 7FF1A02C  FP = 7FF19FEC  SP = 7FF1A068  PC = 80000010
	PSL= 03C00004

%SYSTEM-F-ACCVIO, access violation, reason mask=01, virtual address=E2D440F8, PC=80000010, PSL=03C00004

  Improperly handled condition, image exit forced.

	Signal arguments	      Stack contents

	Number = 00000005		 80264240
	Name   = 0000000C		 00000002
		 00000001		 7FF19F90
		 E2D440F8		 7FF19F78
		 80000010		 00000004
		 03C00004		 00030001
					 00000003
					 000323B0
					 00000015
					 03000001

	Register dump

	R0 = 03C00000  R1 = E2D440F8  R2 = 00010001  R3 = 01455A7C
	R4 = 00010001  R5 = 003171BC  R6 = 01458132  R7 = 7FFEE1A8
	R8 = 0030BFB4  R9 = 0030BFC0  R10= 0030BFBC  R11= 003171BC
	AP = 7FF19F2C  FP = 7FF19EEC  SP = 7FF19F68  PC = 80000010
	PSL= 03C00004

$! lis{ys khp 30.1.92 alkaa
$ show symbol $status
  $STATUS == "%X1000000C"
$ reply/all/bell " ******  MRMEMO Server lopetti hommat  ****** "
$! mail nl: system /subj="MRMEMO Server exit !!!!!!!!!!!!!!!!!!"
$! Kellonaika viestiin /tea 8.4.92
$ mail nl: system /subj="MRMEMO Server exit klo 01:06:07.06 !!!!!!!!"
$! lis{ys khp 30.1.92 loppuu
$ logout
  MRMEMO       job terminated at 12-MAY-1993 01:06:11.73

  Accounting information:
  Buffered I/O count:          153607         Peak working set size:    5774
  Direct I/O count:            167363         Peak page file size:     44280
  Page faults:                  54645         Mounted volumes:             0
  Charged CPU time:           0 01:40:18.63   Elapsed time:     1 18:04:20.79
179.3Laaaaarrrge message disturbs MRMEMOSTKHLM::OLSSONAnders Olsson, SIP SwedenMon Jun 07 1993 16:5161
    Some background on message size restrictions imposed by different
    subsystems between MR and MEMO:

      o MEMO has previously (before 3.1) handled a maximum message size
        of 75 KB. Starting with 3.1, MEMO can handle up to 500 KB messages
        (the upper limit is configurable, I believe).

      o In the same time frame as the arrival of MEMO 3.1, a new MEMO/GWY
        (1.1) was released, which can handle large (500 KB) messages.

      o There has however been no new release of MEMO/MRGATE (still at 1.0) to
        handle large messages.

      o MRMEMO V2.1 can handle 500 KB messages.

    So, the only component that has not been changed to match MEMOs ability
    to handle large message is MEMO/MRGATE. This is a piece of Verimation
    software so if you want to be able to pass larger messages through
    MRMEMO, take the request to Verimation.

    Now to the MRMEMO problems:

.2>	Another customer of ours, VTKK has obviously got the same
.2>	problem which Ari mentioned about. I just got a log from
.2>	them, which starts when MRMEMO receives a 470 blocks long
.2>	message. It is obviously a message, that is too big ! 
.2>	The thing which a little confuses me is that the
.2>	first log is stamped at 21:30 and the stack dump log message
.2>	is stamped first at 23:33. 
.2>	Is it the long message that causes the stack dump later or what ??
.2>	
.2>	Here is the log from our customer:
.2>
.2> Time: 1993-05-11 21:30:31.28; message from server MRMEMO1:
.2> %MRMEMO-I-DISREJ, distribution rejected by MEMO system:
.2> -MRMEMO-S-RCERRTXT_02,   Memo is too big.
.2> Time: 1993-05-11 23:33:30.76; message from server MRMEMO1:
.2> %MRMEMO-F-MESSED, internal error in VAX MAILGATE for MEMO
    .
    .
    .

    The error that occured 21:30 was caused by a message that is larger
    than the 75 KB limit that is imposed by MEMO/MRGATE but smaller than
    the MRMEMO 500 KB limit. The message has been formatted by MRMEMO and
    shipped to the IBM, where something (probably MEMO/MRGATE) rejects the
    message because it's too big.

    The second error at 23:33 is probably caused by another message that is
    larger than (or close to) the MRMEMO limit 500 KB. MRMEMO doesn't handle
    this situation well - overflows its buffer and crashes. There is a patch
    for this (note 5.13) that you could try.

    Anders

PS
    In case someone wonders how 470 blocks of message file can become 500 KB
    of MEMO message, it's due to the not very space efficient MEMO protocol.
    Every message line is padded with spaces to the defined message line
    length (specified by MRMMan> DEFINE/TEXT_LINES=WIDTH=n). Also, any tab
    characters in the MR message are converted to spaces.