segfaults / backend crashing - Mailing list pgsql-general

From Jeffrey Melloy
Subject segfaults / backend crashing
Date
Msg-id E05262D9-DD5D-11D7-8221-000393C78AC0@visualdistortion.org
Whole thread Raw
Responses Re: segfaults / backend crashing  ("Joshua D. Drake" <jd@commandprompt.com>)
Re: segfaults / backend crashing  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-general
I'm having a problem with the backend occasionally crashing.

I have interfaces with the database in two different applications -- a
web viewer using JDBC and an insertion routine written using the c
libraries.  It makes fairly heavy use of the tsearch protocol.  (I
haven't been able to figure out anything repetitious, it happens at
different times, and after I get a new connection the exact search
query or insertion string works fine).

All that's printed to the postgres log is this:
2003-09-02 10:41:18 [428]    LOG:  server process (pid 479) was
terminated by signal 11
2003-09-02 10:41:18 [428]    LOG:  terminating any other active server
processes
2003-09-02 10:41:18 [428]    LOG:  all server processes terminated;
reinitializing shared memory and semaphores
2003-09-02 10:41:18 [510]    LOG:  database system was interrupted at
2003-09-02 10:17:38 CDT
2003-09-02 10:41:18 [428]    LOG:  all server processes terminated;
reinitializing shared memory and semaphores
2003-09-02 10:41:18 [510]    LOG:  database system was interrupted at
2003-09-02 10:17:38 CDT
2003-09-02 10:41:18 [510]    LOG:  checkpoint record is at 0/53E94490
2003-09-02 10:41:18 [510]    LOG:  redo record is at 0/53E94490; undo
record is at 0/0; shutdown TRUE
2003-09-02 10:41:18 [510]    LOG:  next transaction id: 127192; next
oid: 1562034
2003-09-02 10:41:18 [510]    LOG:  database system was not properly
shut down; automatic recovery in progress
2003-09-02 10:41:18 [510]    LOG:  redo starts at 0/53E944D0
2003-09-02 10:41:18 [510]    LOG:  ReadRecord: record with zero length
at 0/53EA28E0
2003-09-02 10:41:18 [510]    LOG:  redo done at 0/53EA282C
2003-09-02 10:41:20 [510]    LOG:  database system is ready

The crash log associated with that crash isn't particularly useful, at
least to me:
Date/Time:  2003-09-02 10:41:17 -0500
OS Version: 10.2.6 (Build 6L60)
Host:       dahak.local.

Command:    postmaster
PID:        479

Exception:  EXC_BAD_ACCESS (0x0001)
Codes:      KERN_INVALID_ADDRESS (0x0001) at 0xffffe640

Thread 0 Crashed:

PPC Thread State:
   srr0: 0xbfffe5e0 srr1: 0x0200f930                vrsave: 0x00000000
    xer: 0x00000000   lr: 0xbfffe5e0  ctr: 0x90010000   mq: 0x00000000
     r0: 0xbfffe5e0   r1: 0xbfffe5d0   r2: 0x8fe4e3e0   r3: 0x00000000
     r4: 0x0088ec1c   r5: 0x00000007   r6: 0x00000006   r7: 0x0088eba4
     r8: 0x0000000c   r9: 0x00000000  r10: 0x004594d4  r11: 0x008a0150
    r12: 0x90010000  r13: 0x00000000  r14: 0x00000000  r15: 0x00000000
    r16: 0x00000000  r17: 0x00000000  r18: 0x00000000  r19: 0x00000000
    r20: 0x00000000  r21: 0x00000000  r22: 0x00000000  r23: 0x00000000
    r24: 0x00000000  r25: 0x00000000  r26: 0xbffffbd4  r27: 0x0000000c
    r28: 0x00000002  r29: 0x00459534  r30: 0x00000000  r31: 0x00000000

One of the crashlogs for the JDBC connection looks like this:
Date/Time:  2003-08-13 01:10:21 -0500
OS Version: 10.2.6 (Build 6L60)
Host:       dahak.local.

Command:    postmaster
PID:        460

Exception:  EXC_BAD_ACCESS (0x0001)
Codes:      KERN_PROTECTION_FAILURE (0x0002) at 0x00daec05

Thread 0 Crashed:
  #0   0x90010070 in __swsetup
  #1   0x00d9f8d8 in compare
  #2   0x00d9f5a0 in merge
  #3   0x00d9f788 in sort
  #4   0x00d9fbc0 in create_pool
  #5   0x00da1b40 in setup_english_stemmer
  #6   0x00da23f4 in initmorph
  #7   0x00da43c4 in tsearch
  #8   0x00064904 in ExecCallTriggerFunc (trigger.c:1120)
  #9   0x00064a7c in ExecBRInsertTriggers (trigger.c:1171)
  #10  0x00071018 in ExecInsert (execMain.c:1216)
  #11  0x00070ee4 in ExecutePlan (execMain.c:1102)
  #12  0x000701fc in ExecutorRun (execMain.c:195)
  #13  0x000d31dc in ProcessQuery (pquery.c:247)
  #14  0x000d0f8c in pg_exec_query_string (postgres.c:839)
  #15  0x000d24d4 in PostgresMain (postgres.c:2020)
  #16  0x000b379c in DoBackend (postmaster.c:2293)
  #17  0x000b2ff4 in BackendStartup (postmaster.c:1916)
  #18  0x000b1ec0 in ServerLoop (postmaster.c:1006)
  #19  0x000b19f8 in PostmasterMain (postmaster.c:785)
  #20  0x00086670 in main (main.c:210)
  #21  0x00001b58 in _start (crt.c:267)
  #22  0x000019d8 in start

PPC Thread State:
   srr0: 0x90010070 srr1: 0x0200f930                vrsave: 0x00000000
    xer: 0x00000000   lr: 0x90010070  ctr: 0x90003900   mq: 0x00000000
     r0: 0x90010070   r1: 0xbfffe5f0   r2: 0x00000000   r3: 0x000000e3
     r4: 0x00000000   r5: 0x000000e3   r6: 0x0000000a   r7: 0x00000020
     r8: 0x00000030   r9: 0x000000e3  r10: 0x00000060  r11: 0xa00043ac
    r12: 0x90003900  r13: 0x001a0c10  r14: 0x009b4f58  r15: 0xbfffed60
    r16: 0x00d37e00  r17: 0x00000000  r18: 0x00000000  r19: 0x00000000
    r20: 0x00d367e8  r21: 0x00d385f0  r22: 0x00d36958  r23: 0x00d36af8
    r24: 0x00000001  r25: 0x00d36f28  r26: 0x00d36a48  r27: 0x00d385f0
    r28: 0xbfffeaa0  r29: 0x00000000  r30: 0x00daebd5  r31: 0x9001000c

Any help would be appreciated.
Jeffrey Melloy
jmelloy@visualdistortion.org


pgsql-general by date:

Previous
From: Gianni Mariani
Date:
Subject: Re: About GPL and proprietary software
Next
From: Ron Johnson
Date:
Subject: Re: delivering database stand-alone