Re: pqReadData() -- backend closed the channel unexpectedly. - Mailing list pgsql-general

I got an email saying that my message did not make it to the list, so I
am posting again. Hopefully it will go through this time.

 I am running PostgreSQL 7.1.2 on FreeBSD 4.3
 This is an old 486 with only 12MB memory.

 I am able to create databases and insert and retrieve data,
 but when I try to query on the structure of the database
 I am getting this error:

 signin=# \d person_personid_seq
 pqReadData() -- backend closed the channel unexpectedly.
         This probably means the backend terminated abnormally
         before or while processing the request.
 The connection to the server was lost. Attempting reset: Failed.
 !# \q


 I have never seen that prompt before (!#)

 I turned up debugging and here is (some) of the output:

<snip>
 DEBUG:  database system is in production state
 DEBUG:  proc_exit(0)
 DEBUG:  shmem_exit(0)
 DEBUG:  exit(0)
 postmaster: reaping dead processes...
 postmaster: ServerLoop:         handling reading 4
 postmaster: ServerLoop:         handling reading 4
<snip>
 postmaster: ServerLoop:         handling writing 4
 postmaster: BackendStartup: pid 635 user lee db lee socket 4
 postmaster child[635]: starting with (postgres -d5 -d 5 -v131072 -p lee )
 FindExec: searching PATH ...
 FindExec: found "/usr/local/bin/postgres" using PATH
 DEBUG:  connection: host=[local] user=lee database=lee
 DEBUG:  InitPostgres
 DEBUG:  StartTransactionCommand
 DEBUG:  query: SELECT usesuper FROM pg_user WHERE usename = 'lee'
<snip>
 DEBUG:  ProcessQuery
 DEBUG:  CommitTransactionCommand
 DEBUG:  proc_exit(0)
 DEBUG:  shmem_exit(0)
 DEBUG:  exit(0)
 postmaster: reaping dead processes...
 postmaster: CleanupProc: pid 635 exited with status 0
 postmaster: ServerLoop:         handling reading 4
<snip>
 postmaster: ServerLoop:         handling writing 4
 postmaster: BackendStartup: pid 642 user lee db signin socket 4
 postmaster child[642]: starting with (postgres -d5 -d 5 -v131072 -p signin )
 FindExec: searching PATH ...
 FindExec: found "/usr/local/bin/postgres" using PATH
 DEBUG:  connection: host=[local] user=lee database=signin
 DEBUG:  InitPostgres
 DEBUG:  proc_exit(0)
 DEBUG:  shmem_exit(0)
 DEBUG:  exit(0)
 postmaster: reaping dead processes...
 postmaster: CleanupProc: pid 639 exited with status 0
<snip>
 postmaster: reaping dead processes...
 postmaster: CleanupProc: pid 642 exited with status 132
 Server process (pid 642) exited with status 132 at Mon Jul 23 09:46:26 2001
 Terminating any active server processes...
 Server processes were terminated at Mon Jul 23 09:46:26 2001
 Reinitializing shared memory and semaphores
 invoking IpcMemoryCreate(size=1236992)
 DEBUG:  database system was interrupted at 2001-07-23 09:44:48 EDT
 postmaster: ServerLoop:         handling reading 4
 postmaster: ServerLoop:         handling reading 4
 postmaster: ServerLoop:         handling reading 4
 postmaster: ServerLoop:         handling reading 4
 The Data Base System is starting up
 postmaster: ServerLoop:         handling writing 4
 DEBUG:  CheckPoint record at (0, 3580356)
 DEBUG:  Redo record at (0, 3580356); Undo record at (0, 0); Shutdown TRUE
 DEBUG:  NextTransactionId: 1362; NextOid: 101105
 DEBUG:  database system was not properly shut down; automatic recovery in progr
 DEBUG:  ReadRecord: record with zero len at (0, 3580420)
 DEBUG:  redo is not required
 DEBUG:  database system is in production state
 DEBUG:  proc_exit(0)
 DEBUG:  shmem_exit(0)
 DEBUG:  exit(0)
 postmaster: reaping dead processes...
 postmaster: ServerLoop:         handling reading 4
 postmaster: ServerLoop:         handling reading 4
<snip>
 DEBUG:  StartTransactionCommand
 DEBUG:  query: SELECT relhasindex, relkind, relchecks, reltriggers, relhasrules
 FROM pg_class WHERE relname='person_personid_seq'
 DEBUG:  parse tree: { QUERY :command 1  :utility <> :resultRelation 0 :into <>
<snip>
<> :sortClause ({ SORTCLAUSE :tleSortGroupRef 1 :sortop 95 }) :limitOffset <> :l
 imitCount <> :setOperations <> :resultRelations ()}
 postmaster: reaping dead processes...
 postmaster: CleanupProc: pid 654 exited with status 132
 Server process (pid 654) exited with status 132 at Mon Jul 23 09:47:37 2001
 Terminating any active server processes...
 Server processes were terminated at Mon Jul 23 09:47:37 2001
 Reinitializing shared memory and semaphores
 invoking IpcMemoryCreate(size=1236992)
 postmaster: ServerLoop:         handling reading 4
 postmaster: ServerLoop:         handling reading 4
 postmaster: ServerLoop:         handling reading 4
 postmaster: ServerLoop:         handling reading 4
 The Data Base System is starting up
 postmaster: ServerLoop:         handling writing 4
 DEBUG:  database system was interrupted at 2001-07-23 09:46:29 EDT
 DEBUG:  CheckPoint record at (0, 3580420)
 DEBUG:  Redo record at (0, 3580420); Undo record at (0, 0); Shutdown TRUE
 DEBUG:  NextTransactionId: 1362; NextOid: 101105
 DEBUG:  database system was not properly shut down; automatic recovery in progr
 DEBUG:  ReadRecord: record with zero len at (0, 3580484)
 DEBUG:  redo is not required
 DEBUG:  database system is in production state
 DEBUG:  proc_exit(0)
 DEBUG:  shmem_exit(0)
 DEBUG:  exit(0)
 postmaster: reaping dead processes...



Could running out of memory cause the backend to die like that? I have the
exact same schema and data loaded on another 486 (with 64M RAM) and it
runs without any hitches.

If it is a memory problem, any hints on tuning the system so that it
can run without crashing?