hot standby failed to start

From: Rural Hunter <ruralhunter(at)gmail(dot)com>
To: pgsql-admin(at)postgresql(dot)org
Subject: hot standby failed to start
Date: 2011-12-09 07:19:42
Message-ID: 4EE1B68E.8000209@gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

Hi,

I'm trying to set up a hot standby server with both primary and standby
are at latest version 9.1.2. I used pg_basebackup to create the backup
and there was no problem in the log:
NOTICE: pg_stop_backup complete, all required WAL segments have been
archived
-rw-r--r-- 1 postgres postgres 52203357921 2011-12-09 02:53
/dbbk/postgres/base/01/base_20111208.tar.gz

After I set up all configuration files at standby, I started the standby
but it failed with this:
2011-12-09 14:40:28 CST LOG: restored log file
"000000020000081B000000F1" from archive
2011-12-09 14:40:29 CST LOG: restored log file
"000000020000081B000000F2" from archive
2011-12-09 14:40:30 CST LOG: restored log file
"000000020000081B000000F3" from archive
2011-12-09 14:40:32 CST LOG: restored log file
"000000020000081B000000F4" from archive
2011-12-09 14:40:33 CST LOG: restored log file
"000000020000081B000000F5" from archive
2011-12-09 14:40:34 CST LOG: restored log file
"000000020000081B000000F6" from archive
2011-12-09 14:40:34 CST PANIC: heap_update_redo: invalid lp
2011-12-09 14:40:34 CST CONTEXT: xlog redo update: rel
1663/16405/8694402; tid 1529971/91; new 1529971/92
2011-12-09 14:40:34 CST LOG: startup process (PID 7728) was terminated
by signal 6: Aborted
2011-12-09 14:40:34 CST LOG: terminating any other active server processes

Here are the wal archives near the fail log:
-rw------- 1 postgres postgres 5464161 2011-12-09 02:53
000000020000081B000000F0.gz
-rw------- 1 postgres postgres 5744461 2011-12-09 02:53
000000020000081B000000F1.gz
-rw------- 1 postgres postgres 5637256 2011-12-09 02:53
000000020000081B000000F2.gz
-rw------- 1 postgres postgres 5685037 2011-12-09 02:53
000000020000081B000000F3.gz
-rw------- 1 postgres postgres 5612374 2011-12-09 02:53
000000020000081B000000F4.gz
-rw------- 1 postgres postgres 228 2011-12-09 02:53
000000020000081200000049.000000C0.backup.gz
-rw------- 1 postgres postgres 5591283 2011-12-09 02:53
000000020000081B000000F5.gz
-rw------- 1 postgres postgres 6557821 2011-12-09 02:54
000000020000081B000000F6.gz
-rw------- 1 postgres postgres 6186619 2011-12-09 02:55
000000020000081B000000F7.gz
-rw------- 1 postgres postgres 5980735 2011-12-09 02:56
000000020000081B000000F8.gz

What's wrong with the standby and what can I do to find the root cause
of this?
btw, I set full_page_writes off on primary server because it creates too
many wal files(around 20-30 files every min).

Browse pgsql-admin by date

  From Date Subject
Next Message Jasen Betts 2011-12-10 09:00:26 Re: How could I find the last modified procedure in the database?
Previous Message fanlijing 2011-12-09 03:17:36 Re: Is there any support like "for update wait N" in PostgreSQL?