Re: warm standby server stops doing checkpointsafterawhile

From: Frank Wittig <fw(at)weisshuhn(dot)de>
To: Simon Riggs <simon(at)2ndquadrant(dot)com>
Cc: Teodor Sigaev <teodor(at)sigaev(dot)ru>, Postgres General <pgsql-general(at)postgresql(dot)org>
Subject: Re: warm standby server stops doing checkpointsafterawhile
Date: 2007-06-01 14:48:30
Message-ID: 466031BE.30508@weisshuhn.de
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Hi Simon,

> The attached patch should show which of these it is.

As I thought, I can trigger the issue by letting the database recreate
and reindex some ten thousand data sets.
Here is what happened in the log (I cleaned it to the lines your patch
provides):

<2007-06-01 16:28:51.708 CEST:%> LOG: GIN incomplete splits=1
<2007-06-01 16:28:51.708 CEST:%> LOG: GIN incomplete split root:11118
l:45303 r:111740 at redo CA/C8243C28
<2007-06-01 16:29:13.912 CEST:%> LOG: GIN incomplete splits=1
<2007-06-01 16:29:13.912 CEST:%> LOG: GIN incomplete split root:11118
l:45303 r:111740 at redo CA/C8243C28
<2007-06-01 16:29:51.594 CEST:%> LOG: GIN incomplete splits=1
<2007-06-01 16:29:51.594 CEST:%> LOG: GIN incomplete split root:11118
l:45303 r:111740 at redo CA/C8243C28
<2007-06-01 16:30:19.351 CEST:%> LOG: GIN incomplete splits=2
<2007-06-01 16:30:19.351 CEST:%> LOG: GIN incomplete split root:11118
l:45303 r:111740 at redo CA/C8243C28
<2007-06-01 16:30:19.351 CEST:%> LOG: GIN incomplete split root:4269
l:102123 r:111741 at redo CA/EEAD8B80
<2007-06-01 16:31:41.991 CEST:%> LOG: GIN incomplete splits=2
<2007-06-01 16:31:41.991 CEST:%> LOG: GIN incomplete split root:11118
l:45303 r:111740 at redo CA/C8243C28
<2007-06-01 16:31:41.991 CEST:%> LOG: GIN incomplete split root:4269
l:102123 r:111741 at redo CA/EEAD8B80
<2007-06-01 16:34:09.217 CEST:%> LOG: GIN incomplete splits=2
<2007-06-01 16:34:09.218 CEST:%> LOG: GIN incomplete split root:11118
l:45303 r:111740 at redo CA/C8243C28
<2007-06-01 16:34:09.218 CEST:%> LOG: GIN incomplete split root:4269
l:102123 r:111741 at redo CA/EEAD8B80
<2007-06-01 16:35:31.933 CEST:%> LOG: GIN incomplete splits=2
<2007-06-01 16:35:31.933 CEST:%> LOG: GIN incomplete split root:11118
l:45303 r:111740 at redo CA/C8243C28
<2007-06-01 16:35:31.933 CEST:%> LOG: GIN incomplete split root:4269
l:102123 r:111741 at redo CA/EEAD8B80
<2007-06-01 16:36:56.391 CEST:%> LOG: GIN incomplete splits=2
<2007-06-01 16:36:56.391 CEST:%> LOG: GIN incomplete split root:11118
l:45303 r:111740 at redo CA/C8243C28
<2007-06-01 16:36:56.392 CEST:%> LOG: GIN incomplete split root:4269
l:102123 r:111741 at redo CA/EEAD8B80
<2007-06-01 16:38:23.133 CEST:%> LOG: GIN incomplete splits=2
<2007-06-01 16:38:23.133 CEST:%> LOG: GIN incomplete split root:11118
l:45303 r:111740 at redo CA/C8243C28
<2007-06-01 16:38:23.133 CEST:%> LOG: GIN incomplete split root:4269
l:102123 r:111741 at redo CA/EEAD8B80

It didnt do us the favor to produce more incomplete splits. But these
two are enough to keep the standby server from doing checkpoints.

Greetings,
Frank Wittig

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Csaba Nagy 2007-06-01 14:49:22 Re: Autovacuum keeps vacuuming a table disabled in pg_autovacuum
Previous Message Erik Jones 2007-06-01 14:43:51 Re: Autovacuum keeps vacuuming a table disabled in pg_autovacuum