Postgres Pro
Downloads
Home   >   mailing lists

Re: Pg_xlog increase due to postgres crash (disk full) - Mailing list pgsql-general

From Cliff de Carteret
Subject Re: Pg_xlog increase due to postgres crash (disk full)
Date
Msg-id CAC+bnxk_yetuwdLU+MsWv-BOcg5Qgj81DrB7xqLnWmsJ=rpr1g@mail.gmail.com
Whole thread Raw
In response to Re: Pg_xlog increase due to postgres crash (disk full)  (Adrian Klaver <adrian.klaver@gmail.com>)
Responses Re: Pg_xlog increase due to postgres crash (disk full)  (Adrian Klaver <adrian.klaver@gmail.com>)
List pgsql-general
Tree view
There are no files in the pg_xlog directory from before the database crash except one file which ends in .backup, there are 759 files. In the pg_xlog/archive_status folder there are 755 files. 
The local ../wal_archive folder has no files in it at all whereas the remote wal_archive folder has 147 files present.

The remote archive folder is not full and has ~14G whereas the local archive folder was previously full but now it has been moved so it has 100G+

The archive directory is a mount and as the user postgres I can copy a file into it successfully.

The current setup has been working successfully for several years until the recent database crash


On 22 January 2013 15:34, Adrian Klaver <adrian.klaver@gmail.com> wrote:
On 01/22/2013 06:59 AM, Cliff de Carteret wrote:
Hello, thanks for your response.

The xlogs that are filling up are in the original data directory and the
wal_keep_segments is commented out on my master and slave postgresql.conf

It is expected that there may a certain increase in WAL files.
 In particular: checkpoint_segments = 16

http://www.postgresql.org/docs/9.2/static/runtime-config-wal.html#GUC-CHECKPOINT-SEGMENTS
"
checkpoint_segments (integer)
Maximum number of log file segments between automatic WAL checkpoints (each segment is normally 16 megabytes). The default is three segments. Increasing this parameter can increase the amount of time needed for crash recovery. This parameter can only be set in the postgresql.conf file or on the server command line.
"

For more information that explains the above see:
http://www.postgresql.org/docs/9.2/static/wal-configuration.html

More questions:

Does it look like any WAL files in the pg_xlog directory are being recycled?

Re your archive error from your previous post, is your archive directory full?

    Assuming the archive directory is remote, is it reachable?


Cliff




--
Adrian Klaver
adrian.klaver@gmail.com

pgsql-general by date:

Previous
From: Adrian Klaver
Date:
Subject: Re: Pg_xlog increase due to postgres crash (disk full)
Next
From: "Kevin Grittner"
Date:
Subject: Re: Pg_xlog increase due to postgres crash (disk full)