From: | Tomas Vondra <tv(at)fuzzy(dot)cz> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Replication: slave is in permanent startup 'recovery' |
Date: | 2011-04-13 20:25:32 |
Message-ID: | 4DA606BC.8060905@fuzzy.cz |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Dne 13.4.2011 20:42, Henry C. napsal(a):
>
> Forgot to mention recovery.conf on slave:
>
> standby_mode = 'on'
> primary_conninfo = 'host..."
> restore_command = 'cp /home/psql-wal-archive/%f "%p"'
> archive_cleanup_command = 'pg_archivecleanup /home/psql-wal-archive %r'
>
>
> The wiki states "If wal_keep_segments is a high enough number to retain the
> WAL segments required for the standby server, this [restore_command] may not
> be necessary."
>
> Presumably this is referring to wal_keep_segments config on the master, right?
Right. If there are enough WAL segments on the master, the standby may
ask for them and the data will be streamed to the standby. So the
archive mode is not a requirement, although if you already use WAL
archiving, it's a good idea to use it (no additional overhead on the
primary etc.).
regards
Tomas
From | Date | Subject | |
---|---|---|---|
Next Message | Jorge Arévalo | 2011-04-13 20:42:58 | Re: Memory management in Postgres |
Previous Message | Simon Riggs | 2011-04-13 20:14:11 | Re: Memory management in Postgres |