Re: XLogArchivingActive - Mailing list pgsql-hackers

From Jim C. Nasby
Subject Re: XLogArchivingActive
Date
Msg-id 20060526201111.GI59464@pervasive.com
Whole thread Raw
In response to Re: XLogArchivingActive  (Simon Riggs <simon@2ndquadrant.com>)
Responses Re: XLogArchivingActive
List pgsql-hackers
On Fri, May 26, 2006 at 10:59:37AM +0100, Simon Riggs wrote:
> Not sure what the edit commands are offhand, but we would need the
> following program:
> 
> - edit postgresql.conf
> - pg_reload_conf()
> - wait 30
> - pg_start_backup('blah')

Rather than 'wait 30', ISTM it would be better to just leave archiving
enabled, but not actually archiving WAL files.

Or, setup some mechanism so that you can tell if any commands who's
behavior would change based on archiving are running, and if any of
those that are running think archiving is disabled,
pg_start_backup_online blocks on them.

Also, regarding needing to place an archiver command in
pg_start_backup_online, another option would be to depend on the
filesystem backup to copy the WAL files, and just let them pile up in
pg_xlog until pg_stop_backup_online. Of course, that would require a
two-step filesystem copy, since you'd need to first copy everything in
$PGDATA, and then copy $PGDATA/pg_xlog after you have that.
-- 
Jim C. Nasby, Sr. Engineering Consultant      jnasby@pervasive.com
Pervasive Software      http://pervasive.com    work: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf       cell: 512-569-9461


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Inefficient bytea escaping?
Next
From: Andreas Seltenreich
Date:
Subject: Re: GIN stuck in loop during PITR