Thread: Failover on Windows
Hi all I'm testing a warm standby setup using PostgreSQL 9 x64 on Windows 2008 R2. The problem is that when I put the trigger file on the location specified in the parameter 'trigger_file' of the recovery.conf, nothing happens. No log entries, the recovery just continues as if nothing has happened. Any clues of what may be wrong? Thanks for the attention. Norberto
On Fri, Oct 29, 2010 at 9:58 PM, Norberto Delle <betodelle@gmail.com> wrote: > I'm testing a warm standby setup using PostgreSQL 9 x64 on Windows 2008 R2. What command (pg_standby? cp?) is supplied in restore_command for warm-standby? Or you are testing streaming replication + hot standby? > The problem is that when I put the trigger file on the location specified in > the parameter > 'trigger_file' of the recovery.conf, nothing happens. No log entries, the > recovery just continues > as if nothing has happened. > Any clues of what may be wrong? At least if you use pg_standby, you have to create the trigger file on the location specified in -t option of pg_standby. Regards, -- Fujii Masao NIPPON TELEGRAPH AND TELEPHONE CORPORATION NTT Open Source Software Center
Em 1/11/2010 09:00, Fujii Masao escreveu: > On Fri, Oct 29, 2010 at 9:58 PM, Norberto Delle<betodelle@gmail.com> wrote: >> I'm testing a warm standby setup using PostgreSQL 9 x64 on Windows 2008 R2. > What command (pg_standby? cp?) is supplied in restore_command for warm-standby? > Or you are testing streaming replication + hot standby? > >> The problem is that when I put the trigger file on the location specified in >> the parameter >> 'trigger_file' of the recovery.conf, nothing happens. No log entries, the >> recovery just continues >> as if nothing has happened. >> Any clues of what may be wrong? > At least if you use pg_standby, you have to create the trigger file on > the location > specified in -t option of pg_standby. > > Regards, > Hi Masao Yes, I'm using pg_standby in the restore_command. I thought that to specify a trigger_file in the recovery.conf file would be enough to be able to stop the recovery process. So, I ignored the -t option of the pg_standby. By specifying it, now I'm able to stop the recovery process. Thanks for your help. Norberto