On Wed, Mar 14, 2012 at 11:07 AM, Joseph Shraibman <jks@xxxxxxxxxxxxxxx> wrote: > I have twice set up pg hot standbys ala the docs at > http://www.postgresql.org/docs/9.1/interactive/hot-standby.html > > The third time I'm trying this I'm running into trouble. The first two > times were with actual servers. This time I'm trying to set up two pg > instances on my desktop for testing. > > First I create the secondary: > > [jks@jks-desktop ~/work/pgmon]{f15}$ time pg_basebackup -D repl-db -P -h > localhost -U replicator > 59303/59303 kB (100%), 1/1 tablespace > NOTICE: pg_stop_backup complete, all required WAL segments have been > archived > > real 0m1.725s > user 0m0.061s > sys 0m0.265s > > Then I copy in recovery.conf and the replacement postgresql.conf and try to > start up. I get: > > > LOG: database system was interrupted; last known up at 2012-03-13 21:29:32 > EDT > LOG: could not open file "pg_xlog/00000001000000000000003D" (log file 0, > segment 61): No such file or directory > LOG: invalid checkpoint record > FATAL: could not locate required checkpoint record > HINT: If you are not restoring from a backup, try removing the file > "/home/jks/work/pgmon/repl-db/backup_label". > LOG: startup process (PID 28220) exited with exit code 1 > LOG: aborting startup due to startup process failure > > Now the file 00000001000000000000003D does exist in the archive directory, > but it appears that restore_command is not being run. Originally it was: > 'cp /home/jks/work/pgmon/wal_drop/%f %p' > > Then I changed it to: > > restore_command = 'echo f %f p %p >> /tmp/rc.log ; cp > /home/jks/work/pgmon/wal_drop/%f %p' > > /tmp/rc.log was never created, so I assume the whole thing isn't being run > for some reason. Any clues where I should look? Confirm that recovery.conf is properly located under the data directory. If recovery.conf is located properly and standby_mode is enabled there, you should get the following log message at the start of recovery: LOG: entering standby mode But you didn't get such message. So I guess that PostgreSQL failed to read recovery.conf and could not run restore_command because of wrong location of recovery.conf. Regards, -- Fujii Masao NIPPON TELEGRAPH AND TELEPHONE CORPORATION NTT Open Source Software Center -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
