lukesmith / docker-postgres-wale Goto Github PK

I'm looking to migrate servers and was reading this in regards to fetching WAL data: https://gist.github.com/ruckus/2293434

What are your thoughts about passing in a master/slave environment variable which when bringing up containers?

Combined with docker-compose and swarm, the first node would be set to master and additional nodes would come up as slaves and automatically fetch-backup.

I am trying to restore a DB on a slave node, but am always getting a time out. Am I doing anything wrong?

postgres-slave_1 | LOG: restored log file "000000010000000000000023" from archive
postgres-slave_1 | LOG: server process (PID 195) exited with exit code 2
postgres-slave_1 | LOG: terminating any other active server processes
postgres-slave_1 | FATAL: could not restore file "000000010000000000000024" from archive: child process was terminated by signal 3: Quit
postgres-slave_1 | LOG: all server processes terminated; reinitializing
postgres-slave_1 | LOG: database system was interrupted while in recovery at log time 2016-12-08 09:29:29 UTC
postgres-slave_1 | HINT: If this has occurred more than once some data might be corrupted and you might need to choose an earlier recovery target.
postgres-slave_1 | LOG: startup process (PID 306) was terminated by signal 9: Killed
postgres-slave_1 | LOG: aborting startup due to startup process failure

If you stop the container and start a new one, pointing at an existing /postgres/data directory, then setup-wale.sh does not execute so the cron jobs do not get set up resulting in no backups.