We are using two slaves this way:
- Active server moves WAL files to local directory,
doesn't care about slaves
(basically archive_command = 'mv %p /some/arch_dir/%f')
- Slaves pull WAL files via rsync
- Slaves also do full syncs every several hours (pg_start_backup / rsync /
pg_stop_backup)
- Ac
Hello,
I've got following two-tables events queue implementation
(general idea is that multiple writers put events, while multiple readers
retrieve and handle them in order):
Table events:
ev_id: SERIAL
ev_data: bytea -- serialized event details
Table eventsconsumers:
con_name: text UNIQUE -- c
> Quick question, you mentioned LVM snapshots, and I am not aware what
> it does! Is there any doc that explains it? Can you point me to some
> such resources?
Newer LVM versions support read/write snapshots [essentially cheap clone of
all partition data].
http://tldp.org/HOWTO/LVM-HOWTO/snapshot
Hi,
I've got following online-backup setup (v. 8.1.8):
- on master
-- archive_command = 'mv %p //%f
-- rsyncd with access to DB data and
- on slave
-- rsync client running every 10sec. to sync and
directories to slave:
rsync -a --delete master::
rsync -a --delete master::
-- pg_start_backu
