diff options
| -rw-r--r-- | ops/robust-psql.md | 43 |
1 files changed, 34 insertions, 9 deletions
diff --git a/ops/robust-psql.md b/ops/robust-psql.md index 20d0cbd..b378bf8 100644 --- a/ops/robust-psql.md +++ b/ops/robust-psql.md @@ -174,33 +174,58 @@ done # Master demotion procedure -TBD +In this procedure, the server being demoted is referred to as the "old master" +and the new master is referred to as such. -Apparently this is somewhat involved +1. Prerequisite: complete the standby promotion procedure +1. Create a replication slot for the old master: -✗ TODO: Test this process + SELECT * FROM pg_create_physical_replication_slot('name...'); +1. Create a user for the old master with login and replication privs +1. Fill out `primary_slot_name` and `primary_conninfo` in the old master's + config, and set `hot_standby` to on +1. Confirm that the old master has replication access in the new master's + `pg_hba.conf` +1. Start the old master and monitor the logs, it should catch up +1. Sanity check the database -# Database upgrade procedure +The old master is now the standby, and the old standby is now the master. -TBD +✓ Process tested -✗ TODO: Test this process +# Database upgrade procedure ## Minor version upgrades -TBD +This process will incur a short service degredation. The services will be in +read-only mode for up to 10 minutes. + +1. Complete the failover procedure +1. Perform the upgrade and restart the postgresql daemon +1. Reverse the failover procedure +1. Upgrade the standby and restart the standby ✗ TODO: Test this process ## Major version upgrades -TBD +This process will incur a longer service degredation. The services will be in +read-only mode for up to 30 minutes. + +1. Complete the failover procedure +1. Perform the upgrade and reboot the host +1. Sanity check the database +1. Reverse the failover procedure +1. Complete the "Preparing the hot standby" procedure to rebuild the standby + from scratch; do system updates and reboot ✗ TODO: Test this process # Accessing past data -TBD +The easiest way to do this is likely to be mounting an old version of the ZFS +dataset on either the master or standby server, and running a secondary +PostgreSQL instance from it. ✗ TODO: Test this process |