When: During a pre-defined DBA maintenance windows
Affected wikis:: https://noc.wikimedia.org/conf/highlight.php?file=dblists/s3.dblist
Checklist:
NEW primary: db2105
OLD primary: db2127
- Check configuration differences between new and old primary:
sudo pt-config-diff --defaults-file /root/.my.cnf h=db2127.codfw.wmnet h=db2105.codfw.wmnet
Failover prep:
- Silence alerts on all hosts:
sudo cookbook sre.hosts.downtime --hours 1 -r "Primary switchover s3 T324180" 'A:db-section-s3'
- Set NEW primary with weight 0 (and depool it from API or vslow/dump groups if it is present).
sudo dbctl instance db2105 set-weight 0 sudo dbctl config commit -m "Set db2105 with weight 0 T324180"
- Topology changes, move all replicas under NEW primary
sudo db-switchover --replicating-master --read-only-master --timeout=25 --only-slave-move db2127 db2105
- Disable puppet on both nodes
sudo cumin 'db2127* or db2105*' 'disable-puppet "primary switchover T324180"'
- Merge gerrit puppet change to promote NEW primary: FIXME
Failover:
- Log the failover:
!log Starting s3 codfw failover from db2127 to db2105 - T324180
- Switch primaries:
sudo db-switchover --replicating-master --read-only-master --skip-slave-move db2127 db2105 echo "===== db2127 (OLD)"; sudo db-mysql db2127 -e 'show slave status\G' echo "===== db2105 (NEW)"; sudo db-mysql db2105 -e 'show slave status\G'
- Promote NEW primary in dbctl
sudo dbctl --scope codfw section s3 set-master db2105 sudo dbctl config commit -m "Promote db2105 to s3 primary T324180"
- Restart puppet on both hosts:
sudo cumin 'db2127* or db2105*' 'run-puppet-agent -e "primary switchover T324180"'
Clean up tasks:
- Clean up heartbeat table(s).
sudo db-mysql db2105 heartbeat -e "delete from heartbeat where file like 'db2127%';"
- change events for query killer:
events_coredb_master.sql on the new primary db2105 events_coredb_slave.sql on the new slave db2127
- Update candidate primary dbctl and orchestrator notes
sudo dbctl instance db2127 set-candidate-master --section s3 true sudo dbctl instance db2105 set-candidate-master --section s3 false (dborch1001): sudo orchestrator-client -c untag -i db2105 --tag name=candidate (dborch1001): sudo orchestrator-client -c tag -i db2127 --tag name=candidate
- Check zarcillo was updated
- db-switchover should do this. If it fails, do it manually: https://phabricator.wikimedia.org/P13956
sudo db-mysql db1115 zarcillo -e "select * from masters where section = 's3';"
- (If needed): Depool db2127 for maintenance.
sudo dbctl instance db2127 depool sudo dbctl config commit -m "Depool db2127 T324180"
- Change db2127 weight to mimic the previous weight db2105:
sudo dbctl instance db2127 edit
- Apply outstanding schema changes to db2127 (if any)
- Update/resolve this ticket.