When: During a pre-defined DBA maintenance windows
Affected wikis:: https://noc.wikimedia.org/conf/highlight.php?file=dblists/s7.dblist
Checklist:
NEW primary: db2118
OLD primary: db2121
- Check configuration differences between new and old primary:
sudo pt-config-diff --defaults-file /root/.my.cnf h=db2121.codfw.wmnet h=db2118.codfw.wmnet
Failover prep:
- Silence alerts on all hosts:
sudo cookbook sre.hosts.downtime --hours 1 -r "Primary switchover s7 T328000" 'A:db-section-s7'
- Set NEW primary with weight 0 (and depool it from API or vslow/dump groups if it is present).
sudo dbctl instance db2118 set-weight 0 sudo dbctl config commit -m "Set db2118 with weight 0 T328000"
- Topology changes, move all replicas under NEW primary
sudo db-switchover --timeout=25 --replicating-master --read-only-master --only-slave-move db2121 db2118
- Disable puppet on both nodes
sudo cumin 'db2121* or db2118*' 'disable-puppet "primary switchover T328000"'
- Merge gerrit puppet change to promote NEW primary: https://gerrit.wikimedia.org/r/c/operations/puppet/+/883516/
Failover:
- Log the failover:
!log Starting s7 codfw failover from db2121 to db2118 - T328000
- Switch primaries:
sudo db-switchover --replicating-master --read-only-master --skip-slave-move db2121 db2118 echo "===== db2121 (OLD)"; sudo db-mysql db2121 -e 'show slave status\\G' echo "===== db2118 (NEW)"; sudo db-mysql db2118 -e 'show slave status\\G'
- Promote NEW primary in dbctl
sudo dbctl --scope codfw section s7 set-master db2118 sudo dbctl config commit -m "Promote db2118 to s7 primary T328000"
- Restart puppet on both hosts:
sudo cumin 'db2121* or db2118*' 'run-puppet-agent -e "primary switchover T328000"'
Clean up tasks:
- Clean up heartbeat table(s).
sudo db-mysql db2118 heartbeat -e "delete from heartbeat where file like 'db2121%';"
- change events for query killer:
events_coredb_master.sql on the new primary db2118 events_coredb_slave.sql on the new slave db2121
- Update candidate primary dbctl and orchestrator notes
sudo dbctl instance db2121 set-candidate-master --section s7 true sudo dbctl instance db2118 set-candidate-master --section s7 false (dborch1001): sudo orchestrator-client -c untag -i db2118 --tag name=candidate (dborch1001): sudo orchestrator-client -c tag -i db2121 --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 = 's7';"
- (If needed): Depool db2121 for maintenance.
sudo dbctl instance db2121 depool sudo dbctl config commit -m "Depool db2121 T328000"
- Change db2121 weight to mimic the previous weight db2118:
sudo dbctl instance db2121 edit
- Update/resolve this ticket.