Page MenuHomePhabricator

Switchover s8 master (db1193 -> db1209)
Closed, ResolvedPublic

Description

When: During a pre-defined DBA maintenance windows

Affected wikis:: https://noc.wikimedia.org/conf/highlight.php?file=dblists/s8.dblist

Checklist:

NEW primary: db1209
OLD primary: db1193

  • Check configuration differences between new and old primary:
sudo pt-config-diff --defaults-file /root/.my.cnf h=db1193.eqiad.wmnet h=db1209.eqiad.wmnet

Failover prep:

  • Silence alerts on all hosts:
sudo cookbook sre.hosts.downtime --hours 1 -r "Primary switchover s8 T381993" 'A:db-section-s8'
  • Set NEW primary with weight 0
sudo dbctl instance db1209 set-weight 0
sudo dbctl config commit -m "Set db1209 with weight 0 T381993"
  • Depool NEW from any specific group (API, vslow, dump) if present.
sudo dbctl instance db1209 edit
# If some changes were made:
sudo dbctl config commit -m "Remove db1209 from API/vslow/dump T381993"
  • Topology changes, move all replicas under NEW primary
sudo db-switchover --timeout=25 --replicating-master --read-only-master --only-slave-move db1193 db1209
  • Disable puppet on both nodes
sudo cumin 'db1193* or db1209*' 'disable-puppet "primary switchover T381993"'

Failover:

  • Log the failover:
!log Starting s8 eqiad failover from db1193 to db1209 - T381993
  • Switch primaries:
sudo db-switchover --replicating-master --read-only-master --skip-slave-move db1193 db1209
echo "===== db1193 (OLD)"; sudo db-mysql db1193 -e 'show slave status\G'
echo "===== db1209 (NEW)"; sudo db-mysql db1209 -e 'show slave status\G'
  • Promote NEW primary in dbctl
sudo dbctl --scope eqiad section s8 set-master db1209
sudo dbctl config commit -m "Promote db1209 to s8 primary T381993"
  • Restart puppet on both hosts:
sudo cumin 'db1193* or db1209*' 'run-puppet-agent -e "primary switchover T381993"'

Clean up tasks:

  • Clean up heartbeat table(s).
sudo db-mysql db1209 heartbeat -e "delete from heartbeat where file like 'db1193%';"
  • change events for query killer:
curl -sS 'https://gerrit.wikimedia.org/r/plugins/gitiles/operations/software/+/refs/heads/master/dbtools/events_coredb_master.sql?format=TEXT' | base64 -d | sudo db-mysql db1209
curl -sS 'https://gerrit.wikimedia.org/r/plugins/gitiles/operations/software/+/refs/heads/master/dbtools/events_coredb_slave.sql?format=TEXT' | base64 -d | sudo db-mysql db1193
  • Update candidate primary dbctl and orchestrator notes
sudo dbctl instance db1193 set-candidate-master --section s8 true
sudo dbctl instance db1209 set-candidate-master --section s8 false
sudo cumin 'dborch*' 'orchestrator-client -c untag -i db1209 --tag name=candidate'
sudo cumin 'dborch*' 'orchestrator-client -c tag -i db1193 --tag name=candidate'
sudo db-mysql db1215 zarcillo -e "select * from masters where section = 's8';"
  • (If needed): Depool db1193 for maintenance.
sudo dbctl instance db1193 depool
sudo dbctl config commit -m "Depool db1193 T381993"
  • Change db1193 weight to mimic the previous weight db1209:
sudo dbctl instance db1193 edit
  • Update/resolve this ticket.

Event Timeline

Change #1102333 had a related patch set uploaded (by Gerrit maintenance bot; author: Gerrit maintenance bot):

[operations/puppet@production] mariadb: Promote db1209 to s8 master

https://gerrit.wikimedia.org/r/1102333

Marostegui triaged this task as Medium priority.
Marostegui moved this task from Triage to Ready on the DBA board.

Mentioned in SAL (#wikimedia-operations) [2025-01-02T11:00:09Z] <root@cumin1002> START - Cookbook sre.hosts.downtime for 1:00:00 on 32 hosts with reason: Primary switchover s8 T381993

Mentioned in SAL (#wikimedia-operations) [2025-01-02T11:00:40Z] <root@cumin1002> END (PASS) - Cookbook sre.hosts.downtime (exit_code=0) for 1:00:00 on 32 hosts with reason: Primary switchover s8 T381993

Weight was 300 and 100 in API

Mentioned in SAL (#wikimedia-operations) [2025-01-02T11:02:33Z] <marostegui@cumin1002> dbctl commit (dc=all): 'Set db1209 with weight 0 T381993', diff saved to https://phabricator.wikimedia.org/P71763 and previous config saved to /var/cache/conftool/dbconfig/20250102-110232-root.json

Mentioned in SAL (#wikimedia-operations) [2025-01-02T11:03:06Z] <marostegui@cumin1002> dbctl commit (dc=all): 'Remove db1209 from API/vslow/dump T381993', diff saved to https://phabricator.wikimedia.org/P71764 and previous config saved to /var/cache/conftool/dbconfig/20250102-110305-root.json

Change #1102333 merged by Marostegui:

[operations/puppet@production] mariadb: Promote db1209 to s8 master

https://gerrit.wikimedia.org/r/1102333

Mentioned in SAL (#wikimedia-operations) [2025-01-02T11:09:03Z] <marostegui> Starting s8 eqiad failover from db1193 to db1209 - T381993

Mentioned in SAL (#wikimedia-operations) [2025-01-02T11:09:24Z] <marostegui@cumin1002> dbctl commit (dc=all): 'Promote db1209 to s8 primary T381993', diff saved to https://phabricator.wikimedia.org/P71765 and previous config saved to /var/cache/conftool/dbconfig/20250102-110923-marostegui.json

Mentioned in SAL (#wikimedia-operations) [2025-01-02T11:11:06Z] <marostegui@cumin1002> dbctl commit (dc=all): 'Depool db1193 T381993', diff saved to https://phabricator.wikimedia.org/P71766 and previous config saved to /var/cache/conftool/dbconfig/20250102-111105-marostegui.json

Marostegui updated the task description. (Show Details)

This is done

Marostegui added a parent task: Restricted Task.Jan 2 2025, 11:21 AM