Page MenuHomePhabricator

Switchover s5 master (db2123 -> db2213)
Closed, ResolvedPublic

Description

When: During a pre-defined DBA maintenance windows

Affected wikis:: https://noc.wikimedia.org/conf/highlight.php?file=dblists/s5.dblist

Checklist:

NEW primary: db2213
OLD primary: db2123

  • Check configuration differences between new and old primary:
sudo pt-config-diff --defaults-file /root/.my.cnf h=db2123.codfw.wmnet h=db2213.codfw.wmnet

Failover prep:

  • Silence alerts on all hosts:
sudo cookbook sre.hosts.downtime --hours 1 -r "Primary switchover s5 T367145" 'A:db-section-s5'
  • Set NEW primary with weight 0
sudo dbctl instance db2213 set-weight 0
sudo dbctl config commit -m "Set db2213 with weight 0 T367145"
  • Depool NEW from any specific group (API, vslow, dump) if present.
sudo dbctl instance db2213 edit
# If some changes were made:
sudo dbctl config commit -m "Remove db2213 from API/vslow/dump T367145"
  • Topology changes, move all replicas under NEW primary
sudo db-switchover --timeout=25 --replicating-master --read-only-master --only-slave-move db2123 db2213
  • Disable puppet on both nodes
sudo cumin 'db2123* or db2213*' 'disable-puppet "primary switchover T367145"'

Failover:

  • Log the failover:
!log Starting s5 codfw failover from db2123 to db2213 - T367145
  • Switch primaries:
sudo db-switchover --replicating-master --read-only-master --skip-slave-move db2123 db2213
echo "===== db2123 (OLD)"; sudo db-mysql db2123 -e 'show slave status\G'
echo "===== db2213 (NEW)"; sudo db-mysql db2213 -e 'show slave status\G'
  • Promote NEW primary in dbctl
sudo dbctl --scope codfw section s5 set-master db2213
sudo dbctl config commit -m "Promote db2213 to s5 primary T367145"
  • Restart puppet on both hosts:
sudo cumin 'db2123* or db2213*' 'run-puppet-agent -e "primary switchover T367145"'

Clean up tasks:

  • Clean up heartbeat table(s).
sudo db-mysql db2213 heartbeat -e "delete from heartbeat where file like 'db2123%';"
  • change events for query killer:
events_coredb_master.sql on the new primary db2213
events_coredb_slave.sql on the new slave db2123
  • Update candidate primary dbctl and orchestrator notes
sudo dbctl instance db2123 set-candidate-master --section s5 true
sudo dbctl instance db2213 set-candidate-master --section s5 false
(dborch1001): sudo orchestrator-client -c untag -i db2213 --tag name=candidate
(dborch1001): sudo orchestrator-client -c tag -i db2123 --tag name=candidate
sudo db-mysql db1215 zarcillo -e "select * from masters where section = 's5';"
  • (If needed): Depool db2123 for maintenance.
sudo dbctl instance db2123 depool
sudo dbctl config commit -m "Depool db2123 T367145"
  • Change db2123 weight to mimic the previous weight db2213:
sudo dbctl instance db2123 edit
  • Update/resolve this ticket.

Event Timeline

Change #1041534 had a related patch set uploaded (by Gerrit maintenance bot; author: Gerrit maintenance bot):

[operations/puppet@production] mariadb: Promote db2213 to s5 master

https://gerrit.wikimedia.org/r/1041534

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

Mentioned in SAL (#wikimedia-operations) [2024-06-11T10:28:14Z] <marostegui@cumin1002> START - Cookbook sre.hosts.downtime for 1:00:00 on 24 hosts with reason: Primary switchover s5 T367145

Mentioned in SAL (#wikimedia-operations) [2024-06-11T10:28:21Z] <marostegui@cumin1002> dbctl commit (dc=all): 'Set db2213 with weight 0 T367145', diff saved to https://phabricator.wikimedia.org/P64609 and previous config saved to /var/cache/conftool/dbconfig/20240611-102820-root.json

Mentioned in SAL (#wikimedia-operations) [2024-06-11T10:28:33Z] <marostegui@cumin1002> END (PASS) - Cookbook sre.hosts.downtime (exit_code=0) for 1:00:00 on 24 hosts with reason: Primary switchover s5 T367145

Old 300
vslow, dump 100

sections:
  s5:
    candidate_master: true
    groups:
      dump:
        pooled: true
        weight: 100
      vslow:
        pooled: true
        weight: 100
    percentage: 100
    pooled: true
    weight: 0

Mentioned in SAL (#wikimedia-operations) [2024-06-11T10:29:00Z] <marostegui@cumin1002> dbctl commit (dc=all): 'Remove db2213 from API/vslow/dump T367145', diff saved to https://phabricator.wikimedia.org/P64610 and previous config saved to /var/cache/conftool/dbconfig/20240611-102900-root.json

Change #1041534 merged by Marostegui:

[operations/puppet@production] mariadb: Promote db2213 to s5 master

https://gerrit.wikimedia.org/r/1041534

Mentioned in SAL (#wikimedia-operations) [2024-06-11T10:42:06Z] <marostegui> Starting s5 codfw failover from db2123 to db2213 - T367145

Mentioned in SAL (#wikimedia-operations) [2024-06-11T10:42:32Z] <marostegui@cumin1002> dbctl commit (dc=all): 'Promote db2213 to s5 primary T367145', diff saved to https://phabricator.wikimedia.org/P64611 and previous config saved to /var/cache/conftool/dbconfig/20240611-104232-root.json

Mentioned in SAL (#wikimedia-operations) [2024-06-11T10:43:37Z] <marostegui@cumin1002> dbctl commit (dc=all): 'Depool db2123 T367145', diff saved to https://phabricator.wikimedia.org/P64612 and previous config saved to /var/cache/conftool/dbconfig/20240611-104336-root.json