Change the active master of es3 shard from es1014 to es1017.
This will allow to:
- Upgrade the es3 master to stretch/10.1
- Finally change the socket location
- Migrate the SPOF away from DC row B
Change the active master of es3 shard from es1014 to es1017.
This will allow to:
Status | Subtype | Assigned | Task | ||
---|---|---|---|---|---|
Invalid | ayounsi | T199142 Increase network capacity (2018-19 Q1 Goal) | |||
Resolved | • Cmjohnson | T183585 Rack/cable/configure asw2-b-eqiad switch stack | |||
Resolved | • jcrespo | T197073 switchover es1014 to es1017 | |||
Resolved | • Cmjohnson | T197072 Physically move es1017 from D to C row | |||
Resolved | • jcrespo | T199224 Test database master switchover script on codfw |
es1017 has been successfully moved to row C.
Once it has been working fine for a few days (not the first time we see old hardware failing after a few days), we should schedule the failover.
Change 446551 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/puppet@production] mariadb: Allow reimage of es1019
Change 446553 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/mediawiki-config@master] mariadb: Depool es1018 for reimage
Change 446551 merged by Jcrespo:
[operations/puppet@production] mariadb: Allow reimage of es1019
Change 446553 merged by Jcrespo:
[operations/mediawiki-config@master] mariadb: Depool es1019 for reimage
Change 446752 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/mediawiki-config@master] mariadb: Repool es1019 with low load
Change 446755 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/mediawiki-config@master] mariadb: Repool es1019 full after maintenance
Change 446752 merged by jenkins-bot:
[operations/mediawiki-config@master] mariadb: Repool es1019 with low load
Change 446827 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/puppet@production] mariadb: Productionize db1095 and db1102 into s1-test
Change 446827 merged by Jcrespo:
[operations/puppet@production] mariadb: Productionize db1095 and db1102 into test-s1
Change 446846 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/mediawiki-config@master] mariadb: Depool db1099 for cloning and upgrade
Change 446846 merged by Jcrespo:
[operations/mediawiki-config@master] mariadb: Depool db1099 for cloning and upgrade
Change 446849 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/mediawiki-config@master] mariadb: Fully repool db1099, including db1099:s8
Change 446849 merged by jenkins-bot:
[operations/mediawiki-config@master] mariadb: Fully depool db1099, including db1099:s8
Change 446875 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/mediawiki-config@master] mariadb: Repool db1099 (both instances) with low load
Change 446875 merged by Jcrespo:
[operations/mediawiki-config@master] mariadb: Repool db1099 (both instances) with low load
Change 446902 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/mediawiki-config@master] mariadb: Repool db1099 fully after warmup
Change 446755 merged by jenkins-bot:
[operations/mediawiki-config@master] mariadb: Repool es1019 fully after maintenance
Change 446902 merged by jenkins-bot:
[operations/mediawiki-config@master] mariadb: Repool db1099 fully after warmup
Change 447584 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/puppet@production] mariadb: Promote es1017 as the master of es3-eqiad (instead of es1014)
Change 447586 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/mediawiki-config@master] mariadb: Promote es1017 as the master of es3-eqiad
Change 447587 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/dns@master] Setup es1017 as the backend for the es3-eqiad master
Change 447584 merged by Marostegui:
[operations/puppet@production] mariadb: Promote es1017 as the master of es3-eqiad (instead of es1014)
Change 447586 merged by jenkins-bot:
[operations/mediawiki-config@master] mariadb: Promote es1017 as the master of es3-eqiad
Congratulations @jcrespo for doing the first automated failover with the new script!
The errors lasted:
First error at 06:01:27 and last error 06:02:05
Very impressive! :)
Change 447587 merged by Jcrespo:
[operations/dns@master] Setup es1017 as the backend for the es3-eqiad master
Mentioned in SAL (#wikimedia-operations) [2018-07-25T06:33:54Z] <jynus> finished es1014 -> es1017 switch T197073
This is now done, and while there are things pending to do related to es1014 maintenance, the main task T183585 is unblocked from all SPOF db hosts.
Change 452637 had a related patch set uploaded (by Jcrespo; owner: Jcrespo):
[operations/puppet@production] mariadb: Failover db1066 (eqiad s2 master) to db1122
Change 452637 abandoned by Jcrespo:
mariadb: Failover db1066 (eqiad s2 master) to db1122
Reason:
Not needed.