Page MenuHomePhabricator

Reboot all Analytics hosts for kernel + openjdk upgrades
Closed, ResolvedPublic13 Estimated Story Points


Clusters/hosts to reboot:

  • Hadoop worker nodes - Analytics cluster (an-worker* are the only ones remaining to reboot)
  • Hadoop masters - Analytics cluster
  • Hadoop coordinator - Analytics cluster
  • Hadoop worker nodes - testing cluster
  • Hadoop masters - testing cluster
  • Hadoop coordinator - Test cluster
  • Kafka Jumbo
  • Druid public
  • Druid private
  • eventlog1002
  • archiva1001
  • matomo1001
  • analytics-tool100[1,2,4], an-tool1005
  • stat100[4,5,6,7]
  • notebook100[3,4]
  • thorium
  • AQS (needs only a roll restart of cassandra, new kernel already in place)

Event Timeline

elukey triaged this task as High priority.Jun 7 2019, 2:30 PM
elukey created this task.

Note that the Ganeti reboots are a little different here as we need to load the new QEMU along for the new instruction. To reboot a Ganeti instance one needs to log to the respective Ganeti master (ganeti2003.codfw.wmnet for codfw, ganeti1001.eqiad.wmnet for eqiad) and run

sudo gnt-instance reboot FQDN

This applies to matomo1001, archiva1001 and analytics-tool*

elukey changed the point value for this task from 0 to 13.
elukey moved this task from In Progress to Done on the Analytics-Kanban board.