Page MenuHomePhabricator

Create separate Kibana dashboards for production Cassandra clusters
Closed, ResolvedPublic

Description

It makes sense to separate the dashboards on a per-project/cluster basis.

Event Timeline

Restricted Application added a subscriber: Aklapper. · View Herald Transcript

Change 278330 had a related patch set uploaded (by Eevans):
make logstash messages separable by cluster

https://gerrit.wikimedia.org/r/278330

I've queued up https://gerrit.wikimedia.org/r/278330 for next Tuesdays Puppet SWAT (16:00 UTC).

NOTE: When merged this will go out to all clusters. Cassandra should automagically reload the logback config when the file is changed, but a rolling restart of Cassandra is still recommended.

Change 278330 merged by Muehlenhoff:
make logstash messages separable by cluster

https://gerrit.wikimedia.org/r/278330

Mentioned in SAL [2016-03-22T16:21:05Z] <urandom> Restarting Cassandra on restbase1007.eqiad.wmnet (canary) : T130393, T128787

Mentioned in SAL [2016-03-22T16:54:55Z] <urandom> Performing rolling restart of AQS Cassandra cluster : T130393, T128787

Mentioned in SAL [2016-03-22T17:03:18Z] <urandom> Rolling restart of AQS Cassandra cluster complete : : T130393, T128787

Mentioned in SAL [2016-03-22T18:39:39Z] <urandom> Rolling restart of Cassandra on maps cluster : T130393, T128787

Mentioned in SAL [2016-03-22T18:41:11Z] <urandom> Aborting restart of Cassandra on maps cluster : T130393, T128787

Mentioned in SAL [2016-03-22T19:53:30Z] <urandom> Rolling restart of RESTBase Cassandra cluster : T130393, T128787

Mentioned in SAL [2016-03-22T20:45:38Z] <urandom> Rolling restart of RESTBase Cassandra cluster complete : T130393, T128787

Eevans triaged this task as Medium priority.Mar 23 2016, 3:13 PM