CirrusSearch: Our Elasticsearch monitoring complains while we're doing normal maintenance
Closed, ResolvedPublic
Actions

Assigned To

None

Authored By

	• Manybubbles
	Mar 26 2014, 7:59 PM

Description

When we do a reindex we end up with unassigned shards for some time. That puts the cluster yellow. That is expected and OK but the monitoring complains about it.

Version: unspecified
Severity: normal

Details

Reference: bz63137

Event Timeline

• bzimport raised the priority of this task from to High.Nov 22 2014, 3:05 AM

• bzimport added a project: CirrusSearch.

• bzimport set Reference to bz63137.

• bzimport added a subscriber: Unknown Object (MLST).

• Manybubbles created this task.Mar 26 2014, 7:59 PM

https://gerrit.wikimedia.org/r/#/c/121399/

Nothing to review at the moment, resetting status.

I think we can still do better.

https://gerrit.wikimedia.org/r/#/c/121399/ got merged.
I assume this is not high priority anymore.

Restricted Application added a project: Discovery-ARCHIVED. · View Herald TranscriptDec 11 2015, 7:50 AM

Tested yesterday while creating some indices, the cluster went yellow or even red and no alert were sent.

dcausse closed this task as Resolved.Dec 11 2015, 8:23 AM

dcausse set Security to None.

• Deskana moved this task from Inbox to Resolved/Invalid/Declined/Legacy on the CirrusSearch board.Dec 31 2015, 5:10 AM

CirrusSearch: Our Elasticsearch monitoring complains while we're doing normal maintenanceClosed, ResolvedPublicActions

Description

Details

Event Timeline

CirrusSearch: Our Elasticsearch monitoring complains while we're doing normal maintenance
Closed, ResolvedPublic
Actions