BryanDavis has uploaded a new change for review.
https://gerrit.wikimedia.org/r/231704
Change subject: Add icinga alert for anomalous
logstash.rate.mediawiki.memcached.ERROR.count
......................................................................
Add icinga alert for anomalous logstash.rate.mediawiki.memcached.ERROR.count
Setup an icinga alert to complain if more than 15% of the last 100
memcached errors per minute counts from MediaWiki are above the
Holt-Winters predictive model rate.
Bug: T100735
Change-Id: I011f583b51862e54e653e9a900635c9ddf1e06e6
---
M manifests/role/graphite.pp
1 file changed, 16 insertions(+), 0 deletions(-)
git pull ssh://gerrit.wikimedia.org:29418/operations/puppet
refs/changes/04/231704/1
diff --git a/manifests/role/graphite.pp b/manifests/role/graphite.pp
index dd9ebd2..a07c063 100644
--- a/manifests/role/graphite.pp
+++ b/manifests/role/graphite.pp
@@ -277,6 +277,22 @@
under => true,
group => 'analytics_eqiad',
}
+
+ # Monitor memcached error rate from MediaWiki. This is commonly a sign of
+ # a failing nutcracker instance that can be tracked down via
+ # https://logstash.wikimedia.org/#/dashboard/elasticsearch/memcached
+ monitoring::graphite_anomaly { 'mediawiki-memcached-anomoly':
+ description => 'MediaWiki memcached error rate',
+ metric => 'logstash.rate.mediawiki.memcached.ERROR.count',
+ # Check over the last 100 samples and:
+ # - alert warn if more than 5% are over the confidence band
+ # - alert critical if more than 15% are over the confidence band
+ check_window => 100,
+ warning => 5,
+ critical => 15,
+ over => true,
+ }
+
}
# == Class: role::graphite::labmon
--
To view, visit https://gerrit.wikimedia.org/r/231704
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: I011f583b51862e54e653e9a900635c9ddf1e06e6
Gerrit-PatchSet: 1
Gerrit-Project: operations/puppet
Gerrit-Branch: production
Gerrit-Owner: BryanDavis <[email protected]>
_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits