Giuseppe Lavagetto has uploaded a new change for review. (
https://gerrit.wikimedia.org/r/350555 )
Change subject: graphite::alerts: add alerting on session loss
......................................................................
graphite::alerts: add alerting on session loss
These two alerts, respectively on session loss on edits and on
CentralAuth error rate, should help us be notified on issues with
sessions in general and redis in particular.
Change-Id: I2859a6beba51af8f2e0a69d6c74be843a2c4e987
---
M modules/role/manifests/graphite/alerts.pp
1 file changed, 20 insertions(+), 1 deletion(-)
git pull ssh://gerrit.wikimedia.org:29418/operations/puppet
refs/changes/55/350555/1
diff --git a/modules/role/manifests/graphite/alerts.pp
b/modules/role/manifests/graphite/alerts.pp
index 2d1bc4e..3296dd0 100644
--- a/modules/role/manifests/graphite/alerts.pp
+++ b/modules/role/manifests/graphite/alerts.pp
@@ -56,6 +56,26 @@
percentage => 70,
}
+ # Monitor MediaWiki session loss
+ monitoring::graphite_threshold { 'mediawiki_session_loss':
+ description => 'MediaWiki edit session loss',
+ metric =>
'scale(consolidateBy(MediaWiki.edit.failures.bad_token.rate, 'max'), 60)',
+ warning => 10,
+ critical => 50,
+ from => '15min',
+ percentage => 30,
+ }
+
+ # Monitor MediaWiki CentralAuth bad tokens
+ monitoring::graphite_threshold { 'mediawiki_centralauth_errors':
+ description => 'MediaWiki centralauth errors',
+ metric => 'sum(MediaWiki.centralauth.centrallogin_errors.*.rate)',
+ warning => 0.5,
+ critical => 1,
+ from => '15min',
+ percentage => 30,
+ }
+
# Monitor EventBus 4xx and 5xx HTTP response rate.
monitoring::graphite_threshold { 'eventbus_http_error_rate':
description => 'EventBus HTTP Error Rate (4xx + 5xx)',
@@ -67,4 +87,3 @@
percentage => 50,
}
}
-
--
To view, visit https://gerrit.wikimedia.org/r/350555
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: I2859a6beba51af8f2e0a69d6c74be843a2c4e987
Gerrit-PatchSet: 1
Gerrit-Project: operations/puppet
Gerrit-Branch: production
Gerrit-Owner: Giuseppe Lavagetto <[email protected]>
_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits