Operating Jenkins at Scale
summary
Describes a working Jenkins Operations Center that centralizes metrics and build archives from many masters, shipping JVM and build metrics via the Metrics plugin into Elasticsearch + Kibana for dashboards and historical queries. Shows DIY alternatives (collectd -> Graphite, Nagios) and claims automated bulk operations (Groovy steps, safe restart, prepare-for-shutdown) across masters, including running backups and plugin rollouts. Demonstrates actionable numbers — queue spikes up to ~165 minutes and a 75% success rate — and says the system is functional but early, with more controls and overlays coming.
AI-written summary · about
chapters (8)
- 0:00Start
- 1:19CloudBees in a Minute
- 3:39Monitoring
- 5:51State of the Art - DIY
- 6:39Collectd - Graphite - Nagios
- 8:04Jenkins Metrics Plugin
- 24:20As your Jenkins grows larger and critical...
- 28:05Adding more masters
related talks
38mOperating Jenkins at Scale
55mContinuous Integration on Steroids
47mJenkins in the Enterprise: Building Resilient CI Infrastructure
50mJenkins in the Enterprise: Building Resilient CI Infrastructure
33mJenkins 101
46mMonitoring a Cloud CI System: The Jenkins Example
42mSeven Habits of Highly Effective Jenkins Users
41mScaling Jenkins