An introduction to nodepool
summary
Nodepool is a VM provisioning system that builds and uploads consistent disk images, boots and retires clouds VMs on demand, and hands them to CI via Gearman and the Jenkins API; it uses Disimage-Builder, Zookeeper, SQLAlchemy/MySQL, ZMQ and reports to statsd/Grafana. Deployment splits into nodepool-daemon (scheduler) and nodepool-builder (image builds), supports per-cloud flavor selection and labels/targets in nodepool.yaml, handles thousands of jobs per hour, and is already usable with plans for pluggable drivers and horizontal scheduler scaling.
AI-written summary · about
chapters (15)
- 0:00Intro
- 0:32Call for papers
- 0:47Agenda
- 1:30Types of testing
- 5:46Limitations
- 6:15What is nodepool
- 6:37Features
- 7:53Continuous Integration
- 11:20Deployment
- 13:34Statistics
- 14:09Configuration
- 19:30Notebook CLI
- 21:35Using nodepool without Jenkins
- 23:26Future nodepool
- 24:22Questions
related talks
30mWorks, so now what?
50mBest Practices for Python in the Cloud
42mBuilding a hosting platform with Python
23mDevOps with Python
34mManaging the cloud with libcloud
34mManaging the cloud with libcloud
40mHomemade scientific clusters in Python
34mAutomating infrastructure at SA Home Loans with Python (and friends)