Live Migrating Production Clusters From Calico to Cilium
Talkat Cilium + eBPF Day North America 2024 · on Tuesday, 2024-11-12· at 14:40· Salt Palace | Level 1 | Grand Ballroom B·
summary
Live in-place migration of ~20 clusters (many production, up to 500 nodes) swaps Calico for Cilium using Cilium node config CRD and a per-node labeled takeover: cordon/drain nodes, label, restart Cilium agents, validate CNI, uncordon, repeat. Tried multis/secondary interfaces and abandoned it. Key config: whitelist pod CIDRs, set Calico IP autodetection, separate pod CIDR pool, enable portmap for hostPorts. Outcome: no service interruptions, reduced tenant PRs by ~50%, but expect iptables cleanup, IPv6 caveats, and later kube-proxy replacement work.
AI-written summary · about
chapters (11)
- 0:00Introduction and agenda
- 0:37Migration objectives
- 2:17Reasons for choosing Cilium
- 4:38Evaluating migration methods
- 6:58Lessons from trial efforts
- 10:24The hybrid migration strategy
- 11:26Running the migration steps
- 13:51Live migration demonstration
- 21:04Considerations and learnings
- 22:10Closing thoughts
- 25:03Q&A session
related talks
32mSwitching your CNI — Simone Sciarrati & Federico Hernandez
25mMore Churn No Problem: Lessons Learned Running Cilium in Production
26mThe Road To Cilium: Migrating 150+ Kubernetes Clusters at Airbnb
28mConnecting the World: Your Hands-on Guide To Cilium ClusterMesh and MCS-API
16mCilium Wars: Scaling Up Your Network with Jedi-like Cilium Handling Techniques - Yash Tandon
20mTackling Container Networking at Scale
24mHigh-Scale Networking for ML Workloads With Cilium
10mLightning Talk: Mastering Zero Downtime Migrations Between Cilium Identity Alloca...