v0.8.0-beta.3
Pre-releaseDrain the nodes before applying the deploy core script
Changelog
- Updated dependencies
- Multus: 4.0.0-unstable (not yet tracked by version tracker)
- ArgoCD: 5.3.6
- "Deploy core" script deploys CoreDNS
- "Deploy core" script only waits for specific deployments instead of all
Breaking changes
The new major version of Multus CNI introduces heavy changes (k8snetworkplumbingwg/multus-cni#893)
The migration is seamless; but needs attention.
Multus CNI 4.0 thick daemonset introduces a new architecture which is the server/client architecture. As there is a server that handles all the network attachments, the process is quite slow.
As soon as you apply the Multus CNI, each pod will be killed to reattach the networks. Be aware that the process is REALLY slow. You might see some Failed to create pod sandbox: rpc error: code = Unknown desc = failed to setup network for sandbox "<id>": plugin type="multus-shim" name="multus-cni-network" failed (add): CNI request failed with status 400: <...>
. This means that the multus-shim
is working hard.
To accelerate the process and to be safe, drain the nodes.
Rebooting a node won't change anything and may break your setup.
Full Changelog: v0.8.0-beta.2...v0.8.0-beta.3