Kubernetes/Container Runtimes OCI
Jump to navigation
Jump to search
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.
Overview
CRI Container Runtime Interface
Introduced in K8s in 2016, uses gRPC to expose the interface, can talk to different runtimes through the same API
containerd
Running in Kubernetes, execute on every node. You point kubelet to unix socket, that listens for gRPC/CRI calls.
systemctl start containerd
kubelet --containre-runtime=remote --runtime-request-timeout=15m \
--containre-runtime-endpoint=unix:///run/containerd/containerd.sock
# Service unit
[Service]
Environment="KUBELET_EXTRA_ARGS=--containre-runtime=remote --runtime-request-timeout=15m
--containre-runtime-endpoint=unix:///run/containerd/containerd.sock"
Worth to note:
- when you restart
containerddaemon all your containers also get restarted
Resources
- Container runtimes part1 Read it first!, published in 2017
- Runtimes
CRI-O
Start it up, by pointing to a socket that listens for gRPC/CRI calls.
systemctl start crio
kubelet --container-runtime=remote --runtime-request-timeout=15m \
--container-runtime-endpoint /var/run/crio/crio.sock ...
Worth to note:
- restarting the cri-o daemon won't cause container restart