Update cri-dockerd switchover docs for v1.24

This commit is contained in:
Tim Bannister
2022-03-30 22:18:36 +01:00
parent 5edbf56ec5
commit b9a7ec6f35
@@ -7,17 +7,20 @@ content_type: task
{{% thirdparty-content %}} {{% thirdparty-content %}}
This page shows you how to migrate your Docker Engine nodes to use `cri-dockerd` This page shows you how to migrate your Docker Engine nodes to use `cri-dockerd`
instead of dockershim. Follow these steps if your clusters run Kubernetes 1.23 instead of dockershim. If you want to switch away from using dockershim and still
or earlier and you want to continue using Docker Engine after use Docker Engine to run containers in Kubernetes, you can follow these steps.
you upgrade to Kubernetes 1.24 and later, or if you just want to move off the If you want to upgrade to Kubernetes v{{< skew currentVersion >}} and your
dockershim component. existing cluster relies on dockershim, you will have to migrate away and
`cri-dockerd` is one of your options.
To learn more about the removal of dockershim, read the [FAQ page](/dockershim).
## What is cri-dockerd? {#what-is-cri-dockerd} ## What is cri-dockerd? {#what-is-cri-dockerd}
In Kubernetes 1.23 and earlier, Docker Engine used a component called the In Kubernetes 1.23 and earlier, you could use Docker Engine with Kubernetes,
dockershim to interact with Kubernetes system components such as the kubelet. relying on a built-in component of Kubernetes named _dockershim_.
The dockershim component is deprecated and will be removed in Kubernetes 1.24. A The dockershim component was removed in the Kubernetes 1.24 release; however,
third-party replacement, `cri-dockerd`, is available. The `cri-dockerd` adapter a third-party replacement, `cri-dockerd`, is available. The `cri-dockerd` adapter
lets you use Docker Engine through the {{<glossary_tooltip term_id="cri" text="Container Runtime Interface">}}. lets you use Docker Engine through the {{<glossary_tooltip term_id="cri" text="Container Runtime Interface">}}.
{{<note>}} {{<note>}}
@@ -93,7 +96,7 @@ in the control plane. To modify this socket for each affected node:
## Restart the kubelet ## Restart the kubelet
```shell ```shell
systemctl start kubelet systemctl restart kubelet
``` ```
## Verify that the node is healthy ## Verify that the node is healthy