Docker daemon either stops abruptly or is not recognized until a restart

[thirumalaicb]

I am using a Mac M1 chip, upgraded the OS to Monterey and installed Rancher desktop (v1.0.1). I was able to setup the container. I followed the FAQ section and disabled the Kubernetes cluster as answered for the FAQ question "I don't need the Kubernetes cluster deployed by Rancher Desktop, how do I disable it to save resources?" This causes the docker deamon to either stop abruptly or not recognized causing disconnection to the container from the host.

Have any of you faced/experienced this?

Note: I now have the Kubernetes cluster as-is and currently validating to see if this issue happens.

[thirumalaicb]

Update: It did not really matter whether the Kubernetes cluster was enabled/disabled. This issue still exists. While running the kubectl get nodes gave the result as below, the docker daemon is no longer recognized.
NAME STATUS ROLES AGE VERSION
lima-rancher-desktop Ready control-plane,master 24h v1.22.6+k3s1

[jandubois]

As for stopping and starting Kubernetes, the workaround is not really robust. The new 1.1.0 release coming in the next few days will have a checkbox to enable/disable Kubernetes properly.

The docker socket suddenly disappearing is likely covered by #1274 (the title is misleading, it happens even when the machine is not sleeping). Unfortunately there is still no solution to that problem.

[thirumalaicb]

@jandubois, thank you for your response.

Fortunately for me, the docker socket works even after at least 12 hours of sleep and I have tested this multiple times. However, after the "Reset" is done, the issue happens, and is consistent, after ~23-24 hours.

[thirumalaicb]

I see #1274 is tracked for everything that happens for docker socket issue. I am marking this as closed and would follow the respective ticket for a resolution.