[k8s] Large CPU requests cause ray to fail to start

[romilbhardwaj]

From a user:

Seeing this error consistently on one of the sky-launched k8s pods, but on other k8s clusters it works:

2024-05-25 21:38:32,524 INFO worker.py:1715 -- Connected to Ray cluster. View the dashboard at http://127.0.0.1:36821/
[2024-05-25 21:38:39,655 E 1262748 1262748] core_worker.cc:215: Failed to register worker 01000000ffffffffffffffffffffffffffffffffffffffffffffffff to Raylet. IOError: [RayletClient] Unable to register worker with raylet. No such file or directory

...
it has 256 cores

Looks related to ray-project/ray#30012 (comment)

[github-actions]

This issue is stale because it has been open 120 days with no activity. Remove stale label or comment or this will be closed in 10 days.

[github-actions]

This issue is stale because it has been open 120 days with no activity. Remove stale label or comment or this will be closed in 10 days.

[github-actions]

This issue was closed because it has been stalled for 10 days with no activity.