You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We've run into problems where the disk id on Openstack is often truncated to 20 characters, but newer versions of Openstack may have the full disk id. This makes it impossible using the current logic for the CPI or the Agent to make the correct decision.
It would be better if the stemcell could provide a list of transformations to try in a preferred order and the agent smartly chooses the best match.
This is a bit tricky because you have to be careful of the partitions that are also listed in /dev/disk/by-id as well as the current retry logic which might make it difficult to be sure you are using the correct matching pattern.
The text was updated successfully, but these errors were encountered:
The current logic is very rigid and inflexible: https://github.com/cloudfoundry/bosh-agent/blob/232da0ab1cbc633351996a52f34424f5989e54c9/infrastructure/devicepathresolver/id_device_path_resolver.go#L71-L97
We've run into problems where the disk id on Openstack is often truncated to 20 characters, but newer versions of Openstack may have the full disk id. This makes it impossible using the current logic for the CPI or the Agent to make the correct decision.
It would be better if the stemcell could provide a list of transformations to try in a preferred order and the agent smartly chooses the best match.
This is a bit tricky because you have to be careful of the partitions that are also listed in
/dev/disk/by-id
as well as the current retry logic which might make it difficult to be sure you are using the correct matching pattern.The text was updated successfully, but these errors were encountered: