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
When a replicationsource or replicationdestination has a name that is too long, this can produce a failure as the subsequent created resources can have names that are too long.
First thing I think that is hit is when we create a mover job. Jobs will use labels based on the job name, and labels have a key length limit of 63 characters. If we create a job with volsync-src-<cr name> then we can easily hit this limit
After this, do we hit any other errors? Not sure what other resources can be affected - we could also hit a next level of name length limit when objects we create (volume snapshots, service accounts, pvcs etc) hit the kubernetes object name length limit.
Discussion:
Possibly generate a name for the job using a hash
Save the job name somewhere in status so users can find it
What about migration? Existing jobs may still use the old name and still be running when volsync is updated to a new version with this fix
What about other objects?
The text was updated successfully, but these errors were encountered:
Describe the bug
When a
replicationsource
orreplicationdestination
has a name that is too long, this can produce a failure as the subsequent created resources can have names that are too long.volsync-src-<cr name>
then we can easily hit this limitDiscussion:
The text was updated successfully, but these errors were encountered: