Open smlx opened 3 years ago
Just noting that adding a random suffix to the bucket name in the schedule object does indeed fix this issue.
(as long as we don't add the same random suffix to the same project name in two clusters :scream:)
Once we get better project naming (to include full cluster details) all projects will be truly unique.
Describe the bug Schedules with the same project name on different clusters will create buckets with the same name. This is a problem, one of the baas operators will succeed in claiming the bucket name and the other will return errors like:
To Reproduce Steps to reproduce the behavior:
schedule
object with the same bucket name on clusters in different regions.Expected behavior Bucket names should be globally unique. Maybe a random suffix?
Screenshots n/a
Additional context n/a