Join GitHub today
GitHub is home to over 40 million developers working together to host and review code, manage projects, and build software together.Sign up
Enable recovery using existing bucket #28
If the managed-velero-operator namespace or CR gets deleted,
This commit adds the ability for managed-velero-operator to recover
Tested in stage.
It successfully recovers a bucket that has the right tags, and ignores the other buckets.
Tested the latest commit in stage.
If it doesn't find a bucket that matches the infraName, it will choose a bucket that is marked for velero backups. This will allow it to recover a bucket even if it doesn't have the infraName tag yet. (Since, when we roll this out, the tags won't exist on the buckets yet). It just chooses the first bucket it sees.
And this only happens in cases where the CR was deleted and re-created without a bucket name specified. Here is an empty CR being created, and how the reconcile loop handles it. (I'm also showing the list of buckets in this account prior to CR creation.)
cblecker left a comment
This looks really great! Mostly some style nits, but nothing major.
After reviewing, yeah, let's drop the last commit. 1 hour after deploying this, all buckets should have the right tags, and we shouldn't need to recover in that time frame.
If the managed-velero-operator namespace or CR gets deleted, usually we would have no way of knowing which S3 bucket is associated with the operator. We would be left with unmanaged S3 resources. This commit add the ability for managed-velero-operator to recover from namespace deletion by searching for an existing S3 bucket associated with velero backups for the cluster on which it is running. https://issues.redhat.com/browse/OSD-2602
[APPROVALNOTIFIER] This PR is APPROVED
The full list of commands accepted by this bot can be found here.
The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing