-
Notifications
You must be signed in to change notification settings - Fork 1.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
FSB backup on EFS Volume I asking for a Volume snapshot Class for provider #8278
Comments
Backup desctbie attached here: |
I have this wrong. For FSB backups the storage class does not need a label and a volumesnapshotclass is not meeded either. I made the change to my example andit backupd and restores efs. I reverted the manual changes to the clioudbees storageclass (no label and there is no volume snapshot class. But the cloudbees backup is still complaining about a missing VolumeSnapshotClass for provider efs. I don't know why this happening. |
I see this in the log but these pvc don't exist.
|
On the phone with Amazon and ran the cycle again:
IS THIS A BUG??? |
Are there some orphan PVCs (PVCs that are not bind to any pods) in your case? |
Wenkai, thank you for your reply. I have no orphaned pvcs anywhere in that cluster that I can see:
Although, the first in the list shown (the 5 GB) isn't really attached to any pod directly and seems to be the offending pod |
Orphaned PVCs are PVCs that not attach to any pod, are these PVCs in the list attached to any pod? Could you show me the yaml of pods that these PVCs are attached? |
All of the cloudbees pvcs are attached to cloudbees contoller pods except fro the first one although it shows as bound. The pods are create from a statefulset. Attached are the pods for these controllers NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE |
Let's focus on the PVC The PVC
So snapshot is used for the PVCs. Could you check whether the pods are created when you taking the backup? It's possible that the pods aren't running yet when you taking the backup. |
Wenkai, then the backup: Describe is: One thing in he describe that is new are these warnings:
I am not sure what these jobs are or how to prevent them from appearing In the backup. Should I not be backing up the velero namespace? Are they from the schedule? If I exclude the velero namepace they go away. |
In the describe I see the statefulset, pod, PVC , and PV but still have the error
The rare a whole lot of PVCs mention in th e backup. I don't where they are coming from:
|
According to the latest output of backup description you provided, seems everything works as expected now.
There is only one error this time, the PVC
The
BTW, if you only want to back up a specific namespace, use |
Wenkai, Thank you very much for your reply and follow-ups. I really appreciate it. cloudbees-efs-cloudbees-ci-persistence is an EFS instance which is shared across all of the CB controllers. It is unclear why one out there is bound but not attached to no pod. I have a question in with Cloudbees support and let's see what they come back with I did drop Velero from the backups and the repo messages went away. I am aware the --include-namespace option. |
But just out of curiosity, what are the repo jobs from. Are they from the schudule? And is it still true you cannot use FSB and VolumeSnaps together? It must be one or the other? |
What steps did you take and what happened:
Hello, me again. and tI thank you for your reply in advance. I am now working on FSB. Backups are failing with a message:
When I was debugging this under tickets 8269/8262, I received a similar message and it turned out it could not find a volumesnapshotclass due to a typo. But my understanding is FSB isn't supported by the EFS CSI so why is it asking for a VolumeSnapshotClass? I have a small sample project that works.
I found the storage class for cloudbees was missing the annotation so I added it manually using
kubectl edit
. Will that work or do I need to bring the whole application down and reprovision it?What did you expect to happen:
The backup describe looks like this worked for the k8s objects and there is data under /kopia. I noticed that not all namespaces are backed up, there are a couple of cloudbees namespaces that are not included in the list.
If you are using velero v1.7.0+:
Please use
velero debug --backup <backupname> --restore <restorename>
to generate the support bundle, and attach to this issue, more options please refer tovelero debug --help
bundle-2024-10-08-11-35-57.tar.gz
Anything else you would like to add:
My configuration is as follows. Should I remove the EnabeCSI and volume snapshot location. But again my example works
Environment:
velero version
): 1.14.velero client config get features
): EnableCSIkubectl version
): 1.28.1/etc/os-release
): Amazon Linux 2Vote on this issue!
This is an invitation to the Velero community to vote on issues, you can see the project's top voted issues listed here.
Use the "reaction smiley face" up to the right of this comment to vote.
The text was updated successfully, but these errors were encountered: