[INVALID] Garbage Collection fails for non-existent VM

[Haabda]

EDIT June 15, 2022 - I was looking at old logs. This is NOT an issue. Board admins, please delete this entire thread as it is completely invalid.

I manually deleted chunks. Because I did that, I was unable to perform syncs. To fix that, I deleted my Datastore and recreated it. (See https://forum.proxmox.com/threads/sync-fails-with-os-error-2.110703/)

()
2022-06-07T14:17:07-05:00: starting garbage collection on store Backups
2022-06-07T14:17:07-05:00: Start GC phase1 (mark used chunks)
2022-06-07T14:30:08-05:00: marked 1% (2 of 132 index files)
2022-06-07T14:30:51-05:00: marked 2% (3 of 132 index files)
2022-06-07T14:31:33-05:00: marked 3% (4 of 132 index files)
2022-06-07T14:32:25-05:00: marked 4% (6 of 132 index files)
2022-06-07T14:32:48-05:00: marked 5% (7 of 132 index files)
2022-06-07T14:33:13-05:00: marked 6% (8 of 132 index files)
2022-06-07T14:48:47-05:00: WARN: warning: unable to access non-existent chunk f7024338a28d78a22ee1b919d68305d68c5a0c5b4f1962619ee7c4adfb56e62e, required by "/mnt/datastore/Backups/vm/105/2021-10-08T00:08:03Z/drive-sata0.img.fidx"
2022-06-07T14:51:24-05:00: received abort request ...
2022-06-07T14:51:24-05:00: TASK ERROR: abort requested - aborting task

I have not even synced the VM105 backup yet so how could it know a non-existent chunk is required for it?

 

[dcsapak]

I have not even synced the VM105 backup yet so how could it know a non-existent chunk is required for it?

what do you mean exactly?
the log you posted shows that a chunk from an index in the datastore is non-existant...?

 

[Haabda]

There is not a backup for VM105. So to me, the error message doesn't make any sense.

 

[dcsapak]

can you post the output of

cat /etc/proxmox-backup/datastore.cfg
ls -lhR /mnt/datastore/Backups/vm/105

?

 

[Haabda]

root@alcaeus:~# cat /etc/proxmox-backup/datastore.cfg
datastore: Backups
        path /mnt/datastore/Backups
root@alcaeus:~# ls -lhR /mnt/datastore/Backups/vm/105
ls: cannot access '/mnt/datastore/Backups/vm/105': No such file or directory

 

[Haabda]

I just realized I was looking at the Garbage Collection logs from before I rebuilt the datastore. It can't even do a garbage collection yet because it isn't finishing syncing my VMs. If possible, please delete this thread as it is completely invalid.