According to current sheepdog's rule, deleting a vdi, no matter it is
snapshot or not, its vdi object, in another word, ledger object, will
not be removed from disks.
I think it is necessary for collecting allocated vdi in node recovery.
Otherwise, the deleted vdi id might be reused. However, because data
objects are copy-on-write, the allocated vdi id probably belongs to
other vdi so that the deleted vdi id must not be reused.
Therefore, it is not a bug. But it is still a good question.
On 2014年10月24日 06:01, Valerio Pachera wrote:
Hi, this isn't urgent but I opened a bug as a reminder:
https://bugs.launchpad.net/sheepdog-project/+bug/1384951
Correct me if I'm wrong: as a workaround, when there are no snapshots
I can safely remove all ledger objects but the last.
cd /mnt/sheep/0
ls 8*
807c2b2500000000
807c2b2700000000
807c2b2900000000
807c2b2a00000000
807c2b2b00000000
807c2b2c00000000
807c2b2e00000000
807c2b2f00000000
ls 8* | grep -v 807c2b2f00000000 | xargs rm
Thank you.
--
sheepdog mailing list
[email protected]
http://lists.wpkg.org/mailman/listinfo/sheepdog
