Also, I noticed when I initiate the snapshot from the Cloudstack side, an image
does get allocated from Ceph’s perspective but then it ultimately fails but the
snapshot remains on the ceph side.
-jeremy
> On Jun 15, 2021, at 5:52 PM, Jeremy Hansen wrote:
>
> Thank you. Creating the snap
Thank you. Creating the snapshot on the Ceph side seems to be no problem and
pretty much instantaneous. It’s just odd since I’m able to snapshot other VM
root volumes also residing on the ceph rbd pool.
-jeremy
> On Jun 15, 2021, at 3:24 AM, Andrija Panic wrote:
>
> I would focus on the ac
I would focus on the actual qemu-img error (i.e. nothing to do with
CloudStack, but something to do with Ceph+RBD - try to do internal Ceph
snapshot, see if the image is corrupted or not, etc, etc):
QemuImgException: qemu-img: error while writing sector 7921663:
Input/output errorqemu-img: Failed
For some reason a particular snapshot I’m trying to create for a volume fails
and I see this error:
2021-06-15 00:28:03,416 ERROR [o.a.c.s.v.VolumeServiceImpl]
(Work-Job-Executor-11:ctx-a1408657 job-673/job-674 ctx-5db754de)
(logid:da060faf) Take snapshot: 49 failed
com.cloud.utils.exception.C
