[moving to ceph-users ...] You should be able to use the rados CLI to list all the objects in your pool, excluding all objects associated with known, valid image ids: rados ls -p rbd | grep -vE "($(rados -p rbd ls | grep rbd_header | grep -o "\.[0-9a-f]*" | sed -e :a -e '$!N; s/\n/|/; ta' -e 's/\./\\./g'))" | grep -E '(rbd_data|journal|rbd_object_map)' Once you tweak / verify the list, you can pipe it to the rados rm command. On Wed, Dec 21, 2016 at 11:17 AM, Ruben Kerkhof <ruben@xxxxxxxxxxxxxxxx> wrote: > Hi Jason, > > On Wed, Dec 21, 2016 at 4:53 PM, Jason Dillaman <jdillama@xxxxxxxxxx> wrote: >> On Wed, Dec 21, 2016 at 10:06 AM, Ruben Kerkhof <ruben@xxxxxxxxxxxxxxxx> wrote: >>> $ sudo rbd rm test-root >>> 2016-12-21 14:46:38.479281 7fa1e17d8760 -1 librbd: Image uses >>> unsupported features: 48 >>> Removing image: 100% complete...done. >>> $ echo $? >>> $ 0 >> >> This isn't an known issue and is an odd corner case of the image >> removal logic. The older client isn't able to open the image created >> with new features, so it just assumes that the image was already >> (partially) removed and that it just needs to clean up the rbd image >> directory. > > Thanks for the confirmation. >> >> I'll open a tracker ticket against the hammer branch, but in the >> meantime, either don't use a hammer rbd CLI against images using jewel >> features or disable all features that are incompatible w/ the older >> releases. > > Will do. But first I'll have to get rid of all these orphaned objects. > I was thinking of creating a new pool, rbd cp all the images and then > rbd rm them from the old pool. > After that's done I could drop the old pool. Or do you perhaps know of > a better way? > >> >> >> -- >> Jason > > Kind regards, > > Ruben -- Jason _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |