I went ahead and removed the assert and conditionalized the future use of the obc variable on its being non-null. And linked that into a custom ceph-osd binary for use on the most problematic node (8). That got the osd up and running again! I took the opportunity to use the standard "remove an osd from the cluster" procedure, which also worked. So now I have my ceph cluster healthy again. ;)
Thanks,
Blade.
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
