Hi, Me and a friend of mine were doing some recovery tests (among others) using radosgw. We did exactly the same tests described in the link below, and we got the same results. http://comments.gmane.org/gmane.comp.file-systems.ceph.devel/9412 After finding this thread I tried what Gregory said, and intentionally changed the file in a non primary copy. Trying to retrieve the file from radosgw triggered an automatically recover of the file and radosgw did not crash. Next test I did was a upgrade to 0.55 and increase the size of the copies to 3, I was wondering if with 2 good copies it would do some kind of election to recover the file. When you know which file you crashed, it is easy to fix, otherwise it is not simple to do. My question is: what is the best way to avoid a problem like this to happen ? bests, -lorieri -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
