Dear Cephers, I have a Ceph cluster with 5 nodes, 5*24=120 OSDs all running Octopus 15.2.8 nowĄŁI encountered the same problem freqeuntly,but i can not find the reason currently. https://tracker.ceph.com/issues/20742 I find that the state is 'can not reproduce'. Maybe,my business can help to reproduce. 3 ceph-fuse clients sequence write 3 dirs separately, when pg remaps. if there is no pg remaps, no problem. https://tracker.ceph.com/issues/20742 _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |