Hi, Is it normal that I deleted files from the cephfs and ceph didn't delete the back objects a day later? Until I restart the mds deamon then it started to release the storage space. I noticed the doc(http://docs.ceph.com/docs/mimic/dev/delayed-delete/) says the file is marked as deleted on the MDS, and deleted lazily. What is the condition to trigger the back object deletion? If it's normal the deletion delayed that much, is there any way to make it faster? Since the cluster is near full. I'm using jewel 10.2.3 both for ceph-fuse and mds. Thanks. _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |