>We've hit issues (twice now) that seem (have not >figured out exactly how to confirm this yet) to be related to kernel >dentry slab cache exhaustion - symptoms were a major slow down in >performance and slow requests all over the place on writes, watching >OSD iostat would show a single drive hitting 90+% util for ~15s with a >bunch of small reads and no writes.
Not specifically ceph related but I hit the same issue with Swift some time ago. I created a blogpost then about the issue and how to troubleshoot it (there are some useful xfs stats) http://engineering.spilgames.com/openstack-swift-lots-small-files/ tldr: In the end we moved to nodes with smaller disks and more memory Cheers, Robert _______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
