I am looking for any materials which can help me to track and troubleshoot performance of my cluster, and particulary rados gateway. I am using command "ceph daemon 'daemon-name' perf dump", and in summary getting a ridiculous count of various metrics, but where i can find their description ? I found some poor description on redhat.com but it is not enough, for good comprehension. Could someone help me found any additional materials or share own experience ? For example what does mean op_laggy, osd_laggy, linger_* and etc not obvious metrics? How to troubleshoot when values is bad ?
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
