Hi, I performed an upgrade of our ceph cluster from Giant to the latest Hammer (hammer branch in git). And although it seemed to work fine at first, looking at the graphs this morning, I've noticed a much increased write activity on the drives, mostly the ones storing RGW buckets (although that bias could be due just because we have much more objects in RGW than in RBD). This graphs shows this pretty clearly : http://i.imgur.com/0ENApdX.png I assume that the very high spike at the beginning was the on disk format upgrade. But then the activity stayed much higher than it was before. We don't have that much write to RGW, and certainly not more than before. This even yielded a couple of 'slow operations' warning with few RGW ops that have blocked for more than 32 seconds. Cheers, Sylvain _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |