Hi Marius, Dan already hit the nail on the head with the pglog usage. One other thing I wanted to poke at wrt your configuration: > There are 3 pools of 256, 128 and 32 PGs (total = 416 PGs across 8 OSDs). > The pool size is 2 Does this mean that all 3 pools are replicated with size=2? If so, that's 416 * 2 / 8 = 104PGs/OSD, which is totally reasonable, though I wonder how much imbalance you have in the system and whether some OSDs have a lot of PGs on them since the configuration is so small. (You can check via "ceph osd df tree".) Josh _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx