Hi, some weeks ago I opened this bug ticket: https://tracker.ceph.com/issues/51463 with ceph octopus I have issues with slow requests while starting/stopping OSDs. Slow requests in this case are blocked for longer than 6 seconds. With nautilus this wasn‘t any issue. Can someone have a look to this issue? As far as I saw, I can mitigate this issue with setting the parameter „osd_pool_default_read_lease_ratio“ from default 0.8 to 0.2. Can at least someone say, if this could have any unwanted side effect? Unfortunately this parameter is not really well documented. Manuel _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |