Hello, This is Nautilus 14.2.21 with 9 OSD hosts and 3 MGR/MON hosts. I've just added 3 new OSD hosts, deploying the OSDs with Ceph-Ansible, and adjusting the backfill process with pgremapper as recently described. I noticed today that for the 3 new OSD hosts the Dashboard doesn't have any data in the CPU/RAM/Network or other graph panes, although it does have although it does report the number of OSDs and the total raw capacity. In investigating this I also noticed a variation in the keyrings in /etc/ceph across the 9 OSD hosts, although /etc/ceph/ceph.conf is consistent across all. Please advise on how to correct. Should I copy the missing keyrings? And which ones? On the older OSD hosts I see ceph.client.admin, ceph.client.crash, and ceph.mgr.xyz for the 3 MGR nodes, but not all are on every OSD host. It's also not clear if this will correct the graphs missing from the Dashboard, or do I need to do something else. Also, is there a Ceph-Ansible trick that I missed that could have caused this? Thanks. -Dave -- Dave Hall Binghamton University kdhall@xxxxxxxxxxxxxx _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |