Hi, I removed one of the monitor in the cluster (having 4 monitors previously), after 20 mins, the result of "# ceph -s" still contains "...a mons down...". Is it a common behavior? Does Ceph need so long to remove the monitor thoroughly? # ceph -s cluster 564a10da-5a0e-4a5f-ae64-a769995217c9 health HEALTH_WARN crush map has legacy tunables; 1 mons down, quorum 0,2,3,4 1,12,2,0; clock skew detected on mon.12, mon.2, mon.0 monmap e11: 5 mons at {0=192.168.123.194:6789/0,1=192.168.123.66:6789/0,12=192.168.123.137:6789/0,2=192.168.123.180:6789/0,7=192.168.123.115:6789/0}, election epoch 40, quorum 0,2,3,4 1,12,2,0 mdsmap e4: 1/1/1 up {0=0=up:active} osdmap e126: 20 osds: 15 up, 15 in pgmap v323: 2880 pgs, 3 pools, 1884 bytes data, 20 objects 604 MB used, 76055 MB / 76660 MB avail 2880 active+clean 16:11 Wei Cao (Buddy)