Hi - I have a 4 node cluster, and started to have some odd access issues to my file system "Home" When I started investigating, saw the message "1 MDSs behind on trimming", but I also noticed that I seem to have 2 MDSs running on each server - 3 Daemons up, with 5 standby. Is this expected behavior after the upgrade to 18.2? or did something go wrong? [root@cube ~]# ceph status cluster: id: fe3a7cb0-69ca-11eb-8d45-c86000d08867 health: HEALTH_WARN 1 filesystem is degraded 1 MDSs behind on trimming services: mon: 3 daemons, quorum rhel1,cube,hiho (age 23m) mgr: hiho.bphqff(active, since 23m), standbys: rhel1.owrvaz, cube.sdhftu mds: 3/3 daemons up, 5 standby osd: 16 osds: 16 up (since 23m), 16 in (since 26h) rgw: 4 daemons active (4 hosts, 1 zones) data: volumes: 0/1 healthy, 1 recovering pools: 12 pools, 769 pgs objects: 3.64M objects, 3.1 TiB usage: 17 TiB used, 49 TiB / 65 TiB avail pgs: 765 active+clean 4 active+clean+scrubbing+deep io: client: 154 MiB/s rd, 38 op/s rd, 0 op/s wr [root@cube ~]# ceph health detail HEALTH_WARN 1 filesystem is degraded; 1 MDSs behind on trimming [WRN] FS_DEGRADED: 1 filesystem is degraded fs home is degraded [WRN] MDS_TRIM: 1 MDSs behind on trimming mds.home.story.sodtjs(mds.0): Behind on trimming (5546/128) max_segments: 128, num_segments: 5546 [root@cube ~]# ceph fs status home home - 10 clients ==== RANK STATE MDS ACTIVITY DNS INOS DIRS CAPS 0 replay home.story.sodtjs<http://home.story.sodtjs> 802k 766k 36.7k 0 1 resolve home.cube.xljmfz<http://home.cube.xljmfz> 735k 680k 39.0k 0 2 resolve home.rhel1.nwpmbg<http://home.rhel1.nwpmbg> 322k 316k 17.5k 0 POOL TYPE USED AVAIL home.meta metadata 361G 14.9T home.data data 9206G 14.9T STANDBY MDS home.rhel1.ffrufi<http://home.rhel1.ffrufi> home.hiho.mssdyh<http://home.hiho.mssdyh> home.cube.kmpbku<http://home.cube.kmpbku> home.hiho.cfuswn<http://home.hiho.cfuswn> home.story.gmieio<http://home.story.gmieio> MDS version: ceph version 18.2.0 (5dd24139a1eada541a3bc16b6941c5dde975e26d) reef (stable)