Hi, I'm hunting down a CephFS issue where mds caps become stale for no apparent reason. I might write another email about that when I don't find a solution. But while debugging that, I also tried to prevent clients from being blocklisted when evicted, so I set this: ceph04:~ # ceph config set mds mds_session_blocklist_on_evict false According to the config help, it can be changed at runtime, and it is applied: mds01:~ # cephadm shell -- ceph daemon mds.cephfs.mds01 config get mds_session_blocklist_on_evict ... { "mds_session_blocklist_on_evict": "false" } But after triggering the misbehaving client, it is still blocklisted: 2025-12-03T13:34:52.498+0000 7f5f6cd2b640 1 mds.0.45600 Evicting (and blocklisting) client session 1822981126 (v1:IP:0/1026509947) 2025-12-03T13:34:52.498+0000 7f5f6cd2b640 0 log_channel(cluster) log [INF] : Evicting (and blocklisting) client session 1822981126 (v1:IP:0/1026509947) ceph04:~ # ceph osd blocklist ls IP:0/1026509947 2025-12-03T14:34:52.503570+0000 listed 1 entries Is this a know bug or something? This is a cephadm managed cluster version 19.2.3. Thanks for any pointers! Eugen