Hi, I'm using Ceph Squid installed with cephadm and I'm having an issue with agents. Cluster health constantly goes in warn as agent do not report to the active mgr - on the other side agents report "Failed to send metadata to mgr: non-200 response <-1> from agent endpoint: [Errno 111] Connection refused". Inspecting agent activity with strace, they're all trying to contact monmgr01.example.com/X.Y.Z.1 , while actually the active manager is monmgr02.example.com/X.Y.Z.1. The issue remains if I redeploy the agents with ceph orch. Any suggestion on how to find out more on the issue? Has it happened to someone else? Thanks -- Francesco Di Nucci System Administrator Compute & Networking Service, INFN Naples Email: francesco.dinucci@na.infn.it