We operate a Ceph Nautilus (15.2.13) cluster with ~400 OSDs and 4 dedicated MDS servers. Currently we are running 2 active, 2 standby MDS, but would like to scale the MDS containers horizontally on each of the physical MDS servers.
The MDS servers in question have dual 16-core Xeon processors with 128GB of physical RAM. Active resource utilization is around the expected 1-core and 8GB of RAM based on MDS cache settings. Ceph documentation from prior to cephadm suggest to scale the number of MDS daemons per node, but attempts to work out how to do this on a containerized cluster have not been productive. Has anyone worked out how to run multiple MDS containers on the same physical server? Thanks! _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-le...@ceph.io
