Hi,
we are planning a new cluster as backend for our new bareos backup setup. It currently supports ceph either via S3 or via filesystem (the rados plugin is deprecated in the current release).
Since I would like to avoid the extra S3 hop (storage daemon <-> ceph rgw) our POC will use cephfs. To further isolate the cluster from all other system, I would also like to run the bareos storage daemon colocated to the CEPH OSDs.
This will require mounting the cephfs instance on the hosts. In the past this was considered a problem since deadlocks may occur. Is this still the case? Or is mounting cephfs on the OSD hosts (kernel implementation) considered safe now?
Best regards and happy holidays, Burkhard Linke _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-le...@ceph.io
