a) Description of problem:

ceph-deploy should add the OSD filesystem details in /etc/fstab.

This helps to ensure that the OSDs get mounted automatically in case the node reboots. If not, the cluster may not be in a healthy state, and ceph will try to bring in other OSDs to maintain the replica set (if other OSDs are available). This would cause lots of data movement.

b) Version-Release number of selected component (if applicable):

RHCS1.2.3

c) How reproducible:

Always

d) Steps to Reproduce:

1. Install an RHCS1.2.3 cluster.
2. Add OSDs and bring the cluster to a healthy state.
3. Check /etc/fstab.

e) Actual results:

The OSD disks and mount points are not added in /etc/fstab, which will prevent the OSDs to be mounted on a reboot.

f) Expected results:

ceph-deploy should add in OSD filesystem information in /etc/fstab.