Why are Brick Processes Not Starting After an OCS 3.x Node Reboot?

Issue

• After a restart of an Openshift Container Storage node, the bricks in the Gluster pod running in this node, are not starting. The output of the gluster v statuscommand shows the bricks hosted in the rebooted node as N/A. Taking the volume heketidbstorageas an example, this is the status observed after rebooting OCS node 10.0.0.1:

  sh-4.2# gluster volume status heketidbstorage
  Status of volume: heketidbstorage
  Gluster process                             TCP Port  RDMA Port  Online PID
  ------------------------------------------------------------------------------
  Brick 10.0.0.1:/var/lib/heketi/mounts/vg_XXXXXXXX/brick_XXXXXXXXX/brick       N/A       N/A        N       N/A
  Brick 10.0.0.2:/var/lib/heketi/mounts/vg_XXXXXXXX/brick_XXXXXXXXX/brick       49152     0          Y       251
  Brick 10.0.0.3:/var/lib/heketi/mounts/vg_XXXXXXXX/brick_XXXXXXXXX/brick       49152     0          Y       212
  

  The same thing in happening for the rest of the volumes.

• Reviewing the output of a pscommand in the pod, the matching glusterfsdprocesses are not running.

• As a workaround, restarting glusterd brings the bricks back online: systemctl restart glusterd.
• How to reboot OCS nodes without any manual intervention afterwards, to get the bricks automatically online?