Created attachment 1534423 [details]
Modified version of /etc/ctdb/events.d/60.nfs

Description of problem:
CTDB starts properly, but fails to monitor the nfs-ganesha.service.

Version-Release number of selected component (if applicable):
ctdb-4.8.3-4.el7.x86_64

How reproducible:
Always (happened on 3 gluster nodes)

Steps to Reproduce:
1.Install storhaug , ctdb & nfs-ganesha
2.Deploy via storhaug (https://github.com/gluster/storhaug/wiki):
- yum -y install storhaug-nfs
- ssh-keygen -f /etc/sysconfig/storhaug.d/secret.pem
- ssh-copy-id -i /etc/sysconfig/storhaug.d/secret.pem.pub root@nodeX
Repeat for all nodes
- populate /etc/ctdb/nodes and /etc/ctdb/public_addresses
- Configure /etc/ctdb/ctdbd.conf:
 grep -Ev "^#|^$" /etc/ctdb/ctdbd.conf
CTDB_RECOVERY_LOCK="/run/gluster/shared_storage/lock"
CTDB_NODES=/etc/ctdb/nodes
CTDB_PUBLIC_ADDRESSES=/etc/ctdb/public_addresses
CTDB_MANAGES_NFS=yes
CTDB_NFS_CALLOUT=/etc/ctdb/nfs-ganesha-callout
CTDB_NFS_STATE_FS_TYPE=glusterfs
CTDB_NFS_STATE_MNT=/run/gluster/shared_storage
CTDB_NFS_SKIP_SHARE_CHECK=yes
NFS_HOSTNAME=localhost
CTDB_DEBUG_LEVEL=DEBUG

-touch /etc/ganesha/ganesha.conf
-gluster volume set all cluster.enable-shared-storage enable
-storhaug setup

3.Start the ctdb.service (if not running)


Actual results:
The CTDB starts , but after some time (checked after a minute) goes to UNHEALTHY.

Expected results:
Either the /etc/ctdb/ctdbd.conf should have a way to define the start/stop/monitor commands for /etc/ctdb/events.d/60.nfs , as directly editing is not good:
service_stop_cmd="systemctl stop  nfs-ganesha.service"
service_start_cmd="systemctl start nfs-ganesha.service"
service_check_cmd="systemctl status nfs-ganesha.service"

Additional info:
I'm attaching my /etc/ctdb/events.d/60.nfs - which works for me (for now)