[Gluster-devel] Gluster service check reports fail after before and after volume creation

[Jonathan Cope]

Bug: service gluster status returns fail message but gluster seems to function normally despite this.  Can peer probe, create bricks, create volume, and confirm file distribution over cluster.  Following example uses single brick volume as demonstration.

Situation:

-fresh Fedora 19 VM

-Gluster Version: glusterfs-fuse-3.4.1-1.fc19.x86_64

Steps to Reproduce:

1. Create new Fedora 19 VM

2. yum install:

    glusterfs

    glusterfs-server

    glusterfs-fuse

3. start glusterd service

4. check glusterd service status  --Reports failed

5. create and start a gluster volume

6. check gluster service status   --Reports failed

# yum install glusterfs glusterfs-server glusterfs-fuse
# service glusterd start

Redirecting to /bin/systemctl status  glusterd.service
glusterd.service - GlusterFS an clustered file-system server
   Loaded: loaded (/usr/lib/systemd/system/glusterd.service; enabled)
   Active: failed (Result: exit-code) since Tue 2013-10-22 20:38:11 CDT; 10s ago
  Process: 2199 ExecStart=/usr/sbin/glusterd -p /run/glusterd.pid (code=exited, status=1/FAILURE)

Oct 22 20:38:11 master systemd[1]: glusterd.service: control process exited...=1
Oct 22 20:38:11 master systemd[1]: Failed to start GlusterFS an clustered f...r.
Oct 22 20:38:11 master systemd[1]: Unit glusterd.service entered failed state.

# mkdir /mnt/brick1

# gluster volume create HadoopVol server-1:/mnt/brick1/hadoop
volume create: HadoopVol: success: please start the volume to access data


# gluster volume start HadoopVol
volume start: HadoopVol: success

# gluster volume status
Status of volume: HadoopVol
Gluster process                        Port    Online    Pid
------------------------------------------------------------------------------
Brick master:/mnt/brick1/hadoop                49152    Y    2809
NFS Server on localhost                    2049    Y    2822
 
There are no active volume tasks



# service gluster status
Redirecting to /bin/systemctl status  gluster.service
gluster.service
   Loaded: error (Reason: No such file or directory)
   Active: inactive (dead)