[Linux-cluster] Mount error: gfs2: rhel 7.2: Transport endpoint is not connected

Bob Peterson rpeterso at redhat.com
Tue Apr 4 14:58:52 UTC 2017

----- Original Message -----
| Hello Team,
| I am working on creating cluster configurations using pacemaker on Rhel 7.2
| server connected to a SAN storage controller and currently I am struck with
| the below issue.
| It would be great if you can help me in resolving this issue.
| *Node1: hostname: node1-atto*
| *Node2: hostname: node2-emulex*
| [root at node1-atto ~]# mount -t gfs2 /dev/vol_group/lv /mnt/
| mount: mount /dev/mapper/vol_group-lv on /mnt failed: Transport endpoint is
| not connected
| *Steps completed:*
| 1. Completed the cluster configurations and fencing (scsi_fence) with 2
| nodes.
| 2. Created LVM using pvcreate, vgcreate, lvcreate on one node and other
| node could see the created LVM successfully.
| 3. Created a FS with gfs2 using the below command on both the nodes -
| successful,
|  mkfs.gfs2 -p lock_dlm -t mycluster:gfs -j 2 /dev/vol_group/lv
| *Please find the attachment - File1.xlsx*
| 4. When I try to mount, I see the below message,
| mount: mount /dev/mapper/vol_group-lv on /mnt failed: Transport endpoint is
| not connected
| *Troubleshooting steps:*
| 1. Verified by creating with ext4 FS, mounting is successful on both the
| nodes.
| 2. Verfied by creating with below command gfs2 FS with lock_nolock options,
| mounting is successful on both the nodes
| mkfs.gfs2 -p lock_nolock -t mycluster:gfs -j 2 /dev/vol_group/lv
| * Issue consisterntly occurs only on lock_dlm options *
| *Other steps performed by following Redhat forums:*  Still I am facing the
| same mount issue. Please find the attached document File2.docx
| *Query:*
| 1. Currently I haven't installed lvm2-cluster package on both the nodes due
| to dependancy issues. Does lvm2-cluster package is mandatory on pacemaker
| for mounting gfs2 on rhel 7.2
| 2. We have bought only one subscription for ha packages and we have
| attached on Node1 and installed the related packages and reinstalled from
| Node 1 and installed on Node2.
| Will this have an impact to our above mount issue?
| Please find the attached document File3.docx
| Thanks,
| Munavar.

Hi Munavar,

If this is truly RHEL, please contact Red Hat support and file a ticket.
That message means GFS2 is not communicating through DLM. That could be
caused by a number of things, such as:
1. Lock table not matching your cluster name before ":"
2. Same lock table after ":" as another mounted gfs2 file system.
3. Not enough journals to use


Bob Peterson
Red Hat File Systems

More information about the Linux-cluster mailing list