Unable to add a GFS2 Data store to a 3 node VME Cluster (VME 8.0.5 / Ubuntu 24.04.2)

[Balazs_Szucs]

Hi all,

I currently set up a 3 node VME Cluster with Synergy bodes booting from Alletra 5030 via FC. I've created LUNs on the Alletra and gave data access to all 3 hosts.

When I try to add a Data Store to the cluster (Infrastructure/Clusters/<mycluster>) I can see all the LUNs provided via FC but nothing happens when I select one of them and hit save.

In the Administration/Helath/Morpheus logs page I see this:

One of the hosts (which runs the vmem vm) I see in the log (/var/log/morpheus-node/morhpd/current):

2025-05-19_08:24:07.85810 08:24:07.858 [pool-3-thread-2] INFO com.morpheus.agent.stats.KvmVmStats - Setting Max Storage: 107374182400
2025-05-19_08:24:37.86423 08:24:37.864 [pool-3-thread-3] INFO com.morpheus.agent.stats.KvmVmStats - Setting Max Storage: 107374182400
2025-05-19_08:25:07.87656 08:25:07.876 [pool-3-thread-5] INFO com.morpheus.agent.stats.KvmVmStats - Setting Max Storage: 107374182400

However on an other host (part of the cluster) I see this in the log:

2025-05-19_08:28:43.67917 device-mapper: create ioctl on 231420fd9bf02d5ea6c9ce9007514b84e1 part1-mpath-231420fd9bf02d5ea6c9ce9007514b84e failed: Device or resource busy
2025-05-19_08:28:43.67917 device-mapper: create ioctl on 28b6c00d2db52c5786c9ce9007514b84e1 part1-mpath-28b6c00d2db52c5786c9ce9007514b84e failed: Device or resource busy
2025-05-19_08:28:43.67918 device-mapper: create ioctl on 28b6c00d2db52c5786c9ce9007514b84e2 part2-mpath-28b6c00d2db52c5786c9ce9007514b84e failed: Device or resource busy
2025-05-19_08:28:43.67918 device-mapper: create ioctl on 28b6c00d2db52c5786c9ce9007514b84e3 part3-mpath-28b6c00d2db52c5786c9ce9007514b84e failed: Device or resource busy

Interestingly this device mapper links to the Boot Volume of the host provided from the Alletra via FC.

What should I check to let it work as intended? I'm out of what log files should I check.

Any insight is very welcome here, thanks! 

[Arnout Verbeken]

I would do a wipefs before you try to add the datastore.  So in your case /dev/mapper/2314.......

wipefs -f -o 0x10000 /dev/path/to/dm

[support_s]

Hello,

Let us know if you were able to resolve the issue.

If you are satisfied with the answers then kindly click the "Accept As Solution" button for the most helpful response so that it is beneficial to all community members.

Please click on "Thumbs Up/Kudo" icon to give a "Kudo".

[Balazs_Szucs]

Thanks for the replay unfortunately it didn't help, the sympthoms remained the same...

Meanwhile I reinstalled all the 3 nodes and the VME manager on it this this without FC Boot (instead the hosts boot from their internal disks).

I see some progress, but still, it doesn't work.

Now I can see the added Datastore in the datastore list of the cluster but the status is red.

And the Morpheus log entry also progressed as it shows now the same symptom as before I had to dig in the Ubuntu logs (Device or resource busy).

What I see on the hosts CLI running dmsetup, that the some parts of the volume I try to add are open:

But I cannot identify what process reserve it.

Any idea?

Thanks!