Dear Roland,
sorry for not having replied earlier: due to the COVID emergency, I was able to put my hands on the machine only today.
Good news: I was able to get the nfs filesystems on the nodes! Thank you so much for your help.
I think it can be useful to tell you about what I have done and what did not work. You were right about the entries in /etc/exports: no path was exported on the IB network even if this network was listed in the menu of qluman-qt Filesystem exports under the submenu Network priorities. Then I tried to copy and paste all the ethernet (Boot) exports and change the IP entries to reflect the IB network. After a restart of the full cluster (I was not sure about which services needed to be restarted) I got a different behavior trying to login on the nodes: the login was unsuccessful because of a timeout on the nfs mount of the home directory. By the way, I was unable to find this modification reflected in the qluman-qt entry "Network NFS mounts". Then I checked the entry NEED_RDMA="yes" in file /etc/default/nfs-kernel-server and it was missed even if the proper check was activated in qluman-qt . I did not find any way to edit this configuration within qluman-qt. At this point, I choose to avoid having nfs on IB, so I edited again the /etc/exports file deleting the entry about IB and in qluman-qt I deleted the reference to IB in the Network priorities sub-menu of the Filesystem exports menu. Et voilà, after a new restart of the cluster, users are now able to login to the compute nodes landing in their home directory.
Your suggestions were essential to solve this issue, thank you again. If I can be of any help for a better analysis of the reported behavior, do not hesitate to contact me.
A side info: I get continuous disconnections from the qluman-qt interface, while all connections to the server console or to the nodes are perfectly stable.
My best regards,
Franco