Dear starcluster team,<div><br></div><div>I tripped over what might be an error with the NFS setup when adding new nodes to a cluster.</div><div><br></div><div>I set up my cluster with initially one root node only and then first added one node and subsequently 4 more nodes.</div>
<div>I noticed that my ebsvolume wasn't getting mounted correctly on the nodes, calling 'df' reported 'stale filehandle' for</div><div>/home /opt/sge6 and /data</div><div><br></div><div>My impression is that as nodes get added, the /etc/exports file which is responsible for allowing NFS access gets overwritten. Therefore only the last added node can access the shared file systems.</div>
<div><br></div><div>Here is how I resloved the issue. First I unmounted all the volumes:</div><div> </div><div><div>root@node001:~# umount -f /data</div><div><br></div><div>At this point remounting doesn't work:</div>
<div><br></div><div>root@node001:~# mount -t nfs master:/data /data</div></div><div><br></div><div><div>mount.nfs: access denied by server while mounting master:/data</div></div><div><br></div><div><br></div><div>I then edited /etc/exports on the master node. Here only the last node was listed:</div>
<div><br></div><div>/home node005(async,no_root_squash,no_subtree_check,rw) </div><div><div><div><div>/opt/sge6 node005(async,no_root_squash,no_subtree_check,rw) </div><div>/data node005(async,no_root_squash,no_subtree_check,rw)</div>
</div></div></div><div><br></div><div>I changed this to</div><div><div>/home node001(async,no_root_squash,no_subtree_check,rw)</div><div>/opt/sge6 node001(async,no_root_squash,no_subtree_check,rw)</div><div>/data node001(async,no_root_squash,no_subtree_check,rw)</div>
<div>/home node002(async,no_root_squash,no_subtree_check,rw)</div><div>/opt/sge6 node002(async,no_root_squash,no_subtree_check,rw)</div><div>/data node002(async,no_root_squash,no_subtree_check,rw)</div><div>/home node003(async,no_root_squash,no_subtree_check,rw)</div>
<div>/opt/sge6 node003(async,no_root_squash,no_subtree_check,rw)</div><div>/data node003(async,no_root_squash,no_subtree_check,rw)</div><div>/home node004(async,no_root_squash,no_subtree_check,rw)</div><div>/opt/sge6 node004(async,no_root_squash,no_subtree_check,rw)</div>
<div>/data node004(async,no_root_squash,no_subtree_check,rw)</div><div>/home node005(async,no_root_squash,no_subtree_check,rw)</div><div>/opt/sge6 node005(async,no_root_squash,no_subtree_check,rw)</div><div>/data node005(async,no_root_squash,no_subtree_check,rw)</div>
</div><div><br></div><div>then restart the nfs-server:</div><div><br></div><div>$ /etc/init.d/nfs-kernel-server restart</div><div><br></div><div>After that running 'df' on each node showed the nfs now working correctly.</div>
<div><br></div><div>kind regards</div><div>Paul</div>