lost connection to NFS datastores

[fizzer]

Hi All, hope you can help.
we have 2 x ts-ec879-rp units providing an NFS shared folder to 2 x Dell R620 esxi4.1 hosts

These have been running OK for a couple of weeks and we have gone live.

Last friday there was a problem (still don't know the cause) where one of the esxi hosts dropped off the network for some time. while we were working on the host we restarted the VM's from the second host. which was still working fine.

Eventually we got the host back up and running and on the network, all seems fine - except...
we can no longer browse the qnap nfs shares from that host - they are greyed out and 'unmounted'

so, we've removed them using esxcfg-nas and are now trying to reconnect them. unfortunately we get the following errors:

/var/log/messages:Jul 23 11:55:47 vmkernel: 2:19:56:15.973 cpu0:519687)NFS: 149: Command: (mount) Server: (snas02) IP: (x.x.x.x) Path: (/nfs-vmstore02) Label: (store02) Options: (None)
/var/log/messages:Jul 23 11:55:47 vmkernel: 2:19:56:15.975 cpu0:519687)WARNING: NFS: 1138: Server (x.x.x.x) does not support NFS Program (100003) Version (3) Protocol (TCP)
/var/log/messages:Jul 23 11:55:47 vmkernel: 2:19:56:15.975 cpu0:519687)NFS: 171: NFS mount snas02:/nfs-vmstore02 status: The NFS server does not support NFS version 3 over TCP

I know that this box should support NFS version 3 over tcpip, can anyone help find a solution please? vmware support are pointing squarely at the QNAPs - however BOTH are down at the moment on this host and working fine on the other host.

At the moment, the site is just about running on one host and I'm really reluctant to reboot either nas or the host at the moment, if we lose any more connections then the whole site is screwed till I fix it.

[fizzer]

here's the error from the gui:
Call "HostDatastoreSystem.CreateNasDatastore" for object "datastoreSystem-63" on vCenter Server "hostname" failed.
Operation failed, diagnostics report: Unable to complete Sysinfo operation. Please see the VMkernel log file for more details.

[fizzer]

investigating this thread. looks promising:
viewtopic.php?p=249871

[fizzer]

looks like the NFS service is broken (on both) - it's serving NFS data to existing connections but not allowing any new connections.
We will be performaing a restart tomorrow when we can get some downtime on the virtual machines.

Does anyone have andy experience of restarting NFS and its effect on datastores? If this is going to happen regularly we might want to schedule a daily NFS restart, but not if it will disrupt the VMs