NFS mount for patch repository very slow or hanging

[atc]

I'm serving NFS from a RHEL4 server and mounted the NFS share on my ESX 3.5 host as /var/updates/patches. Using WinSCP, I was able to copy all my ESX 3.5 Jan. 17 patches from the ESX 3.5 host to the ESX350 depot but when I go to run esxupdate, it would just hang.

Anyone has better luck setting up an NFS mount for patching?

Thanks in advance.

[Schorschi]

If you map the service console to NFS share and just do simple file copying back and forth, does it hang, or is it slow? If so, you may want to consider changing your NFS configuration on RHEL, increase buffer size, etc. This feels like a NFS configuration issue, not specific to esxupdate. I know the NFS on service console is really picky in establishing connections, but once working, works fine.

[atc]

The thing is I was able to use WinSCP and copy the files from ESX to the NFS share with no delay problem. It's only when I go to run esxupdate info or scan, service console return a message saying the NFS server is not responding. The only way to get out is start another session and reboot ESX host.

Any recommendation on buffer size or other non-default config changes? Thank you.