[Solved]TS-809U and TS-410 issue with NFS on ESX 4

[mmorel]

Hi,

I am experiencing intermittent lost of contact with new TS-809U or its little brother TS-410. They are connected NFS on a brand new IBM server. I can't seem to find the source of this issue.

This setup is intended for production. It needs to be rock solid reliable.

[setup]
ESX 4.0.0.208167 with NIC teaming enabled for VMkernel and vConsole. ESX name is in DNS.
QNAP TS-809U connected thru NFS with port trunking enabled.
f/w 3.2.5 Build 0410T. Has fixed IP and uses DNS server.
QNAP TS-410 connected thru NFS with port trunking enabled.
f/w 3.2.5 Build 0409T. Same IP setup.
HP Procurve switch trunked on appropriate ports.

[problem]
NFS storage is unavailable intermittent. ESX then tags the storage unavailable and access to the QNAP contents and its VMs is lost. It comes back available after a while.

[troubleshooting steps]
ping -t <IP of TS-809U or TS-410> times out intermittent every 10 seconds-1 minute.
Ping <ESX name.domain> from SSH session in either QNAP does the same.
Added ESX name in etc/hosts file on either QNAP as seen on other posts.
Had f/w 3.2.4 until this morning on both. Upgrade didn't solve.
Scan with QNAP Finder seems to wake the QNAPs and revives the ping for a few positive replys.

Any ideas?


Thanks.

Michel

[mmorel]

>>> UPDATE <<<

I have decided to reformat using EXT3 instead of EXT4 as suggested in other posts. Will see when drive sync is finished if this issue is resolved.

Michel

[mmorel]

>>> UPDATE <<<

Problem resolved.

The HP Procurve switch had lost it's config. Ports weren't trucked any more. That's why I had unexpected results with ping. Redoing the trunks on the switch solved the problem.

Michel