On Mon, 25 Feb 2008, Jos Vos wrote:
On my system (with 5 Ethernet interfaces), the interface that causes the problem is the most heavily used interface
In my case, the NFS traffic is the most heavy type of traffic and it goes over this "problem" interface. There are 3 NICs present in the system; one other interface is also driven by e1000, but it's a different model (82541GI). I have planned to switch the IPs and functions of some of the NICs to check if the messages would go away, but didn't find the proper moment yet.
I don't see how Ethernet driver kernel messages could relate to NFS issues.
If packets are truncated or lost, any kind of connection would be disturbed. The NFS issues on the clients are only temporary, the same operations that spit the 'stale fh' messages work again after some short time.
-- Bogdan Costescu IWR, University of Heidelberg, INF 368, D-69120 Heidelberg, Germany Phone: +49 6221 54 8869/8240, Fax: +49 6221 54 8868/8850 E-mail: [EMAIL PROTECTED] _______________________________________________ rhelv5-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/rhelv5-list
