Date: Sat, 08 Aug 1998 22:50:53 +0200 From: Mattias Pantzare <pantzer@ludd.luth.se> To: Morgan Davis <mdavis@cts.com> Cc: stable@FreeBSD.ORG Subject: Re: tuning nfs Message-ID: <199808082050.WAA08160@zed.ludd.luth.se> In-Reply-To: Your message of "Sat, 08 Aug 1998 11:54:31 PDT." <199808081854.LAA03637@io.cts.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> We have a server that is causing these messages to appear on clients:
>
> nfs server nsfserv-i2:/data: not responding
> nfs server nsfserv-i2:/data: is alive again
> nfs server nsfserv-i2:/data: not responding
> nfs server nsfserv-i2:/data: is alive again
> nfs server nsfserv-i2:/data: not responding
> nfs server nsfserv-i2:/data: is alive again
> nfs server nsfserv-i2:/data: not responding
> nfs server nsfserv-i2:/data: is alive again
> nfs server nsfserv-i2:/data: not responding
> nfs server nsfserv-i2:/data: is alive again
>
> The pairs ("not responding" and "is alive again") seem to occur in the
> same second, according to timestamps in /var/log/messages. Is this
> from having some timeout threshhold set too low, not enough nfsds on
> the server, not enough nfsiods on the client, a network problem, or
> what? Is it a real problem or just an annoying diagnostic?
A network problem or to slow server (or to low timeout).
Check the network for lost packets, ping with a big packet size for example.
If it is on a local ethernet something is wrong if you lose packets. If it is
loosing packets and you can't do anyting about it, try NFS over TCP instead of
UDP (not all NFS implementations support that).
As to tuning, the only thing that I know of is the number of NFS servers, try
a higher number and se if it helps. Check if the CPU or disks are overloaded.
The problem is real in the sense that you don't get all the performance, but
no data is lost.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199808082050.WAA08160>