>>>>> " " == Luke Miller <millerlu@peak.org> writes:
> I have an NFS server running 2.2.13 (smp) and a client running
> 2.2.12-20smp. I am seeing error like this on the client
> system:
> nfs: server khan2 not responding, still trying nfs: server
> khan2 OK
> I put a note on Redhat's bug tracking system thinking it was
> something with NFS, they came back and said:
>> there is some heavy networking going on there that makes the
>> kernel on the client side to run out of available sockets. I
>> doubt this is related to the NFS server.
> Could somone point me in the right direction on changing the
> number of kernel sockets.
There exists no such option under Linux. The design is for 1 socket
per mount. What you're probably seeing is rather congestion of the
existing socket. This can have one of 3 causes:
Either:
- Server is getting congested, and is replying too slowly to
requests. In this case a 'tcpdump' will show large delays between
the client sending out a request, and the server replying to it.
or
- Actual network is congested, and you are running out of socket
buffer space. You'll probably need to wait for linux-2.4 before a
general fix is available (it has already been included in
linux-2.3.x).
or
- The RPC congestion avoidance algorithm is messing up. Try
fiddling with increasing the 'timeo', and 'retrans' NFS mount
option values.
Cheers,
Trond
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:11 EST