I ran into a problem with NFS eariler tonight that I thought I would
share. This is something that might have been fixed recently, but
perhaps not:
client:
running 2.1.131-ac13 (newly compiled)
server:
running 2.1.131-ac8
user space nfsd from nfs-server-2.2beta37-1
I mount a nfs partiton with the word perfect 8 tar on it on the
client. I unpack it, I run the "Runme" install script.
It unpacks fine and lots of nfs traffic goes by. Then it locks up.
Server starts getting:
Dec 17 20:58:00 jelerak kernel: UDP: bad checksum. From 207.174.18.202:800 to 207.174.18.194:2049 ulen 1956
(lots and lots of them).
Client starts getting:
Dec 17 20:56:41 nightfall kernel: nfs: task 16926 can't get a request slot
Dec 17 20:56:41 nightfall kernel: nfs: task 16927 can't get a request slot
Dec 17 20:56:43 nightfall kernel: nfs: task 16929 can't get a request slot
I can repeat this at will.
Also tried a 2.0.36 server. Same thing (except no UDP checksum errors
on the server)
Tried 2.1.131 stock on the laptop also, and it did the same thing.
I can create tcpdump/strace logs if anyone has interest...
Hope this helps someone clean up the nfs stuff more.
kevin
-
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/