What happens exactly I don't know. Running nfsd with debugging switched on
looks like the calls are coming through OK. tcpdump on the client seems to
indicate that the responses are coming in as well, but my packet-level
knowledge of NFS is rusty. The result is processes hanging in Disk-wait; lots
of signal 9's gets them killed and things are normal after that. Sometimes it
works, it seems to be related with the amount of calls (eg., a 'du' over the
mounted directory will certainly hang, while a simple 'ls' often comes through
fine).
My test results:
2.0.35 -> 2.2.x user : OK
2.0.35 -> 2.2.x kernel: not tested
2.2.x -> 2.0.35 user: OK
2.2.x -> 2.2.x user : BAD
2.2.x -> 2.2.x kernel: BAD
2.2.x + v3 patches -> 2.2.x user: BAD
Seems that something in 2.2.x fs/nfs is broken so it doesn't talk too well to
2.2.x servers?
Hth,
Cees
Lines like these are in my logfile:
May 13 09:06:25 dev-ltws-2 kernel: nfs: server gaia.cdg.acriter.nl not
responding, still trying
May 13 09:08:37 dev-ltws-2 kernel: nfs: RPC call returned error 111
May 13 09:08:37 dev-ltws-2 kernel: RPC: task of released request still queued!
May 13 09:08:37 dev-ltws-2 kernel: RPC: (task is on xprt_pending)
--
Cees de Groot http://www.cdegroot.com <cg@cdegroot.com>
http://www.sgmltools.org <cg@sgmltools.org>
ICQ#38145238
-
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/