Original Post:
> We are running 4.0D and are getting errors from a server we use to server
> out home directories to other systems (via mount and automount).
>
> The errors:
> Jun 7 02:09:07 name vmunix: NFS3 RFS3_WRITE failed for server
> panda: RPC: Timed out
> Jun 7 02:09:07 name vmunix: NFS3 write error 60 on host panda
>
> Is there someplace where these error numbers are documented? Would
> increasing the number of threads for the UDP nfsd daemon help this
(increase
> from 7 to 16?).
>
Thanks to all that responded. From Dr. Thomas P. Blinn:
Have you looked at /usr/include/errno.h ? It's the standard UNIX error
message number file; error number 60 is:
#define ETIMEDOUT 60 /* Connection timed out */
which fits very nicely with the text error message:
NFS3 RFS3_WRITE failed for server panda: RPC: Timed out
In other words, the server "panda" (which is presumably in your local
network somewhere) isn't responding.
Increasing the number of threads on panda might help. There may be other
things you can do to improve the performance of the server. Perhaps panda
is not up to the load you're imposing, but that seems unlikely. Make sure
you've got a reliable network -- check for saturation, lots of retries, all
of
the things that can make network connections time out. Also make sure that
panda can support enough network connections and processes/threads.
Received on Tue Jun 08 1999 - 14:30:05 NZST