On Fri, Sep 23, 2005 at 06:53:09AM +0000, Martijn van Buul wrote:
> It occurred to me that Andy Ruhl wrote in gmane.os.netbsd.general:
> > Yeah but all bets are off if you have to sit next to your access
> > point for the thing to actually work. There's only so much the
> > protocol can do if the hardware doesn't cooperate.
> 
> The hardware was cooperating enough; just at a low speed. But NFS
> dying on me, even when using nfs-over-tcp, indicates a more
> fundamental problem.  Lowering the blocksizes helped, which indicates
> that there's something fishy with large packets.

there's something fishy with NFS in the 2.0 branch in general.  I've
ended up going back to 4kB UDP mounts at home as a workaround.  at work
I'm doing 8kB TCP without problems yet, but the fact that hitting NFS
hard gives me dozens of "not responding" / "is alive again" makes me
wonder if some timer either needs tweaking or isn't being auto-adjusted
properly.

random packet loss seems to be what triggers the issue.  at home I have
a cheapass realtek 100Mbit switch which tends to reset itself during
minor power glitches, and my home problems seem to relate to that.  the
client just completely gets lost, like retries simply aren't happening.

-- 
  Aaron J. Grier | "Not your ordinary poofy goof." | agrier@poofygoof.com
              "silly brewer, saaz are for pils!"  --  virt