Re: Cache invalidation bug in NFS v3 - trivially reproducible

[Jakob Oestergaard <jakob@unthought.net>]

On Tue, Oct 11, 2005 at 11:09:27AM +0200, Leif Nixon wrote:
...
> 
> Now client n2 is stuck in a state where it uses its old cached data
> forever (or at least for several hours):
> 
>   NFS client n1                NFS client n2
> 
>   $ cat f
>   2
> 			       $ cat f
> 			       1

I can confirm this on NFSv3 UDP export from patched 2.6.11.11 server
(dual opteron 64-bit kernel) to two different SMP (32-bit x86) clients
with 2.6.12.4 and 2.6.11.11 kernels.

There are definitely timing issues - in order to reproduce the problem I
had to use "touch . ; echo 2 > r" and "touch r; cat r" on the clients -
I couldn't type the commands quickly enough one-by-one.

But right now I have:

[phoenix:joe] $ cat r
2

[raven:joe] $ cat r
1

Beautiful :)

-- 

 / jakob

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/