[Gluster-devel] atomic operations fails_

[Fredrik Widlund]

"--entry-timeout=0" did not make any visible change.

Regards,
Fredrik

-----Original Message-----
From: anand.avati at gmail.com [mailto:anand.avati at gmail.com] On Behalf Of Anand Avati
Sent: den 10 februari 2010 05:00
To: Fredrik Widlund
Cc: gluster-devel at nongnu.org
Subject: Re: [Gluster-devel] atomic operations fails_

On Wed, Feb 10, 2010 at 12:48 AM, Fredrik Widlund
<fredrik.widlund at qbrick.com> wrote:
>
> Hi,
>
>
>
> I'll try to make myself clearer than in the earlier thread, since I need
> some help here. I'm not sure if I am missing something.
>
>
>
> I am not able to use atomic operations using glusterfs 3.0.2, on a Arch
> Linux 2.6.32.7-1 x86_64 server.
>
>
>
> I've stripped everything down to the most simple configuration
>
> Server: a posix store with locks
>
> Client: a tcp client
>
>
>
> I mount the glusterfs on /gfs, and use a "writer" and a "reader" as below:
>
> [writer]                root at frustration> while :; do date > /gfs/.file; mv
> /gfs/.file /gfs/file; done
>
> [reader]               root at frustration> while :; do cat /gfs/file >
> /dev/null; done
>
>
>
> Leaving this running, I get a flood of
>
> cat: /gfs/file: No such file or directory
>
>
>
> This should not happen on a filesystem that supports atomic operations?!

Please try turning off FUSE entry caching by passing --entry-timeout=0
to your mount command on the client side and let us know what
difference that makes.

Thanks,
Avati