[Gluster-devel] Segmentation fault with afr on dapper using fuse 2.6.3 kernel module
avati at zresearch.com
Sun Apr 29 23:22:33 UTC 2007
I could not find the attachments with your post. can you also tell
what version you are running? If it is not the latest TLA checkout, is
it possilbe for you to try with the latest tla checkout (fixes have
gone into AFR since the last pre-release)
On Mon, Apr 30, 2007 at 12:45:33AM +0200, Bernhard J. M. Grün wrote:
> Hello list!
> I have a quite strange problem here.
> Our servers are using ubuntu feisty and we have some clients running
> on feisty too. This works great together. Unfortunately we also have
> some dapper servers that we have to integrate in our scenario.
> The problem is now that we get an error on dapper with a configuration
> file that works fine on feisty.
> After some debugging we also know that the problem only occurs when we
> are using afr. It does not occur with unify. Therefore we think that
> the problem is somewhere inside afr.
> Let me describe our dapper installation:
> First I compiled fuse 2.6.3 including the kernel module to /opt/fuse
> Then I compiled glusterfs from exactly the same sources that we used
> on feisty on a dapper host. To use the fuse in /opt/fuse I used the
> following configure line:
> ./configure CFLAGS="-I/opt/fuse/include -L/opt/fuse/lib"
> LDFLAGS="-L/opt/fuse/lib" CPPFLAGS="-I/opt/fuse/include"
> LIBS="-L/opt/fuse/lib" --prefix=/opt/glusterfs --disable-server
> I think this is correct up to this point.
> When I try to mount the volume using afr df shows that the volume is
> mounted and the size is also correctly displayed. But when I try to
> enter the mount point glusterfs segfaults.
> I've attached some config stuff and logs. I really hope someone of you
> can help us.
> Many thanks
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
sleep (years2secs (7500000));
More information about the Gluster-devel