[Gluster-devel] split brain with all-zero pending matrix

[Emmanuel Dreyfus]

Pranith Kumar Karampuri <pkarampu at redhat.com> wrote:

>    Were the bricks down at any point in time before you hit this?.

Yes, I have two bricks on the same machine that crashed because they ran
out of memory. While being there: there are probably memory leaks, and
it would be nice if glusterd could detect and restart when glusterfsd
crashes

>  Do you have any logs available that give us hints about this issue?

Here is client log: http://ftp.espci.fr/shadow/many/nullmatrix.log

> Is it a 1 X 2 X 2 (striped replicate) configuration?

Yes.

-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
manu at netbsd.org