[Gluster-devel] brick stops responding

[Emmanuel Dreyfus]

Emmanuel Dreyfus <manu at netbsd.org> wrote:

> Now the statedump command completes, but since I restarted the daemons
> in the meantime, the bug is not there anymore. I will restart my tests.

It came rather quickly from another brick. I assume the condition comes after
some time, and it was ready to happen here.

[2014-01-19 04:08:21.648828] C [client-handshake.c:127:rpc_client_ping_timer_expired] 0-gfs35b1-client-1: server 192.0.2.98:49152 has not responded in the last 42 seconds, disconnecting.

The offending brick state dump is here:
http://ftp.espci.fr/shadow/manu/statedump

Do I read it correctly? It seems full of pending locks. What does that mean?

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