[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ceph-devel
Subject:    Re: mds goes stale when downloading a large file
From:       Sage Weil <sage () newdream ! net>
Date:       2010-03-31 16:25:21
Message-ID: Pine.LNX.4.64.1003310919370.7923 () cobra ! newdream ! net
[Download RAW message or body]

On Wed, 31 Mar 2010, Wido den Hollander wrote:
> Hi,
> 
> I was using the 0.19 kclient release, but i just switched to the GIT
> release.
> 
> Mounting goes fine, but when running the "sync" command, i got:
> 
> wido@wido-desktop:~$ dmesg 
> [ 7314.160201] BUG: unable to handle kernel NULL pointer dereference at
> 0000000000000010
> [ 7314.160215] IP: [<ffffffffa0c8a856>] ceph_write_inode+0x26/0x1c0
> [ceph]

Did you get a compilation warning about pointer types?  I think this is 
due to an API change that happend in 2.6.34.  If you're using 
ceph-client-standalone.git, make sure you use the master-backport or 
unstable-backport branch, which compile on kernels back through 2.6.27.

Your available bandwidth isn't the source of the problem... things should 
work (albeit more slowly) regardless of the speed of the network.

Could you see about how big the file got before things stalled out?  
Also, can you check for a /var/log/ceph/mds0.0 log file (or any mds0*)?  
The 'reconnect' message suggests that the MDS is crashing (and another is 
taking over), and the log file often has a stack trace indicating where 
things went wrong.

Was the ceph package version 0.19.1?

Thanks-
sage

--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
[prev in list] [next in list] [prev in thread] [next in thread]