[Gluster-devel] [Gluster-users] Possible memory leak ?
lmohanty at redhat.com
Thu Sep 12 20:44:05 UTC 2013
On 09/12/2013 07:01 AM, haiwei.xie-soulinfo wrote:
> We meet memory leak in 3.4.0 & sambavfs.
> With mounting cifs, our application runs,'VIRT' of smbd process will increase untill oom-kill.
> using 'fsync/sync' or 'echo 1 > /proc/sys/vm/drop_caches' can't resolve it. With mounting fuse, no memory leaks.
> I guess fuse API or samba-gluster-vfs has bug, any advise?
Do you have any back trace for the crash, log messages or anything that
would help us debug the issue?
>> I'm aware of 2 different kinds of memory leaks on 3.3.1, one is in
>> geo-replication and another one is native client side memory leak.
>> Sadly both got mixed in https://bugzilla.redhat.com/show_bug.cgi?id=841617
>> I can tell you that geo-replication leak is still present in 3.4.0 and
>> native client leak isn't but I don't know what patch you need to backport :(
>> On Wed, Sep 11, 2013 at 1:16 PM, John Ewing <johnewing1 at gmail.com> wrote:
>>> I am using gluster 3.3.1 on Centos 6, installed from
>>> the glusterfs-3.3.1-1.el6.x86_64.rpm rpms.
>>> I am seeing the Committed_AS memory continually increasing and the
>>> processes using the memory are glusterfsd instances.
>>> see http://imgur.com/K3dalTW for graph.
>>> Both nodes are exhibiting the same behaviour, I have tried the suggested
>>> echo 2 > /proc/sys/vm/drop_caches
>>> but it made no difference. It there a known issue with 3.3.1 ?
>>> Gluster-users mailing list
>>> Gluster-users at gluster.org
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
More information about the Gluster-devel