Re: [Gluster-devel] Glusterfs distributed volume stops a long time with bailing out frame type(GlusterFS 3.2.7) op(INODELK(29)

[Joe Julian]

On 01/22/2013 07:03 PM, Song wrote:

Hi,

 

When application access a directory of glusterfs volume, glusterfs client spend about 1 hour, from 19:44:05 to 20:44:25.

The gluster volume is DHT + AFR, native mount to /xmail/gfs1

 

Access(/xmail/gfs1/xmail_dedup/gfs1_000/011/204/)

 

 

The following message is displayed in client log:

 

Are your bricks ext4? http://joejulian.name/blog/glusterfs-bit-by-ext4-structure-change/