[Gluster-users] I am very confused about strip Stripe what way it hold space?

Brian Foster bfoster at redhat.com
Tue Nov 6 12:34:55 UTC 2012


On 11/05/2012 08:38 PM, 肖力 wrote:
> I have 4 dell 2970 server , three server  harddisk is 146Gx6 ,one hard
> disk is 72Gx6:
> 
> each server mount info is
> //dev/sda4 on /exp1 type xfs (rw)/
> //dev/sdb1 on /exp2 type xfs (rw)/
> //dev/sdc1 on /exp3 type xfs (rw)/
> //dev/sdd1 on /exp4 type xfs (rw)/
> //dev/sde1 on /exp5 type xfs (rw)/
> //dev/sdf1 on /exp6 type xfs (rw)/
> 
> I create a gluster volume  have 4 stripe 
> /gluster volume create test-volume3 stripe 4 transport tcp \/
> /172.16.20.231:/exp4 \/
> /172.16.20.232:/exp4  \/
> /172.16.20.233:/exp4  \/
> /172.16.20.235:/exp4 \/
> 
> then i mount volume on client 172.16.20.230
> /mount -t glusterfs 192.168.106.231:/test-volume3 /gfs3/
> and i dd 10G file in gfs3
> /dd if=/dev/zero of=/gfs3/3 bs=1M count=10240/
> /10240+0 records in/
> /10240+0 records out/
> /10737418240 bytes (11 GB) copied, 119.515 s, 89.8 MB/s/
> I am very confused about this
> /[root at node231 ~]# du -hs /exp4/
> /10G     /exp4/
> /[root at node232 ~]# du -hs /exp4/
> /10G     /exp4/
> /[root at node233 ~]# du -hs /exp4/
> /10G     /exp4/
> /[root at node235 ~]# du -hs /exp4/
> /10G     /exp4/
> i *understand **Stripe 4 should hold 1/4 space on each brick,why is 10G
> on each brick?*
> *can someon e explain it,think you.*
> 

The default stripe data layout conflicts with XFS default speculative
preallocation behavior. XFS preallocates space beyond the end of files
and the stripe translator continuously seeks past this space, making it
permanent.

You can address this by 1.) enabling the cluster.stripe-coalesce
translator option in gluster or 2.) setting the allocsize mount option
(i.e., allocsize=128k) in XFS. Note that using the latter option will
increase the likelihood of fragmentation on the backend filesystem.

Brian

> 
> 
> 
> 
> 
> 
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://supercolony.gluster.org/mailman/listinfo/gluster-users
> 




More information about the Gluster-users mailing list