<div dir="ltr"><br>| <br>| Citējot Vijay Bellur <<span class="" id="OBJ_PREFIX_DWT2162_com_zimbra_email"><span class="" id="OBJ_PREFIX_DWT2163_com_zimbra_email"><span class="" id="OBJ_PREFIX_DWT2168_com_zimbra_email"><a href="mailto:vbellur@redhat.com">vbellur@redhat.com</a></span></span></span>> :<br>
| > On <span class="" id="OBJ_PREFIX_DWT2164_com_zimbra_date"><span class="" id="OBJ_PREFIX_DWT2165_com_zimbra_date"><span class="" id="OBJ_PREFIX_DWT2169_com_zimbra_date">05/11/2014</span></span></span> 02:04 AM, Vadims Korsaks wrote:<br>
| > > HI!<br>| > ><br>| > > Created 2 node setup with oVirt 3.4 and<br>| CentOS 6.5, for storage created<br>| > > 2 node replicated gluster (3.5) fs on same<br>| hosts with oVirt.<br>| > > mount looks like this:<br>
| > > 127.0.0.1:/gluster01 on<br>| > ><br>| /rhev/data-center/mnt/glusterSD/127.0.0.1:_gluster01<br>| type fuse.glusterfs<br>| > ><br>| (rw,default_permissions,allow_other,max_read=131072)<br>| > ><br>
| > > when i making gluster test with dd, something<br>| like<br>| > > dd if=/dev/zero bs=1M count=20000<br>| > ><br>| of=/rhev/data-center/mnt/glusterSD/<a href="http://127.0.0.1">127.0.0.1</a>\:_gluster01/kaka<br>
| > > i'm gettting speed ~ 110 MB/s, so this is<br>| 1Gbps speed of ethernet adapter<br>| > ><br>| > > but with in VM created in oVirt speed is<br>| lower than 20 MB/s<br>| > ><br>| > > why there is so huge difference?<br>
| > > how can improve VMs disks speed?<br>| > ><br>| ><br>| > What are your gluster volume settings? Have you<br>| applied the following<br>| > performance tunables in gluster's virt profile:<br>
| ><br>| > eager-lock=enable<br>| > remote-dio=enable<br>| ><br>| > Regards,<br>| > Vijay<br>| ><br>| setting were:<br>| [root@centos155 ~]# gluster volume info gluster01<br>| <br>| Volume Name: gluster01<br>
| Type: Replicate<br>| Volume ID: 436edaa3-ac8b-421f-aa35-68b5bd7064b6<br>| Status: Started<br>| Number of Bricks: 1 x 2 = 2<br>| Transport-type: tcp<br>| Bricks:<br>| Brick1: 10.2.75.152:/mnt/gluster01/brick<br>| Brick2: 10.2.75.155:/mnt/gluster01/brick<br>
| Options Reconfigured:<br>| storage.owner-gid: 36<br>| storage.owner-uid: 36<br>| <br>| <br>| add your settings settings now it looks<br>| <br>| [root@centos155 ~]# gluster volume info gluster01<br>| <br>| Volume Name: gluster01<br>
| Type: Replicate<br>| Volume ID: 436edaa3-ac8b-421f-aa35-68b5bd7064b6<br>| Status: Started<br>| Number of Bricks: 1 x 2 = 2<br>| Transport-type: tcp<br>| Bricks:<br>| Brick1: 10.2.75.152:/mnt/gluster01/brick<br>| Brick2: 10.2.75.155:/mnt/gluster01/brick<br>
| Options Reconfigured:<br>| network.remote-dio: enable<br>| cluster.eager-lock: enable<br>| storage.owner-gid: 36<br>| storage.owner-uid: 36<br>| <br>| <br>| but this didn't affected performace in any big way<br>| should hosts to be restarted?<br>
| <br><br>glusterfs
storage domain configuration GUI got ""Optimize for virt. store" option
which have to be enabled when configuring for virt store.<br><br>Ref#<span class="" id="OBJ_PREFIX_DWT2166_com_zimbra_url"><span class="" id="OBJ_PREFIX_DWT2167_com_zimbra_url"><span class="" id="OBJ_PREFIX_DWT2170_com_zimbra_url"><a target="_blank" href="http://www.ovirt.org/Features/GlusterFS_Storage_Domain">http://www.ovirt.org/Features/GlusterFS_Storage_Domain</a></span></span></span><br>
<br>If the configuration is manual, you need to set the group to 'virt' as shown below<br><br>#gluster volume set VOLNAME group virt<br><br>This will enable below options on gluster volume :<br><br>quick-read=off<br>
read-ahead=off<br>io-cache=off<br>stat-prefetch=off<br>eager-lock=enable<br>remote-dio=on<br><br><br>Can you please make sure group has set properly ?<br><br>Also, invoke "dd" with oflag=direct option and check whether it helps ..<br>
<br><br>--Humble<br></div>