<div> I change the envirnment :</div><div> glusterfs versions: glusterfs-3.7.10</div><div style="line-height: 21px;"> ovirt-engine: 3.6.4</div><div style="line-height: 21px;"> gluster volume brick use SSD disk </div><div style="line-height: 21px;"><br></div><div style="line-height: 21px;"><br></div><div style="line-height: 21px;"> hosted-engine volume : replicate 3 arbit 1 (only for ovirt-engine vm)</div><div style="line-height: 21px;"> data volume : replicate 2 (for other vms ,replicate 2 write performance is goole than replicate 3 )</div><div style="line-height: 21px;"><br></div><div style="line-height: 21px;"> and vm memory is show normal value , not up to the <span style="line-height: 1.5;">high value suddenly .</span></div><div style="line-height: 21px;"><span style="line-height: 1.5;"><br></span></div><div style="line-height: 21px;"> I think there are several possible :</div><div style="line-height: 21px;"> 1. gluster volume(<span style="line-height: 1.5;"> </span><span style="line-height: 1.5;">replicate 3 </span>) on HDD disk (and I only one HDD disk on each server as gluster volume brick ) <span style="line-height: 1.5;">performance is pool </span></div><div style="line-height: 21px;"> 2. possible is versions(glusterfs:3.6.7 /ovirt:3.6.1 ) problem ,the glusterfs and ovirt-engine on host , memory or cpu resource management is not very google .</div><div style="line-height: 21px;"> 3. maybe only is ovirt show problem . </div><div style="line-height: 21px;"><br></div><div><includetail><div> </div><div> </div><div style="font:Verdana normal 14px;color:#000;"><div style="FONT-SIZE: 12px;FONT-FAMILY: Arial Narrow;padding:2px 0 2px 0;">------------------ Original ------------------</div><div style="FONT-SIZE: 12px;background:#efefef;padding:8px;"><div id="menu_sender"><b>From: </b> "胡茂荣"<maorong.hu@horebdata.cn>;</div><div><b>Date: </b> Tue, Apr 12, 2016 10:47 AM</div><div><b>To: </b> "Sahina Bose"<sabose@redhat.com>; <wbr></div><div></div><div><b>Subject: </b> Re: [ovirt-users] on oVirt and Gluster hyper-converged environment vmmemory show up to 97~99%</div></div><div> </div><div> glusterfs versions: glusterfs-3.7.6 </div><div> ovirt-engine: 3.6.1 </div><div> </div><div> glusterfs volume : hosted-engine replicate 3 arbit 1 </div><div> <span style="line-height: 1.5;"> </span><span style="line-height: 1.5;">features.shard is on </span></div><div> config as :</div><div><div>gluster volume set hosted-engine group virt</div><div>gluster volume set hosted-engine storage.owner-uid 36</div><div>gluster volume set hosted-engine storage.owner-gid 36</div><div>gluster volume set hosted-engine cluster.quorum-type auto</div><div>gluster volume set hosted-engine network.ping-timeout 10</div><div>gluster volume set hosted-engine auth.allow \*</div><div>gluster volume set hosted-engine server.allow-insecure on</div><div>gluster volume set hosted-engine features.shard on</div><div>gluster volume set hosted-engine cluster.data-self-heal-algorithm full</div><div>gluster volume set hosted-engine nfs.disable on</div></div><div><br></div><div><div> and three servers memory 、cpu usage is low , cpu% is low to 40% (my host: memory is 16GB ,4 cpu cores). only start 1~3 vms , these problem also occure .</div><div><br></div><div><br></div><div> </div><div style="font:Verdana normal 14px;color:#000;"><div style="FONT-SIZE: 12px;FONT-FAMILY: Arial Narrow;padding:2px 0 2px 0;">------------------ Original ------------------</div><div style="FONT-SIZE: 12px;background:#efefef;padding:8px;"><div id="menu_sender"><b>From: </b> "Sahina Bose"<sabose@redhat.com>;</div><div><b>Date: </b> Mon, Apr 11, 2016 10:25 PM</div><div><b>To: </b> "胡茂荣"<maorong.hu@horebdata.cn>; "users"<users@ovirt.org>; <wbr></div><div></div><div><b>Subject: </b> Re: [ovirt-users] on oVirt and Gluster hyper-converged environment vmmemory show up to 97~99%</div></div><div> </div>
<br>
<br>
<div class="moz-cite-prefix">On 04/06/2016 02:53 PM, 胡茂荣 wrote:<br>
</div>
<blockquote cite="mid:tencent_7F76DE96294FD295701831CD@qq.com" type="cite">
<div>
<div> my problem :</div>
<div> On oVirt and Gluster hyper-converged environment (
which is glusterfs and ovirt on one host ,and there are three
hosts ,glusterfs volume is replication 3 abrit 1 )</div>
<div><br>
</div>
<div> storage domain use glusterfs volume which is used
glusterfs mount or nfs mount , vms running , happenned vm
memory top to 97%~99% suddenly , this current high </div>
<div><br>
</div>
<div>probability on vms which on the same host , but this time ,
vm'memory not used 97%~99% actually ,and memory down to the
actual value latterly , all above problem can see by <br>
</div>
</div>
</blockquote>
<br>
Is sharding enabled on the gluster volume?<br>
Which version of glusterfs are you using?<br>
<br>
Is the CPU usage on the servers consistently high? <br>
<br>
<blockquote cite="mid:tencent_7F76DE96294FD295701831CD@qq.com" type="cite">
<div>
<div><br>
</div>
<div>ovirt-engine webmin UI </div>
</div>
<div><br>
</div>
<div><img src="cid:A9EA30E2@FA2D8B33.88A50C57"></div>
<div><br>
</div>
<div></div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
Users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Users@ovirt.org">Users@ovirt.org</a>
<a class="moz-txt-link-freetext" href="http://lists.ovirt.org/mailman/listinfo/users">http://lists.ovirt.org/mailman/listinfo/users</a>
</pre>
</blockquote>
<br>
</div></div></div><!--<![endif]--></includetail></div>