<div style="line-height:1.7;color:#000000;font-size:14px;font-family:Arial">Could you please tell me your glusterfs version and the mount command that you have used?? My GlusterFS version is 3.3.0, different versions may be exits different results.<br><br><br><br><br><div style="position:relative;zoom:1"></div><div id="divNeteaseMailCard"></div><br>At 2016-09-06 12:35:19, "Ravishankar N" <ravishankar@redhat.com> wrote:<br> <blockquote id="isReplyContent" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div class="moz-cite-prefix">That is strange. I tried the experiment
on a volume with a million files. The client node's memory usage
did grow, as I observed from the output of free(1)
<a class="moz-txt-link-freetext" href="http://paste.fedoraproject.org/422551/">http://paste.fedoraproject.org/422551/</a> when I did a `ls`.<br>
-Ravi<br>
<br>
On 09/02/2016 07:31 AM, Keiviw wrote:<br>
</div>
<blockquote cite="mid:41fc21e8.642a.156e8a129e6.Coremail.keiviw@163.com" type="cite">
<div id="contentDescription" style="line-height:1.5;text-align:justify;text-justify:inter-ideograph">
<div>Exactly, I mounted the volume in a no-brick node(nodeB),
and nodeA was the server. I have set different timeout, but
when I excute "ls /mnt/glusterfs(about 3 million small files,
in other words, about 3 million dentries)", the result was the
same, memory usage in nodeB didn't change at all while nodeA's
memory usage was changed about 4GB!</div>
<div><br>
</div>
<div class="NETEASEMAILMASTERLOCALSIGNATURE"><span style="color:#888;font-size:15px;">发自 </span><a moz-do-not-send="true" href="http://u.163.com/signature" style="font-size:15px;color:#2e90eb;" target="_blank">网易邮箱大师</a></div>
<div class="borderFixWidth iMailDoNotReScale" style="background-color:#f2f2f2;color:black;padding-top:6px;padding-bottom:6px;border-radius:3px;-moz-border-radius:3px;-webkit-border-radius:3px;margin-top:45px;margin-bottom:20px;">
<div style="font-size:14px;line-height:1.5;word-break:break-all;margin-left:10px;margin-right:10px">On
09/02/2016 09:45, <a moz-do-not-send="true" style="text-decoration:none;color:#2a97ff;" href="mailto:ravishankar@redhat.com">Ravishankar N</a>
wrote:</div>
</div>
<blockquote id="ntes-iosmail-quote" style="margin:0">
<div class="moz-cite-prefix">On 09/02/2016 05:42 AM, Keiviw
wrote:<br>
</div>
<blockquote cite="mid:6cbf1881.1050.156e83d1e68.Coremail.keiviw@163.com" type="cite">
<div style="line-height:1.7;color:#000000;font-size:14px;font-family:Arial">Even
if I set the attribute-timeout and entry-timeout to
3600s(1h), in the nodeB, it didn't cache any metadata
because the memory usage didn't change. So I was confused
that why did the client not cache dentries and inodes.<br>
<br>
</div>
</blockquote>
<tt>If you only want to test fuse's caching, I would try
mounting the volume on a separate machine (not on the brick
node itself), disable all gluster performance xlators, do a
find.|xargs stat on the mount 2 times in succession and see
what free(1) reports the 1st and 2nd time. You could do this
experiment with various attr/entry timeout values. Make sure
your volume has a lot of small files.<br>
-Ravi<br>
</tt>
<blockquote cite="mid:6cbf1881.1050.156e83d1e68.Coremail.keiviw@163.com" type="cite">
<div style="line-height:1.7;color:#000000;font-size:14px;font-family:Arial"><br>
<br>
在 2016-09-01 16:37:00,"Ravishankar N" <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:ravishankar@redhat.com"><ravishankar@redhat.com></a>
写道:<br>
<blockquote id="isReplyContent" style="PADDING-LEFT: 1ex;
MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div class="moz-cite-prefix">On 09/01/2016 01:04 PM,
Keiviw wrote:<br>
</div>
<blockquote cite="mid:14087c4e.3f73.156e4ab683e.Coremail.keiviw@163.com" type="cite">
<div style="line-height:1.7;color:#000000;font-size:14px;font-family:Arial">
<div>Hi,</div>
<div> I have found that GlusterFS client(mounted
by FUSE) didn't cache metadata like dentries and
inodes. I have installed GlusterFS 3.6.0 in nodeA
and nodeB, and the brick1 and brick2 was in nodeA,
then in nodeB, I mounted the volume to
/mnt/glusterfs by FUSE. From my test, I excuted
'ls /mnt/glusterfs' in nodeB, and found that the
memory didn't use at all. Here are my questions:</div>
<div> 1. In fuse kernel, the author set some
attributes to control the time-out about dentry
and inode, in other words, the fuse kernel
supports metadata cache, but in my test, dentries
and inodes were not cached. WHY?</div>
<div> 2. Were there some options in GlusterFS
mounted to local to enable the metadata cache in
fuse kernel? </div>
</div>
<br>
<br>
</blockquote>
<tt>You can tweak the attribute-timeout and
entry-timeout seconds while mounting the volume.
Default is 1 second for both. `man mount.glusterfs`
lists various mount options.<br>
-Ravi<br>
</tt>
<blockquote cite="mid:14087c4e.3f73.156e4ab683e.Coremail.keiviw@163.com" type="cite"><span title="neteasefooter">
<p> </p>
</span><br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
Gluster-devel mailing list
<a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:Gluster-devel@gluster.org">Gluster-devel@gluster.org</a>
<a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://www.gluster.org/mailman/listinfo/gluster-devel">http://www.gluster.org/mailman/listinfo/gluster-devel</a></pre>
</blockquote>
<p><br>
</p>
</blockquote>
</div>
<br>
<br>
<span title="neteasefooter">
<p> </p>
</span> </blockquote>
<p><br>
</p>
</blockquote>
</div>
<br>
<br>
</blockquote>
<p><br>
</p>
</blockquote></div><br><br><span title="neteasefooter"><p> </p></span>