<p dir="ltr">We can't turn off the memory balloon option because we are running the 3.3.2 with the bug of the memory balloon. Thx for the help!</p>
<div class="gmail_quote">On Feb 13, 2014 10:51 AM, "Doron Fediuck" <<a href="mailto:dfediuck@redhat.com">dfediuck@redhat.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
<br>
----- Original Message -----<br>
> From: "René Koch" <<a href="mailto:rkoch@linuxland.at">rkoch@linuxland.at</a>><br>
> To: "Doron Fediuck" <<a href="mailto:dfediuck@redhat.com">dfediuck@redhat.com</a>><br>
> Cc: "Koen Vanoppen" <<a href="mailto:vanoppen.koen@gmail.com">vanoppen.koen@gmail.com</a>>, <a href="mailto:users@ovirt.org">users@ovirt.org</a>, "Martin Sivak" <<a href="mailto:msivak@redhat.com">msivak@redhat.com</a>><br>
> Sent: Thursday, February 13, 2014 11:16:27 AM<br>
> Subject: Re: [Users] Memory usage<br>
><br>
> On Thu, 2014-02-13 at 03:49 -0500, Doron Fediuck wrote:<br>
> ><br>
> > ----- Original Message -----<br>
> > > From: "René Koch" <<a href="mailto:rkoch@linuxland.at">rkoch@linuxland.at</a>><br>
> > > To: "Doron Fediuck" <<a href="mailto:dfediuck@redhat.com">dfediuck@redhat.com</a>><br>
> > > Cc: "Koen Vanoppen" <<a href="mailto:vanoppen.koen@gmail.com">vanoppen.koen@gmail.com</a>>, <a href="mailto:users@ovirt.org">users@ovirt.org</a>, "Martin<br>
> > > Sivak" <<a href="mailto:msivak@redhat.com">msivak@redhat.com</a>><br>
> > > Sent: Wednesday, February 12, 2014 7:31:47 PM<br>
> > > Subject: Re: [Users] Memory usage<br>
> > ><br>
> > > On Wed, 2014-02-12 at 11:22 -0500, Doron Fediuck wrote:<br>
> > > ><br>
> > > > ----- Original Message -----<br>
> > > > > From: "René Koch" <<a href="mailto:rkoch@linuxland.at">rkoch@linuxland.at</a>><br>
> > > > > To: "Koen Vanoppen" <<a href="mailto:vanoppen.koen@gmail.com">vanoppen.koen@gmail.com</a>><br>
> > > > > Cc: <a href="mailto:users@ovirt.org">users@ovirt.org</a><br>
> > > > > Sent: Wednesday, February 12, 2014 4:18:37 PM<br>
> > > > > Subject: Re: [Users] Memory usage<br>
> > > > ><br>
> > > > > On Wed, 2014-02-12 at 15:14 +0100, Koen Vanoppen wrote:<br>
> > > > > > In The GUI, it says it's using 25% of the memory.<br>
> > > > ><br>
> > > > ><br>
> > > > > I guess it's the real value, right?<br>
> > > > > The same happened for the memcached vm, someone reported to me -<br>
> > > > > negative value in REST-API, but correct graph in oVirt webadmin GUI.<br>
> > > > ><br>
> > > > > I fear I have no idea how this can happen - so maybe someone else can<br>
> > > > > help you troubleshoot this issue.<br>
> > > > > ><br>
> > > > > ><br>
> > > > > ><br>
> > > > > > 2014-02-12 15:10 GMT+01:00 Koen Vanoppen <<a href="mailto:vanoppen.koen@gmail.com">vanoppen.koen@gmail.com</a>>:<br>
> > > > > > Thanks for the quick respons, but there is no memcached<br>
> > > > > > running on that VM.<br>
> > > > > ><br>
> > > > > ><br>
> > > > > > Kind regards<br>
> > > > > ><br>
> > > > > ><br>
> > > > > ><br>
> > > > > > 2014-02-12 15:06 GMT+01:00 René Koch <<a href="mailto:rkoch@linuxland.at">rkoch@linuxland.at</a>>:<br>
> > > > > ><br>
> > > > > > On Wed, 2014-02-12 at 14:55 +0100, Koen Vanoppen<br>
> > > > > > wrote:<br>
> > > > > > > Dear all,<br>
> > > > > > ><br>
> > > > > > ><br>
> > > > > > > When we monitor one of our machines, we noticed<br>
> > > > > > > that<br>
> > > > > > there was one vm<br>
> > > > > > > that was constantly giving a error of memory<br>
> > > > > > > usage.<br>
> > > > > > But when we took a<br>
> > > > > > > look at it, there is actually nothing wrong with<br>
> > > > > > > it.<br>
> > > > > > Now we looked<br>
> > > > > > > furhter then that. We looked at the API of the<br>
> > > > > > machine and noticed<br>
> > > > > > > something very strange:<br>
> > > > > > ><br>
> > > > > > > <statistic<br>
> > > > > > ><br>
> > > > > > href="/api/vms/3b9aa245-75ff-42e8-b921-1c9ce61826bf/statistics/b7499508-c1c3-32f0-8174-c1783e57bb08"<br>
> > > > > > id="b7499508-c1c3-32f0-8174-c1783e57bb08"><name>memory.used</name><description>Memory<br>
> > > > > > used (agent)</description><values<br>
> > > > > > type="INTEGER"><value><datum>-944892806</datum></value></values><type>GAUGE</type><unit>BYTES</unit><br>
> > > > > > ><br>
> > > > > > ><br>
> > > > > > > It's a negative...<br>
> > > > > ><br>
> > > > > ><br>
> > > > > ><br>
> > > > > > Do you have memcached running in this vm?<br>
> > > > > ><br>
> > > > > > I heard about this issue with memcached, but never<br>
> > > > > > tested memcached in<br>
> > > > > > my oVirt environment. You get the real usage value<br>
> > > > > > with<br>
> > > > > > memory.used = memory.installed + memory.used<br>
> > > > > ><br>
> > > > > ><br>
> > > > > > Regards,<br>
> > > > > > René<br>
> > > > > ><br>
> > > > > > ><br>
> > > > > > ><br>
> > > > > > > What could be the problem?<br>
> > > > > > ><br>
> > > > > > ><br>
> > > > > > > Kind regards,<br>
> > > > > > ><br>
> > > > > > > koen<br>
> > > > > > ><br>
> > > > > ><br>
> > > ><br>
> > > > Guys,<br>
> > > > these values are usually a result of overcommitment mechanism usage.<br>
> > > > For example, if KSM is effective, it will free a lot of memory pages,<br>
> > > > and total-free-committed becomes negative.<br>
> > ><br>
> > ><br>
> > > Thanks a lot for the information.<br>
> > ><br>
> > ><br>
> > > ><br>
> > > > This was reported in <a href="https://bugzilla.redhat.com/show_bug.cgi?id=977758" target="_blank">https://bugzilla.redhat.com/show_bug.cgi?id=977758</a><br>
> > > > and the engine is using memFree reported by vdsm, which is more<br>
> > > > accurate.<br>
> > > ><br>
> > > > The API reports the old version due to backwards compatibility.<br>
> > ><br>
> > ><br>
> > > I just had a look at the bugzilla report and the RHEV documentation<br>
> > > which says "Current memory in bytes used by the virtual machine.".<br>
> > > So this means, the reported values are totally useless for monitoring<br>
> > > memory usage of a virtual machine if KSM is active. I would expect to<br>
> > > get the memory usage of a virtual machine and not how much memory is<br>
> > > consumed on the hypervisor (this is pretty useless information for me).<br>
> > ><br>
> > > Is it planned to report the memory usage in a virtual machine in the API<br>
> > > as well?<br>
> > ><br>
> > ><br>
> > > Regards,<br>
> > > René<br>
> > ><br>
> > ><br>
> ><br>
> > René,<br>
> > started digging into this, but it may take some time.<br>
> ><br>
> > Note that there are 2 memory usage reports; one for the host,<br>
> > and another for each VM.<br>
> > My response is related to the host, while the original question was<br>
> > for the VM. I still think the root cause is the same (ie- related to<br>
> > overcommitment; for example the VM may have a balloon inflated), but<br>
> > would like to properly check and fix if needed.<br>
><br>
><br>
> Yes, I know. So my question was if we can add another value for the<br>
> usage in the vm in addition to the usage of the vm on the host.<br>
><br>
We'll know once we understand the cause- ie is it a bug needs fixing or<br>
should we indeed add another value<br>
<br>
If you have a setup with this issue, try removing the balloon device<br>
and see if it helps / changes.<br>
<br>
><br>
> ><br>
> > In order to make sure we keep track of it, please open a bug with all the<br>
> > relevant info.<br>
> ><br>
><br>
> I'll open a bug for this.<br>
><br>
Thanks<br>
<br>
><br>
> > Thanks and keep reporting!<br>
> > Doron<br>
><br>
><br>
><br>
</blockquote></div>