Hi,<div><br></div><div>Capacity planning is based on MEMORY and CPU. In your case 8 VMs of 512MBs completely fills the 4GB (=8*512Mb) capacity of the hosts. Note that CPU and MEMORY can be use to guide CPU/MEMORY over-commitment (i.e. you can pack more VMs in a server by setting up sizes i.e. 1 VM of CPU = 4 is equal to 2 VMs of CPU=2, equal to 4 VMs of CPU = 1, equalt to 8 VMS of CPU = 0.5....). Note that in this way you can pre-define sizes and plan capacity allocation based on those pre-define VM types...</div>
<div><br></div><div>Note that this does not take into account the monitor information. In this way, we do not have to move out any VM if the host become overloaded...</div><div><br></div><div>Cheers,</div><div><br></div>
<div>
Ruben<br><div><br><div class="gmail_quote">On Tue, Jul 31, 2012 at 3:12 PM, Jan Benadik <span dir="ltr"><<a href="mailto:jan.benadik@atos.net" target="_blank">jan.benadik@atos.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div text="#000000" bgcolor="#FFFFFF">
Hi all,<br>
<br>
for test purposes I have 2 OpenNbeula hosts (4GB RAM, DC CPU each)
and frontend (10GB, QC CPU, 410GB HDD shared by NFS), which is node
too.<br>
On those nodes there is 8 running VM's (512MB RAM, 1CPU each).<br>
<br>
I've tried to start another VM, but it is still in PENDING mode, in
the /var/log/one/sched.log, there is message:<br>
<small>Tue Jul 31 17:04:32 2012 [HOST][D]: Discovered Hosts
(enabled):<br>
0 3 4<br>
Tue Jul 31 17:04:32 2012 [VM][D]: Pending and rescheduling VMs:<br>
46<br>
Tue Jul 31 17:04:33 2012 [SCHED][D]: Host 0 filtered out. Not
enough capacity. <br>
Tue Jul 31 17:04:33 2012 [SCHED][D]: Host 3 filtered out. Not
enough capacity. <br>
Tue Jul 31 17:04:33 2012 [SCHED][D]: Host 4 filtered out. Not
enough capacity. <br>
Tue Jul 31 17:04:33 2012 [SCHED][I]: Selected hosts:<br>
PRI HID VM: 46</small><br>
-----------------------<br>
<br>
If I check capacity by monitoring (see an attached pictures), there
has to be enough sources. <br>
<br>
Why VM is still in PENDING mode?<span><font color="#888888"><br>
<br>
Jan<br>
<br>
<div>-- <br>
<font size="3" color="black" face="Verdana"> <img alt="" height="15" width="252"><br>
<b>Ján Beňadik</b><br>
<font face="Verdana"> Managed Services - Solution
Design Architect<br>
<a href="tel:%2B421%2046%205151%20332" value="+421465151332" target="_blank">+421 46 5151 332</a><br>
<a href="tel:%2B421%20903%20691%20634" value="+421903691634" target="_blank">+421 903 691 634</a><br>
<a href="mailto://jan.benadik@atos.net" target="_blank">jan.benadik@atos.net</a><br>
Vinohradnícka 6, 971 01 Prievidza<br>
<a href="http://www.sk.atos.net" target="_blank">www.sk.atos.net</a><br>
__________________________________<br>
<br>
<img alt="" height="58" width="261"><br>
</font> </font></div>
</font></span></div>
<br>_______________________________________________<br>
Users mailing list<br>
<a href="mailto:Users@lists.opennebula.org" target="_blank">Users@lists.opennebula.org</a><br>
<a href="http://lists.opennebula.org/listinfo.cgi/users-opennebula.org" target="_blank">http://lists.opennebula.org/listinfo.cgi/users-opennebula.org<br clear="all"><div><br></div>-- <br>Ruben S. Montero, PhD<br>Project co-Lead and Chief Architect<br>
OpenNebula - The Open Source Solution for Data Center Virtualization<br></a><a href="http://www.OpenNebula.org" target="_blank">www.OpenNebula.org</a> | <a href="mailto:rsmontero@opennebula.org" target="_blank">rsmontero@opennebula.org</a> | @OpenNebula<br>
<br>
</blockquote></div></div>
</div>