Hi, I'm a student currently doing my final study project based around Ovirt. I played a lot with Ovirt to have an overview of what could or could not be done, but I still have a few questions and stuff I want to clarify. My project consist of a HA architecture using Ovirt, powered by 8 reasonably powerful servers (4 x Quad core Opteron, 32Go RAM, 2x10k RPM HDD) equipped with 2 NIC each. I followed this tutorial to use Ovirt with GlusterFS : http://community.redhat.com/blog/2014/05/ovirt-3-4-glusterized/ Worked well ! I installed 2 self hosted engine on 2 different servers and a GlusterFS using 2 other servers. But since we had only 2 x 1Gb NIC per server, we decided to go with bonding and VLAN to separate each networks inspired by this blog post : http://captainkvm.com/2013/04/maximizing-your-10gb-ethernet-in-rhev/ Unfortunately, it seems like Ovirt 3.4 does not support installing self-hosted engine on bond + vlan. I tried 3.5 but there were to much bug to be usable and the project is set to be deployed in 2 months. A colleague suggested me a workaround using OpenvSwitch between Ovirt and NIC bond to "translate tagged packet into non tagged packet" and hide the bond from Ovirt. Does this have a chance to work ? Since the GlusterFS is accessed by NFS, I was able to bond the two servers. A few questions : - What is the purpose of ovirtmgmt network ? I did a lot of search by haven't found any clear explanation. Does it need to be publicly accessible or a private ip is fine too ? - Is the display network used for SPICE/VNC connection ? - How do Ovirt differentiate VM network from a storage network ? They both have the same vm role in the interface. Both networks could (should) be on private ip range right ? - How can I add the storage network after (or before, would be better) self hosted engine is installed ? (Since self hosted engine is stored on Gluster, I will loose connection to the engine) - Using self hosted engine, does all my nodes need to be installed with hosted-engine --deploy, or can I have only 2 self hosted engine nodes and 4 classic nodes ? - I'm trying to cleanly re-install the second hosted engine after some experiments, but the behavior is strange : [ INFO ] Stage: Transaction commit [ INFO ] Stage: Closing up To continue make a selection from the options below: (1) Continue setup - engine installation is complete (2) Power off and restart the VM (3) Abort setup Isn't this supposed to give me information to connect to the vm so i can install the engine ? Thanks a lot for this truly well made software !