Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I've been playing around with RHEV. I was wondering what is the best practise for the RHEV Manager in terms of installation.

In a big production environment is it best for the Manager to be installed on a separate physical server-blade alone (I guess not because in terms of requirements it's a complete overkill), as a guest on a separate physical server, or it makes sense to be installed as a guest machine under a hypervisor that is managed by it (is that even possible?).

Thanks in advance.

share|improve this question
Just note that while a RHEV-M can manage multiple data centers, you can quick run into problems if those data centers are geographically diverse and the links is high latency and or low bandwidth. I've had heaps of problems with RHEV-H nodes being fenced because they couldn't talk to the RHEV-M with-in the required time frame due to network problems between the sites. I've logged an RFE to get the timeouts to be customizable per data center – user1174838 May 1 '14 at 8:13
up vote 1 down vote accepted

Self hosted engine is the way to go if you want to save up on the engine host. If you are running a large environment (and RHEV can easily scale to hundreds of managed hosts, even thousands), I would definitely recommend building a separate cluster for the engine, which is easily clusterable in at least three different ways.

For self-hosted you need to use RHEL hosts, RHEV-H, afaik, doesn't support it yet (due to lots of extra packages that will have to make their way into the locked down hypervisor distribution.)

share|improve this answer

Current stable RHEV version is 3.3.

The upstream project (OVirt) has released version 3.4 as of March 2014, with RHEV 3.4 being available as a BETA release.

From OVirt's 3.4 release notes:

Hosted Engine

oVirt 3.4 features hosted engine, which enables oVirt engine to be run as a virtual machine (VM) on the host it manages. Hosted engine solves the chicken-and-the egg problem for users: the basic challenge of deploying and running an oVirt engine inside a VM. This clustered solution enables users to configure multiple hosts to run the hosted engine, ensuring the engine still runs in the event of any one host failure.

If you want to try this feature, this howto explains the procedure to install the hosted engine.

share|improve this answer
Thanks for the useful info. As I can see RHEV 3.3 has also this self hosted engine. link But I read that: "The self-hosted engine currently only runs on Red Hat Enterprise Linux 6.5 hosts. Hypervisors and older versions of Red Hat Enterprise Linux are not recommended for use with a self-hosted engine." I guess with hypervisors means out of the box rhev-h images? – Cobra Kai Dojo Apr 4 '14 at 13:38

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.