OpenStack Compute (nova)

VMWareAPI or libvirt > 0.8

Asked by Laurent Borensky

Hello,

OK, VMWareAPI is planned for Cactus or Diablo? Why did you create VMWareAPI as XenAPI, while 0.8 libvirt can control the resources VMWare (ESX, but vCenter Server) ? We want to use Nova with VMWare Server via vCenter and also have one or more hypervisors OpenSource (XCP, KVM, ...)

Thank you for your response.

Regards,

Question information

Language:
English Edit question
Status:
Solved
For:
OpenStack Compute (nova) Edit question
Assignee:
No assignee Edit question
Solved by:
Laurent Borensky
Solved:
Last query:
Last reply:
Revision history for this message
Armando Migliaccio (armando-migliaccio) said : 		#1

VMWareAPI is not yet in trunk, so I assume that there will be room for this discussion when it is proposed for merge.

However, you can find reasons for providing Xen support via XenAPI here: http://wiki.openstack.org/XenAPI

I guess the same applies to the support for vmware.

Hope this help!
Cheers,
Armando

Revision history for this message
Laurent Borensky (laurent-borensky) said : 		#2

Hi Armando,

Thank you. For now, we prefer to use VMWare over Citrix for its advanced features (ie: Fault Tolerance, Storage VMotion, ...). But why was it not possible to use libvirt (> 0.8 with vpx:///) to manage the ESX Server via vCenter?

Regards

Revision history for this message
Ewan Mellor (ewanmellor) said : 		#3

Libvirt is a lowest-common-denominator API, and wouldn't allow us to control advanced features in the way that we will need to. Going direct to the vSphere API will allow us that control. (libvirt's ESX support is also a long way from finished -- see http://www.jedi.be/blog/2010/12/08/libvirt-0-8-6-and-vmware-esx/ for a discussion).

Note that in Cactus these advanced features will not be present in any case. OpenStack's models are not rich enough to represent Fault Tolerance set-ups, or Storage vMotion. We expect to extend this over time.

Revision history for this message
Laurent Borensky (laurent-borensky) said : 		#4

Thank you, very much for your return.