[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-API] enabling plain HVM migration in xapi


  • To: "xen-api@xxxxxxxxxxxxxxxxxxx" <xen-api@xxxxxxxxxxxxxxxxxxx>
  • From: Dave Scott <Dave.Scott@xxxxxxxxxxxxx>
  • Date: Thu, 28 Jan 2010 11:16:06 +0000
  • Accept-language: en-US
  • Acceptlanguage: en-US
  • Delivery-date: Thu, 28 Jan 2010 03:18:39 -0800
  • List-id: Discussion of API issues surrounding Xen <xen-api.lists.xensource.com>
  • Thread-index: AcqgC6WB9ELhEw4gS1m3HAouxqBh2A==
  • Thread-topic: enabling plain HVM migration in xapi

Hi,

Xapi currently insists that VMs have PV drivers installed before allowing 
migration. The rationale for this is: (i) you get the best performance with 
drivers installed, so you should always install them; and (ii) we focus our 
testing on this case :-)

However, sometimes it is convenient to bypass this particular check and just do 
it anyway. I've written a wiki page containing a simple bash script which 
writes keys to xenstore, fooling xapi into believing that PV tools are 
functioning. Let me know if you find it useful!

http://wiki.xensource.com/xenwiki/XAPI_fake_presence_of_pv_drivers

In the future we probably should add some better mechanism to xapi to 
selectively bypass this check; either something global ("enable experimental 
feature:x") or something per-VM (e.g. an "other-config" key). Any thoughts? My 
own thinking was that if migrate should always work (if a little slowly) then a 
global option is better; but if it's guest dependent (i.e. crashes guest type 
X, works ok on Y) then a per-VM flag is better.

Cheers,
Dave

_______________________________________________
xen-api mailing list
xen-api@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/mailman/listinfo/xen-api


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.