Re: [Qemu-devel] [RFC][PATCH v6 00/23] virtagent: host/guest RPC communication agent

[Anthony Liguori <aliguori@linux.vnet.ibm.com>]

On 02/18/2011 08:30 AM, Jes Sorensen wrote:
> On 02/18/11 15:07, Anthony Liguori wrote:
>    
>> On 02/18/2011 06:45 AM, Jes Sorensen wrote:
>>      
>>> It may not be so fundamental, but it still makes me wary. XMLRPC
>>> handling is quite high level and introduces the potential of errors that
>>> are outside of our direct control. Personally I don't see the big
>>> benefit of having virtagent terminate in QEMU,
>>>        
>> Live migration.  If it's a separate daemon, then live migration gets fugly.
>>
>> If xmlrpc-c is a PoS, then we ought to look at using something else.
>> But let's understand what's happening first before drawing any conclusions.
>>      
> Urgh, I always do my best to pretend that there is no such thing as live
> migration :) Never seem to work though :(
>
> However if there's an agent connection, it could be arranged in a way
> allowing the host to reconnect to the guest agent. In that way it really
> shouldn't be a big deal as long as our agent commands aren't too complex.
>    

Oh, but they'll be nice and complex :-)  What happens if you do a live 
migration in the middle of doing a live backup?  You'll end up with the 
guest waiting to be told that it's okay to unfreeze itself only to never 
be told.

Terminating in QEMU means we can do intelligent things like delay live 
migration convergence, save state, etc.

Regards,

Anthony Liguori

> xmlrpc-c is probably fine, but it introduces a layer of complexity which
> always makes me worried.
>
> Cheers,
> Jes
>