Re: libxl: error handling before xenstored runs

[Vincent Hanquez <vincent.hanquez@eu.citrix.com>]

On 11/02/11 09:49, Tim Deegan wrote:
> At 21:55 +0000 on 10 Feb (1297374910), Vincent Hanquez wrote:
>> On 10/02/11 11:24, Ian Campbell wrote:
>>> Right but this approach doesn't work with xenstored in a stubdomain.
>>
>> yeah I know. xenstored in a stubdom is just an experiment, when it
>> become a serious feature, this argument would hold. however it's not
>> going to be use in 4.1, and in any production settings.
>
> You seem to be arguing that we shouldn't fix a bug in the kernel.  I
> don't understand that.  How is it going to become a "serious feature" if
> we don't fix the bugs that affect it?

If you want to fix the behaviour which is present since xen 3.0 (you can 
understand why i'm not holding my breath anymore), all the best.

What I'm arguing is that behaviour should not exists in the next stable 
version of xen, specially for a feature that is not serious yet (at 
least upstream).

> In any case, I can think of three projects off the top of my head that
> are using stub domains aggressively, one of which I know is using
> xenstore stub domains in particular.  I'm sure there are others.

Good for them. I'm sure they can carry a 2 liner patch to connect to the 
ring instead of the unix socket, until they actually submit upstream 
their stuff that make xenstored stubdomain great.

-- 
Vincent