Re: [xen-unstable test] 128240: regressions - FAIL

[Juergen Gross <jgross@suse.com>]

On 02/10/2018 10:29, Jan Beulich wrote:
>>>> On 01.10.18 at 18:07, <jgross@suse.com> wrote:
>> On 01/10/2018 17:48, George Dunlap wrote:
>>> On 10/01/2018 04:40 PM, Andrew Cooper wrote:
>>>> If there isn't an obvious fix, then the switch of default scheduler
>>>> needs reverting until there is a fix present.  This is currently
>>>> blocking master.
>>>
>>> Agreed.  I'd argue for ignoring failures to set scheduler parameters on
>>> migrate, on the grounds that this will be less risk to the project as a
>>> whole than reverting credit2 again.  But either way we should do
>>> something quickly.
>>
>> We should ignore a mismatch of the scheduler. Failures when setting
>> parameters for a matching scheduler should not be ignored IMO.
> 
> Well, depends on whether the scheduler was explicitly chosen.
> I don't think migration should succeed when e.g. RTDS was used
> and isn't available on the destination host.

The only way I could think of to tell an explicit vs. an implicit
scheduler selection would be to specify a cpupool in the domain's
config file.

So what about the following:

A mismatch of the scheduler should be ignored on the receiving side
if no cpupool was specified explicitly for the domain.

I have checked that config.c_info.pool_name in JSON data is available
only if the cpupool is specified in the domain config.


Juergen


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel