On 04/29/2014 05:09 PM, Jakub Hrozek wrote:
On Tue, Apr 29, 2014 at 03:53:37PM +0200, Sumit Bose wrote:
> On Tue, Apr 29, 2014 at 03:45:36PM +0200, Pavel Březina wrote:
>> On 04/29/2014 03:24 PM, Jakub Hrozek wrote:
>>> Hi,
>>>
>>> we've been keeping the sssd-1-11 branch fairly conservative recently,
>>> mostly as a precaution to not break downstreams that consume sssd-1-11,
>>> like RHEL.
>>>
>>> But some relatively larger features, such as the InfoPipe responder,
>>> need to be backported to sssd-1-11 (and by extension to RHEL-6) now. One
>>> large difference between master and sssd-1-11 are the DEBUG messages and
>>> I'd like to propose that we backport the new DEBUG messages to sssd-1-11
>>> and mass-convert the callers the same way we did in master.
>>>
>>> I think backporting the DEBUG messages change would be relatively
>>> low-risk, as the messages don't generally affect the behaviour of the
>>> daemon.
>>>
>>> The big downside of course is that we break git-blame and cause a bit of
>>> churn in sssd-1-11. But I think this downside it outweighted by not
>>> having to backport each and every DEBUG message from patches we'll be
>>> pulling in from master to sssd-1-11, which is all the SBUS changes Stef
>>> did recently and also all the work on the IFP responder.
>>>
>>> The next step would be backporting the SBUS changes, but I'd prefer a
>>> separate thread about those.
>>>
>>> Thoughts?
>>
>> +1
>
> +1
>
> bye,
> Sumit
Nikolai, do you have time to do the backport before you leave for
vacation or would you like some help?
I'll send the first version of the backported patches later today.
Sincerely,
Nick