On Thu, Sep 21, 2017 at 02:20:46PM +0300, Timo Aaltonen wrote:
On 21.09.2017 14:16, Jakub Hrozek wrote:
> On Thu, Sep 21, 2017 at 01:04:04PM +0200, Lukas Slebodnik wrote:
>> On (19/09/17 20:50), Jakub Hrozek wrote:
>>> Hi,
>>>
>>> Timo mentioned last week on IRC that he would appreciate if we released
>>> 1.13.5.
>>>
>>> Does anyone have some patches to merge in sssd-1-13 or can we release
>>> the tarball?
>>>
>>> I know there are some pending PRs with backports and some patches for
RHEL-6
>>> bugs were proposed in
bugzilla.redhat.com, but there are already quite a
>>> few patches on top of 1.13.4 so I would prefer to release the tarball now
>>> and then, around the time of RHEL-6.10 development freeze, release 1.13.6.
>>>
>> There are patches for UPN on review.
>>
>> But sssd-1-13 master have more failures in ad_forest test then default
>> sssd in el6. I noticed even crashes
>
> Can you report those issues or send me links to test runs with those
> failures? We can't fix bugs that are not reported..
>
>>
>> So we should not release 1.13.5 which is worse then previous release
>
> Yes, regressions are blocking the release.
>
> Timo, how much time do you have to upload the new tarball? Which patches
> were you most interested in?
It's this
https://pagure.io/SSSD/sssd/issue/2751
Ah, in that case, I would suggest to cherry-pick the patches into Debian
in the meantime, sorry..
Note that the sssd patch itself was just using additional capability
enabled by several ding-libs patches, so updating sssd on its own won't
fix the bug.
I'm in no rush, take your time fixing the issues or revert the offending
commits from the branch.
OK, I still think it would be a good idea to release the tarball, but
not with regressions..