Re: KDE Gear projects with failing CI (master + stable) (11 February 2026)

On 2/13/26 10:15 PM, Nicolas Fella wrote:
> On 2/13/26 9:53 PM, Nicolas Fella wrote:
>> On 2/13/26 9:24 PM, Ben Cooksley wrote:
>>> On Thu, Feb 12, 2026 at 1:09 PM Albert Astals Cid <[email protected]> wrote:
>>>
>>>     Please work on fixing them, otherwise i will remove the failing
>>>     CI jobs on
>>>     their 4th failing week, it is very important that CI is passing
>>>     for multiple
>>>     reasons.
>>>
>>>     Good news: 2 repo fixed
>>>
>>>
>>>     Bad news: 5 repo started failing, 1 keeps failing
>>>
>>>
>>>     kontact - LAST WEEK BEFORE REMOVAL
>>>      * https://invent.kde.org/pim/kontact/-/pipelines/1162355
>>>       * Craft Windows job running out of memory
>>>
>>>
>>> PIM folks, could we resolve this for now by dropping KItinerary 
>>> support in Craft Windows builds, or at the very least only target a 
>>> release build of KItinerary?
>>> In a cache build we are running in a High Memory VM which will have 
>>> the resource to get through building this so the fix to this is stop 
>>> rebuilding the whole PIM stack....
>>>
>>>
>>>
>>>     kdenlive - NEW
>>>      * https://invent.kde.org/multimedia/kdenlive/-/pipelines/1162146
>>>       * macos fails to compile
>>>
>>>
>>>     kosmindoormap - NEW
>>>      *
>>>     https://invent.kde.org/libraries/kosmindoormap/-/pipelines/1161455
>>>      *
>>>     https://invent.kde.org/libraries/kosmindoormap/-/pipelines/1161715
>>>     (stable)
>>>       * Android build seems to be using the wrong binary format?
>>>     cross-compilation issue?
>>>
>>>
>>>     neochat - NEW
>>>      * https://invent.kde.org/network/neochat/-/pipelines/1162323
>>>     (stable)
>>>       * flatpak build fails
>>>
>>>
>>> Not sure why a stable branch is building against master of 
>>> libquotient, but seems like 
>>> https://invent.kde.org/network/neochat/-/commit/8ca1b8b1d387838fde1c1c448e5d2fe3a077a6e2 
>>> needs to be cherry picked if that is intended?
>>>
>>>
>>>
>>>     kgpg - NEW
>>>      * https://invent.kde.org/utilities/kgpg/-/pipelines/1161149
>>>       * Timeout in kgpg-disable and kgpg-export
>>>
>>>
>>>     kmbox - NEW
>>>      * https://invent.kde.org/pim/kmbox/-/pipelines/1162163
>>>       * Timeout in mbox-mboxbenchmark
>>>
>>>
>>> This test seems to be sitting on the edge of whether it will pass or 
>>> not - Qt 6.11 only just passed by running in 57 seconds vs. the 
>>> timeout limit of 60 seconds.
>>> What is it trying to achieve and why so many iterations?
>> I don't follow what's going on there.
>>
>> Locally that benchmark finishes in ~1 second here, so no problem 
>> whatsoever.
>>
>> On CI the first two sub-benchmarks seem completely fine, but 
>> voidTestMD5Performance seems to hand/time out. It's not clear to me 
>> how many iterations it does.
>>
>> It looks like QBENCHMARK automatically adjusts the number of 
>> iterations until the results are "good" (whatever that exactly 
>> means). Possibly some CI conditions trigger that to go haywire.
>>
> That said, all this seems to do is benchmark the performance of 
> QCryptographicHash, which is nonsensical here.
>
> https://invent.kde.org/pim/kmbox/-/merge_requests/18
>
Doesn't seem to actually help :/