# Pipeline issues with latest changes

**URL:** <https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097>\
**Category:** Development\
**Created:** [October 9, 2026, 10:49am UTC](https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097 "2026-10-09T10:49:04Z")\
**Posts on this page:** 9\
**Page:** 1

<div class="post-metadata">

**Author:** ![dicaveman](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/dicaveman/32/9710_2.png) [@dicaveman](https://gramps.discourse.group/u/dicaveman)\
**Post date:** [October 9, 2026, 10:49am UTC](https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097/1 "2026-10-09T10:49:04Z")

</div>

I have been working on a pipeline change in my fork and a few days ago everything went South. It seems like there were some MSys updates which changed some packages, obsoleted others, etc., and now my Windows AIO pipelines are failing. I see that the main gramps AIO pipeline hasn’t run since September - could someone please take a look and see if everything is OK?

Thanks,

-Kevin

---

<div class="post-metadata">

**Author:** ![Nick-Hall](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/nick-hall/32/95_2.png) [@Nick-Hall](https://gramps.discourse.group/u/Nick-Hall)\
**Post date:** [October 9, 2026, 10:54am UTC](https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097/2 "2026-10-09T10:54:49Z")

</div>

I’ve just started a Windows AIO build.

---

<div class="post-metadata">

**Author:** ![dicaveman](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/dicaveman/32/9710_2.png) [@dicaveman](https://gramps.discourse.group/u/dicaveman)\
**Post date:** [October 10, 2026, 2:07pm UTC](https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097/3 "2026-10-10T14:07:03Z")

</div>

Thanks. It fails, probably for the same reason mine does.

---

<div class="post-metadata">

**Author:** ![Nick-Hall](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/nick-hall/32/95_2.png) [@Nick-Hall](https://gramps.discourse.group/u/Nick-Hall)\
**Post date:** [October 10, 2026, 2:21pm UTC](https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097/4 "2026-10-10T14:21:24Z")

</div>

OK. I’ll remove _mingw-w64-ucrt-x86\_64-python-nose_ and try again.

Update: This is still failing. One of our Windows maintainers will need to take a look at it.

---

<div class="post-metadata">

**Author:** ![emyoulation](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/emyoulation/32/67_2.png) [@emyoulation](https://gramps.discourse.group/u/emyoulation)\
**Post date:** [October 10, 2026, 3:25pm UTC](https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097/5 "2026-10-10T15:25:11Z")

</div>

Gemini claims that “_Nose is currently in maintenance mode; newer projects typically use pytest or `nose2`._ [[1](https://nose.readthedocs.io/), [2](https://docs.nose2.io/), [3](https://pypi.org/project/nose/)]”

Which is strange since this Nose inclusion must be new. (Nose is an extension of Python’s built-in `unittest` module that automatically discovers and runs tests, supporting plugins for coverage, output capture, and profiling.) Because the massive expansion in testing coverage is new.

---

<div class="post-metadata">

**Author:** ![Nick-Hall](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/nick-hall/32/95_2.png) [@Nick-Hall](https://gramps.discourse.group/u/Nick-Hall)\
**Post date:** [October 10, 2026, 3:28pm UTC](https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097/6 "2026-10-10T15:28:48Z")

</div>

We have never used _nose_. I’m not sure why it was in the build.

---

<div class="post-metadata">

**Author:** ![emyoulation](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/emyoulation/32/67_2.png) [@emyoulation](https://gramps.discourse.group/u/emyoulation)\
**Post date:** [October 10, 2026, 3:38pm UTC](https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097/7 "2026-10-10T15:38:40Z")

</div>

Then it seems like the bigger questions are: who changed the build; and, what other changes were made?

---

<div class="post-metadata">

**Author:** ![dicaveman](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/dicaveman/32/9710_2.png) [@dicaveman](https://gramps.discourse.group/u/dicaveman)\
**Post date:** [October 10, 2026, 3:38pm UTC](https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097/8 "2026-10-10T15:38:57Z")

</div>

Not sure - this prompted me to update my MSys locally and it took about 20 minutes, so it seems like it was a pretty substantial change.

---

<div class="post-metadata">

**Author:** ![Nick-Hall](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/nick-hall/32/95_2.png) [@Nick-Hall](https://gramps.discourse.group/u/Nick-Hall)\
**Post date:** [October 10, 2026, 4:04pm UTC](https://gramps.discourse.group/t/pipeline-issues-with-latest-changes/10097/9 "2026-10-10T16:04:01Z")

</div>

It looks like the DLL path in the NSIS Unicode plugin changed. I’ll attempt to fix it.

Update: Fixed in pull request [#2544](https://github.com/gramps-project/gramps/pull/2544).
