Bug 2371770
| Summary: | F43FailsToInstall: pcs, pcs-web-ui | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Fedora Fails To Install <fti-bugs> |
| Component: | pcs | Assignee: | Michal Pospisil <mpospisi> |
| Status: | CLOSED ERRATA | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | rawhide | CC: | anprice, cfeist, idevat, mlisik, mpospisi, omular, tojeline, yselkowi |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | pcs-0.12.1-1.fc43 | Doc Type: | --- |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2025-07-22 21:01:13 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
| Bug Depends On: | |||
| Bug Blocks: | 2339435, 2322407 | ||
|
Description
Fedora Fails To Install
2025-06-11 08:05:42 UTC
I'm not quite sure why this is not a FTBFS bug since pcs-0.12.0-2.fc43 failed to build. pcs-0.12.0-1.fc42.noarch is not a rebuilt package and that is the reason why it fails to install. The "python(abi) = 3.13" requirement is generated by RPM during build time depending on the Python interpreter present on the build machine. This is not a real error. Here are the failed builds: https://koji.fedoraproject.org/koji/taskinfo?taskID=133516909 https://koji.fedoraproject.org/koji/taskinfo?taskID=133622677 I had a quick look and didn't find much because the build.log has over 33GB! Something seems to be really wrong. I'll try to resolve the issue next week. Upstream 0.12.1 includes fixes for Python 3.14. Yes, mlisik was faster than me and discovered that the issue was caused by a switch to the forkserver method in multiprocessing.Pool [1]. The issue was that the executors in the test suite no longer inherited context changed in other parts of pcs done outside of the test code (like modifying of sys.path). We decided to explicitly use the fork method because properly fixing this would make running tests harder. We already encountered the issue with deadlocks that is being solved in Python 3.14 - this was in pcsd, and we already use forkserver to deal with it. Since I'm done with the upstream releases now, I'll rebase pcs in Fedora very soon and that will fix this issue. [1] https://docs.python.org/3.14/whatsnew/3.14.html#incompatible-changes Any chance the version bump can be done before the F43 mass rebuild starts on Wednesday? Otherwise we're going to end up with an attempted rebuild of 0.12.0 which ends up in an (endless?) loop of retries. Yes, already on it: https://src.fedoraproject.org/rpms/pcs/pull-request/56 FEDORA-2025-6e6f4af8f4 (pcs-0.12.1-1.fc43) has been submitted as an update to Fedora 43. https://bodhi.fedoraproject.org/updates/FEDORA-2025-6e6f4af8f4 FEDORA-2025-6e6f4af8f4 (pcs-0.12.1-1.fc43) has been pushed to the Fedora 43 stable repository. If problem still persists, please make note of it in this bug report. |