https://koji.fedoraproject.org/koji/taskinfo?taskID=118234666
It failed right at the end with this mysterious error:
GenericError: srpm mismatch for /mnt/koji/work/tasks/4737/118234737/ocaml-ocamldoc-debuginfo-5.2.0-1.fc41.x86_64.rpm: (none) (expected ocaml-5.2.0-1.fc41.src.rpm)
I have just now kicked off another build in case this was a one-off, but anyone got ideas about this?
Rich.
On 29. 05. 24 13:38, Richard W.M. Jones wrote:
https://koji.fedoraproject.org/koji/taskinfo?taskID=118234666
It failed right at the end with this mysterious error:
GenericError: srpm mismatch for /mnt/koji/work/tasks/4737/118234737/ocaml-ocamldoc-debuginfo-5.2.0-1.fc41.x86_64.rpm: (none) (expected ocaml-5.2.0-1.fc41.src.rpm)
I have just now kicked off another build in case this was a one-off, but anyone got ideas about this?
RPM 4.20 regression. Fix is being cooked at https://src.fedoraproject.org/rpms/rpm/c/9042b409567e8479d6ddafc26e33badbe3b...
On Wed, May 29, 2024 at 01:46:43PM +0200, Miro Hrončok wrote:
On 29. 05. 24 13:38, Richard W.M. Jones wrote:
https://koji.fedoraproject.org/koji/taskinfo?taskID=118234666
It failed right at the end with this mysterious error:
GenericError: srpm mismatch for /mnt/koji/work/tasks/4737/118234737/ocaml-ocamldoc-debuginfo-5.2.0-1.fc41.x86_64.rpm: (none) (expected ocaml-5.2.0-1.fc41.src.rpm)
I have just now kicked off another build in case this was a one-off, but anyone got ideas about this?
RPM 4.20 regression. Fix is being cooked at https://src.fedoraproject.org/rpms/rpm/c/9042b409567e8479d6ddafc26e33badbe3b...
Thanks! I would never have worked that one out :-)
Rich.
On Wed, May 29, 2024 at 01:46:43PM +0200, Miro Hrončok wrote:
On 29. 05. 24 13:38, Richard W.M. Jones wrote:
https://koji.fedoraproject.org/koji/taskinfo?taskID=118234666
It failed right at the end with this mysterious error:
GenericError: srpm mismatch for /mnt/koji/work/tasks/4737/118234737/ocaml-ocamldoc-debuginfo-5.2.0-1.fc41.x86_64.rpm: (none) (expected ocaml-5.2.0-1.fc41.src.rpm)
I have just now kicked off another build in case this was a one-off, but anyone got ideas about this?
RPM 4.20 regression. Fix is being cooked at https://src.fedoraproject.org/rpms/rpm/c/9042b409567e8479d6ddafc26e33badbe3b...
I see the rpm package build containing this one failed, with the same failure in debuginfo generation :-( (https://koji.fedoraproject.org/koji/taskinfo?taskID=118237702)
Is the bad rpm package going to be untagged?
I was planning to do the OCaml 5.2 rebuild today, but if RPM is full of regressions maybe that's not such a good idea. What do you think?
Rich.
On 29. 05. 24 13:59, Richard W.M. Jones wrote:
On Wed, May 29, 2024 at 01:46:43PM +0200, Miro Hrončok wrote:
On 29. 05. 24 13:38, Richard W.M. Jones wrote:
https://koji.fedoraproject.org/koji/taskinfo?taskID=118234666
It failed right at the end with this mysterious error:
GenericError: srpm mismatch for /mnt/koji/work/tasks/4737/118234737/ocaml-ocamldoc-debuginfo-5.2.0-1.fc41.x86_64.rpm: (none) (expected ocaml-5.2.0-1.fc41.src.rpm)
I have just now kicked off another build in case this was a one-off, but anyone got ideas about this?
RPM 4.20 regression. Fix is being cooked at https://src.fedoraproject.org/rpms/rpm/c/9042b409567e8479d6ddafc26e33badbe3b...
I see the rpm package build containing this one failed, with the same failure in debuginfo generation :-( (https://koji.fedoraproject.org/koji/taskinfo?taskID=118237702)
Is the bad rpm package going to be untagged?
We'll use side tags to build it with old rpm.
I was planning to do the OCaml 5.2 rebuild today, but if RPM is full of regressions maybe that's not such a good idea. What do you think?
I suggest waiting a day.
On 5/29/24 14:59, Richard W.M. Jones wrote:
On Wed, May 29, 2024 at 01:46:43PM +0200, Miro Hrončok wrote:
On 29. 05. 24 13:38, Richard W.M. Jones wrote:
https://koji.fedoraproject.org/koji/taskinfo?taskID=118234666
It failed right at the end with this mysterious error:
GenericError: srpm mismatch for /mnt/koji/work/tasks/4737/118234737/ocaml-ocamldoc-debuginfo-5.2.0-1.fc41.x86_64.rpm: (none) (expected ocaml-5.2.0-1.fc41.src.rpm)
I have just now kicked off another build in case this was a one-off, but anyone got ideas about this?
RPM 4.20 regression. Fix is being cooked at https://src.fedoraproject.org/rpms/rpm/c/9042b409567e8479d6ddafc26e33badbe3b...
I see the rpm package build containing this one failed, with the same failure in debuginfo generation :-( (https://koji.fedoraproject.org/koji/taskinfo?taskID=118237702)
Is the bad rpm package going to be untagged?
I was planning to do the OCaml 5.2 rebuild today, but if RPM is full of regressions maybe that's not such a good idea. What do you think?
I should hope not "full of" regression :)
This particular issue is a highly specific one: sub-package debuginfo packages are missing their RPMTAG_SOURCERPM tag due to internal hysterics. Who knew Koji checks for such a thing? Well, good thing it did and the regression was caught, it would've probably taken a long, long time for anybody to notice that otherwise.
The fix is simple code-wise, but because rpm itself has sub-packages with debuginfo (eg rpm-plugin-dbus-announce-debuginfo) we need to rebuild it with an older rpm to get the fix through koji.
- Panu -
- Panu -
On Wed, May 29, 2024 at 12:38:31PM +0100, Richard W.M. Jones wrote:
It failed right at the end with this mysterious error:
GenericError: srpm mismatch for /mnt/koji/work/tasks/4737/118234737/ocaml-ocamldoc-debuginfo-5.2.0-1.fc41.x86_64.rpm: (none) (expected ocaml-5.2.0-1.fc41.src.rpm)
OK, this should be fixed now:
https://bodhi.fedoraproject.org/updates/FEDORA-2024-0cdd01deef
We've been anxiously counting every minute here while the Fedora CI was taking its time to progress through the large test-suite. Sorry for the inconvenience again!
Cheers,
On Wed, May 29, 2024 at 04:35:57PM +0200, Michal Domonkos wrote:
On Wed, May 29, 2024 at 12:38:31PM +0100, Richard W.M. Jones wrote:
It failed right at the end with this mysterious error:
GenericError: srpm mismatch for /mnt/koji/work/tasks/4737/118234737/ocaml-ocamldoc-debuginfo-5.2.0-1.fc41.x86_64.rpm: (none) (expected ocaml-5.2.0-1.fc41.src.rpm)
OK, this should be fixed now:
https://bodhi.fedoraproject.org/updates/FEDORA-2024-0cdd01deef
Thanks!
Do you plan to submit rebuilds for the failed packages are should packagers do that individually?
Zbyszek
On Wed, May 29, 2024 at 02:38:59PM +0000, Zbigniew Jędrzejewski-Szmek wrote:
Do you plan to submit rebuilds for the failed packages are should packagers do that individually?
Yes, I just went through the failed builds from today and restarted those that were "almost" successful (i.e. were actually built) but just failed in Koji with the known "GenericError: srpm mismatch ..." error, namely:
composefs https://koji.fedoraproject.org/koji/taskinfo?taskID=118243204 boost https://koji.fedoraproject.org/koji/taskinfo?taskID=118243374 xpdf https://koji.fedoraproject.org/koji/taskinfo?taskID=118243422 initscripts https://koji.fedoraproject.org/koji/taskinfo?taskID=118243434 glibc https://koji.fedoraproject.org/koji/taskinfo?taskID=118243459 scipy https://koji.fedoraproject.org/koji/taskinfo?taskID=118243758
I might have missed some so apologies to those folks, please resubmit them at your discretion.
Thanks,
On Wed, May 29, 2024 at 05:27:18PM +0200, Michal Domonkos wrote:
I might have missed some so apologies to those folks, please resubmit them at your discretion.
Just a note:
For one, I skipped over those build that were done against a side tag, to avoid messing up somebody's work-in-progress.
-
Michal Domonkos -
Miro Hrončok -
Panu Matilainen -
Richard W.M. Jones -
Zbigniew Jędrzejewski-Szmek