pandoc causes FTBFS on armhf and s390x (Kinetic) for other packages

[Christian Ehrhardt ]

As identified by Steve [1] and suggested for a removal by me [2] this is broken in current kinetic-proposed.

The problem are FTBFS on armhf and s390x [3].
And by now this starts to block other packages to build in proposed [4]

Worth an update-excuse bug for sure (=this bug).

But I'm not at all into the matter to suggest any great solution except - for now - to remove the proposed pandoc to get back unblocking others that depend on it.

Steves summary was a non hopeful "no solution seems to be forthcoming"

But it is a good candidate for someone on +1 duty to have a deeper look.
Until then tagging this update-excuse and one might add other PKGs blocked by it.

[1]: https://lists.ubuntu.com/archives/ubuntu-devel/2022-July/042197.html
[2]: https://lists.ubuntu.com/archives/ubuntu-devel/2022-July/042216.html
[3]: https://launchpad.net/ubuntu/+source/pandoc/2.9.2.1-3ubuntu3
[4]: https://launchpad.net/ubuntu/+source/xen/4.16.1-1/+build/24186367

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Luís Infante da Câmara (luis220413)]

The FTBFS on armhf is due to be due to Haskell build dependencies that are present, but are not going to be installed.

[Luís Infante da Câmara (luis220413)]

On s390x, the FTBFS is due to the same issue. However, there are 30 less build dependencies that are not going to be installed, but 2 that are incompatible with the version in Kinetic (including the proposed pocket).

[Luís Infante da Câmara (luis220413)]

Running a test build of haskell-aeson on all architectures except riscv64 in my PPA: https://launchpad.net/~luis220413/+archive/ubuntu/test-builds.

[Luís Infante da Câmara (luis220413)]

The build failed for haskell-aeson, haskell-aeson-pretty, haskell-attoparsec and haskell-base64-bytestring.

I uploaded all the other Haskell dependencies (with the versions in the release pocket) to my PPA to test whether they build: https://launchpad.net/~luis220413/+archive/ubuntu/test-builds

[Luís Infante da Câmara (luis220413)]

I opened bug #1982554 for the Haskell build failures.

[Christian Ehrhardt  (paelzer)]

Updates for comment #8 to #11 are actually around a xen FTFBS that was blocked behind pandoc.
I'll transfer those to a new bug and clean them here.

[Luís Infante da Câmara (luis220413)]

(Please ignore) Christian Ehrhardt, please remove the bug task for xen.

[Luís Infante da Câmara (luis220413)]

This issue was caused by the combination of the version of pandoc in the release pocket (as the version in the proposed pocket failed to build on armhf) with the version of pandoc-data in the release pocket (because it is architecture-independent, and therefore listed in the Packages files for all architectures, and was built on amd64). To avoid this issue, architecture-independent binary packages built from a given source package that also builds architecture-dependent packages should only be published on architecture X if and only if the build of that source package for architecture X succeeds.

Now that the version in kinetic-proposed is in dependency wait for all architectures, I am setting the status of the bug task for xen to Fix Committed.

[Luís Infante da Câmara (luis220413)]

The android-platform-tools source package builds successfully on all architectures.

[Luís Infante da Câmara (luis220413)]

The build of xen 4.16.1-1 in Kinetic was retried on Thursday by Julian Andres Klode and succeeded.

[Christian Ehrhardt  (paelzer)]

\o/ stars aligned and it migrated :-)
Thanks to all that helped!