Hi, On Thu, 16 Mar 2023 at 13:44:11 +0100, Paul Gevers wrote: >> As I already mentioned on this or some related bug, I would find it nice >> for #1014110 to be fixed in bookworm (threaded argon2 executable) but I >> do not insist on it. > > cryptsetup can only migrate when argon2 migrates, I see that in the excuse page now but don't understand the reason why, there was no ABI change in the src:argon2 upload and libcryptsetup12 has ‘Depends: libargon2-1 (>= 0~20171227)’. I'm not really familiar with britney's logic but don't understand what it tries to enforce that ordering. > which leaves me two options: […] Apologies for the mess and extra work, and many thanks for trying to find a way forward! > While writing this up and discussing with others, I realized that the error > is coming from one of glibc's binaries. It has been stated that the issue > started with with a change there, but is that change done on purpose, or a > bug? I.e. is one of glibc's binaries missing a dependency? It was intentional, see the article https://developers.redhat.com/articles/2021/12/17/why-glibc-234-removed-libpthread . Unfortunately that change broke initramfs-tools' fix for https://bugs.debian.org/950254 which we (src:cryptsetup maintainers) relied on for cryptsetup-initramfs. Until last week src:argon2 had never been rebuilt with the newer glibc, so it's just unfortunate that we missed that at the time. I wonder if other packages are affected. Another fix would be to change initramfs-tools' inclusion logic for LIBGCC_S_SO. (I don't know if autodetection is still doable, otherwise it could still copy the library unconditionally at the expense of a larger initramfs image.) -- Guilhem.
Attachment:
signature.asc
Description: PGP signature