I think I've seen this before, and indeed: [bug 1789539]. Can something be done about this apparently fragile lack of explicit lockstep at times (for when one updates packages selectively, which for something as exposed as web browser could be pretty justified)? Where casual SONAME/symbol versioning fails us? Is it because of the tight firefox-nss development coupling that some best practices on the ABI/API compat front are neglected? Coincidentally, I've just noticed that automatic dependencies generation seems to receive a new wave of attention if that might help in this use case as well: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/EZFX5ARQWOXBUXID4HH74ETD2QBF2DPB/ * * * In particular: * firefox-110.0-2.fc39.x86_64 + nss-3.87.0-1.fc38.x86_64 -> OK * firefox-110.0-3.fc39.x86_64 + nss-3.87.0-1.fc38.x86_64 -> no page loading * firefox-110.0-2.fc39.x86_64 + nss-3.88.1-2.fc39.x86_64 -> OK
* firefox-110.0-3.fc39.x86_64 + nss-3.88.1-2.fc39.x86_64 -> OK
This bug appears to have been reported against 'rawhide' during the Fedora Linux 39 development cycle. Changing version to 39.