Here is my take on moving to modules (speaking from dogtag side of things): Pros: * We would be version-focused instead of fedora-release focused [1]* One source for multiple fedora-release [2]* Dogtag's future would be as described in [3] Cons: * Dogtag's CI runs on Travis. Moving into modules would break the CI workflow leading to a major restructure of upstream CI* In addition to official releases via koji/bodhi, we do maintain an official COPR repo (@pki/10.x). This would need to be reconfigured* Dogtag's nightlies run on COPR. This would again need to be reconfigured. Though in future modules MIGHT help us, the effort required to move into modules (and to make it work) is releatively VERY HIGH. [1] https://docs.fedoraproject.org/en-US/modularity/architecture/building/#_modular_package_builds[2 ] https://docs.fedoraproject.org/en-US/modularity/architecture/building/#_building_one_source_for_multiple_releases[3 ] https://i.ibb.co/23nbNWK/module-branching.jpg > Of course, we can move FreeIPA to a module relatively fast in > Rawhide. > But first we need to understand whether the same is possible for > Dogtag. > > -- > / Alexander Bokovoy > Sr. Principal Software Engineer > Security / Identity Management Engineering > Red Hat Limited, Finland > > _______________________________________________ > Pki-devel mailing list > [email protected] > https://www.redhat.com/mailman/listinfo/pki-devel
_______________________________________________ Pki-devel mailing list [email protected] https://www.redhat.com/mailman/listinfo/pki-devel
