Hi! > On 15-05-2023 07:55, Otto Kekäläinen wrote: > > This pre-unblock request is to get a decision from the Bookworm > > release team if you prefer to accept this 10.11.3 into Bookworm, or if > > you wish it to be postponed to a stable update in Bookworm some time > > later in fall 2023. > > I just discussed with the security team (live, here in Hamburg), and > mariadb is on their list for taking upstream releases, so I'm of the > opinion that the release team should follow that. Having said that,
Thanks for the guidance. The new upstream import pending at https://salsa.debian.org/mariadb-team/mariadb-server/-/merge_requests/46 has already been polished by me and reviewed by several people, including MariaDB.org staff and the MySQL maintainer in Ubuntu. The CI detected a couple days ago a regression in Piuparts, potentially due to recent adduser 1.133 upload, which I still need to debug and decide what to do on. Hopefully tomorrow I can upload this to Debian experimental for an extra round of QA, and then to Debian unstable on Friday if nothing new shows up. I know we have no time left, so I am trying to move as fast as it is safe and reasonable to do. > we're now so close to the release date that I want you to have *no* > changes in the debian sub-directory that are not absolutely required to > support the new upstream version. It's just too late. Based on feedback in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1033811#34 I opened https://salsa.debian.org/mariadb-team/mariadb-server/-/merge_requests/45, which is now pending feedback on what to revert. I understand that you highly value stability and absolutely do not want anything to regress this late in the release cycle. I have the same concern, and thus I have only done changes that aim to fix bugs or docs/translations (in February/March). I would argue that the bugfixes have been well tested by now and by reverting those bugfixes the chance of a regression actually might go up. I will try to collect more feedback on what commits to revert from a larger pool of peope who hopefully have bandwidth to read the commits and assess them more in !45.
