Package: release.debian.org Severity: normal User: release.debian....@packages.debian.org Usertags: unblock
Hi. On behalf of the Debian Edu team, I would like to ask if the current set of Debian Edu packages in unstable can be made to propagate quickly into testing? The source packages in question is debian-edu - our metapackages used when installing Debian Edu to install a bit more than 2500 packages depending on profile selected during installation. debian-edu-install - our installation preseeding and d-i hooks/overrides. debian-edu-config - our configuration and modifications for the installed systems. We have tried to work out of unstable, but bugs in packages in unstable keep blocking the installation of our profiles, and we are thus forced to test using testing/jessie instead where we get a more consistent set of packages. The current propagation delay make it hard to get feedback from testers on our current set of packages and configuration, as the testers test packages in testing and we can thus only get feedback on the setup already in testing. The current 10 day delay slow down the feedback loop so much that it make it unlikely we will be able to find and fix many issues before the freeze, which is a goal for us. The Debian Edu setup is a bit special from other packages in the archive, as we depend on the around 2500 packages to be installable together and at the same time, and without any broken recommended packages. Our main server + thin client server profile install that many packages during the d-i run. When I asked on #debian-release about this, jmw said he might consider to reduce the migration time for these packages from 10 to 5 days (but not 2 days). Such step would improve the situation, as it would half the duration of the feedback loop before the freeze. :) I notice someone have been wondering why I have kept using urgency=high also after the migration delay was raised to 10 days. This is to document which urgency I believe it is to get the packages into testing. It is mostly for documentation to the rest of the people involved, as I know the value is ignored by the archive processing scripts. :) -- Happy hacking Petter Reinholdtsen -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org