=== emma is now known as Em === nigelb_ is now known as nigelb [12:48] tumbleweed: i got back to my room and gave up on udd. when i rewrote the backport security/sru checker script in launchpadlib, it took me about 30 minutes and runs in 11 :) [12:48] http://pastebin.ubuntu.com/727221/ [13:00] broder: :) === Quintasan_ is now known as Quintasan === bulldog98_ is now known as bulldog98 [13:41] broder: source or it didn't happen! ;) [13:41] kees: http://paste.ubuntu.com/727242/ [13:42] i'm working on reformatting it to feed the output into harvest [13:54] Laney: do you know why DEP-8 doesn't add any fields to the .dsc file? [13:55] dude [13:55] I had never even looked at that page before this session [20:26] angelabad: congrats :P [20:28] tumbleweed, thanks! finally... :-D [20:31] broder: re backport security/sru checker> oh! excellent, can you ping me when you start using it regularly? [20:31] * jdstrand hugs broder :) [20:32] jdstrand: he made that my problem... next on my todo list [20:32] heh [20:33] * jdstrand hugs tumbleweed instead of broder [20:33] broder: ;) [20:33] tumbleweed: thanks! [20:33] lucky you [20:33] tumbleweed: so, would you mind pinging me when the backporters start using it regularly? [20:46] jdstrand: I won't know that, but I can tell you when it's up === yofel_ is now known as yofel [20:59] tumbleweed: that would be great. thanks again [21:11] jdstrand: did you see the preliminary output? it doesn't look like there are very many things that need to be re-backported [21:12] (i've cleaned up some of the false positives since that edition of the code, but i don't think i've found any false negatives) [21:14] and of the packages listed, quassel and clamav have already been handled, so that just leaves libvirt (jaunty->hardy), pidgin (intrepid->hardy), and tomcat6 (intrepid->hardy) [21:14] and i think all of those will require some policy discussion from backports [21:23] broder: any desktop backports for hardy can be ignored [21:24] ok, so we don't need to worry about pidgin, but we do need an answer for tomcat and libvirt [21:26] given that they were backported from releases which are now desupported, i don't know what the right answer is. as a straw man, i propose reviving the security/SRU packages from the desupported releases as being no worse than what's in backports currently [21:27] right, you could also try to backport fixes from lucid if appropriate (or apply ones from hardy as appropriate) [21:27] i'm concerned that creates more friction than is necessary [21:29] i should probably check my backport-helper script and see if it will even agree to backport from a dead release... [21:29] well, if want the backport properly patched, it would be necessary, if you want the path of least resistance, then you could just take what was done before [21:30] backports doesn't provide security support. we shouldn't get into the business of backporting security updates that aren't directly derived from the package that was originally backported [21:30] tracking security/SRUs is already stretching that a good degree from what we've done traditionally [21:31] valid_series = set(s.name for s in lp.distributions['ubuntu'].series if s.active)> bah. too clever for my own good [21:37] right, it depends how much effort you want to put into it :) [21:38] I'd suggest showing them (maybe aggregated at the bottom), but flagging that they need extra patches from elsewhere (possibly surrounding supported releases) [21:38] *might need [21:42] i'll leave them in the report. i'm planning to just spit out a machine-readable report and slurp that into harvest, and we can hide the ones we don't care about solving there === jtechidna is now known as JontheEchidna