Hi Andreas, On Samstag, 28. Januar 2012, Andreas Beckmann wrote: > To recall the problem: piuparts test of failing-package fails, but the > problem is actually in a different package: buggy-package (and there is > some (indirect) dependency of failing-package in buggy-package). The > piuparts test of buggy-package by itself succeeds (probably because it > is not getting "used" in a way that fails).
right :( > Looks like we need another category "affected" for logs that need to be > ignored by the humans doing log processing, but that gets rescheduled > regularily to see if buggy-package has been fixed. we could move those to a new directory, "affected", and reschedule those regularily. > Abusing usertags is not an option due to the restricted charset ... > Hmm, we could use base64 encoded info: $ echo 1.2.3+x~_by_buggy/2:3.4-5|base64 -w0 MS4yLjMreH5fYnlfYnVnZ3kvMjozLjQtNQo= Hm, no, I dont think we should do this :-) Either we use another directory to move these logs too, or, we use our own database. Whether sqlite or a real sql or udd... cheers, Holger -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org