Hello Aaron M. Ucko. On Mon, Aug 01, 2016 at 11:17:20AM -0400, Aaron M. Ucko wrote: > > Closing this bug report now that we have a succesful build. > > Thanks, and sorry for the noise. I'm not a buildd maintainer or > official porter, so I can't directly schedule retries either.
Me neither, but we need to ask the people who can... I guess in this case a mail to arm buildd maintainers would have been suitable. > (I just look at build logs and report bugs, focusing primarily on > sources corresponding to new binary packages that have shown up for > amd64 but not i386 or vice versa.) Thanks for doing that, I bet it's very useful to get a bug report in most cases where you file. (In theory a good maintainer ofcourse follows up to check that his upload has made it into testing as planned, but I certainly know there are cases where I've missed this. Specially for the lesser loved packages.) > That said, in cases where the problem looks like it could have simply > been a glitch, I suppose I could retry on a porterbox and indicate > whether I could reproduce the error. I don't think you really should be wasting your time on a porterbox either. Atleast not for architectures which has enough buildd resources (and hopefully all/most have that). A simple automatic rebuild to check if the problem is reproducible would be best before investing manual developer time. Even better would be if buildds where improved to automatically reattempt when they're idle and for example failed a build for the first time in several uploads and all other arches succeded or something like that. Would save us all valuable time. :) Have a nice day! Regards, Andreas Henriksson
