Santiago, there was some improvements on the upstream side, especially concerning the failures due to insufficient memory and also some tests are skipped because the time taken is fixed on all arch (and it will be improved soon). Did you see some improvement in your tests ?
F. On Tue, 31 May 2016 16:13:38 +0200, Santiago Vila <sanv...@unex.es> wrote: > These are the last tests I performed: > > Virtual machine with 1GB RAM and 2GB swap, sbuild and eatmydata. > Build seems to hang. > > Virtual machine with 1GB RAM and 2GB swap, sbuild and without eatmydata. > Build seems to hang. > > Virtual machine with 2GB RAM and 2GB swap, sbuild and eatmydata. > Build seems to hang. > > Real machine with 4GB RAM and 2GB swap and eatmydata. > dpkg-buildpackage by hand from unpriviledged user using a chroot. > Build finishes successfully. > > When using sbuild, it's triggered by a cron job. > > All machines, real or virtual, are running stretch. >
