Lucas Nussbaum wrote: > On 27/01/10 at 09:42 +0100, Ludovic Brenta wrote: >> >> Lucas Nussbaum wrote: >>> What about testing with the network down to find which is the >>> affected test, and disable that test? >> >> "the network down" is much too general. Remember: this is about >> distributed applications; they obviously require "the network". >> What that means is they must at least be able to open sockets to >> localhost. Maybe they make some wild assumptions, e.g. that >> 127.0.0.1 is reachable or that there is some sort of DNS available >> that will resolve "localhost". Maybe they need to be able to listen >> on some specific UDP or TCP ports. I don't know and that's why I >> tagged this bug with "help". > > All of this is true on my build nodes. The loopback network is fully > functional. You just can't talk to the Internet.
That statement is of little help. Posting the logs of a rebuild of 2.6.0~20090423-3 or -4 on those "internet-less" machines would be much more helpful. Reto has changed the test harness so it sends more details to the build log for analysis. Maybe one of us could then pinpoint the problem. -- Ludovic Brenta. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
