On Thu, Feb 22, 2007 at 10:52:38PM +0000, Ben Hutchings wrote: > "peter green" <[EMAIL PROTECTED]> wrote: > > would it be possible to reduce the test to only cause a warning until > > it is worked out what about the build environment is causing the false > > positive?
> I think that may be a reasonable thing to do here. This test is trying > to check that "pwd" can read very long paths, specifically 256 levels of > directories below the test's current directory. The error message > indicates that it is failing after going 265 levels up (i.e. outside of > the build directory, closer to the root). > The error seems to suggest that lstat() failed, though none of the > documented reasons for failure seem plausible. Another possibility is > that the filesystem on the buildd is corrupt and one ".." link does not > point to the correct directory. (Lack of read permission on a directory > is another possibility I thought of, but that results in a different > error message.) Bastian, do you have any explanation for this? coreutils now builds successfully in the test environment Manoj set up using bind mounts, but still fails on the s390 buildd. What is special/different still about the bind mount setup on the s390 buildd? The autobuildability requirement should certainly apply to core packages like coreutils, but at this point it seems as likely to be the buildd environment that's broken as the testsuite. -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. [EMAIL PROTECTED] http://www.debian.org/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
