[lfs-dev] LFS SVN urrent build results

Matt Burgess matthew at linuxfromscratch.org
Mon Dec 24 15:03:29 PST 2012


On Mon, 2012-12-24 at 23:24 +0100, g.esp at free.fr wrote:

> the remaining process are
> 23812 pts/0    SN     0:00 /bin/sh ./test_one ../../tests/f_mmp
> 23813 pts/0    SN    52:27 ../debugfs/debugfs -w f_mmp.tmp
> 23814 pts/0    SN    13:49 cat /dev/zero
> 
> So the issue is in tests/f_mmp.tmp
> 
> [chroot-i486] root:/$ ls -hl /usr/src/e2fs*/build/tests/f_mmp*.log
> -rw-r--r-- 1 root root 1.5G Dec 24 21:03 /usr/src/e2fsprogs-1.42.6/build/tests/f_mmp.log
> -rw-r--r-- 1 root root  322 Dec 24 19:03 /usr/src/e2fsprogs-1.42.6/build/tests/f_mmp_garbage.1.log
> -rw-r--r-- 1 root root  277 Dec 24 19:03 /usr/src/e2fsprogs-1.42.6/build/tests/f_mmp_garbage.2.log
> -rw-r--r-- 1 root root  122 Dec 24 19:03 /usr/src/e2fsprogs-1.42.6/build/tests/f_mmp_garbage.log
> 
> f_mmp.log is filled in a continuous loop until the disk is full.

Thanks for the analysis. I too spotted this a while ago, but wondered
whether it was just my environment.  It also explains (I think) Bruce's
gzip failure; I've had the same test fail but only when the e2fsprogs
caused particularly quick exhaustion of disk space.

I'm happy enough to revert back to E2fsprogs-1.42.5, but it will
probably have to wait until around the 28th-29th.

Thanks,

Matt.




More information about the lfs-dev mailing list