Karel Zak wrote:
On Tue, Feb 02, 2016 at 09:14:48PM +0100, Stanislav Brabec wrote:
Stanislav Brabec wrote:
+mkdir -p s3/bind-mnt
So, this is enough to fix the testcase, right? I'd like to merge it.
Karel
Well, not.
There are missing umount commands in the middle of the test, which
trigger following problem:
truncate -s10M source.img
mkdir -p /source/subdir
mkdir -p /dest
/sbin/mkfs.ext4 source.img
losetup /dev/loop0 $PWD/source.img
mount -text4 /dev/loop0 /source
mkdir -p /source/subdir
mount -o bind /source /dest
mount -o bind /source/subdir /dest
Mounting a bind mount over a root of existing bind mount causes mounting
over the bind mount source. I expected mounting over the target
directory I specified.
I am not sure, whether it is correct behavior, but for sure it is
surprising.
Additionally, I am working on a reproducer of an invalid "mount -a"
behavior triggered by kernel bug reporting false subvol. Now I have only
a reproducer outside testsuite:
#/btrfs_test.img /btrfs_mnt_root auto loop,subvol=/ 0 0
#/btrfs_test.img /btrfs_mnt auto loop,subvolid=257 0 0
#/btrfs_mnt_root/d0/dd0/ddd0/s1/bind-mnt /btrfs_bind auto bind 0 0
Round 1:
Uncomment 1st and 3rd line and call mount -a.
Round 2:
Uncomment 2nd line and call mount -a.
Result:
Only two mounts exist.
I am working on a subtest that will trigger it inside the test suite.
I already have a possible fix for it, but I want to confirm that it is
correct before sending it upstream.
--
Best Regards / S pozdravem,
Stanislav Brabec
software developer
---------------------------------------------------------------------
SUSE LINUX, s. r. o. e-mail: sbrabec@xxxxxxxx
Lihovarská 1060/12 tel: +49 911 7405384547
190 00 Praha 9 fax: +420 284 084 001
Czech Republic http://www.suse.cz/
PGP: 830B 40D5 9E05 35D8 5E27 6FA3 717C 209F A04F CD76
--
To unsubscribe from this list: send the line "unsubscribe util-linux" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
