Closed tdenof closed 2 years ago
The failing tests should have failed when the fallback on mountpoint
(which uses /proc/self/mountinfo
) was introduced, but due to the -n
bug, they were still passing. Fixing
All tests concerning mountpoints for which retvalshouldbe
was set to 1 are now switched to 0 as now we're correctly going through the code calling mountpoint
when fstab is empty or doesn't exists. Also removed the dismiss_count_for_test
for tests that should pass now.
Other tests were passing, but generating stderr as mountpoint
couldn't find some mounpoints (which was the expected behavior), so calling it now with quiet parameter which is enough for the needed test without generating stderr.
The test was checking if fstab is not empty, so we were always falling back on /proc/mounts Before
After