On Fri, Feb 3, 2017 at 11:35 AM, Eryu Guan <eguan@xxxxxxxxxx> wrote: > On Fri, Jan 27, 2017 at 12:39:49PM +0200, Amir Goldstein wrote: >> Test generic/007 was failing after running test overlay/007, >> because the latter left behind a directory named 007 in test dir >> and the former failed to mkdir a directory with the same name. >> >> Greping the tests for the pattern $TEST_DIR/$seq* found some more >> files/dirs of this sort that were not being cleaned up. >> >> Clean those files/dir on _cleanup trap to fix the 007 tests collision >> and avoid similar collisions in future tests. >> >> Left the directories $TEST_DIR/$seq.mnt in tact, because they are >> always empty and created with mkdir -p. >> >> There are more files left behind as can be seen in any aged test dir. > > TEST_DIR is supposed to be aged over multiple runs, so leaving test > files in it should be just fine, unless these files are consuming a lot > of free space and would block subsequent tests. > Good point, but aging doesn't mean that the files need to stay there create + delete is perfectly good aging. As a matter of fact, all these test take care of removing the test files *before* running the test, so why not cleanup *after* the test as well? I can understand why it makes sense to leave behind $seq.$$ files to fill up TEST_DIR with junk over time, but I cannot understand the reasoning to leave behind $seq.<const> files, which are going to be removed before next run anyway. > For tests that need an empty dir or a newly created file in TEST_DIR, > they could just remove the dir or file in test setup phase before doing > any actual setups. > > So I think adding "-p" to mkdir command in generic/007 looks sufficient. > Fair enough, but please give this cleanup issue a second thought. > Thanks, > Eryu -- To unsubscribe from this list: send the line "unsubscribe fstests" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
