| commit | b6e6379514ad49e798a80b8464ebfc2b77f86574 | |
| author | Tim Beale <timbeale@catalyst.net.nz> | |
| Thu, 17 Aug 2017 00:30:30 +0000 (17 12:30 +1200) | ||
| committer | Andrew Bartlett <abartlet@samba.org> | |
| Tue, 29 Aug 2017 05:23:28 +0000 (29 07:23 +0200) | ||
| tree | 8b2e01ec0e19fe81e64b0981f841470d2978adba | treesnapshot (tar.gz zip) |
| parent | d6a384b24bb762abc340158bbcd3aad828a4b490 | commitdiff |
selftest: Use a unique(ish) OU for every run of getnc_unpriv
An intermittent problem I noticed with tests in the past is that the
setup can fail to create the base OU because it already exists.
I believe this is because the previous testenv DC has replicated out the
test object, but not its deletion at the point that the next testenv DC
starts running the test.
This only seemed to happen very occassionally (I haven't seen it
happen with getnc_unpriv yet, but I also haven't run it through the
autobuild yet).
Using same randomness in the test OU should help avoid this sort of
problem, and it matches what some other replication tests do.
Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>
An intermittent problem I noticed with tests in the past is that the
setup can fail to create the base OU because it already exists.
I believe this is because the previous testenv DC has replicated out the
test object, but not its deletion at the point that the next testenv DC
starts running the test.
This only seemed to happen very occassionally (I haven't seen it
happen with getnc_unpriv yet, but I also haven't run it through the
autobuild yet).
Using same randomness in the test OU should help avoid this sort of
problem, and it matches what some other replication tests do.
Signed-off-by: Tim Beale <timbeale@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Reviewed-by: Garming Sam <garming@catalyst.net.nz>
| source4/torture/drs/python/getnc_unpriv.py | diffblobblamehistory |