summaryrefslogtreecommitdiff
path: root/tests/ipf
AgeCommit message (Collapse)Author
2021-11-10s/endianess/endianness/msaitoh
2015-12-26Remove the "expected failure" from test n12, and change it not to usepgoyette
the "-v" command line option to ipf_test. The -v option causes it to display additional information, some of which is not fixed (looks like an internal data structure address). One fewer on the expected-failures list. :)
2015-11-05Ooops, pilot error - revert lastmartin
2015-11-05Fix expected output to match -current NetBSD reality (port "255" is notmartin
mapped to "reserved")
2014-12-06Null commit, to note that the previous commit by darrenr (which saiddholland
just "PR kern/47462") didn't even have the right PR number. The correct number, as far as I can tell, was PR 47262. /brick
2014-12-01Remove the "broken on BE" mark from the N14 testcase - ozaki-r fixed itmartin
recently.
2014-11-25Fix wrong expected outputsozaki-r
IP addresses in the expected outputs looks reversed (in byte order). It seems the expected outputs had been generated based on source code with a bug and the bug was fixed by a commit titled "#534 destination list hashing not endian neutral" by darrenr@.
2014-07-15Darren fixed n8 and n9 recentlymartin
2014-07-15NetBSD has an entry for protocol #255 in /etc/protocols labeled "reserved"martin
so expect that in the parsing output instead of 255.
2014-07-12Test n17_6 does not yet exist so remove unrequired files and test casedarrenr
2014-06-30New IPv6 tests for ipnat parsing need to be built by make so that they candarrenr
be executed.
2014-06-30NetBSD has an entry for protocol #255 in /etc/protocols labeled "reserved"darrenr
so expect that in the parsing output instead of 255.
2014-06-29PR kern/47462darrenr
2014-06-29Add in test cases for testing ipnat's parsing of IPv6 NAT rulesdarrenr
2014-06-29in2 is no longer a broken test casedarrenr
2013-05-16Mark two more test case as only failing for BEmartin
2013-05-16Assume that the NAT failures that happen on some architectures are endianessmartin
bugs, so do not expect failure on LE archs for now.
2013-03-19Fix argument order for "n14"martin
2013-03-18Mark a few more test cases known to only work by chance as failingmartin
and point to PR 47665. Julio: how do we make such cases not fail when they actually happen to work (by pure luck)?
2013-03-17Mark some long-standing failures as known failures.jmmv
The offending tests are these: - t_filter_exec: f26, f27. - t_filter_parse: i17. - t_nat_exec: n12. These tests are confirmed to fail in NetBSD/current under amd64, i386 and sparc as reported by the continuous testing systems.
2012-12-03Christos fixed the crash in ipftest, so re-enable the failing tests.pgoyette
2012-12-03Add a couple more inet6 tests to verify correct rule parsingpgoyette
2012-12-03Include an explicit test for correctly inferring family inet6 when therepgoyette
is an IPv6 address in the rule
2012-12-03Now that expected output files have been adjusted, we can remove thepgoyette
expectation that tests i14 and i19 will fail. Should address kern/47262 which will be closed.
2012-12-03Without explicit address, the address family is not inferred and notpgoyette
displayed in the output.
2012-12-03For rules without any address, don't infer an address family.pgoyette
2012-12-02Test-case n3_6 is not applicable to NetBSD (according to Darren R.)pgoyette
2012-12-01More test clean-uppgoyette
2012-12-01Put the reason for expected failure in the correct positionpgoyette
2012-12-01Use proper quoting for the expected-failure reasonpgoyette
2012-12-01Now that we have a PR for the brokenness in "ipftest -P -N", mark thepgoyette
related tests as Expected Failure rather than just skipping the tests.
2012-12-01Based on the error messages, the test is not broken, but ipftest probablypgoyette
in broken. Once we can analyze the error further, we can open a PR in Gnats and then move this to Expected-Failure. But we shouldn't skip it because we'll never come back to fix it.
2012-11-30If we're going to expect a failure, we need to use the correct shellpgoyette
function! "atf_expect_fail()" does not have a "_tc" in its name!
2012-11-30Rather than providing a custom Atffile that requires separatepgoyette
maintenance, just don't build or install the t_bpf.sh test.
2012-11-29fix typo in previous - missing quotepgoyette
2012-11-29Update a couple of tests to indicate that they are broken/failing becausepgoyette
of missing input/expected data files
2012-11-29Now that I've created PR/47262, reference it in the tests that failpgoyette
because of the PR
2012-11-29Add another test option for expected failures, rather than simplypgoyette
skipping tests
2012-11-29Even though this test is currently disabled, fix it so it looks for thepgoyette
correct input and expected-output files for test case bpf_f1
2012-11-29Since all of the expected-results files provided from upstream includepgoyette
the "-D" output (Dump additional internal tables), update the tests to actually generate the data! This lets us re-enable 28 of the 30 "skipped" tests. XXX Still need to work on tests n12 and n17_6
2012-07-22test cases n201 and n202 were removed on import of 5.1.2darrenr
2012-07-22Merged IPFilter 5.1.2 into HEADdarrenr
2012-07-22Import IPFilter 5.1.2darrenr
2012-07-08Restore the rest of the ipfilter test framework, even if most of thesepgoyette
tests are broken. The "golden output" files for these tests need to be updated.
2012-07-08Restore one more group of tests.pgoyette
XXX Test i17 was broken previously, and remains broken.
2012-07-08Restore some more tests from ipfilter 5.1.1pgoyette
XXX Test in2 was broken before the ipfilter 5.1.1 import, and is still XXX broken.
2012-07-08Restore some more of the ipfilter 5.1.1 tests.pgoyette
XXX It appears that ipftest is not handling the policy {,src-,dst-}hash XXX correctly, so these tests are still marked broken/skipped.
2012-07-07Re-enable some of the ipf tests. The only thing wrong with these testspgoyette
is that the tests were not re-imported with the rest of ipfilter 5.1.1 XXX There are still some more ipf test failures that need to get resolved, XXX but at least we'll get some of them back just in case we have a XXX regression.
2012-03-27Mark the failing tests as broken. XXX: If no one is willing to maintainjruoho
the ipf tests, these should be removed.
2012-03-23Update to the tests from 5.1.1christos