I've never experienced the issue locally (I see it on Jenkins quite a
lot), but I suspect a stale registrar process left from another test may
be stopping the socket from closing. Not that registrars are also
simulated for discovery tests, so it may not necessarily be Reggie.
The code is duplicated in two places in superclasses of the tests that
are failing, the method portInUse(int port) is supposed to check if the
ports available, but only selects from a list of LookupLocator's known
to have started, so doesn't actually check the port's available.
Perhaps you can find the stale test process responsible?
Regards,
Peter.
Greg Trasuk wrote:
On Mon, 2013-04-08 at 15:59, Peter wrote:
Greg,
I apologise again for my outburst, I was wrong about your leadership skills.
I couldn't help but notice the number of test failures, did they have an error
message like the number of services started != number of services wanted?
If so, this is because port 4160 is in use.
Yes, that's it. Definitely a configuration issue. It's not that port
4160 is in use by anything else - the problem is that the test framework
is trying to start up multiple Reggies, which requires port overrides
that are currently not getting through. Perhaps if someone recognizes
the issue they can clue me in and save some time. If not, then it's
just a matter of slogging through the code to understand the test
configurations.
There's a file called 'reggie3_2Ports.properties' in
qa/src/com/sun/jini/test/share' that seems to call out these overrides,
but I haven't yet figured out how it's supposed to get included in the
test configuration. I suspect it's something fairly trivial that needs
to be setup locally.
Anybody recognize the problem? On the bright side, every failure I see
in the 2.2 branch that I'm testing locally appears to be caused by this
configuration problem, so everything looks very promising.
Cheers,
Greg.
See the latest arm test failure on jenkins for an example, exactly 280 tests
failed there too.
Regards,
Peter.
----- Original message -----
The tests appear to be awfully sensitive to their environment. Insofar as when
I run them locally on an untouched source tree, I get 280 failures.
