Well, how about the following situation I faced a number of times: the build is broken and I try to fix it. But I don't succeed.. so how should I know about that? Manually go&check in hudson? Not that handy. Of course, Hudson spamming is much worse. I'd rather we disable Derby builds until those two tests can work.
2009/12/11 Aristedes Maniatis <[email protected]> > On 11/12/09 12:55 PM, Andrus Adamchik wrote: > >> >> On Dec 10, 2009, at 6:14 PM, Aristedes Maniatis wrote: >> >>> >>> >>> How about something really simple like >>> >>> if (! System.getProperty("CayenneTestConnection").equals("derby")) { >>> run test... >>> } >>> >>> >>> That is, if these test failures are actually for real and SHOULD not >>> pass on Derby. Otherwise, I'll just configure Hudson to only send >>> emails when the test failures change from run to run, not when they >>> remain the same. >>> >> >> I think the later option is better for now. IIRC these should work on >> Derby, but the right implementation is just too quirky and waiting its >> time. >> >> Andrus >> > > > OK, I've done that now. We will not get emails after the first broken > build. > > Do we need Jira(s) for the broken tests on Derby? Are they the same issue > or two different things? > > > Ari Maniatis > > > -- > > --------------------------> > Aristedes Maniatis > GPG fingerprint CBFB 84B4 738D 4E87 5E5C 5EFA EF6A 7D2E 3E49 102A > -- Andrey
