| === daek_ is now known as Daekdroom | ||
| === mpt_ is now known as mpt | ||
| smspillaz | Trevinho: dunno what's up with those autolanding failures | 11:52 |
|---|---|---|
| smspillaz | Trevinho: either the autolander is running the test-requiring-x when it shouldn't be or its running them through xvfb and this doesn't work on armhf | 11:53 |
| Trevinho | smspillaz: no, I think it's just a problem of the armhf builder, but it seems fixed on the last run | 11:53 |
| smspillaz | Trevinho: BTW, there's a trick you can use to test that something *wont* crash that I posted here: http://smspillaz.wordpress.com/2013/05/18/regression-testing-for-crashes-on-google-test/ | 11:53 |
| smspillaz | Trevinho: well, the odd thing is that CI is green but the autolander is not | 11:53 |
| Trevinho | smspillaz: so retrying is generally the only thing we can do... | 11:54 |
| Trevinho | smspillaz: yes, I've seen your test... but it wasn't crashing there :( | 11:54 |
| Trevinho | smspillaz: very nice, though | 11:54 |
| smspillaz | Trevinho: ah, so the point is that the test passes because it doesn't crash :) | 11:54 |
| Trevinho | smspillaz: yeah, I know, but it didn't crash on the forked process as well | 11:55 |
| smspillaz | Trevinho: right, it wasn't meant to | 11:55 |
| smspillaz | Trevinho: so, google test doesn't provide an EXPECT_NDEATH | 11:55 |
| Trevinho | smspillaz: also the crash is something that generally happen but could also not happen in some rare cases... | 11:55 |
| Trevinho | checking later.. need to have lunch | 11:55 |
| Trevinho | see you | 11:56 |
| smspillaz | Trevinho: anyways, not too much of a big deal, just a tip | 11:56 |
| smspillaz | Trevinho: BTW, the -ci builds are running test-gtest-xless and the autolander builds are running test-gtest | 11:56 |
| smspillaz | that's probably where the problems are coming from | 11:56 |
| smspillaz | actually, never mind, I'm wrong | 11:57 |
| xylon_ | Hello people | 12:34 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!