Android mochitest-2 failure log, which was considered "busted"/red: http://tinderbox.mozilla.org/showlog.cgi?log=Mozilla-Inbound/1308946128.1308947406.6238.gz The failure was: { ======== BuildStep ended ======== ======== BuildStep started ======== Reboot Device slave lost === Output === python /builds/sut_tools/reboot.py 10.250.49.29 in dir /builds/tegra-042/test/build (timeout 1200 secs) watching logfiles {} argv: ['python', '/builds/sut_tools/reboot.py', '10.250.49.29'] environment: PATH=/opt/local/Library/Frameworks/Python.framework/Versions/2.6/bin:/opt/local/bin:/opt/local/sbin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin PWD=/builds/tegra-042/test/build SUT_IP=10.250.49.29 SUT_NAME=tegra-042 VERSIONER_PYTHON_PREFER_32_BIT=yes VERSIONER_PYTHON_VERSION=2.6 __CF_USER_TEXT_ENCODING=0x1F6:0:0 closing stdin using PTY: False connecting to: 10.250.49.29 reconnecting socket unable to connect socket reconnecting socket unable to connect socket reconnecting socket unable to connect socket reconnecting socket unable to connect socket reconnecting socket unable to connect socket /builds/tegra-042/test/build/../proxy.flg reconnecting socket unable to connect socket Traceback (most recent call last): File "/builds/sut_tools/reboot.py", line 25, in <module> dm.getInfo('process') File "/builds/tools/sut_tools/devicemanager.py", line 1082, in getInfo data = self.verifySendCMD(['info ' + d]) File "/builds/tools/sut_tools/devicemanager.py", line 139, in verifySendCMD return self.sendCMD(cmdline, newline, False) File "/builds/tools/sut_tools/devicemanager.py", line 166, in sendCMD raise DMError("unable to connect to %s after %s attempts" % (self.host, self.retrylimit)) devicemanager.DMError: unable to connect to 10.250.49.29 after 5 attempts program finished with exit code 1 elapsedTime=30.237428 === Output ended === ======== BuildStep ended ======== } That's an infra problem, and should ideally be recognized as "status: exception" (purple) rather than busted/red.

That does know that it's an infra problem (and a problem that follows some other problem): see, e.g., the purple http://tinderbox.mozilla.org/showlog.cgi?log=Mozilla-Aurora/1308965272.1308965392.21163.gz The actual problem in your log was one buildstep above that, bug 660480, which does already know that it's the wrong color.