Philip,
You are correct. I inadvertently introduced a JUnit failure on 3/19,
but due to other build failures on that day, my "build breaker" didn't
propagate until last night.
The reason the JUnit test failed in CppUnit was due to some debugging
code that I did not remove before committing. Essentially, I was
writing a JDom document to an XML file to verify its structure.
However, the output file I specified exists on my local machine, but not
on HackyDev. Thus an IOException was thrown in the JUnit test, causing
the build failure.
I'm running a freshStart all.junit now and will commit the changes as
soon as they are successful.
Best regards,
Mike
Philip Johnson wrote:
--On Tuesday, March 21, 2006 6:00 AM -1000 Hackystat Administrator
<[EMAIL PROTECTED]> wrote:
Integration build for Hackystat-7 project on 21-Mar-2006 FAILED.
Module: hackySensor_CppUnit
Failure Types: JUnit
Plausible Culprit: Unknown
How culprit is identified: Build failure reason is unclear, and no one
has made any commit in hackySensor_CppUnit. Perhaps hackystat sensor
data
is incomplete, or it is caused by external error on the integration
build
box. Failure Messages:
* org.hackystat.sensor.cppunit.TestCppUnitSensor
The failure is in the test case:
testCppUnitSensorOnTestDataSetFiles
And the actual failure is:
C:\svnHackystat\hackySensor_CppUnit\out.xml (The system cannot find
the path specified)
I seem to recall that Mike was working in this module. Mike, can you
investigate this?
From a build failure analysis perspective, I'm not positive the "unknown
culprit" designation is warranted. This might be the same build
failure as yesterday; we just didn't actually deal with it for some
reason. It could be that the build failure analysis should be able to
look back more than 24 hours. (However, I don't know yet if looking
back 48 or 72 hours would reveal Mike as the culprit or not.)
Thanks,
Philip
