--On Wednesday, April 05, 2006 6:00 AM -1000 Hackystat Administrator <[EMAIL PROTECTED]>
wrote:
The following alert(s) were triggered today:
* Build analysis for project Hackystat-7:
Integration build for Hackystat-7 project on 05-Apr-2006 FAILED.
Module: hackyApp_Cgqm
Failure Types: JUnit
Plausible Culprit: [EMAIL PROTECTED]
How culprit is identified: No local build/test immediately before the last
commit of
the day. Failure Messages:
* org.hackystat.app.cgqm.plugin.buildFailures.TestQNightlyBuildFailures
Culprit was, indeed, me. What happened was the following: I was working with my
workstation configured to think it was in Germany in order to fix some
localization-related issues. Then, after doing a _complete_ build and test (20 minutes),
and finding everything was OK, I then tried to commit, and discovered that there was a
merge conflict (involving 1 line in hackyApp_Cgqm--the line that caused this JUnit
failure). So, I did an update, invoked the merge resolution tool, and ended up picking
the wrong version of the line to commit!
Of course, it was so "simple", just a _one line_ change, and I had _just_ done a 20
minute test, so was I really gonna do another 20 minute test to check my one line change
before committing? Of course not, its gotta work this time! (Famous last words.)
Sorry about that.
Cheers,
Philip
