more symlink failures on Jenkins

2011-08-01 Thread Brett Porter
I committed a fix for MNG-3951's test case that stopped it breaking on Mac (and 
passed the first Jenkins ubuntu run).

A later build threw up this:
https://builds.apache.org/job/core-integration-testing-maven-3-trunk-jdk-1.6/7/testReport/

It's full of path related issues, but my change didn't touch any Maven code or 
other test cases. Is anyone familiar with what was previously broken here able 
to shed any light before I take a closer look?

- Brett

--
Brett Porter
br...@apache.org
http://brettporter.wordpress.com/
http://au.linkedin.com/in/brettporter





-
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org



Re: more symlink failures on Jenkins

2011-08-01 Thread Andrew Bayer
Looks like they pass on ubuntu1 but fail on ubuntu3 - I'd guess that
/home/jenkins is a symlink on ubuntu3 but not on ubuntu1, and that something
in the tests is resolving the symlink on one side but not resolving it on
the other. Just a guess, from the Jenkins side of things.

A.

On Mon, Aug 1, 2011 at 3:22 PM, Brett Porter br...@apache.org wrote:

 I committed a fix for MNG-3951's test case that stopped it breaking on Mac
 (and passed the first Jenkins ubuntu run).

 A later build threw up this:

 https://builds.apache.org/job/core-integration-testing-maven-3-trunk-jdk-1.6/7/testReport/

 It's full of path related issues, but my change didn't touch any Maven code
 or other test cases. Is anyone familiar with what was previously broken here
 able to shed any light before I take a closer look?

 - Brett

 --
 Brett Porter
 br...@apache.org
 http://brettporter.wordpress.com/
 http://au.linkedin.com/in/brettporter





 -
 To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
 For additional commands, e-mail: dev-h...@maven.apache.org




Re: more symlink failures on Jenkins

2011-08-01 Thread Brett Porter
Yup, that was it. I've fixed it for those tests, and it seems to be ok now.

All things going well it may make sense to have the verifier return the 
extracted resources as a canonical path so it's consistent across test cases, 
since that's what ends up being compared against from maven.

- Brett

On 02/08/2011, at 1:51 AM, Andrew Bayer wrote:

 Looks like they pass on ubuntu1 but fail on ubuntu3 - I'd guess that
 /home/jenkins is a symlink on ubuntu3 but not on ubuntu1, and that something
 in the tests is resolving the symlink on one side but not resolving it on
 the other. Just a guess, from the Jenkins side of things.
 
 A.
 
 On Mon, Aug 1, 2011 at 3:22 PM, Brett Porter br...@apache.org wrote:
 
 I committed a fix for MNG-3951's test case that stopped it breaking on Mac
 (and passed the first Jenkins ubuntu run).
 
 A later build threw up this:
 
 https://builds.apache.org/job/core-integration-testing-maven-3-trunk-jdk-1.6/7/testReport/
 
 It's full of path related issues, but my change didn't touch any Maven code
 or other test cases. Is anyone familiar with what was previously broken here
 able to shed any light before I take a closer look?
 
 - Brett
 
 --
 Brett Porter
 br...@apache.org
 http://brettporter.wordpress.com/
 http://au.linkedin.com/in/brettporter
 
 
 
 
 
 -
 To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
 For additional commands, e-mail: dev-h...@maven.apache.org
 
 

--
Brett Porter
br...@apache.org
http://brettporter.wordpress.com/
http://au.linkedin.com/in/brettporter





-
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org