Re: second fix for MNG-6330 [regression] Parents relativePath not verified anymore

2017-12-28 Thread Karl Heinz Marbaise 

Hi Robert,

+1 from me now after the master green(pipeline) (..sorry blue ball) again..

Kind regards
Karl Heinz Marbaise


On 28/12/17 19:41, Karl Heinz Marbaise wrote:

Hi Robert,

So I have restarted the master build based on that and see if everything 
is fine..(which I expect otherwise the branch should have shown the two 
test errors as well)..


Kind regards
Karl Heinz
On 28/12/17 19:34, Robert Scholte wrote:

Hi Karl Heinz,

AFAIK you're hitting https://issues.apache.org/jira/browse/INFRA-15681

thanks,
Robert

On Thu, 28 Dec 2017 19:31:35 +0100, Karl Heinz Marbaise 
 wrote:



Hi Robert,

On 28/12/17 19:10, Robert Scholte wrote:

Hi Karl Heinz,
 just looking for a +1 from another committer before I merge it back 
to the master.


Ah ok...So the first question is why is the master at the moment 
yellow with two failed tests ? Do you have any clue about that ?



Kind regards
Karl Heinz






 thanks,
Robert
 On Thu, 28 Dec 2017 18:55:15 +0100, Karl Heinz Marbaise 
 wrote:



Hi Robert,


On 28/12/17 17:58, Robert Scholte wrote:

Hi,
 Looking for someone to confirm and second MNG-6330[1]


how can I help here?

Karl Heinz Marbaise

While working on the maven-release-plugin I discovered regression 
introduced in Maven 3.5.0 by a change to the ReactorModelCache[2]
This only occurs when a parent was cached and another module 
points to this parent, even though the relativePath for this 
parent results in a non-existing pom file.

Change is a simple null-check[3]
Regression covered by integration test[4]
Jenkins is happy[5]
 thanks,
Robert
 [1] https://issues.apache.org/jira/browse/MNG-6330
[2] https://issues.apache.org/jira/browse/MNG-6030
[3] 
https://github.com/apache/maven/commit/4eb5747e8229d0201b707052ce7f8ae2fe24eec2 
 [4] 
https://git-wip-us.apache.org/repos/asf?p=maven-integration-testing.git;a=commit;h=cfd0cc629c2dd496554c59c1680fb1d90b1311b3 
 [5] 
https://builds.apache.org/job/maven-wip/job/maven/job/MNG-6330_RelativePath/ 


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

Re: second fix for MNG-6330 [regression] Parents relativePath not verified anymore

2017-12-28 Thread Karl Heinz Marbaise 

Hi Robert,

So I have restarted the master build based on that and see if everything 
is fine..(which I expect otherwise the branch should have shown the two 
test errors as well)..


Kind regards
Karl Heinz
On 28/12/17 19:34, Robert Scholte wrote:

Hi Karl Heinz,

AFAIK you're hitting https://issues.apache.org/jira/browse/INFRA-15681

thanks,
Robert

On Thu, 28 Dec 2017 19:31:35 +0100, Karl Heinz Marbaise 
 wrote:



Hi Robert,

On 28/12/17 19:10, Robert Scholte wrote:

Hi Karl Heinz,
 just looking for a +1 from another committer before I merge it back 
to the master.


Ah ok...So the first question is why is the master at the moment 
yellow with two failed tests ? Do you have any clue about that ?



Kind regards
Karl Heinz






 thanks,
Robert
 On Thu, 28 Dec 2017 18:55:15 +0100, Karl Heinz Marbaise 
 wrote:



Hi Robert,


On 28/12/17 17:58, Robert Scholte wrote:

Hi,
 Looking for someone to confirm and second MNG-6330[1]


how can I help here?

Karl Heinz Marbaise

While working on the maven-release-plugin I discovered regression 
introduced in Maven 3.5.0 by a change to the ReactorModelCache[2]
This only occurs when a parent was cached and another module points 
to this parent, even though the relativePath for this parent 
results in a non-existing pom file.

Change is a simple null-check[3]
Regression covered by integration test[4]
Jenkins is happy[5]
 thanks,
Robert
 [1] https://issues.apache.org/jira/browse/MNG-6330
[2] https://issues.apache.org/jira/browse/MNG-6030
[3] 
https://github.com/apache/maven/commit/4eb5747e8229d0201b707052ce7f8ae2fe24eec2 
 [4] 
https://git-wip-us.apache.org/repos/asf?p=maven-integration-testing.git;a=commit;h=cfd0cc629c2dd496554c59c1680fb1d90b1311b3 
 [5] 
https://builds.apache.org/job/maven-wip/job/maven/job/MNG-6330_RelativePath/ 


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

Re: second fix for MNG-6330 [regression] Parents relativePath not verified anymore

2017-12-28 Thread Robert Scholte 

Hi Karl Heinz,

AFAIK you're hitting https://issues.apache.org/jira/browse/INFRA-15681

thanks,
Robert

On Thu, 28 Dec 2017 19:31:35 +0100, Karl Heinz Marbaise  
 wrote:



Hi Robert,

On 28/12/17 19:10, Robert Scholte wrote:

Hi Karl Heinz,
 just looking for a +1 from another committer before I merge it back to  
the master.


Ah ok...So the first question is why is the master at the moment yellow  
with two failed tests ? Do you have any clue about that ?



Kind regards
Karl Heinz






 thanks,
Robert
 On Thu, 28 Dec 2017 18:55:15 +0100, Karl Heinz Marbaise  
 wrote:



Hi Robert,


On 28/12/17 17:58, Robert Scholte wrote:

Hi,
 Looking for someone to confirm and second MNG-6330[1]


how can I help here?

Karl Heinz Marbaise

While working on the maven-release-plugin I discovered regression  
introduced in Maven 3.5.0 by a change to the ReactorModelCache[2]
This only occurs when a parent was cached and another module points  
to this parent, even though the relativePath for this parent results  
in a non-existing pom file.

Change is a simple null-check[3]
Regression covered by integration test[4]
Jenkins is happy[5]
 thanks,
Robert
 [1] https://issues.apache.org/jira/browse/MNG-6330
[2] https://issues.apache.org/jira/browse/MNG-6030
[3]  
https://github.com/apache/maven/commit/4eb5747e8229d0201b707052ce7f8ae2fe24eec2  
 [4]  
https://git-wip-us.apache.org/repos/asf?p=maven-integration-testing.git;a=commit;h=cfd0cc629c2dd496554c59c1680fb1d90b1311b3  
 [5]  
https://builds.apache.org/job/maven-wip/job/maven/job/MNG-6330_RelativePath/  
   
-

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


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


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

Re: second fix for MNG-6330 [regression] Parents relativePath not verified anymore

2017-12-28 Thread Karl Heinz Marbaise 

Hi Robert,

On 28/12/17 19:10, Robert Scholte wrote:

Hi Karl Heinz,

just looking for a +1 from another committer before I merge it back to 
the master.


Ah ok...So the first question is why is the master at the moment yellow 
with two failed tests ? Do you have any clue about that ?



Kind regards
Karl Heinz







thanks,
Robert

On Thu, 28 Dec 2017 18:55:15 +0100, Karl Heinz Marbaise 
 wrote:



Hi Robert,


On 28/12/17 17:58, Robert Scholte wrote:

Hi,
 Looking for someone to confirm and second MNG-6330[1]


how can I help here?

Karl Heinz Marbaise

While working on the maven-release-plugin I discovered regression 
introduced in Maven 3.5.0 by a change to the ReactorModelCache[2]
This only occurs when a parent was cached and another module points 
to this parent, even though the relativePath for this parent results 
in a non-existing pom file.

Change is a simple null-check[3]
Regression covered by integration test[4]
Jenkins is happy[5]
 thanks,
Robert
 [1] https://issues.apache.org/jira/browse/MNG-6330
[2] https://issues.apache.org/jira/browse/MNG-6030
[3] 
https://github.com/apache/maven/commit/4eb5747e8229d0201b707052ce7f8ae2fe24eec2 
 [4] 
https://git-wip-us.apache.org/repos/asf?p=maven-integration-testing.git;a=commit;h=cfd0cc629c2dd496554c59c1680fb1d90b1311b3 
 [5] 
https://builds.apache.org/job/maven-wip/job/maven/job/MNG-6330_RelativePath/ 
  -

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


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

Re: second fix for MNG-6330 [regression] Parents relativePath not verified anymore

2017-12-28 Thread Robert Scholte 

Hi Karl Heinz,

just looking for a +1 from another committer before I merge it back to the  
master.


thanks,
Robert

On Thu, 28 Dec 2017 18:55:15 +0100, Karl Heinz Marbaise  
 wrote:



Hi Robert,


On 28/12/17 17:58, Robert Scholte wrote:

Hi,
 Looking for someone to confirm and second MNG-6330[1]


how can I help here?

Karl Heinz Marbaise

While working on the maven-release-plugin I discovered regression  
introduced in Maven 3.5.0 by a change to the ReactorModelCache[2]
This only occurs when a parent was cached and another module points to  
this parent, even though the relativePath for this parent results in a  
non-existing pom file.

Change is a simple null-check[3]
Regression covered by integration test[4]
Jenkins is happy[5]
 thanks,
Robert
 [1] https://issues.apache.org/jira/browse/MNG-6330
[2] https://issues.apache.org/jira/browse/MNG-6030
[3]  
https://github.com/apache/maven/commit/4eb5747e8229d0201b707052ce7f8ae2fe24eec2  
 [4]  
https://git-wip-us.apache.org/repos/asf?p=maven-integration-testing.git;a=commit;h=cfd0cc629c2dd496554c59c1680fb1d90b1311b3  
 [5]  
https://builds.apache.org/job/maven-wip/job/maven/job/MNG-6330_RelativePath/  
  -

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



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


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

Re: second fix for MNG-6330 [regression] Parents relativePath not verified anymore

2017-12-28 Thread Karl Heinz Marbaise 

Hi Robert,


On 28/12/17 17:58, Robert Scholte wrote:

Hi,

Looking for someone to confirm and second MNG-6330[1]


how can I help here?

Karl Heinz Marbaise

While working on the maven-release-plugin I discovered regression 
introduced in Maven 3.5.0 by a change to the ReactorModelCache[2]
This only occurs when a parent was cached and another module points to 
this parent, even though the relativePath for this parent results in a 
non-existing pom file.

Change is a simple null-check[3]
Regression covered by integration test[4]
Jenkins is happy[5]

thanks,
Robert

[1] https://issues.apache.org/jira/browse/MNG-6330
[2] https://issues.apache.org/jira/browse/MNG-6030
[3] 
https://github.com/apache/maven/commit/4eb5747e8229d0201b707052ce7f8ae2fe24eec2 

[4] 
https://git-wip-us.apache.org/repos/asf?p=maven-integration-testing.git;a=commit;h=cfd0cc629c2dd496554c59c1680fb1d90b1311b3 

[5] 
https://builds.apache.org/job/maven-wip/job/maven/job/MNG-6330_RelativePath/ 



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




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

second fix for MNG-6330 [regression] Parents relativePath not verified anymore

2017-12-28 Thread Robert Scholte 

Hi,

Looking for someone to confirm and second MNG-6330[1]
While working on the maven-release-plugin I discovered regression  
introduced in Maven 3.5.0 by a change to the ReactorModelCache[2]
This only occurs when a parent was cached and another module points to  
this parent, even though the relativePath for this parent results in a  
non-existing pom file.

Change is a simple null-check[3]
Regression covered by integration test[4]
Jenkins is happy[5]

thanks,
Robert

[1] https://issues.apache.org/jira/browse/MNG-6330
[2] https://issues.apache.org/jira/browse/MNG-6030
[3]  
https://github.com/apache/maven/commit/4eb5747e8229d0201b707052ce7f8ae2fe24eec2
[4]  
https://git-wip-us.apache.org/repos/asf?p=maven-integration-testing.git;a=commit;h=cfd0cc629c2dd496554c59c1680fb1d90b1311b3
[5]  
https://builds.apache.org/job/maven-wip/job/maven/job/MNG-6330_RelativePath/


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