Re: SCM polling after build failure question...

2013-02-21 Thread Jeff
So my current theory is that the SCM version isn't getting updated after the 'mvn clean deploy -U' fails with a compile error. Here is the polling log for one build: Started on Feb 19, 2013 8:01:24 AM Received SCM poll call on for ResultsService (trunk) on Feb 19, 2013 8:01:26 AM http://svn/S

Re: SCM polling after build failure question...

2013-02-19 Thread Sami Tikka
The Jenkins SVN plugin has some dependency on time. To achieve best results you should have both Jenkins and SVN server clocks synchronized using NTP or similar. -- Sami Jeff kirjoitti 19.2.2013 kello 18.44: > We are using SVN (SVNBridge to TFS) for most of our projects. It doesn't > seem t

Re: SCM polling after build failure question...

2013-02-19 Thread William Soula
Check the polling log to see if there is any more information there. I've had the same thing happen to me and it was building constantly because the workspace was gone and it needed to create a new one but couldn't because the workspace was locked. What was put in the log for reason for the bu

Re: SCM polling after build failure question...

2013-02-19 Thread Jeff
We are using SVN (SVNBridge to TFS) for most of our projects. It doesn't seem to happen on all of our projects, just sometimes and it seems to be related to type of build failure (compile vs. test vs. Sonar failure). On Tue, Feb 19, 2013 at 9:39 AM, Michael Clarke wrote: > Which SCM are you usi

Re: SCM polling after build failure question...

2013-02-19 Thread Michael Clarke
Which SCM are you using? Polling will continue regardless of build status, but it's up to each SCM plugin to decide if it sees changes. Thanks Michael On 19 Feb 2013, at 16:28, Jeff wrote: We had a check-in for a project that caused a compile error last night but it didn't get fixed until this

SCM polling after build failure question...

2013-02-19 Thread Jeff
We had a check-in for a project that caused a compile error last night but it didn't get fixed until this morning. However, Jenkins polling (every 4 minutes) detected an SCM change and attempted to rebuild. This repeated every 4 minutes, even though the SCM revision was the same for every build w