J K commented on Bug JENKINS-15881

One of our products is a mesh networking device comprised of three microprocessors, each of which maps to a different development team, each of which maintains their own SVN repo and such. Our team is using Jenkins to manage a CI process with build management and unit testing (all specific to our microprocessor), but also integration testing and release management. The integration testing involves pulling and building the code for all three processors and deploying it into a testbed environment that runs automated tests taking anywhere from a few minutes to a few days depending on needed scope. Generally we want to pull from the head of trunk for all three projects.

However, there have been times when the independent teams get into a state where there would be incompatibilities running the most recent code for all three. For instance, if the highest priority for teams A and B is to get a new interprocessor communication protocol working, but for team C is to get a mesh networking component ready, then trunk@HEAD processor C can no longer talk with trunk@HEAD A or B. When this happens we've used the ability to go back revisions on A and B to do integration testing and (internal) release management when our systems test (human-driven) resources were available.

I created a groovy script that monitors SVN commit comments for certain key word/value pairs, one of which specifies if a particular revision of processor A and/or B code is needed. This, plus the patch I submitted on github for issue #14116, took care of the problem. We're still quite new to Jenkins and CI in general, so there may be a more elegant solution. But so far this has been working for us.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators.
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to