1

I'm trying to build system tests into Jenkins. The trouble is I need to test distributed deployments where the multiple instances of the software under test automatically connect to each other. The steps should be as follows:

  1. A Jenkins machine handles the build job (maven).
  2. On successful completion, it deploys the artifacts on a number of machines (Windows 7, XP and Linux)
  3. Those machines kill the previous execution of the software under test and start it again (with the recently deployed version)
  4. The Jenkins machine then runs a set of tests that make use of the remote deployments

What I am currently considering is using Fabric (with Cygwin on the Windows machines), and use ant tasks in maven to orchestrate it all.

Before I commit to this, though, I'd like to hear some of you testing experts in case I'm making my life harder than it should be ;)

Thanks!

Improve this question

1 Answer 1

Reset to default
1

I’m not an expert (not even almost), but I have made some experience with build-engines and automatic deployment. That's what we've made:

We are using IBM's RTC for the build process and STAF for the deployment (on Windows and on Linux). Everything is controlled via ANT - the build, the tests and the deployment. That works pretty good. STAF is a small tool (from IBM, too, but open source), that runs on Windows and Linux. It provides methods to interact with remote machines (e.g. copy files or start tasks). you can also write an own Java program for STAF (some kind of a plug-in) to do what ever you want on a remote machine.

If you are already working with Java, this could be a good solution.

Improve this answer
Sign up to request clarification or add additional context in comments.

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.