You're on your way to the next level! Join the Kudos program to earn points and save your progress.
Level 1: Seed
25 / 150 points
1 badge earned
Challenges come and go, but your rewards stay with you. Do more to earn more!
What goes around comes around! Share the love by gifting kudos to your peers.
Keep earning points to reach the top of the leaderboard. It resets every quarter so you always have a chance!
Join now to unlock these features and more
I have a given plan in Bamboo that I want to run on different agents. This is useful because I need to test that my software can compile and run on different platforms. How do I achieve this ?
I have configured 2 agents but the plan is always ran only on the first one. I don't think that concurrent run correspond to what I am looking for.
Cloning jobs may save a couple steps when you do the clone, but it doesn't address the maintenance issue raised by Barthelemy von Haller. That's why I suggested (below) changing the requirements without cloning the plan. It is a little bit ugly, but note that the unique agent names only have to be defined once. After that, the ugly part is quite small and easy to manage.
Changing the requirements doesn't solve the problem of ensuring that every build runs on all supported platforms. If you only have one job, then Bamboo will select just one agent to run the build.
This is a big issue for C and C++ projects, where running every build on every supported platforms is critical.
Not sure why you say that. Bamboo will only select an agent that matches the requirements. When one requirement is to match a unique agent name, that's the only agent that will be selected.
The more compelling issue is that you have to go through this process every time you modify the build.
Hi Ed and David,
Ed, what David and I meant is that a single plan should be able to automatically run on multiple/all agents, not only one. And this is not achievable at the moment. Therefore, I have to have 1 plan per agent on which I want to test. It becomes a maintenance problem very quickly. Hudson and buildbot allow it for example (and TeamCity apparently).
Please see (and vote!) these issues I created for this missing feature : https://jira.atlassian.com/browse/BAM-9188 and https://jira.atlassian.com/browse/BAM-9189
+1 for any feature that makes this easier. We have several plans/jobs which are identical, except for their dependencies, just to get things to build across all of our supported platforms. It's a minor pain to set up, but the real hurt comes when you have to change anything, b/c you have to make the same change across all your clones plans/jobs.
1. Define, for each agent that you want to test, a custom property that is a unique agent name, e.g. <servername>.<agentname>.
2. For each agent under test,
2.1 Modify the plan to require this agent's unique name.
2.2 Run the build and evaluate for success vs failure.
2.3 Fix as needed.
3. After all agents have been successfully tested, remove the requirement. Future builds will now be able to build on any of the tested agents.
I am looking for a feature that JetBrain's TeamCity has where we can define a plan and have it run on all required agents. e.g. we have a plan that installs a software on all of our remote agents (12 of them right now). If I have to define 12 jobs to do this, it is not scalable.