Server side artifact clean-up

We've had a few occaisions where build triggers have been configured in such a way that builds trigger in a loop and generate a large number of builds and a correspondingly large volume of artifacts to be stored on the server.  This often results in the disk filling up and causes the server to fail.  This can happen in a time span that is less than the daily cycle you can set the artifact clean up process to run at.  What is the recommended practice for ensuring that the caches for messages and artifacts don't exceed a given size?

1 comment
Comment actions Permalink

I think it is better to move artifacts directory to separate disk. In this case the server will continue to work properly, even if artifacts are filled up. Note that you cannot configure path to artifacts directory in TeamCity, but on Windows and Linux you can use links for this purpose (for Windows see: mklink, or SysInternals junction commands).


Please sign in to leave a comment.