I have a very strange problem. On one of my boxes with Hudson deployment, the working copy folder (which is basically an SVN checkout) continually gets deleted every night. Not a huge issue; mostly strange and annoying. Because of this, every morning, I have to perform a full checkout instead of update/switch. It's not a configuration issue in Hudson as far as I know, and I've checked every crontab file to verify that it's not that.
In Linux, is there a bash script or some command line magic to watch a folder, save history about when it is deleted, and who deletes it? This way, I can narrow down what is performing the actual deletion.
You might be running into JENKINS-8880 in which custom workspaces in the node's workspace/ directory are deleted nightly. The fix is to upgrade to 1.427 (or later), or put custom workspaces outside the workspace/ directory.