Hi,

Not sure if this really warrants a bug report, but I tried to ask the question in other channels (gitter) and did not receive a response.

We rely on preventive node monitoring to disable nodes based on free disk space, but because the check frequency is 1 hour, even with a relatively high threshold, it never really catches low space and we constantly have nodes running out of space during the hour between checks.

We have dozens to hundreds of jobs and most with external triggers. If we get a bunch of builds triggered back to back, it will push the nodes over the edge within an hour.

We already have cleanups scheduled, however most of our jobs involve building docker images, and there are certain things we cannot clean up after each build due to docker image layers being shared between different builds.

Is there a way for the user to modify the frequency of the checks?

Thanks.