Not specifically a Meteor question - but what could cause an app to spend a significant amount of time (e.g., 25% ) of CPU time in garbage collection, if it was not close to it’s memory limit?
In the last couple of days I’ve noticed the CPU usage of one server hits 100% and stays there until the server is restarted. I’ve found no pattern as to which method calls or publications could cause this, but today I ran a profiler on the server in question when it happened - and the only obvious problem was that garbage collection took 25% of the time, 10% of the time was just “program” the rest was distributed across a couple hundred function calls.
What could cause this amount of garbage collection? The server has 2GB of memory - and is the only thing running, it was using around 600MB at the time.