About once an hour, when
TraceLedgerEvent.TookSnapshot is written to the log, there is a CPU spike and at the same time the
slotMissedNum is increasing. Is this a bug in the node? Or is there a fix?
That CPU usage spike is only at most 8% of the CPU capacity, which is extremely small. I get spikes on my laptop CPU of 100% for longer time intervals than the spikes here, when just casually browsing the web.
If anything is being written anywhere on the computer, or anything is happening such as
slotMissedNum increasing, I would expect a CPU usage spike of around the size of the extremely small spikes you're seeing here.
I hope that takes care of the CPU spikes part of the question. If you believe that the increasing of
slotMissedNum is a bug, then that's a different matter, for which I'm not immediately able to offer a solution, but that bug can be addressed entirely independently since the CPU spikes do not create any additional concerns.