cp3065 went down yesterday 2019-11-11 at 21:54:24 showing the same symptoms as described in T237348 for cp3057.
The server has been power-cycled by @ema and is currently reachable but depooled.
Mentioned in SAL (#wikimedia-operations) [2019-11-12T10:06:59Z] <ema> repool cp3065, nothing interesting in kern.log and SEL T238032
Perhaps interestingly, or maybe entirely unrelated: a couple of hours before crashing the host had a spike in cache write errors:
Tracking the issue on the parent task: T238305