Workshop. Post-mortem check-out: Crashes, hangs, and high CPU (part 2)

Crashing and hanging processes, resource starvation, and other fun stuff are not very hard to deal with in development, but what can we do if it happens in production, especially after a week of normal service activity.

Black magic rituals, pentagrams, and server restarts might work occasionally, but it is a temporary solution at best.

In this workshop we will discuss possible symptoms of the aforementioned issues and also Mikhail will show how in practice such issues can be investigated and root causes found.