On 01/07/2021 23:41, Elliott Sims wrote:
Also for narrowing down performance issues, I've had good luck with the "ttop" module of Swiss Java Knife and with the async-profiler tool: https://github.com/jvm-profiling-tools/async-profiler <https://github.com/jvm-profiling-tools/async-profiler>
I like the SJK's ttop too, it's a very useful tool indeed. However, I have to admit that very often I find the "SharedPool-Worker-##" uses the most CPU or allocates the most heap memory, and that makes the troubleshooting more complicated, because those workers do a wide range of things. Sometimes I really wish Cassandra uses separate thread pools for those.
Anyway, the general advise for anyone running a Cassandra 2.x cluster and having issues with it is to upgrade. There's many improvements and bugfixes in the newer versions, and even the upgrade itself didn't fix the issues you have, it will at least make troubleshooting easier.
