Please read the first part of this series and then come again. :) Also Network experiments. I did It delays
Fail #2: Do not trust the docs, just try it
One of the core principles of chaos engineering is to run experiments with a clear hypothesis. After that, you can create your own. Use your creativity — like Yoda said, “Do or do not, there is no try.”
I was thinking, what’s the worst that could happen in my database? Then I knew — losing inserted records would be the real nightmare. So, I created my own experiment. Let’s check what happens if the writer goes down. I read an article about how a read instance can be resilient and swap to another DB replica. But imagine if the writer reboots… yeah, cue dramatic music.
The base experiment is from this 👉blog . It explains everything, but let’s see what happens if we change something — like targeting the writer instead of the reader.💥
📽️ let check the results
Some pictures to analyze what happens.
In conclusion, I made a basic experiment that could’ve been answered by the docs but maybe not apply to my case, but honestly, I like to try it myself — that’s how I really learn.