Show HN: Weekly troubleshooting challenges for DevOps and SREs

circuitops.com

1 point·hkh··3 comments

Feedback: having done your first one, I find myself wondering how the thing got into that state in the first place, and what can be done about it. That is, [Jub be jung fperjrq hc gur crezvffvbaf/bjarefuvcf ba gur erqvf cngu] (spoilers).

Basically, this is the lowest level of "if this then that" fixing, and people need to be able to do that, but they also need to then go "we can't keep operating in this mode".

I mean, yeah, here, it's artificial for the purposes of the exercise. But, if you were to see this in real life, there would be a bunch more questions to ask and things to dig into to keep it from happening again. Randos with root? Bad script somewhere? Terrible system management stuff? RPM getting a little too frisky with how it manages certain directories?

I'm not sure how you get that part into a thing like this, though.

Great point! I wonder if there are things we can do to break the scenario while the user is logged in to simulate operational issues such as OOMKilled, network glitches or disk filled. That might enable a better root cause discussion during interviews

Hi HN, Hassan here - one of the co-founders of CircuitOps. While building CircuitOps, we have had so much fun watching people try to diagnose and fix issues in production like environments, so we made it into a public challenge!

We are planning on releasing one challenge a week.

Would love to hear your feedback and see some of you on the leaderboard!

Cheers, Hassan

Reply on news.ycombinator.com