News for hackers. Stuff that matters.
Re:Unpacking Observability: The Observability Stack by enobrev on Friday September 10, 2021 @09:12AM (#28481041)
I appreciate the goals and tenets of observability, but I really wish these tools didn't absolutely insist upon being integrated into my software.

Give me a log format and parse my logs and then do what you will with those logs to provide observability. All my logs are already json, and they're already relying upon convention for quick and easy parsing elsewhere, so adjusting the format isn't a big lift.

The last thing I want in between the innards of my application is communication with some external system (even if it's a sidecar) - especially if there is any sort of networking involved. If that sidecar drops or slows down, so does my application, which is a variable I'd rather not have to consider, and isn't necessarily tracked by the tools themselves.

Re:Unpacking Observability: The Observability Stack by sagichmal on Friday September 10, 2021 @09:59AM (#28481671)
Parsing logs has a performance ceiling that locks you out of ~most of the value that good observability systems should be delivering.
Re:Unpacking Observability: The Observability Stack by enobrev on Friday September 10, 2021 @11:06AM (#28482609)
I've been finding that vector.dev keeps up well. Before I moved to containers, I was using rsyslog, which also kept up incredibly well. I've had both set up to pour directly into elasticsearch, which, if tuned, can keep up as well.