We had a legacy endpoint crawling under load. Two years ago I would've spent a week fixing it myself.
Instead I looked at the logs, saw an team was hammering it with a cron job, and sent their dev a Slack message asking if they still needed that data.
Got an answer the next day, a simple no and they turned off the job. Latency dropped. Problem solved
The embarrassing part is how many times I've probably dove headfirst into a problem that could be solved with communication
Has anyone else noticed this becoming more obvious the longer they're in the industry?
where people try to fix communication problem with code?
[link] [comments]