Monitoring for devs. The necessity or self-indulgence
summary
Argues that developers must own application monitoring, not just admins, and shows concrete fixes: detect a memory leak that forces a restart every 15–16 days, spot growing log volume that reveals a string-construction hotspot, and find one slow instance among 35. Recommends JMX and JVM agents, Spring interceptors and simple grep-based log aggregation, push metrics from containers (Graphite/Influx/ELK/Papertrail/New Relic), and focus on 95/99th-percentile latency rather than averages to catch real user-impacting tails.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 7:36System vs app monitoring
- 12:12App monitoring by devs
- 18:38Cloud infra and logs
- 26:38Devs' role in monitoring
- 31:40Access limits and tools
related talks
21mMonitoring Your Apps is not About Devops
01h 10mObservability For (Java) Devs - 2023 Edition
05mA 5 Minute Checklist for Application Monitoring
30mHow to collect application metrics if it needs to be done yesterday
50mYou build it, you run it (why developers should also be on call)
39mPanel Discussion: Application Monitoring
42mEffective App Monitoring in Production
31mObservability for Developers: How to Get There