Monitoring and Troubleshooting Spring Boot Microservices Architecture
summary
A Spring Boot deployment on PCF crashes under load; logs, App Manager and PCF Metrics plus New Relic and a JMeter load test reveal heap dumps look normal but native memory exhaustion occurs. Apache MAT and JVM heap analysis fail to show the problem; fixing requires raising container memory, increasing thread count, reducing per-thread stack and redistributing memory allocations so the app stops crashing. PCF Metrics 1.10 aims to unify metrics, logs, events, alerts and tracing out of the box and auto-correlates traces via Spring Cloud Sleuth.
AI-written summary · about
chapters (25)
- 0:00Introduction
- 1:12New Application
- 2:24Musthaves
- 4:54Growth
- 7:30PCF Metrics
- 8:10Root Cause
- 8:43Memory Analyzer
- 9:09heap usage
- 9:36native memory
- 9:58running out of native memory
- 11:16conclusion
- 11:55recap
- 12:20pivotal mission
- 12:56metrics mission
- 16:29observability unified
- 17:21observability
- 17:39container and router
- 18:16logs
- 18:43events
- 19:11alerts
- 20:35Distribute tracing
- 22:01Multidimensional metrics
- 24:09DevOps
- 24:37Golden Performance Signals
- 26:03Metric Starter
related talks
37mJava Microservices at Scale on Cloud Foundry, Today and Tomorrow
49mMonitor Your Microservices with Logs, Metrics, Pings and Traces
44mWelcome to the Metrics
42mMonitoring your Spring Boot & Micronaut microservices with Micrometer
51mI Can See Clearly Now: Observability of JVM & Spring Boot 2-3-4 apps
53mScaling Spring Boot Microservices
29mIntro to Development of Java Spring Microservices with Cloud Foundry
36mTools that will help you survive in the micro-service architecture