2
votes

Currently, we have a clustered app running Spring Boot(1.5.2). On occasion, the health checks hitting the /health actuator endpoint responds with a 503. However, there is nothing in the logs to indicate why the service is unavailable. Is there any way to get a more accurate reason on why the 503 is being produced?

1
Have you looked at this: stackoverflow.com/questions/30736514/… - jny
Yup, It is hard to capture what exactly is going on. I tried changing the logging level for the health package to debug but nothing is logged. Perhaps the load balancer can capture some more information. I looked at the source code for Actuator on GitHub and did not see any logging in the health check endpoint either. - Andrew Wanczowski

1 Answers

0
votes

So, long story short. I set up a small shell script that pings the health endpoint over and over. This caused the actuator endpoint to log to debug. It turns out we had a small connection timeout with one of our connectors. Hope this helps.