Observability tips with HAProxy
summary
Covers practical observability for HAProxy by focusing on four actionable metrics: connection errors, processing times, termination codes, and status distribution. Recommends using the HAProxy stats page, built-in logs, stick tables and selective sampling, enabling proxy processing-time and shipping-time logs to locate slow services, and gives concrete operational tips such as correlating connection times with MACs to find a bad switch and retaining a week of compressed logs for investigating intermittent faults.
AI-written summary · about
chapters (12)
- 0:00HAProxy observability
- 1:49Load balancers for observability
- 3:17HAProxy logs and metrics
- 5:51Request times in HAProxy
- 7:15Anomalies via HAProxy status
- 9:02Load balancing fairness
- 10:26Custom logs for tracking
- 11:32Logging volume vs sampling
- 12:35HAProxy troubleshooting
- 14:00HAProxy logs case studies
- 15:45Log retention best practices
- 16:49Summary
related talks
39mHow to Use Common Investigation Practices in Troubleshooting
24mGremlins Exposed: Shining a Light on Mischievous Systems
33mWhat Developers Should Know About Observability
38mPractical Guide to Application Observability in 10 Steps
50mObservability and the Development Process
38mOff the Beaten Path: Moving Observability Focus from Your Service
55mBeginner's guide to observability
20mNetwork Monitor: A Tale of ACKnowledging an Observability Gap