Download to read offline
Uploaded byYan Cui
Troubleshooting serverless applications
The document discusses the challenges and solutions related to troubleshooting serverless applications. It emphasizes the benefits of serverless architectures such as cost efficiency and reduced management overhead while outlining different monitoring and logging options like CloudWatch, custom solutions, and third-party services. Key takeaways include understanding the importance of identifying issues rapidly to maintain user experience and exploring various tools to effectively manage serverless environments.
More Related Content
Similar to Troubleshooting serverless applications
Troubleshooting serverless applications
- 1.
- 2. I watch alot of TV shows…
- 3.
- 4. 3 hours earlier…protagonist is shot
- 5. 3 hours earlier…protagonist is shot
- 6. 3 hours earlier…protagonist emerges victoriously protagonist is shot
- 7.
- 8.
- 9.
- 10. happened system repaireduserimpact goal: to fail without users noticing
- 11. happened system repaireduserimpact reduce MTTR
- 12. Yan Cui http://theburningmonk.com @theburningmonk Developer Advocate@ Independent Consultant AWS user since 2009 since 2018 yan@lumigo.io
- 13. What do youmean by ‘serverless’?
- 14.
- 16. Gojko Adzic It isserverless the same way WiFi is wireless. http://bit.ly/2yQgwwb
- 17. Serverless means… don’t payfor it if no-one uses it don’t need to worry about scaling don’t need to provision and manage servers
- 18. in other words,it’s a lot like taking a cab
- 19. Ownership Fuel Navigate To get there! Focuson getting there!
- 20. HW Ownership OS Runtime &Scale Code Focus on getting there! Physical Servers Virtual Machines Containers Serverless
- 21. Nano Services SelfManaged Cost Paradigm ChangeAsync Dynamic agile env
- 22. happened system repaireduserimpact reduce MTTR
- 23.
- 24.
- 25. happened system repaireduserimpact MTTDiscovery
- 27. “What alerts shouldI have?”
- 28. It depends onwhat you’re building…
- 29. But, this isa good starting point
- 30. Lambda error rate % throttlecount DLR error count iterator age regional concurrency
- 31. Lambda error rate % throttlecount DLR error count iterator age regional concurrency API Gateway p90/95/99 latency success rate % 4xx rate % 5xx rate %
- 32. API Gateway p90/95/99 latency successrate % 4xx rate % 5xx rate % SQS message age Lambda error rate % throttle count DLR error count iterator age regional concurrency
- 33. API Gateway p90/95/99 latency successrate % 4xx rate % 5xx rate % SQS message age Step Functions failed count throttle count timed out count Lambda error rate % throttle count DLR error count iterator age regional concurrency
- 34. SQS message age Step Functions failedcount throttle count timed out count API Gateway p90/95/99 latency success rate % 4xx rate % 5xx rate % Lambda error rate % throttle count DLR error count iterator age regional concurrency
- 35. “Can’t you codifythese?”
- 37.
- 38. happened system repaireduserimpact finding root cause
- 39. option 1: CloudWatch& friends
- 41.
- 42. Pros Out of thebox No overhead Comparatively cheap AWS support
- 43. Pros Out of thebox No overhead Comparatively cheap AWS support Cons Complicated
- 44.
- 45. Pros Out of thebox No overhead Comparatively cheap AWS support Cons Complicated Hard to query* * Insights improved things drastically, but still a gap to ELK
- 48.
- 49. Pros Out of thebox Source of truth No overhead* Comparatively cheap AWS support * unless you record custom metrics synchronously
- 50. Pros Out of thebox Source of truth No overhead* Comparatively cheap AWS support * unless you record custom metrics synchronously ** can compensate with custom metrics/metric filters, etc. Cons Missing metrics** Lambda percentile latencies don’t work Only granular to 1 min No query language
- 52. Pros Out of thebox SDK No overhead Comparatively cheap AWS support
- 53. Pros Out of thebox SDK No overhead Comparatively cheap AWS support Cons Poor async support
- 55. Pros Out of thebox SDK No overhead Comparatively cheap AWS support Cons Poor async support No auto- instrumentation Bad DX (for node.js) Poor documentation
- 56. option 2: custombuilt solutions
- 57.
- 59.
- 60.
- 61.
- 62. Structured Logging Sampling Correlation IDs Auto“instrumentation”
- 63. Structured Logging Sampling Correlation IDs Auto“instrumentation” Support async events
- 64. enrich the usefulnessof your logs
- 65.
- 66.
- 68.
- 69. Pros Tailor fit Free! Cons Very high-touch Notall services are supported equally Tailor fit (for someone else…)
- 70. option 3: serverlessmonitoring solutions
- 74. Pros SAAS Serverless focus More thanjust tracing Very low touch Cons Yet another 3rd party More than just tracing
- 75. Takeaways Serverless is agame-changer Serverless has challenges Options for troubleshooting serverless applications
- 76.
- 78.