Jupyter as Incident Response Tool
summary
Jupyter is used as an incident response console that queries Prometheus for five minutes of data at 10s resolution, finds the largest amplitude among four instances (identifies object 4), maps that label to an EC2 instance via boto3 and Paramiko, SSHs in, finds and removes a core file, and verifies p90 latency returns to normal. The notebook and screenshots get duplicated, annotated, and export notebook to HTML for the retrospective; automation and reproducible queries work and the fix succeeds.
AI-written summary · about
chapters (16)
- 0:00Intro
- 0:24Acknowledgement of Country
- 1:02Jupyter Kernels
- 1:51Data Science Origins
- 3:11Imaginary Stack
- 4:09Symptom
- 4:37Spoiler
- 7:27Query Data
- 8:20Analyze Data
- 9:44Boto3
- 11:04We Know The Culprit
- 11:34Connect to Host
- 13:01Confirm the Problem
- 13:22Fix the Problem
- 14:42Post-Incident
- 18:58Final Thoughts
related talks
24mJupyter Notebooks for Ops
49mBest tools for Jupyter notebooks
01h 00mInteractive Computing with Jupyter: Past, Present, and Future
36mThe Lifecycle of a Jupyter Environment
01h 16mMake your research interactive with Jupyter Dashboards
19mJupyterHub: A "things explainer overview"
17mNotes on Notebooks: Is Jupyter the Bringer of Jollity?
33mSignificant Roadblocks to Usefulness for Jupyter Notebooks and a Recipe to Fix them