Lightning Talk: Using Jupyter Notebooks To Gain Insight Of Your Cluster
summary
Uses JupyterLab notebooks and the Kubernetes Python client to stream kube-apiserver logs, normalizes them with simple Python string ops and stores rows in SQLite, then loads into pandas for analysis and bokeh for visuals. Builds a ~30-line ETL that processes about 122,000 log lines from synthetic tests, finds replica-set and namespace thrash produce roughly six times more errors than other tests, and demonstrates interactive debugging and performance-insight workflows that empirically work.
AI-written summary · about
chapters (6)
related talks
24mJupyter Notebooks for Ops
05m⚡Lightning Talk: Learning Kubernetes From Logs: Building a Foundation for Future…
49mExperimenting with Large Data Volumes Using Notebooks
50mBeyond Interactive: Empowering Everyone with Jupyter Notebooks
28mScheduling Jupyter Notebooks Using Airflow on Kubernetes
45mData Science Apps: Beyond Notebooks
26mBuilding analytical microservices powered by Jupyter Kernels
05mLightning Talk: Flexible Logging Pipelines with Fluentd and Kubernetes