Kubernetes Deployment Validation in Two Minutes
Sai Joshitha Kathari joins Alan Shimel to explain how embedding pod-readiness and workload-health checks in Jenkins reduced her Kubernetes deployment validation time from 45 minutes to two. The post Kubernetes Deployment Validation in Two Minutes appeared first on Cloud Native Now .
A Kubernetes deployment's green result does not assure a healthy application. Pods might enter crash loops, fail to become ready, or destabilize the workload even after the rollout reports success. Manually verifying these conditions involves shifting between clusters, dashboards, and logs. Across numerous clusters, this redundant task can consume engineers' time that would be better spent investigating failures or enhancing reliability.
Sai Joshitha Kathari, a senior site reliability engineer at Visa, joins Alan Shimel to elucidate how she integrated those validations into the deployment pipeline. Building upon her Cloud Native Now article on automating deployment validation, Kathari outlines how she cut validation time from 45 minutes to mere two minutes. Her method begins with the checks engineers already carry out, rather than assuming a completed deployment signifies the application is functioning optimally.
It incorporates validation into the release workflow rather than treating it as a distinct manual exercise post-deployment. The Jenkins pipeline records the cluster state prior to deployment, subsequently scrutinizes pod readiness, failure rates, and workload stability post the rollout of the new version. Instances of crash loops, failed pods, and workloads that fail to emerge are documented in the pipeline logs.
This shift in approach alters where troubleshooting commences. Instead of visiting each cluster to ascertain whether something went amiss, an engineer can pinpoint the affected cluster and utilize the available logs to delve into the issue. Successful deployment and successful validation remain separate outcomes, both of which are crucial before affirming the application's health.
Kathari presents this work as a starting point for other SRE and DevOps teams to adapt, and she encourages enhancements from engineers grappling with similar challenges. The advantage extends beyond merely completing a checklist faster. Embedding validation into CI/CD offers engineers a more insightful account of what transpired during a rollout, directing their focus towards the systems that require attention.
The time saved from repetitive inspections can be directed towards troubleshooting, further automation, and bolstering application reliability.
Written by urgent.news from Container Journal's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
