Log Router Design and Data Flow with Fluentd

5th chunk of `content/en/blog/_posts/2015-06-00-Cluster-Level-Logging-With-Kubernetes.md`

50ceacc7352c168ac0d5cf0357328ca0a31a837e9dbeead50000000100000169

![](https://4.bp.blogspot.com/-UXOxauNy8FQ/Vxf0SaGujNI/AAAAAAAAAb0/Pnf6e_iiUfoKkooGyrF3Gmd8wh0vPrteQCLcB/s400/pk22.png)](https://4.bp.blogspot.com/-UXOxauNy8FQ/Vxf0SaGujNI/AAAAAAAAAb0/Pnf6e_iiUfoKkooGyrF3Gmd8wh0vPrteQCLcB/s1600/pk22.png)  

[![](https://2.bp.blogspot.com/-UgpwCx4BNwQ/Vxf0Wc8-HwI/AAAAAAAAAb4/g3D1bE74FQA2k9uwc9ZbZuB1N7MTU7swgCLcB/s400/20ej.png)

Title: Log Router Design and Data Flow with Fluentd

Summary

The first image illustrates a log router design using Fluentd, where multiple Fluentd instances forward logs to a primary Fluentd aggregator, which then sends the aggregated logs to Stackdriver. The second image visually represents how data flows through Fluentd, showing the input, buffer, and output stages involved in log processing and routing.