This package assembled a logging path from Kubernetes to Elasticsearch. Fluentd collected the records, used its plugins to parse them and add context, then sent them to the search database where users could investigate them. It was useful for centralizing logs without changing every application. The bundled image included particular plugins, so reproducing that behavior requires checking the whole pipeline rather than simply choosing another image named Fluentd.
Deployment and operating notes
The old stable chart bundled a particular Fluentd image with Elasticsearch output, Kubernetes metadata, exception detection and other plugins. Fluentd’s current project chart and Elasticsearch output documentation are the appropriate starting points, but they do not guarantee that every historical plugin is installed or configured identically.
Inventory Ruby and plugin versions, parsers, tags, index naming, authentication, TLS, buffer paths and retry settings. Check the output plugin against the target Elasticsearch version and its index or data-stream expectations. Test multiline exceptions, rotated container logs and rejected records using the actual runtime log format. Preserve tail-state and buffered chunks through a controlled transition or plan an explicit replay window. Compare indexed document counts and mappings while watching for duplicates and ingestion errors. Keep credentials scoped to the intended indices. Removing the old DaemonSet before draining its buffers can lose records, while leaving both collectors active indefinitely can double ingestion; choose and verify the cutover behavior.
Historical upstream link check · 2026-10-09
The recorded upstream address responded successfully (HTTP 200) on 2026-10-09. GitHub confirms that helm/charts is archived: this is a historical chart distribution, not evidence that the application itself is retired. Link availability does not certify the historical installation instructions or current security support.
Website availability is separate from project, chart and image support. Use the current guidance and primary sources on this page to assess the distribution.
Historical Kubedex content
Preserved for context. Commands, versions, prices and results below reflect the original research.
This chart bootstraps a Fluentd daemonset on a Kubernetes cluster using the Helm package manager. It’s meant to be a drop in replacement for fluentd-gcp on GKE which sends logs to Google’s Stackdriver service but can also be used in other places where logging to ElasticSearch is required. The used Docker image also contains Google’s detect exceptions (for Java multiline stacktraces), Prometheus exporter, Kubernetes metadata filter & Systemd plugins.
Fluentd
Fluentd is an open source data collector for unified logging layer.
Fluentd allows you to unify data collection and consumption for a better use and understanding of data.
Elasticsearch Output Plugin
The out_elasticsearch Output plugin writes records into Elasticsearch. By default, it creates records by bulk write operation. This means that when you first import records using the plugin, no record is created immediately.
The record will be created when the chunk_keys condition has been met. To change the output frequency, please specify the time in chunk_keys and specify time key value in conf.
Moreover, Elasticsearch, Fluentd, and Kibana (EFK) allow you to collect, index, search and visualize log data. This is a great alternative to the proprietary software Splunk, which lets you get started for free, but requires a paid license once the data volume increases.
Sources & further reading
Spotted something that needs another look?
Help improve this page →