Enrich your logs with Kubernetes metadata using Fluent Bit

A blue hexagon with a white ship wheel in the center, representing Kubernetes, is overlaid on a background of a green circle and circuit-like lines, capturing essential Kubernetes metadata.
ACF Image Blog

Learn how Kubernetes metadata can enhance traceability and reduce MTTR and how Fluent Bit makes it possible.

A person with curly hair stands in front of a train station platform with arched architecture, illuminated signage, and posters on the walls, their hand clutching a book about Kubernetes and Fluent Bit.
Celalettin Calis Member of Technical Staff | Chronosphere
7 MINS READ

Sometimes you love Kubernetes; sometimes you don’t

If you work in application orchestration (microservices, CI/CD, multi-cloud deployments, data pipelines), Kubernetes is likely an essential tool in your tech stack. This robust platform provides new ways for application deployment and scaling and has revolutionized how we manage and scale containerized apps.

Kubernetes may be your best buddy on most days, but it can also produce an overwhelming amount of event data. This can make it difficult to pinpoint issues, especially when you’ve started getting alerts for your workloads, support tickets, and endless notifications on Slack from the stakeholders about service degradation. That’s where logs come in, helping you to identify your system’s point of failure.

If you are using K8s on a public cloud provider, odds are that you are already running Fluent Bit, an open source high-performance data collection agent. Fluent Bit is deployed by default in major Kubernetes distributions, including Google Kubernetes Engine (GKE), AWS Elastic Kubernetes Service (EKS), and Azure Kubernetes Service (AKS). In many cases, the out-of-the-box Fluent Bit instance is configured to route data to the public cloud provider’s backend (e.g. CloudWatch, Azure Monitor Logs, or Google Cloud Logging). What a lot of developers don’t know is that you can also customize Fluent Bit to enrich logs with K8s metadata and route them to multiple locations.

Below you will learn how to take advantage of Fluent Bit’s ability to tap into K8s metadata to enrich logs before routing them to any of the 100+ backends supported by Fluent Bit.

We’ll also discuss the value of Kubernetes metadata for enhancing traceability and faster diagnostics.

A colorful hummingbird logo followed by the text "fluentbit" and an icon of Kubernetes, with the words "Kubernetes logs" underneath.

Understanding Kubernetes metadata

Metadata, in its simplest form, is the “info about the info.” It’s like looking up your favorite movie on IMDB. The information about who’s in it, who directed it, how long it is, and so on — that’s all metadata about the movie

In the context of Kubernetes, metadata contains the details of your running workloads. It covers names, namespaces, labels, annotations, and more.

To illustrate, let’s consider a simple example. Suppose you have a pod named checkout-service-2 in the checkout namespace. This pod might have a label like app=checkout-service and an annotation like product-owner=J. Doe. This information might seem basic, but this information can be a lifesaver during a disaster. Later in this blog, we’ll share some valuable tips and best practices for creating and managing good metadata, aiming to optimize your operations.

Let’s dig a little deeper.

How metadata can reduce MTTR (Mean Time to Remediate)

When your workloads generate extensive logs without the right tools, it can become quite challenging to manage the data. This is where metadata becomes invaluable.

Let’s say you have received an alert about higher-than-normal response times in your e-commerce application. Naturally, you’d begin to check your logs. Since your e-commerce application runs on multiple pods, all logging simultaneously, it’s not easy to isolate the problem.

This is where the Kubernetes metadata can help. By sorting sample logs by response time, you can determine that the endpoints are related to the checkout service. The app=checkout-service label on the pod will enable you to quickly isolate logs associated with this particular service in your Kubernetes cluster.

The Kubernetes metadata will enable you to dig even deeper. By looking at the metadata, you are able to determine that each slow request is associated with a particular pod, checkout-service-2. Now that you have narrowed down the issue to a specific pod, you uncover that this pod had been scheduled to a node in a bad state. A quick reschedule later, your e-commerce application is back to running properly.

Using metadata to navigate your logs will help your operations to become more efficient and streamlined, reducing downtime and minimizing disruption to the user experience. This proactive approach will enable you to anticipate issues before they become significant problems, further increasing your manager’s confidence in your abilities and the robustness of the system.

Automating Kubernetes metadata with Fluent Bit

Fluent Bit, the lightweight and highly performant log processor and forwarder, excels when dealing with Kubernetes metadata. Fluent Bit is designed to automatically enrich your logs with Kubernetes metadata, tagging each log entry with helpful information such as the pod name, namespace, labels, and more.

In addition to capturing and enriching logs, Fluent will ship your logs, fully loaded with metadata, to your preferred log analysis platform. Fluent Bit has more than 100 built-in integrations, including Elasticsearch, AWS Cloudwatch, and more.

In the context of Fluent Bit, Filters are components that allow you to manipulate and process the data logs before they are sent to output. Fluent Bit Kubernetes Filter allows you to enrich your log files with Kubernetes metadata.

After adding the Kubernetes Filter block to your Fluent Bit configuration, this filter aims to perform the following operations:

  • Analyze the Tag and extract the following metadata:
    • Pod Name
    • Namespace
    • Container Name
    • Container ID
  • Query the Kubernetes API Server to obtain extra metadata for the POD in question:
    • Pod ID
    • Labels
    • Annotations

The data is cached locally in memory and appended to each record.

You may find the example configuration below:

[INPUT]
    Name            tail
    Tag             kube.*
    Path            var/log/containers/*.log
    Parser          docker

[FILTER]
    Name            kubernetes
    Match           kube.*
    Kube_URL        https://kubernetes.default.svc:443
    Kube_CA_File    /var/run/secrets/kubernetes.io/serviceaccount/ca.crt
    Kube_Token_File var/run/secrets/kubernetes.io/serviceaccount/token
    Kube_Tag_Prefix kube.var.log.containers.
    Merge_Log       On
    Merge_Log_Key   log_processed

Metadata best practices: Labels & Annotations

Now that you’ve seen the importance of Kubernetes metadata in action, let’s share some best practices.

Use labels to categorize your Kubernetes objects. Some examples of label best practices include:

  1. Create a label to group all the components of a specific application running across different microservices.
  2. Create a label to distinguish the environment of particular workloads.

Annotations, meanwhile, are like Post-it notes on your workloads. They’re excellent for storing extra details like a description of the workload, the person in charge, etc. In essence, labels are for identifying, and annotations provide additional context.

Conclusion: Embrace the metadata

We’ve discussed Kubernetes metadata and how it brings value to your logging process. We’ve explored how it turns the challenging task of troubleshooting into a straightforward process. Additionally, we’ve seen how Fluent Bit automates the enrichment of your logs with metadata and ships them to your preferred log analysis platform, such as Chronosphere Observability Platform, Datadog, Splunk, or AWS S3.

Just remember, as powerful as Kubernetes metadata is, it’s only as good as your usage. Use labels and annotations effectively. Enrich your logs with as much context as possible. Make certain that your tooling preserves the information as the logs are routed downstream. And above all, embrace the metadata.

Learn more about Fluent Bit and Kubernetes

A promotional image displaying an offer for a free copy of the book “Fluent Bit with Kubernetes.” The book cover shows an illustration of a person and MEAP stamped on it. Logos of Chronosphere and Calyptia are also visible.

We invite you to download a free copy of Fluent Bit with Kubernetes by Phil Wilkins. This practical guide to monitoring cloud-native and traditional environments with Fluent Bit covers the basics of collecting app logs, filtering, routing, enriching, and transforming logs, metrics, and traces.

About Fluent Bit and Chronosphere

With Chronosphere’s acquisition of Calyptia in 2024, Chronosphere became the primary corporate sponsor of Fluent Bit. Eduardo Silva — the original creator of Fluent Bit and co-founder of Calyptia — leads a team of Chronosphere engineers dedicated full-time to the project, ensuring its continuous development and improvement.

Fluent Bit is a graduated project of the Cloud Native Computing Foundation (CNCF) under the umbrella of Fluentd, alongside other foundational technologies such as Kubernetes and Prometheus. Chronosphere is also a silver-level sponsor of the CNCF.

Share This:
Table Of Contents

Ready to see it in action?

Request a demo for an in depth walk through of the platform!