Open Source Solution for Monitoring Kubernetes pod and namespace traffic without heavy-weight scraping tools We have some exciting news to share - we've just released our open source project, Podloxx.
If you use Kubernetes, you're probably familiar with Prometheus or similar tools for collecting traffic data on pods and namespaces. These tools are great, but we wanted to find a more lightweight alternative. As developers and DevOps professionals, we're always trying to figure out which technical decisions to make ourselves and which to leave to third-party vendors. So, we came up with a solution to efficiently gather traffic data on pods and namespaces. We found it so useful that we decided to open source it and share it with the community.
So, what's the problem Podloxx solves? Well, sometimes we want to check the health and data of individual pods or entire namespaces in a Kubernetes cluster. Maybe overall performance is suffering, or we're seeing unusual traffic spikes that we can't explain. That's exactly what happened to us recently. We were struggling to identify the source of the problem because we had limited visibility. It turned out that one of our internal services was creating and provisioning more pods than necessary in a constant loop, causing a huge amount of traffic on our load balancers. But none of our alerting tools had picked up on this issue.
We hope Podloxx will be helpful for other developers and DevOps professionals facing similar challenges. Feel free to check it out and contribute to the project if you'd like!
So we decided to set up something to help us better manage these challenges in the future. And we had a few key goals in mind:
Instant provisioning of traffic data within a Kubernetes cluster Fast and simple - not another complex tool with a long learning curve Low cluster resource consumption No licensing issues (Apache 2.0) That's why we created Podloxx, an open-source web UI for monitoring the traffic of entire namespaces and individual pods in a Kubernetes cluster. It gives you an instant snapshot and visualization of traffic over time. Some of the features include the ability to:
Visualize pod and namespace traffic in a Kubernetes cluster Distinguish between cluster-internal and internet traffic Get instant insight from the first use Filter by namespace, pods, and traffic type to narrow down results API access One of the advantages of Podloxx compared to other tools is that it's very lightweight and doesn't require a lot of storage or compute. It also doesn't rely on regular traffic scraping.
How to install Just download the standalone CLI:
And run it
When the start commands finished successfully, a browser window will open up and you can start using podloxx when it has successfully received the first dataset (you need to wait a few seconds on first startup).
(Manual: http://localhost:1337/podloxx/#/dashboard )
We hope Podloxx will be helpful for other developers and DevOps professionals facing similar challenges. Feel free to check it out and contribute to the project if you'd like!
You can contribute directly in the GitHub repo: https://github.com/mogenius/podloxx
See License for usage here: https://github.com/mogenius/podloxx/blob/main/LICENSE
