Add blog post: Observability at the Edge: OpenTelemetry Support in k8s ingress #7959
Conversation
…bernetes Ingress Controllers Signed-off-by: Kasper Borg Nissen <[email protected]>
Signed-off-by: Kasper Borg Nissen <[email protected]>
There was a problem hiding this comment.
overall LGTM, note that we can not allow links out to your companies' blog. I know that the blog posts provide a deep dive and are enriching the content, but we need to avoid situations that create questions on later blog posts why this is allowed here and not on other blogs.
| Kubernetes has transformed the way applications are deployed and scaled, but one | ||
| component remains especially critical: the ingress controller. Ingress sits at | ||
| the edge of the cluster, where it terminates TLS, applies routing rules, | ||
| enforces policies, and directs requests to the correct backend. |
There was a problem hiding this comment.
I think it would make sense to point out to some k8s docs around ingress controller in general to allow people to read more on the topic if they have questions.
content/en/blog/2025/observability-at-the-edge-otel-support-in-k8s-ingress.md
Show resolved
Hide resolved
| Ingress-NGINX is the veteran among ingress controllers. Maintained under | ||
| Kubernetes SIG Networking, it quickly became the default in many distributions |
There was a problem hiding this comment.
link to the SIG if they have a place to learn more about them.
content/en/blog/2025/observability-at-the-edge-otel-support-in-k8s-ingress.md
Outdated
Show resolved
Hide resolved
content/en/blog/2025/observability-at-the-edge-otel-support-in-k8s-ingress.md
Outdated
Show resolved
Hide resolved
content/en/blog/2025/observability-at-the-edge-otel-support-in-k8s-ingress.md
Outdated
Show resolved
Hide resolved
| _Read the full deep dive on | ||
| [observing Traefik with OpenTelemetry](https://www.dash0.com/blog/observing-traefik-with-opentelemetry-and-dash0). | ||
| While the post uses Dash0 in the examples, the same setup works with any | ||
| OTLP-based backend._ |
There was a problem hiding this comment.
| _Read the full deep dive on | |
| [observing Traefik with OpenTelemetry](https://www.dash0.com/blog/observing-traefik-with-opentelemetry-and-dash0). | |
| While the post uses Dash0 in the examples, the same setup works with any | |
| OTLP-based backend._ |
content/en/blog/2025/observability-at-the-edge-otel-support-in-k8s-ingress.md
Outdated
Show resolved
Hide resolved
| a torrent of Prometheus series, Ingress-NGINX exposes a smaller set of | ||
| NGINX-specific metrics, and Traefik embraces OTLP natively. Logs remain the | ||
| hardest signal. They require customization in Ingress-NGINX, parsing in Envoy, | ||
| and in Traefik they are still maturing as an OTLP feature. |
There was a problem hiding this comment.
What does "OTLP feature" mean here?
content/en/blog/2025/observability-at-the-edge-otel-support-in-k8s-ingress.md
Outdated
Show resolved
Hide resolved
|
@kaspernissen one more remark, you raised this PR from the dash0hq github org, this means we (maintainers) can not edit your PR for final refinement, which means you have to do if needed. |
|
Thank you for all your guidance, @svrnm! Do you think it would make sense to adapt the other posts and make them less vendor specific so that we have don't loose the background for this post? |
…-k8s-ingress.md Co-authored-by: Severin Neumann <[email protected]>
…-k8s-ingress.md Co-authored-by: Severin Neumann <[email protected]>
…-k8s-ingress.md Co-authored-by: Severin Neumann <[email protected]>
…-k8s-ingress.md Co-authored-by: Severin Neumann <[email protected]>
…-k8s-ingress.md Co-authored-by: Severin Neumann <[email protected]>
…-k8s-ingress.md Co-authored-by: Severin Neumann <[email protected]>
…-k8s-ingress.md Co-authored-by: Severin Neumann <[email protected]>
…-k8s-ingress.md Co-authored-by: Severin Neumann <[email protected]>
Hey @kaspernissen sorry for the delay, I was traveling all week and hadn't checked in with all of my github! Yes, that would even be better! We had some "series" of blog posts already, see @jpkrohling's last blog posts about naming things. He did something similar. |
|
No worries. I've just been had 2 conferences back to back as well. I will add it to my todo to open PRs for the other blog posts. Would it be OK to link to this repo (https://github.com/dash0hq/dash0-examples) for the examples? |
Yes, I think so, as it helps people to get started with what you are sharing more easily. I've done similarly in my last blog post. If you add the "and here is how it works with jaeger/prometheus/..." and people can comment out the dash0 specific stuff, that would be great, e.g. having a collector config like this one https://github.com/dash0hq/dash0-examples/blob/main/traefik/collector/otel-collector-deployment.yaml with those extra details. |
|
Perfect! Yes, I can add a OSS observability stack as well. |
What would the order of the blog post be then, first the individual ones and then the summary? |
|
Yes, it would be the individual ones first. That would make most sense. 👍 |
Put it on |
2 participants
This PR adds a blog post on my recent exploration of OpenTelemetry support in k8s ingress controllers.
This post was originally published on the Dash0 blog: https://www.dash0.com/blog/observability-at-the-edge-opentelemetry-support-in-kubernetes-ingress-controllers
But I think it's an interesting write-up for the wider community.