Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Launch HN: Odigos (YC W23) – Instant distributed tracing for Kubernetes clusters

news.ycombinator.com

Post reply on HN
Print view View original

31–40 of 53 posts

Re: Launch HN: Odigos (YC W23) – Instant distributed tracing for Kubernetes clusters

#31

Post by mdaniel »

Congratulations on the launch, and thank you for choosing an awesome license!

For an unrelated reason, today I was reminded about Pixie (https://news.ycombinator.com/item?id=25375170 and https://news.ycombinator.com/item?id=31687978 and https://github.com/pixie-io/pixie#readme ), which says is also an ebpf kubernetes observability tool, also Apache licensed.

I suspect the difference may be your aspirations to move out of just kubernetes, but I wondered if that's the biggest difference between your project and theirs? Or maybe the C++ versus golang?

Re: Launch HN: Odigos (YC W23) – Instant distributed tracing for Kubernetes clusters

#32

Post by edenfed »

yashap wrote: post #30
edenfed wrote: post #29

Earlier quoted context omitted.

We are actually able to handle the long tail of tracing by leveraging the amazing open source community. For languages like Java we use the automatic instrumentation created by the OpenTelemetry community which is really great and support ton of libraries, you can see a list of supported libraries here: https://github.com/open-telemetry/opentelemetry-java-instrum... This also allows us to support async tracing like d…

Ah cool! So like, if I used some Open Telemetry libs for more manual instrumentation, would it "play nice" with the automated instrumentation? Like say: - I instrument a Scala app with Odigos, and it handles say 90% of the metrics, trace spans, etc. that I want - But I want to add some extra spans, extra metrics - If I then explicitly add OpenTelemetry libs as dependencies, will they conflict with the automated OpenT…

Yes exactly. Odigos plays nice with manual instrumentation, meaning distributed traces will include both automatic and manual spans. Currently there is no way to point manual instrumentation to the destination selected in Odigos but we working on it and should release it soon. Most SDKs have a concept of no-op exporter that way Odigos will be able to pick up the manually created traces and deliver them to the chosen backend

Re: Launch HN: Odigos (YC W23) – Instant distributed tracing for Kubernetes clusters

#33

Post by edenfed »

mdaniel wrote: post #31

Congratulations on the launch, and thank you for choosing an awesome license! For an unrelated reason, today I was reminded about Pixie ( https://news.ycombinator.com/item?id=25375170 and https://news.ycombinator.com/item?id=31687978 and https://github.com/pixie-io/pixie#readme ), which says is also an ebpf kubernetes observability tool, also Apache licensed. I suspect the difference may be your aspirations to move o…

As far as I know Pixie use eBPF for generating metrics. Odigos is focused on generating distributed traces which is a different signal that spans across multiple applications

Re: Launch HN: Odigos (YC W23) – Instant distributed tracing for Kubernetes clusters

#36

Post by yashap »

edenfed wrote: post #32
yashap wrote: post #30

Earlier quoted context omitted.

Ah cool! So like, if I used some Open Telemetry libs for more manual instrumentation, would it "play nice" with the automated instrumentation? Like say: - I instrument a Scala app with Odigos, and it handles say 90% of the metrics, trace spans, etc. that I want - But I want to add some extra spans, extra metrics - If I then explicitly add OpenTelemetry libs as dependencies, will they conflict with the automated OpenT…

Yes exactly. Odigos plays nice with manual instrumentation, meaning distributed traces will include both automatic and manual spans. Currently there is no way to point manual instrumentation to the destination selected in Odigos but we working on it and should release it soon. Most SDKs have a concept of no-op exporter that way Odigos will be able to pick up the manually created traces and deliver them to the chosen…

Very cool, ty for the responses!

Re: Launch HN: Odigos (YC W23) – Instant distributed tracing for Kubernetes clusters

#37

Post by edenfed »

allisdust wrote: post #34

Is it possible to use this for non kubernetes setups (for example, a single docker container or a single server).

Not yet, but i can setup some custom docker compose yaml for you depending on the programming language you are using

Re: Launch HN: Odigos (YC W23) – Instant distributed tracing for Kubernetes clusters

#38

Post by allisdust »

edenfed wrote: post #37
allisdust wrote: post #34

Is it possible to use this for non kubernetes setups (for example, a single docker container or a single server).

Not yet, but i can setup some custom docker compose yaml for you depending on the programming language you are using

I'm using Rust (with warp framework if it helps). I can help test if thee is a docker compose :)
Post reply on HN