Prometheus Remote-Write receiver Alpha

[dashpole]

Component(s)

receiver/prometheusremotewrite

Describe the issue you're reporting

• Handle empty otel_scope_name and otel_scope_version
• Handle Timeseries samples
• Handle Counters
• Handle Histograms
• Handle Summaries (maybe too hard and could be left out for Alpha)
• Handle Created Timestamp
• Return number of samples and histograms ingested via response headers
• Handle multiple timeseries with same metric name+type+unit
• Handle target_info

[github-actions]

Pinging code owners for receiver/prometheusremotewrite: @dashpole @ArthurSens. See Adding Labels via Comments if you do not have permissions to add labels yourself. For example, comment '/label priority:p2 -needs-triaged' to set the priority and remove the needs-triaged label.

[adwait-godbole]

Hi @ArthurSens! Hi @dashpole! I am eager to be a part of this project and looking forward to work under your mentorship and guidance. Are there any pre-requisites / pre-tests for the selection process of mentees ?

[ArthurSens]

Hi Adwait, glad to see your interest. There are no pre-requisites or tests, just make sure to apply following the LFX process. Also be aware that our proposal is still just a PR, we still need both CNCF and LFX to approve the mentorship

[PiyushRajDev]

Hello @ArthurSens ! I am looking forward to contribute to this project. Can you please guide me on how to get started?

[ArthurSens]

Hello @ArthurSens ! I am looking forward to contribute to this project. Can you please guide me on how to get started?

A good starting point is understanding Remote-Write and the OTLP->Prometheus compatibility spec. If you few comfortable tackling some of the checkboxes above, that's also nice :)

[hanshal101]

Hey @ArthurSens, Hanshal here. I am an open-source developer contributed to various projects like Buildsafe, Glasskube, Cyclops and a couple of Apache's too. I am currently interning at CloudRaft as a DevOps Intern. I recently published a blog on OTel Auto Instrumentation, and my curiosity towards observability brought me here. I'm looking forward to working on this issue if it gets accepted.
I am pretty much aware of the OTLP -> Prometheus spec; what else should I focus on more to understand the issue in a better way?

Thank you.

[ArthurSens]

Hi Hanshal, this project is quite simple so we're not requiring that much experience, to be honest. There's nothing else to be familiar with besides the two protocols that were already mentioned: Prometheus Remote Write and OTLP.

We'd love to mentor someone who hasn't had a chance with OSS yet and is excited to participate in the community, that's all.

Anyone interested, just make sure you follow the LFX process correctly.

[PiyushRajDev]

Hello @ArthurSens ! I have read through the suggested documentation and familiarized myself with the protocols. As I dive deeper into the project, I may encounter some challenges along the way. Would it be possible to join a Slack channel or another communication platform where I can ask quick clarifying questions when needed? I understand you’re busy, so I really appreciate any help or pointers you can offer.

[pratikgitss]

I agree with @PiyushRajDev, it would be great to have a communication channel as I have also gone through the protocols and I am figuring out how to tackle the requirements mentioned above and will be really helpful if there is a channel to get the questions clarified.

[ArthurSens]

Hello hello 👋 -- This issue is a great communication channel! Feel free to ask questions here :)

But yes, there is also the CNCF slack. We have a channel called #otel-prometheus that can also be used to ask questions. Asking it here would be a lot better though, since the information would be accessible to all.

[sonalgaud12]

Hey everyone, I'm Sonal! Recently, I've been contributing to the OTEL Collector and OTEL Collector Contrib, specifically under the Receiver package. I'm really excited about this project and believe there's so much to learn from it. By the way, are there any upcoming meetings or briefings about the project? I'd love to contribute more!

[PiyushRajDev]

Hello @sonalgaud, there are not any meeting or briefings as such but we need to just apply through the LFX process for this issue.

[Brant-lzh]

Hello, I just need this feature and I am very eager to participate in this project, so what do I need to do now?

[PiyushRajDev]

Hey @Brant-lzh, for getting started you need to refer to one of the messages sent by @ArthurSens up above on this thread where he sent links of the two protcols for getting started.

[Brant-lzh]

Hey @PiyushRajDev ，I understand these contents. I have a certain understanding of prometheus and opentelemetry. How should I start now? Is there a working group?

[sonalgaud12]

Hey @Brant-lzh , this project will be developed under the LFX mentorship program. You’ll need to apply for it, and the selected candidate will be working on it. For communication, you can join the CNCF Slack channel named otel-prometheus.

[Brant-lzh]

Hey @sonalgaud12 , I joined the Slack Channel, and how to apply for this LFX mentorship program? Is there any relevant documentation?

[sonalgaud12]

Yes @Brant-lzh , refer to this: https://github.com/cncf/mentoring/tree/main/programs/lfx-mentorship/2025/01-Mar-May

[Brant-lzh]

@sonalgaud12 thx!

[PiyushRajDev]

Hey @ArthurSens , Could you please point me to any existing implementations or helper functions in the code that follow a similar structure to addCounterDatapoints? It would be helpful to review how similar cases are handled elsewhere in the code for consistency.

[ArthurSens]

It should be very very similar to addGaugeDatapoints, but I wouldn't implement this first though, let's first handle the scope attributes or the samples.

[sonalgaud12]

Hey @ArthurSens , for the first to do I could think of two possible solution.

1. Use Collector Build Tags
2. Use a Default Fallback Value (If Build Tags Are Not Available)
   As the comment suggest the collector build approach do let me know which one will be better I'll raise an PR.

[ArthurSens]

Whoops, this was closed by accident