Jonathan Perry

> I think a change to the build tool repo could definitely help. Actually, in the build pipeline you launched on the branch for the PR I made (thank you...

> btw I think this [slack message](https://cloud-native.slack.com/archives/C02AB15583A/p1713986839962629) is referring to the same issue? for reference, also [this thread on #262](https://github.com/open-telemetry/opentelemetry-network/pull/262#issuecomment-2059266876)

@jakub-racek-swi can you make a PR on the build-tools repo? Now that we merged the DNS patch, we're blocked on the build repo to release. I'm asking since you probably...

Yes there should be an artificial upper limit to the number of CPUs, that is used to allocate some static memory in the kernel collector. There is no inherent limitation...

After some further investigation: - [PerfContainer](https://github.com/open-telemetry/opentelemetry-network/blob/d7441104b8739785466b46f3d9f777c1d39917a2/collector/kernel/perf_reader.h#L83) holds the perf rings in `readers_`. - [PerfContainer.entries_](https://github.com/open-telemetry/opentelemetry-network/blob/d7441104b8739785466b46f3d9f777c1d39917a2/collector/kernel/perf_reader.h#L99C22-L99C34) is a heap that holds indexes of non-empty readers, so messages can be dequeued in-order. It...

cc @open-telemetry/network-maintainers @open-telemetry/network-approvers @open-telemetry/network-triagers

@atoulme fyi

Hi @kubakukis14 , awesome you're tackling this! I see the PR is still in draft but thought I'd see if I can help out. You probably noticed this, but just...

I also triggered a [build run](https://github.com/open-telemetry/opentelemetry-network/actions/runs/8584438547), if helpful to your efforts

Thank you for persevering, sorry for missing your comment. The specification for generated code is in the render directory. This might be the line: https://github.com/open-telemetry/opentelemetry-network/blob/d7441104b8739785466b46f3d9f777c1d39917a2/render/ebpf_net.render#L1328 The render language generated both...