Thomas Eizinger

refactor: use `target_arch = "bpf"` for `panic_handler`

4

Related: https://github.com/aya-rs/book/issues/214

Recommendation around panic-handler and use of default-members

19

Instead of using `default-members`, I opted for the following in my eBPF code: ```rust #[cfg(target_arch = "bpf")] extern crate panic_halt; // Defines a no-op panic handler (we always compile with...

refactor: use `target_arch = "bpf"` for `panic_handler`

1

Related: #214

Bump to opentelemetry 0.29

1

Supersedes #32 & #34.

Replaces https://github.com/NAlexPear/tracing-stackdriver/pull/31.

Should `system.network.dropped` really only be about "no error"?

1

### Area(s) area:network ### What's missing? The description here currently says: > Count of packets that are dropped or discarded even though there was no error Yet, I believe this...

feat(apple): use `.zip` for logs

6

This PR replaces the use of Apple Archive with an API that allows us to zip the log file contents. This API doesn't handle symlinks well so we move the...

Don't use .aar to pack logs

5

The .aar format is propietary and can't be unpacked on other platforms. We should use (g)zip or something else that is well-supported on all platforms (especially Linux) instead.

From #9525: > If a direct connection fails (15s) and we immediately re-establish it, does the TCP connection on top also fail, or does it heal? (i.e. are the NAT...

We need a self-monitoring module that raises an issue as soon as we use more CPU or memory than expected. https://docs.rs/perf_monitor/latest/perf_monitor/index.html