Suppress WorkerTracer warning for duplicate alarm requests

[mar-cf]

No description provided.

[fhanau]

@mar-cf I'm sorry, is this not more or less the same PR as #5577 which has been open for 3 weeks now?

[mar-cf]

@mar-cf I'm sorry, is this not more or less the same PR as #5577 which has been open for 3 weeks now?

It's more or less the same. I never saw tests nor an explanation that made sense. Check the internal PR

[mar-cf]

@fhanau can you either approve this PR or push some changes to the internal PR I opened to support landing the other (update to #5577 + description)

[fhanau]

Code itself LGTM, but there is one complication with downstream usage: We support marking a WorkerTracer as unused, but that does not extend to the RPC equivalent, which is set to just have a no-op markUnused function, it would have to propagate the unused signal. However, that might just be fine if the duplicate alarm scenario cannot happen under the RPC case as you suggested? We'd have to support it in the future though if we end up marking tracers as unused in more cases.

[codspeed-hq]

CodSpeed Performance Report

Merging #5685 will improve performances by 9.56%

_{Comparing mar/stw-alarm (d3ade51) with main (7b5d410)}

Summary

⚡ 1 improvement
✅ 56 untouched
⏩ 34 skipped¹

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
⚡	simpleStringBody[Response]	21.4 µs	19.6 µs	+9.56%

Footnotes

1. 34 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[mar-cf]

However, that might just be fine if the duplicate alarm scenario cannot happen under the RPC case as you suggested? We'd have to support it in the future though if we end up marking tracers as unused in more cases.

Exactly!