Revert "Revert "feat: Experiment using statsdproxy for aggregation (#…

[lynnagara]

…4734)""

I reverted the original PR was reverted since it seems to cause a 6x reduction in throughput. Re-opening in case we want to try it again.

This reverts commit 201e07b.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (257e690) 90.52% compared to head (de4ca37) 90.52%.
Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #5064   +/-   ##
=======================================
  Coverage   90.52%   90.52%           
=======================================
  Files         880      880           
  Lines       43071    43071           
  Branches      288      288           
=======================================
  Hits        38992    38992           
  Misses       4037     4037           
  Partials       42       42           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[untitaker]

I think it's just this mutex causing trouble, and we'd probably be able to fix this by making the entire metrics client thread-local: https://github.com/getsentry/statsdproxy/blob/07b51e05107882226d94c27207bb968c1181e3d9/src/cadence.rs#L38

[untitaker]

getsentry/statsdproxy#38

[getsentry-bot]

PR reverted: b291670