Benchmarks

Benchmark execution time: 2024-08-12 23:14:11

Comparing candidate commit 482b48a in PR branch lograge-rails with baseline commit 6b50227 in branch master.

Found 2 performance improvements and 0 performance regressions! Performance is the same for 21 metrics, 2 unstable metrics.

scenario:profiler - Allocations ()

• 🟩 throughput [+165850.392op/s; +173043.914op/s] or [+4.230%; +4.414%]

scenario:profiler - sample timeline=false

• 🟩 throughput [+0.257op/s; +0.396op/s] or [+3.886%; +5.980%]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.84%. Comparing base (6b50227) to head (482b48a).
⚠️ Report is 5232 commits behind head on master.

@@            Coverage Diff             @@
##           master    #3839      +/-   ##
==========================================
- Coverage   97.84%   97.84%   -0.01%     
==========================================
  Files        1264     1264              
  Lines       75682    75696      +14     
  Branches     3720     3723       +3     
==========================================
+ Hits        74053    74061       +8     
- Misses       1629     1635       +6     

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

We've started getting this error message since this update.

However, we already account for this with customisations to the log formatter to output in JSON and include Datadog tags:
Part of our implementation for this is made public here: https://github.com/armstrjare/rails-json-tagged-logging

Can we add a way to disable this warning to stop polluting our startup logs?

@marcotc @TonyCTHsu

Hey @marcotc just wanted to +1 for a way to disable this warning message.

I'm also confused about the exact behavior that's broken. We set up some log remapping that ensures the datadog trace IDs are properly read as trace IDs and we still have log correlation working. Is there anything else that might be missing with lograge that I'm not seeing?

Thank you!

Hey @armstrjare and @alexevanczuk, thank you for reporting this issue. It makes total sense.

I opened #4268 so we can track the work of adding an option to disable this log message.

Thanks @marcotc !! Also let me know if there's anything that lograge makes not possible besides log correlation (which we 'roll ourselves' by putting the trace ID into the lograge logged message)!

Thanks again!

We've started getting this error message since this update.

However, we already account for this with customisations to the log formatter to output in JSON and include Datadog tags: Part of our implementation for this is made public here: https://github.com/armstrjare/rails-json-tagged-logging

Can we add a way to disable this warning to stop polluting our startup logs?

@marcotc @TonyCTHsu

Warning: If upgrading from Rails 7.1 => 7.2 (even if you have a working JSON tagged logging setup using lograge + rails-json-tagged-logging), your logging setup will break, which seems to also cause a memory leak. I wasn't able to debug exactly what broke; I instead switched everything over to rails_semantic_logger

Warning: If upgrading from Rails 7.1 => 7.2 (even if you have a working JSON tagged logging setup using lograge + rails-json-tagged-logging), your logging setup will break, which seems to also cause a memory leak. I wasn't able to debug exactly what broke; I instead switched everything over to rails_semantic_logger

@tcannonfodder I’m still using this in production without issue with a decent request volume - and also was on 7.2 when it was released. Happy to help if you throw an issue over at my repo, but sounds like you were happy to just move to semantic logger anyway 👍