Lauri Tulmin

Are you sure this is because virtual fields or just guessing? To me it looks more likely that you somehow end up serializing request attributes map.

If things are really slow only when debugger is attached then try deleting all method breakpoints you have, line breakpoints are fine.

> To get there, a first step will be to change the default of `otel.javaagent.experimental.indy` from `false` to `true`. This will allow us to evaluate the approach in a safe...

@PeterF778 do you have any guidance on how the metrics should be named. The existing ones aren't particularly consistent and use `CamelCase`, `camelCase` and `dot.separated`.

With current release workflows it is much easier to do patch releases in 1.x series. We would need to resolve the automation issues if we want to bump minor version...

> Do you mean these metrics should be gathered & exposed automatically - just by using the SDK (I don't think that's happening ;) )? Just using the SDK is...

@vanilla-sundae thanks for reporting, unfortunately the information provided is not enough to understand and fix the issue. You should examine the heap dump and try to answer the following. -...

> Also seeing some error messages Failed to propagate context because previous propagated context is already set; new: {0}, old: {1} java.lang.String - nanosecond timeout value out of range java.lang.String...

@vanilla-sundae https://github.com/open-telemetry/opentelemetry-java-instrumentation/pull/12397 is not enough to fix your issue.

While it is not possible to limit span size it is possible to limit the number of attributes and their size, see https://opentelemetry.io/docs/languages/java/configuration/#properties-general Limiting based on span size doesn't seem...