|
2 | 2 |
|
3 | 3 | ## dev |
4 | 4 |
|
5 | | -- **Feature: Add configation option error_collector.backtrace_truncate_location** |
| 5 | +- **Feature: Introduce Hybrid Agent for OpenTelemetry Tracing Support** |
| 6 | + |
| 7 | + OpenTelemetry Tracing APIs can now be translated into New Relic telemetry with the New Relic Ruby agent's new Hybrid Agent features. This allows the `newrelic_rpm` gem to behave similarly to an OpenTelemetry SDK, accepting OpenTelemetry API calls and turning them into New Relic Transactions and Segments with familiar attributes and names. This also allows users to take advantage of OpenTelemetry instrumentation for libraries not instrumented by the New Relic Ruby agent. |
| 8 | + |
| 9 | + The following configuration options relate to Hybrid Agent features: |
| 10 | + |
| 11 | + | Configuration name | Default | Behavior | |
| 12 | + | ------------------ | ------- |----------| |
| 13 | + | opentelemetry.enabled | `false` | A global configuration option for disabling all OpenTelemetry signals sent through New Relic. | |
| 14 | + | opentelemetry.traces.enabled | `true` | Enables the creation of Transaction Trace segments and timeslice metrics from OpenTelemetry Spans | |
| 15 | + | opentelemetry.traces.include | `''` | A comma-delimited list of OpenTelemetry Tracers, represented as a string (e.g. "AppTracer1,OpenTelemetry::Instrumentation::Bunny::Instrumentation"), that **will** have their trace signals sent to New Relic.| |
| 16 | + | opentelemetry.traces.exclude | `''` | A comma-delimited list of OpenTelemetry Tracers, represented as a string (e.g. "AppTracer1,OpenTelemetry::Instrumentation::Bunny::Instrumentation"), that will **not** have their trace signals sent to New Relic. All known tracers for instrumentation that conflicts with New Relic instrumentation are excluded by default. | |
| 17 | + |
| 18 | + |
| 19 | +- **Feature: Add configuration option error_collector.backtrace_truncate_location** |
6 | 20 |
|
7 | 21 | A new configuration option has been added, `error_collector.backtrace_truncate_location`, which allows the user to specify where in the backtrace to truncate when the number of frames exceeds `error_collector.max_backtrace_frames`. Options are `'top'` (removes frames from the beginning), `'middle'` (removes frames from the middle, preserving the beginning and end), or `'end'` (removes frames from the end). The default is `'middle'`. [PR#3424](https://github.com/newrelic/newrelic-ruby-agent/pull/3424) |
8 | 22 |
|
|
28 | 42 |
|
29 | 43 | The agent now properly initializes health check loops after forking, ensuring each process generates its own health check file. This fix also has the effect of correctly including `entity.guid` values in the health check files. [PR#3409](https://github.com/newrelic/newrelic-ruby-agent/pull/3409) [Issue#3408](https://github.com/newrelic/newrelic-ruby-agent/issues/3408) |
30 | 44 |
|
31 | | -- **Bugfix: Fix `sidekiq.ignore_retry_errors`** |
| 45 | +- **Bugfix: Fix `sidekiq.ignore_retry_errors`** |
32 | 46 |
|
33 | | - The configuration option `sidekiq.ignore_retry_errors: true` was continuing to report retry errors. The agent now correctly ignores retry errors and only reports when jobs permanently fail. [PR#3399](https://github.com/newrelic/newrelic-ruby-agent/pull/3399) |
| 47 | + The configuration option `sidekiq.ignore_retry_errors: true` was continuing to report retry errors. The agent now correctly ignores retry errors and only reports when jobs permanently fail. [PR#3399](https://github.com/newrelic/newrelic-ruby-agent/pull/3399) |
34 | 48 |
|
35 | 49 | ## v10.0.0 |
36 | 50 |
|
|
0 commit comments