opentelemetry-demo
opentelemetry-demo copied to clipboard
Review and enhance metric support for product catalog service (Go)
The following are requirements for metrics in the project. Not all services will meet all requirements.
- [ ] Service metrics should not be duplicative of trace-derived metrics.
- [ ] Services should emit multiple business metrics where appropriate.
- [ ] Services should emit trace exemplars for their metrics.
- [ ] Services should use push metrics where possible.
- [ ] Services should explicitly read baggage where needed.
- [ ] Collectors should perform metric transforms to normalize resource and service metrics.
Referencing issue: https://github.com/open-telemetry/opentelemetry-demo-webstore/issues/43
hi,I want help it. :)
All yours! @Frapschen
@cartersocha I'm a little confuse, what metrics should we add to service? in my view, we can add some span metric by the otel-collector:
- calls_total
- latency_bucket
We haven’t had time to define our metric story yet. We are planning on discussing overall metric requirements in our next sig meeting.
We’ll need to set up a consumption experience too. I’m guessing we’ll add grafana in a container or something similar.
Feel free to open a pr with what metrics you think would be useful. Then we can use that to solicit feedback / input from everyone
The newest implementation instrumentation for the runtime
package is in process, waiting for its finish.
Basic metrics have been added; closing issue.