Yuna Verheyden

Just a note: this will probably be a breaking change. If people were relying on specific Prometheus attributes these might change now. I think this is fine, we just have...

Hi, thanks for opening an issue. We currently do not offer rpm packages for Tempo, a good first step would be to actually build them. Looking at Cortex, they build...

This can be a good first issue if you wan to learn more about packages and build automation

fyi, adding an external label worked for me. Using the Grafana Agent: ```yaml prometheus: global: external_labels: cluster: k3d-local-dev # ... ```

> This was originally written this way intentionally. If we've already done the work to pull additional traces why throw them away? However, we've gotten so many questions about this...

No changes here, multitenancy is only supported when ingesting gRPC, so you should use either OTLP gRPC or Jaeger gRPC in a multitenant setup.

@liguozhong suggested in https://github.com/grafana/tempo/issues/1381#issuecomment-1098877991 that we could use the Authenticator interface to extract this information out of the request coming from the OpenTelemetry receivers.

Hi, can you clarify what you need? Tempo can ingest both OTLP HTTP and OTLP gRPC, you just have to enable the respective receivers [in the distributor config section](https://grafana.com/docs/tempo/latest/configuration/#distributor). If...

Okay, OTLP/HTTP is currently not supported with Tempo multitenancy. For now we only support multitenancy when using OTLP/gRPC, as you noted we rely on the context object to extract the...

Yeah, I have the same issue unfortunately. The synthetic load generator we use has been archived and its images are a couple of years old by now. I noticed older...