[Argus] use ES data streams, enable log correlation#4964
[Argus] use ES data streams, enable log correlation#4964mnaamani merged 4 commits intoJoystream:masterfrom
Conversation
|
@zeeshanakram3 FYI, did the same for Colossus as well |
717a875 to
86a96be
Compare
|
As the integration-test logs suggest that the colossus-1 is not serving any buckets (i.e. storage node during initialization could not find worker/buckets that it's supposed to serve). I think this is mainly because QN is lagging and info about worker/buckets created in the previous step hasn't been processed yet. FixAdd |
| "cli": "./bin/run", | ||
| "start": "./bin/run start", | ||
| "start:with-instrumentation": "export OTEL_APPLICATION=distributor-node; node --require @joystream/opentelemetry ./bin/run start" | ||
| "start:with-instrumentation": "export OTEL_APPLICATION=distributor-node; export OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:8200; export OTEL_RESOURCE_ATTRIBUTES=service.name=argus,deployment.environment=production; node --require @joystream/opentelemetry ./bin/run start" |
There was a problem hiding this comment.
I don't think it is the best place to hardcode the otlp endpoint in this npm script.
But since in production we are using docker compose file which doesn't directly use this script I think its okay, but I just wanted to make a note of it.
config.logs.elastic.indextoconfig.logs.elastic.indexPrefix. Node ID from config will be automatically appended to the index name.--elasticSearchIndexoption and replaced with--elasticSearchIndexPrefix. Node ID from config will be automatically appended to the index name.