I'm not sure the prometheus scraper in otel-collector is working. I'm going off the instructions to...

Dusty McDustins

03/28/2025, 4:34 PM

I'm not sure the prometheus scraper in otel-collector is working. I'm going off the instructions to enable a prometheus receiver. Since I'm going the self-hosted standalone linux binary route I'm instead editing

/opt/signoz-otel-collector/conf/config.yaml

When I add a scraper clause nothing seems to happen. More details in thread..

Dusty McDustins

03/28/2025, 4:36 PM

here is my config

Copy code

receivers:
  prometheus:
    config:
      scrape_configs:
        - job_name: "vllm"
          scrape_interval: 5s
          static_configs:
            - targets: ["acme-tgi:8000"]

I know the scrape target is valid because I can curl

/metrics

there and get the usual prometheus metrics data. I can also install Prometheus and config a scraper and see the metrics I expect

Dusty McDustins

03/28/2025, 4:38 PM

nothing ever shows up when I try to find those metrics using the signoz query builder. even if I set the target to an invalid hostname the otel-collector never seems to care.

Vibhu Pandey

03/29/2025, 5:40 PM

did you enable it in the pipelines?

Vibhu Pandey

03/29/2025, 5:40 PM

Copy code

metrics:
  receivers:
   - prometheus

Dusty McDustins

03/30/2025, 5:32 PM

gah, that was it. thank you! I notice this error gets thrown when I try to create a histogram chart.

Dusty McDustins

03/30/2025, 5:32 PM

Untitled

Vibhu Pandey

03/30/2025, 6:50 PM

Thanks for pointing this out @Dusty McDustins. @Nagesh Bansal we have missed instructions for adding the histogramQuantile in the clickhouse setup for linux.

2 Views

Open in Slack

Previous Next

SigNoz Community

SigNoz is an open-source APM. It helps developers monitor their applications & troubleshoot problems, an open-source alternative to DataDog, NewRelic, etc.