Open amhuber opened 1 year ago
Hi @amhuber so would you rather expect the new metrics to have the bosh_deployment available or to adjust the dashboard accordingly?
Greetings, Benjamin
I think it would be better to have the deployment in the metrics since it's inconsistent right now and some people may have more than one deployment.
Hi, in this case it would make more sense to create the issue in here => https://github.com/bosh-prometheus/firehose_exporter/issues (and maybe reference this on in there.) .
OK, I'll create an issue there and leave this one open for now until it's resolved one way or the other.
It appears that the HTTP related metrics (firehose_http_total and firehose_http_duration_seconds_sum) from the new firehose_exporter don't have the bosh_deployment tag present, so the dashboards for app latency and requests aren't showing any data unless I manually edit the query string to set "deploment" to empty string. The "Apps: System" dashboard does have data since the related metrics (firehose_container_metric_cpu_percentage for example) do have the bosh_deployment tag present.
I've confirmed that the raw metrics emitted on the firehose do appear to have the deployment property present and I can't tell why the tags are missing on just the HTTP metrics. Please let me know if this is something we have misconfigured.