search

prometheus-community / stackdriver_exporter

Google Stackdriver Prometheus exporter
Apache License 2.0
263 stars 99 forks source link

Backends by state metric causes 500 error on exposition #360

Open mbrancato opened 3 months ago

mbrancato commented 3 months ago

The stackdriver_cloudsql_database_cloudsql_googleapis_com_database_postgresql_num_backends_by_state metric appears to have some collection issue when I enabled it. The exposition returns a 500 error.

An error has occurred while serving metrics:\n\n43 error(s) occurred:\n* [from Gatherer #2] collected metric \"stackdriver_cloudsql_database_cloudsql_googleapis_com_database_postgresql_num_backends_by_state\" { label:{name:\"database\" value:\"cloudsqladmin\"} label:{name:\"database_id\" value:\"my-project:mydb\"} label:{name:\"project_id\" value:\"my-project..p_ms:1724089680000} was collected before with the same name and label values\n* [from Gatherer #2] collected metric \"stackdriver_cloudsql_database_cloudsql_googleapis_com_database_uptime\" { label:{name:\"database_id\" value:\"myproject:myotherdb\"} label:{name:\"project_id\" value:\"myproject\"} label:{name:\"region\" value:\"us-central\"} label:{name:\"unit\" value:\"s\"} gauge:{value:60} timestamp_ms:1724089680000} was collected before with the same name and label values\n
mbrancato commented 3 months ago

I'm seeing this with stackdriver_cloudsql_database_cloudsql_googleapis_com_database_uptime as well. Otherwise called cloudsql.googleapis.com/database/uptime.