prometheus_queries.txt

TFS/Components:
---------------

variables:
name=component
    query=metrics(tfs_)
    regex=/tfs_(.+)_rpc_.*/
name=method
    query=metrics(tfs_[[component]]_rpc_)
    regex=/tfs_[[component]]_rpc_(.+)_histogram_duration_bucket/

plots:
tfs_[[component]]_rpc_[[method]]_counter_requests_started_total
tfs_[[component]]_rpc_[[method]]_counter_requests_completed_total
tfs_[[component]]_rpc_[[method]]_counter_requests_failed_total
tfs_[[component]]_rpc_[[method]]_histogram_duration_sum
#sum(increase(tfs_[[component]]_rpc_[[method]]_histogram_duration_bucket[$__rate_interval])) by (le)
sum(
    max_over_time(tfs_[[component]]_rpc_[[method]]_histogram_duration_bucket{pod=~"[[pod]]"}[1m]) -
    min_over_time(tfs_[[component]]_rpc_[[method]]_histogram_duration_bucket{pod=~"[[pod]]"}[1m])
) by (le)


TFS/Device/Driver:
------------------

variables:
name=method
    query=metrics(tfs_device_driver_.+)
    regex=/tfs_device_driver_(.+config)_histogram_duration_bucket/
name=driver
    query=label_values(tfs_device_driver_[[method]]_histogram_duration_bucket, driver)
    regex=

plots:
tfs_device_driver_[[method]]_counter_requests_started_total{driver="[[driver]]"}
tfs_device_driver_[[method]]_counter_requests_completed_total{driver="[[driver]]"}
tfs_device_driver_[[method]]_counter_requests_failed_total{driver="[[driver]]"}
tfs_device_driver_[[method]]_histogram_duration_sum{driver="[[driver]]"}
#sum(increase(tfs_device_driver_[[method]]_histogram_duration_bucket{driver="[[driver]]"}[$__rate_interval])) by (le)
sum(
    max_over_time(tfs_device_driver_[[method]]_histogram_duration_bucket{driver="[[driver]]", pod=~"deviceservice-[[pod]]"}[1m]) -
    min_over_time(tfs_device_driver_[[method]]_histogram_duration_bucket{driver="[[driver]]", pod=~"deviceservice-[[pod]]"}[1m])
) by (le)

variables:
name=method
    query=metrics(tfs_service_handler_.+)
    regex=/tfs_service_handler_(.+config)_histogram_duration_bucket/
name=driver
    query=label_values(tfs_service_handler_[[method]]_histogram_duration_bucket, handler)
    regex=

plots:
tfs_service_handler_[[method]]_counter_requests_started_total{handler="[[handler]]"}
tfs_service_handler_[[method]]_counter_requests_completed_total{handler="[[handler]]"}
tfs_service_handler_[[method]]_counter_requests_failed_total{handler="[[handler]]"}
tfs_service_handler_[[method]]_histogram_duration_sum{handler="[[handler]]"}
#sum(increase(tfs_service_handler_[[method]]_histogram_duration_bucket{handler="[[handler]]"}[$__rate_interval])) by (le)
sum(
    max_over_time(tfs_service_handler_[[method]]_histogram_duration_bucket{handler=~"[[handler]]", pod=~"serviceservice-[[pod]]"}[1m]) -
    min_over_time(tfs_service_handler_[[method]]_histogram_duration_bucket{handler=~"[[handler]]", pod=~"serviceservice-[[pod]]"}[1m])
) by (le)