We’re on Temporal Cloud and using Prometheus+Grafana.
I thought that using the temporal_request SDK metric and filtering down to instances where operation=StartWorkflowExecution
would be the ticket, but there are 2 problems with this approach:
- The temporal_request metric does not include the
workflow_type=
, even whenoperation=StartWorkflowExecution
. - Being a SDK/client-side metric, my understanding is that temporal_request represents request which may or may not have been fulfilled. (Network issues, invalid params, rate limiting, etc might get in the way of that request being fulfilled).
Instead, I’m looking to tally all Workflows of a particular type which were successfully started.
I’m assuming I can use Cloud/Server metrics for this… Would temporal_cloud_v0_frontend_service_request_count filtered by operation=StartWorkflowExecution
be the right approach? Thanks in advance!