It would be nice if the metrics api in fme server had the fme server job queue time in seconds. e.g:
fme_queue_time_total 300
The queue time could also be broken down further by server queue:
fme_queue_time {queue="LargeBatchJobs"} 300 fme_queue_time {queue="boundaryqa"} 0 fme_queue_time {queue="default"} 0
After there are no jobs in the queue, this value would be set back to 0.
What problem would this solve?
This would allow users to take actions if the queue time on fme server gets too large. For example, if the queue time exceeds, 500 seconds, I could use that as a trigger to add additional dynamic engines or even turn on another server which has an fme engine installed on it to help process these jobs faster.
Location of current fme metric endpoint I am referring to:
https://docs.safe.com/fme/html/FME_REST/apidoc/v3/index.html#!/metrics/metrics_get_0
