So a bit of an odd question here maybe, but is there a way to track, get notified or visually see (other than digging into AWS EB logs) when a new tracker starts sending metrics (is subscribed) to the collector - let’s say Clojure Collector (since that’s my case).
I copy metrics also in PostgreSQL, therefore I thought I could query for new app_ids, but since I process data nightly, that might take some time for notifications of such nature…
Any thoughts on this?
P.S. - This one off-topic, but I’ll add it anyway: What path should I use in AWS Application Load Balancer (created by the Elastic Beanstalk) to monitor the health of the Collector. I have currently set it temporarily under / and configured it to listen for 404 responses.
So a bit of an odd question here maybe, but is there a way to track, get notified or visually see (other than digging into AWS EB logs) when a new tracker starts sending metrics (is subscribed) to the collector - let’s say Clojure Collector (since that’s my case).}
v_tracker in the enriched data, which is the tv/tracker version. Here’s a handy reference of the tracker protocol mappings of query params to their fields, and the canonical event model which goes a little way towards adding some meaning to that. They’re a great reference for figuring out this kind of thing.
P.S. - This one off-topic, but I’ll add it anyway: What path should I use in AWS Application Load Balancer (created by the Elastic Beanstalk) to monitor the health of the Collector. I have currently set it temporarily under / and configured it to listen for 404 responses.
You can ping /health for a healthcheck response. If you’re after regular monitoring another option is to send empty requests periodically to the actual collector endpoint that you use for events - it’ll just get filtered into bad rows, but it’ll give you a 200 if everything’s operating as normal.
So as far as I understand, you’re suggesting I should query the enriched data in my db? Sometime after the enrichment process finishes…?
My collector gives a 404 Not Found on /health that’s why I made that workaround. Are you sure that the Clojure Collector has this /health endpoint? I am using clojure-collector-2.1.2-standalone. I’ll just hit /i after all - since that’s filtered in bad rows.
