How about metrics?

[xperimental]

We currently monitor the state of most of our services / nodes using Prometheus. Unfortunately bamboo currently does not provide any metrics or monitoring endpoint other than the /status which just responds with OK.

I was thinking of implementing a Prometheus metrics endpoint but did not want to start if there is no interest in it being merged, so my question is if there is an interest in such an endpoint?

[j1n6]

If you are using HAProxy, you might have to parse the HAProxy stats dump.

Another way is to use the Bamboo state API, ask Prometheus extractor to check application Heath separately based on the app healthcheck endpoint.

Both way has pros/cons depends what you are looking for alerting target.

On Nov 17, 2016, at 14:56, Robert Jacob [email protected] wrote:

We currently monitor the state of most of our services / nodes using Prometheus. Unfortunately bamboo currently does not provide any metrics or monitoring endpoint other than the /status which just responds with OK.

I was thinking of implementing a Prometheus metrics endpoint but did not want to start if there is no interest in it being merged, so my question is if there is an interest in such an endpoint?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

[xperimental]

We have metrics from HAProxy using the haproxy_exporter, but I was more interested in metrics from bamboo itself. The state API does not really help here, because it does not contain any metrics just the current state of the configured ACLs (or if it has metrics, please point me to them).

It is probably possible to synthesize metrics by scraping the API endpoints of bamboo from the outside and create a "bamboo-exporter". But having the monitored service (in this case bamboo) provide the metrics itself is usually more reliable and exact.