We have added standard metrics in the #27. So there are scanning systems that are trying to find a vulnerability in our service. Now, the metrics for incoming requests look too big:
$ curl https://docbot.tarantool.io/metrics > metrics.txt
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 12.6M 100 12.6M 0 0 1061k 0 0:00:12 0:00:12 --:--:-- 1715k
Size of metrics is 12.6M (!)
Prometheus takes such a volume every 10 seconds, which leads to an instant expansion of timeseries storage and the inactivity of Prometheus.
The service has two endpoints:
We have added standard metrics in the #27. So there are scanning systems that are trying to find a vulnerability in our service. Now, the metrics for incoming requests look too big:
Size of metrics is 12.6M (!)
Prometheus takes such a volume every 10 seconds, which leads to an instant expansion of timeseries storage and the inactivity of Prometheus.