Proposal for Prometheus Metric to Monitor Dropped Packets

[misaki-kawakami]

Is your feature request related to a problem? Please describe.
At present, we lack an easy way to monitor number of dropped packets by a logger when the buffer is full. A a new Prometheus metric would improve our monitoring capabilities, allowing us to trigger alerts if a significant number of packets are being dropped.

Describe the solution you'd like
A new Prometheus metric that can indicate the number of dropped packets per logger/collector. This metric provides insight into the number of dropped packets, eliminating the need to search through logs.

Describe alternatives you've considered
Checking dns-collector logs every now and then but with large number of instances, this method is becoming impractical.

[dmachard]

Good idea, I think adding global metrics to DNScollector would be the right way to expose this type of information.

[dmachard]

The following new metric will be available in next release.

# HELP dnscollector_exporter_worker_discarded_traffic_total Discarded traffic associated to each worker
# TYPE dnscollector_exporter_worker_discarded_traffic_total counter
dnscollector_exporter_worker_discarded_traffic_total{worker="console"} 0
dnscollector_exporter_worker_discarded_traffic_total{worker="tap"} 0

[dmachard]