Metrics on cluster of Logstash

[StephaneFr44]

Hey,
I have a swarm replicaset of 8 logstash and I wan't to know if it's possible to identify each logstash in the metrics ?

best regards
Stéphane

[jensskott]

I have the same issue in kubernetes.
For example
logstash_node_mem_heap_max_bytes gives me a combine metric from all my nodes in behind the service in kubernetes without a way to have a selector per cluster node.

[StephaneFr44]

Have you got an issue to analyse each logstash ?

[mpucholblasco]

Maybe this issue could be fixed by adding a custom label (e.g. https://albersdevelopment.net/2019/08/28/prometheus-adding-a-label-to-a-target/) on each scrapping configuration. For instance, we use prometheus operator, so we could add this via ServiceMonitor -> spec -> metricRelabelings.