Error while querying Elasticsearch for nodes stats about elasticsearch_exporter HOT 5 CLOSED

@dominikschulz ok i found it.... i need to set network.host to 0.0.0.0 in elasticsearch config
sorry for the noise

from elasticsearch_exporter.

This doesn't make sense to me. If you can curl it, the exporter should be able to query it as well.

from elasticsearch_exporter.

@dominikschulz exactly, but it happens.....

from elasticsearch_exporter.

@dominikschulz ok thre is a difference between running curl via kubectl exec and via kubectl exec /bin/bash and the use curl..... the second one also does not work.

So the behaviour is the same - but why

from elasticsearch_exporter.

I'm having a similar problem with getting the cluster node stats, I get this error when running the docker container elasticsearch_exporter

$ docker run --rm --net=host -p 9108:9108 justwatch/elasticsearch_exporter:1.0.0 '-es.uri=http://172.23.64.131:9200' '-es.all=true'
level=info msg="starting elasticsearch_exporter" addr=:9108
level=warn msg="failed to fetch and decode node stats" err="failed to get node stats from http://172.23.64.131:9200/_nodes/stats: Get http://172.23.64.131:9200/_nodes/stats: net/http: request canceled (Client.Timeout exceeded while awaiting headers)"

I've verified that the network.host setting of the elasticsearch server at 172.23.64.131 is set to 0.0.0.0 as above. I'm running my elasticsearch_exporter on a separate network from the elasticsearch servers, but since I can curl there and get /_nodes/stats and all the _cluster stats seem to be working, routing is good.

Anyone have any other suggestions on what could be happening, or how I should proceed to debug further, I'd appreciate it.

from elasticsearch_exporter.