# Metricbeat cannot get metrics from kafka running in k8 **URL:** \ **Category:** Beats\ **Tags:** metricbeat\ **Created:** [August 20, 2019, 11:29pm UTC](https://discuss.elastic.co/t/metricbeat-cannot-get-metrics-from-kafka-running-in-k8/196014 "2019-08-20T23:29:58Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![memelet](https://avatars.discourse-cdn.com/v4/letter/m/6bbea6/32.png) [@memelet](https://discuss.elastic.co/u/memelet)\ **Post date:** [August 20, 2019, 11:29pm UTC](https://discuss.elastic.co/t/metricbeat-cannot-get-metrics-from-kafka-running-in-k8/196014/1 "2019-08-20T23:29:58Z")
My configuration contains: ```auto - condition: equals: kubernetes.labels: strimzi_io/name: vikafka-kafka config: - module: kafka metricsets: ["partition","consumergroup"] hosts: ["${data.host}:9092"] enabled: true ``` Which does find the pods. But I get the sequence of repeating logs: ```auto default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.453Z INFO kafka/log.go:53 Connected to broker at 10.12.196.222:9092 (unregistered) default/metricbeat-h4qbr[metricbeat]: default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.454Z INFO kafka/log.go:53 Connected to broker at 10.12.196.222:9092 (unregistered) default/metricbeat-h4qbr[metricbeat]: default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.455Z INFO kafka/log.go:53 Connected to broker at 10.12.196.222:9092 (unregistered) default/metricbeat-h4qbr[metricbeat]: default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.459Z INFO kafka/log.go:53 Connected to broker at 10.12.196.222:9092 (unregistered) default/metricbeat-h4qbr[metricbeat]: default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.471Z INFO kafka/log.go:53 Closed connection to broker 10.12.196.222:9092 default/metricbeat-h4qbr[metricbeat]: default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.471Z INFO module/wrapper.go:247 Error fetching data for metricset kafka.consumergroup: error in connect: No advertised broker with address 10.12.196.222:9092 found default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.471Z INFO kafka/log.go:53 Closed connection to broker 10.12.196.222:9092 default/metricbeat-h4qbr[metricbeat]: default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.471Z INFO module/wrapper.go:247 Error fetching data for metricset kafka.partition: error in connect: No advertised broker with address 10.12.196.222:9092 found default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.472Z INFO kafka/log.go:53 Closed connection to broker 10.12.196.222:9092 default/metricbeat-h4qbr[metricbeat]: default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.472Z INFO module/wrapper.go:247 Error fetching data for metricset kafka.consumergroup: error in connect: No advertised broker with address 10.12.196.222:9092 found default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.476Z INFO kafka/log.go:53 Closed connection to broker 10.12.196.222:9092 default/metricbeat-h4qbr[metricbeat]: default/metricbeat-h4qbr[metricbeat]: 2019-08-20T23:24:57.477Z INFO module/wrapper.go:247 Error fetching data for metricset kafka.partition: error in connect: No advertised broker with address 10.12.196.222:9092 found ``` Not sure what the complaint is. kafkcat shows ```auto $ kafkacat -L -b 10.12.196.222 ops/git/master !+ Metadata for all topics (from broker -1: 10.12.196.222:9092/bootstrap): 3 brokers: broker 0 at vikafka-kafka-0.vikafka-kafka-brokers.default.svc.cluster.local:9092 (controller) broker 2 at vikafka-kafka-2.vikafka-kafka-brokers.default.svc.cluster.local:9092 broker 1 at vikafka-kafka-1.vikafka-kafka-brokers.default.svc.cluster.local:9092 0 topics: ``` And from within the metricbeat container I can ping the IP `10.12.196.222` and the DNS `vikafka-kafka-0.vikafka-kafka-brokers.default.svc.cluster.local`, so at least connectivity exists. ---
**Author:** ![memelet](https://avatars.discourse-cdn.com/v4/letter/m/6bbea6/32.png) [@memelet](https://discuss.elastic.co/u/memelet)\ **Post date:** [August 21, 2019, 2:28pm UTC](https://discuss.elastic.co/t/metricbeat-cannot-get-metrics-from-kafka-running-in-k8/196014/2 "2019-08-21T14:28:09Z")
We were able to get metricbeat to connect only by hard coding the k8 dns names -- not using autodiscover ```auto deployment: overrideModules: kubernetes: enabled: true config: - module: kafka hosts: - "vikafka-kafka-0.vikafka-kafka-brokers.default.svc.cluster.local:9092" - "vikafka-kafka-1.vikafka-kafka-brokers.default.svc.cluster.local:9092" - "vikafka-kafka-2.vikafka-kafka-brokers.default.svc.cluster.local:9092" metricsets: [consumergroup, partition] ``` So still at a loss how to make this work with autodiscover ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [September 18, 2019, 2:28pm UTC](https://discuss.elastic.co/t/metricbeat-cannot-get-metrics-from-kafka-running-in-k8/196014/3 "2019-09-18T14:28:10Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.