CC00327.CloudKubernetes——|KuberNetes&运维.V45|——|白盒监控.v07|kafka.v04|Prometheus获取不到数据|

本文探讨了在Kubernetes环境中实现白盒监控的方案，特别关注了如何将Kafka与Prometheus进行集成，解决Prometheus无法获取到数据的问题。通过详细步骤和实例，展示了如何使用Python和Java进行监控配置，以及在大数据场景下Kafka的重要性。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

附录一：

### --- 报错现象：

~~~     报错现象：Prometheus监控不到数据
~~~     报错现象：http://prom.test.com——>Alerts——>Enable query history——>没有kafka数据

### --- 报错分析：

~~~     # 查看kafka-exporter的po地址
[root@k8s-master01 kafka-exporter]# kubectl get po -n monitoring -owide
NAME                                   READY   STATUS    RESTARTS   AGE     IP               NODE           NOMINATED NODE   READINESS GATES
kafka-exporter-d47b9f76f-k92rm         1/1     Running   0          44m     172.27.14.227    k8s-node02     <none>           <none>

~~~     # 进入容器内部

[root@k8s-master01 kafka-exporter]# kubectl exec -ti kafka-exporter-d47b9f76f-k92rm -n monitoring -- sh
/ # nslookup kafka-0.kafka-headless.public-service      
Server:     10.96.0.10
Address:    10.96.0.10:53

** server can't find kafka-0.kafka-headless.public-service: NXDOMAIN

*** Can't find kafka-0.kafka-headless.public-service: No answer  // nslookup解析不到

/ # ping kafka-0.kafka-headless.public-service                   // ping是没有问题的
PING kafka-0.kafka-headless.public-service (172.27.14.226): 56 data bytes
64 bytes from 172.27.14.226: seq=0 ttl=63 time=0.138 ms
64 bytes from 172.27.14.226: seq=1 ttl=63 time=0.108 ms

~~~     # coredns解析不到
~~~     分析定位为：kafka没有创建topic

[root@k8s-master01 kafka-exporter]# kubectl get po -n public-service -owide // 报error

### --- 解决方案：

~~~     # 进入kafka容器给kafka创建topic
[root@k8s-master01 kafka-exporter]# kubectl exec -ti kafka-0 -n public-service -- bash
I have no name!@kafka-0:/$ kafka-topics.sh --create --zookeeper zookeeper:2181 --replication-factor 1 --partitions 1 --topic mytopic
Created topic mytopic.

~~~     # 再次请求metrics，就会发现有了kafka-topic这个数据
~~~     报错现象：http://prom.test.com——>Alerts——>Enable query history——>数据已经检测到

[root@k8s-master01 kafka-exporter]# curl 10.110.47.163:9308/metrics | grep kafka
# HELP kafka_topic_partition_under_replicated_partition 1 if Topic/Partition is under Replicated
# TYPE kafka_topic_partition_under_replicated_partition gauge
kafka_topic_partition_under_replicated_partition{partition="0",topic="mytopic"} 0