Question

我正在尝试使用Kubernetes 1.7.12流利弹性搜索插件： https://github.com/kubernetes/kubernetes/tree/v1.7.12/cluster/addons/fluentd-elasticsearch

ElasticSearch启动并可以回复：

{
 "name" : "0322714ad5b7",
 "cluster_name" : "kubernetes-logging",
 "cluster_uuid" : "_na_",
 "version" : {
   "number" : "2.4.1",
   "build_hash" : "c67dc32e24162035d18d6fe1e952c4cbcbe79d16",
   "build_timestamp" : "2016-09-27T18:57:55Z",
   "build_snapshot" : false,
   "lucene_version" : "5.5.2"
 },
 "tagline" : "You Know, for Search"
}

但是Kibana仍然无法连接它。连接错误始于：

{"type":"log","@timestamp":"2018-01-23T07:42:06Z","tags":["warning","elasticsearch"],"pid":6,"message":"Unable to revive connection: http://elasticsearch-logging:9200/"}
{"type":"log","@timestamp":"2018-01-23T07:42:06Z","tags":["warning","elasticsearch"],"pid":6,"message":"No living connections"}

在ElasticSearch启动后，错误将更改为：

{"type":"log","@timestamp":"2018-01-23T07:42:08Z","tags":["status","plugin:elasticsearch@1.0.0","error"],"pid":6,"state":"red","message":"Status changed from red to red - Service Unavailable","prevState":"red","prevMsg":"Unable to connect to Elasticsearch at http://elasticsearch-logging:9200."}

所以好像，Kibana最终能够从ElasticSearch获得响应，但仍然无法建立连接。

这就是Kibana仪表板的样子：

我试图让日志输出更多信息，但是没有足够的知识来了解Kibana和ElasticSearch以了解我接下来还能尝试什么。

我可以使用此docker-compose.yml：

在本地重现错误

version: '2'
services:
 elasticsearch-logging:
   image: gcr.io/google_containers/elasticsearch:v2.4.1-2
   ports:
     - "9200:9200"
     - "9300:9300"

 kibana-logging:
   image: gcr.io/google_containers/kibana:v4.6.1-1
   ports:
     - "5601:5601"
   depends_on:
     - elasticsearch-logging
   environment:
     - ELASTICSEARCH_URL=http://elasticsearch-logging:9200

根据我从这个问题中可以看出的内容，看起来不应该涉及太多： Kibana on Docker cannot connect to Elasticsearch 和这个博客：https://gunith.github.io/docker-kibana-elasticsearch/

但我无法弄清楚我错过了什么。

任何想法我还能尝试什么？

感谢您的时间。：）

更新1：

Kubernetes集群上的{p> curl http://elasticsearch-logging导致了相同的输出：

{
  "name" : "elasticsearch-logging-v1-68km4",
  "cluster_name" : "kubernetes-logging",
  "cluster_uuid" : "_na_",
  "version" : {
    "number" : "2.4.1",
    "build_hash" : "c67dc32e24162035d18d6fe1e952c4cbcbe79d16",
    "build_timestamp" : "2016-09-27T18:57:55Z",
    "build_snapshot" : false,
    "lucene_version" : "5.5.2"
  },
  "tagline" : "You Know, for Search"
}

由于代理规则，Kubernetes群集上的

curl http://elasticsearch-logging/_cat/indices?pretty超时。在本地使用docker-compose.yml和curl（例如curl localhost:9200/_cat/indices?pretty）会导致：

{
  "error" : {
    "root_cause" : [ {
      "type" : "master_not_discovered_exception",
      "reason" : null
    } ],
    "type" : "master_not_discovered_exception",
    "reason" : null
  },
  "status" : 503
}

docker-compose日志显示：

[2018-01-23 17:04:39,110][DEBUG][action.admin.cluster.state] [ac1f2a13a637] no known master node, scheduling a retry

[2018-01-23 17:05:09,112][DEBUG][action.admin.cluster.state] [ac1f2a13a637] timed out while retrying [cluster:monitor/state] after failure (timeout [30s])
[2018-01-23 17:05:09,116][WARN ][rest.suppressed          ] path: /_cat/indices, params: {pretty=}
MasterNotDiscoveredException[null]
     at org.elasticsearch.action.support.master.TransportMasterNodeAction$AsyncSingleAction$5.onTimeout(TransportMasterNodeAction.java:234)
     at org.elasticsearch.cluster.ClusterStateObserver$ObserverClusterStateListener.onTimeout(ClusterStateObserver.java:236)
     at org.elasticsearch.cluster.service.InternalClusterService$NotifyTimeout.run(InternalClusterService.java:804)
     at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
     at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
     at java.lang.Thread.run(Thread.java:745)

更新2：正在运行kubectl --namespace kube-system logs -c kubedns po/kube-dns-667321983-dt5lz --tail 50 --follow 产率：

I0124 16:43:33.591112       5 dns.go:264] New service: kibana-logging
I0124 16:43:33.591225       5 dns.go:264] New service: nginx
I0124 16:43:33.591251       5 dns.go:264] New service: registry
I0124 16:43:33.591274       5 dns.go:264] New service: sudoe
I0124 16:43:33.591295       5 dns.go:264] New service: default-http-backend
I0124 16:43:33.591317       5 dns.go:264] New service: kube-dns
I0124 16:43:33.591344       5 dns.go:462] Added SRV record &{Host:kube-dns.kube-system.svc.cluster.local. Port:53 Priority:10 Weight:10 Text: Mail:false Ttl:30 TargetStrip:0 Group: Key:}
I0124 16:43:33.591369       5 dns.go:462] Added SRV record &{Host:kube-dns.kube-system.svc.cluster.local. Port:53 Priority:10 Weight:10 Text: Mail:false Ttl:30 TargetStrip:0 Group: Key:}
I0124 16:43:33.591390       5 dns.go:264] New service: kubernetes
I0124 16:43:33.591409       5 dns.go:462] Added SRV record &{Host:kubernetes.default.svc.cluster.local. Port:443 Priority:10 Weight:10 Text: Mail:false Ttl:30 TargetStrip:0 Group: Key:}
I0124 16:43:33.591429       5 dns.go:264] New service: elasticsearch-logging

更新3：

我仍然试图让一切工作，但在其他人的帮助下，我相信这是一个RBAC问题。我不完全确定，但看起来弹性搜索节点由于权限而无法与主服务器连接（我从来不知道甚至需要它）。

以下是一些有用的步骤，以防其他人开始：

使用RBAC：

# kubectl --kubeconfig kubeconfig.yaml --namespace kube-system logs po/elasticsearch-logging-v1-wkwcs
F0119 00:18:44.285773       9 elasticsearch_logging_discovery.go:60] kube-system namespace doesn't exist: User "system:serviceaccount:kube-system:default" cannot get namespaces in the namespace "kube-system". (get namespaces kube-system)
goroutine 1 [running]:
k8s.io/kubernetes/vendor/github.com/golang/glog.stacks(0x1f7f600, 0xc400000000, 0xee, 0x1b2)
        vendor/github.com/golang/glog/glog.go:766 +0xa5
k8s.io/kubernetes/vendor/github.com/golang/glog.(*loggingT).output(0x1f5f5c0, 0xc400000003, 0xc42006c300, 0x1ef20c8, 0x22, 0x3c, 0x0)
        vendor/github.com/golang/glog/glog.go:717 +0x337
k8s.io/kubernetes/vendor/github.com/golang/glog.(*loggingT).printf(0x1f5f5c0, 0xc400000003, 0x16949d6, 0x1e, 0xc420579ee8, 0x2, 0x2)
        vendor/github.com/golang/glog/glog.go:655 +0x14c
k8s.io/kubernetes/vendor/github.com/golang/glog.Fatalf(0x16949d6, 0x1e, 0xc420579ee8, 0x2, 0x2)
        vendor/github.com/golang/glog/glog.go:1145 +0x67
main.main()
        cluster/addons/fluentd-elasticsearch/es-image/elasticsearch_logging_discovery.go:60 +0xb53
[2018-01-19 00:18:45,273][INFO ][node                     ] [elasticsearch-logging-v1-wkwcs] version[2.4.1], pid[5], build[c67dc32/2016-09-27T18:57:55Z]
[2018-01-19 00:18:45,275][INFO ][node                     ] [elasticsearch-logging-v1-wkwcs] initializing ...

# kubectl --kubeconfig kubeconfig.yaml --namespace kube-system exec kibana-logging-2104905774-69wgv curl elasticsearch-logging.kube-system:9200/_cat/indices?pretty

{
  "error" : {
    "root_cause" : [ {
      "type" : "master_not_discovered_exception",
      "reason" : null
    } ],
    "type" : "master_not_discovered_exception",
    "reason" : null
  },
  "status" : 503
}

关闭RBAC：

#  kubectl --kubeconfig kubeconfig.yaml --namespace kube-system log elasticsearch-logging-v1-7shgk
[2018-01-26 01:19:52,294][INFO ][node                     ] [elasticsearch-logging-v1-7shgk] version[2.4.1], pid[5], build[c67dc32/2016-09-27T18:57:55Z]
[2018-01-26 01:19:52,294][INFO ][node                     ] [elasticsearch-logging-v1-7shgk] initializing ...
[2018-01-26 01:19:53,077][INFO ][plugins                  ] [elasticsearch-logging-v1-7shgk] modules [reindex, lang-expression, lang-groovy], plugins [], sites []

#  kubectl --kubeconfig kubeconfig.yaml --namespace kube-system exec elasticsearch-logging-v1-7shgk curl http://elasticsearch-logging:9200/_cat/indices?pretty
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100    40  100    40    0     0      2      0  0:00:20  0:00:15  0:00:05    10
green open .kibana 1 1 1 0 6.2kb 3.1kb

感谢大家的帮助：）

Answer 1

一些疑难解答提示：

1）确保ElasticSearch正常运行。

输入运行elasticsearch的容器并运行：

curl localhost:9200

你应该得到一个JSON，其中有一些关于elasticsearch的数据。

2）确保可以从kibana容器中访问ElasticSearch

输入kibana容器并运行：

curl <elasticsearch_service_name>:9200

你应该得到与上面相同的输出。

3）确保您的ES指数正常。

从elasticsearch容器运行以下命令：

curl localhost:9200/_cat/indices?pretty

您应该获得一个表格，其中包含ES群集中的所有索引及其状态（如果您只有一个ES副本，则应为绿色或黄色）。

如果上述任何一点失败，请检查ES容器的日志是否有任何错误消息并尝试解决。

Answer 2

此异常表示2配置错误 1. Kubernetes的DNS插件无法正常工作。检查你的dns插件日志 2. Pod 2 Pod通讯无法正常工作。这与你的底层sdn addon cni法兰绒印花布有关。

您可以通过ping另一个pod中的一个pod进行检查。如果它不起作用，请检查您的网络配置，尤其是kube-proxy组件。

Kibana未连接到ElasticSearch

2 个答案: