【kubernetes/k8s源码分析】kube-dns 源码解析之sidecar

github地址:https://github.com/kubernetes/dns

 

sidecar

● 定期检查kubedns和dnsmasq的健康状态(nslookup)

● 为k8s活性检测提供HTTP API

/sidecar 

--v=2 

--logtostderr 

--probe=kubedns,127.0.0.1:10053,kubernetes.default.svc.cluster.local,5,A 

--probe=dnsmasq,127.0.0.1:53,kubernetes.default.svc.cluster.local,5,A

 

一 sidecar启动流程

 

1 main函数

 

解析参数,创建参数配置,启动服务

func main() {
   options := sidecar.NewOptions()
   configureFlags(options, pflag.CommandLine)
   flag.InitFlags()
   // Convinces goflags that we have called Parse() to avoid noisy logs.
   // OSS Issue: kubernetes/kubernetes#17162.
   goflag.CommandLine.Parse([]string{})
   logs.InitLogs()
   defer logs.FlushLogs()

   glog.Infof("Version v%s", version.VERSION)

   version.PrintAndExitIfRequested()

   server := sidecar.NewServer()
   server.Run(options)
}

 

2 Run函数

 

主要是启动l两个健康检查,kubedns,dnsmasq

 

--probe=kubedns,127.0.0.1:10053,kubernetes.default.svc.cluster.local,5,A 

--probe=dnsmasq,127.0.0.1:53,kubernetes.default.svc.cluster.local,5,A

 

// Run the server (does not return)
func (s *server) Run(options *Options) {
   s.options = options
   glog.Infof("Starting server (options %+v)", *s.options)

   for _, probeOption := range options.Probes {
      probe := &dnsProbe{DNSProbeOption: probeOption}
      s.probes = append(s.probes, probe)
      probe.Start(options)
   }

   s.runMetrics(options)
}

3 Start函数

 

注册http API

 

func (p *dnsProbe) Start(options *Options) {
   glog.V(2).Infof("Starting dnsProbe %+v", p.DNSProbeOption)

   p.lastError = fmt.Errorf("waiting for first probe")

   http.HandleFunc("/healthcheck/"+p.Label, p.httpHandler)
   p.registerMetrics(options)

   if p.delayer == nil {
      glog.V(4).Infof("Using defaultLoopDelayer")
      p.delayer = &defaultLoopDelayer{}
   }

   go p.loop()
}

4 loop函数

 

定期探测,发送DNS请求

 

func (p *dnsProbe) loop() {
   glog.V(4).Infof("Starting loop")
   p.delayer.Start(p.Interval)

   dnsClient := &dns.Client{}

   for {
      glog.V(4).Infof("Sending DNS request @%v %v", p.Server, p.Name)
      msg, latency, err := dnsClient.Exchange(p.msg(), p.Server)
      glog.V(4).Infof("Got response, err=%v after %v", err, latency)

      if err == nil && len(msg.Answer) == 0 {
         err = fmt.Errorf("no RRs for domain %q", p.Name)
      }

      p.update(err, latency)
      p.delayer.Sleep(latency)
   }
}

 

 

 

 

e=admin t=2025-05-27T16:03:49.522532125Z level=info msg="Request Completed" method=POST path=/api/ds/query status=400 remote_addr=10.244.1.0 time_ms=30004 duration=30.0044843s size=548 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/ds/query status_source=downstream logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:03:49.54540158Z level=error msg="Exemplar query failed" query="sum(kube_node_status_allocatable{origin_prometheus=~\"\",resource=\"memory\", unit=\"byte\", node=~\"^$\"})" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": dial tcp 10.109.175.244:9090: i/o timeout" logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:04:09.517139742Z level=error msg="Exemplar query failed" query="sum(kube_pod_container_resource_requests{origin_prometheus=~\"\",resource=\"memory\", unit=\"byte\",node=~\"^$\"})" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": dial tcp 10.109.175.244:9090: i/o timeout" logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:04:09.524447242Z level=error msg="Exemplar query failed" query="sum(kube_pod_container_resource_requests{origin_prometheus=~\"\",resource=\"cpu\", unit=\"core\",node=~\"^$\"})" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": dial tcp 10.109.175.244:9090: i/o timeout" logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:04:09.530136946Z level=error msg="Exemplar query failed" query="sum(kube_node_status_allocatable{origin_prometheus=~\"\",resource=\"cpu\", unit=\"core\", node=~\"^$\"})" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": dial tcp 10.109.175.244:9090: i/o timeout" logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:04:09.551153764Z level=error msg="Exemplar query failed" query="sum (container_memory_working_set_bytes{origin_prometheus=~\"\",container!=\"\",node=~\"^$\"})" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": dial tcp 10.109.175.244:9090: i/o timeout" logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:04:09.551261227Z level=info msg="Request Completed" method=POST path=/api/ds/query status=400 remote_addr=10.244.1.0 time_ms=40010 duration=40.010425856s size=406 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/ds/query status_source=downstream logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:04:12.007179521Z level=error msg="Exemplar query failed" query="sum (irate(container_cpu_usage_seconds_total{origin_prometheus=~\"\",id=\"/\",node=~\"^$\"}[2m]))" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": context canceled" logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:04:12.007522403Z level=info msg="Request Completed" method=POST path=/api/ds/query status=400 remote_addr=10.244.1.0 time_ms=22481 duration=22.481742859s size=381 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/ds/query status_source=downstream logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:04:12.007750053Z level=error msg="Exemplar query failed" query="count(kube_node_info{origin_prometheus=~\"\"})" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": context canceled" logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:04:12.00727399Z level=info msg="Request Completed" method=POST path=/api/ds/query status=400 remote_addr=10.244.1.0 time_ms=42482 duration=42.482681596s size=1032 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/ds/query status_source=downstream logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:04:12.008305258Z level=error msg="Exemplar query failed" query="sum(kube_pod_container_resource_limits{origin_prometheus=~\"\",resource=\"cpu\", unit=\"core\",node=~\"^$\"})" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": context canceled" logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:04:12.008338346Z level=error msg="Exemplar query failed" query="count(kube_pod_info{origin_prometheus=~\"\",created_by_kind!~\"<none>|Job\",node=~\"^$\"})" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": context canceled" logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:04:12.008580027Z level=info msg="Request Completed" method=POST path=/api/ds/query status=400 remote_addr=10.244.1.0 time_ms=62492 duration=1m2.492137094s size=701 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/ds/query status_source=downstream logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:04:12.008745526Z level=error msg="Exemplar query failed" query="sum(kube_node_status_allocatable{origin_prometheus=~\"\",resource=\"pods\", unit=\"integer\",node=~\"^$\"})" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": context canceled" logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:04:12.008810034Z level=info msg="Request Completed" method=POST path=/api/ds/query status=400 remote_addr=10.244.1.0 time_ms=2454 duration=2.454267935s size=527 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/ds/query status_source=downstream logger=tsdb.prometheus endpoint=queryData pluginId=prometheus dsName=Prometheus dsUID=prometheus uname=admin t=2025-05-27T16:04:12.009153219Z level=error msg="Exemplar query failed" query="sum(kube_pod_container_resource_limits{origin_prometheus=~\"\",resource=\"memory\", unit=\"byte\",node=~\"^$\"})" err="Post \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/query_exemplars\": context canceled" logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:04:12.009254653Z level=info msg="Request Completed" method=POST path=/api/ds/query status=400 remote_addr=10.244.1.0 time_ms=62501 duration=1m2.501457741s size=701 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/ds/query status_source=downstream logger=infra.usagestats t=2025-05-27T16:04:33.193272755Z level=info msg="Usage stats are ready to report" logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:05:52.404511678Z level=error msg="no admin configuration available" error="no admin configuration available" remote_addr=10.244.1.0 traceID= logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:05:52.40475163Z level=info msg="Request Completed" method=GET path=/api/v1/ngalert/admin_config status=404 remote_addr=10.244.1.0 time_ms=1 duration=1.763324ms size=59 referer=http://172.25.254.50/alerting/admin handler=/api/v1/ngalert/admin_config status_source=server logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:09:26.491031204Z level=error msg="no admin configuration available" error="no admin configuration available" remote_addr=10.244.1.0 traceID= logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:09:26.491245851Z level=info msg="Request Completed" method=GET path=/api/v1/ngalert/admin_config status=404 remote_addr=10.244.1.0 time_ms=2 duration=2.983592ms size=59 referer=http://172.25.254.50/alerting/admin handler=/api/v1/ngalert/admin_config status_source=server logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:09:37.75798914Z level=info msg="Request Completed" method=GET path=/api/datasources/proxy/uid/alertmanager/api/v1/status/buildinfo status=404 remote_addr=10.244.1.0 time_ms=8 duration=8.517576ms size=19 referer=http://172.25.254.50/connections/datasources/edit/alertmanager handler=/api/datasources/proxy/uid/:uid/* status_source=server logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:09:37.780046916Z level=info msg="Request Completed" method=GET path=/api/datasources/proxy/uid/alertmanager/alertmanager/api/v2/status status=404 remote_addr=10.244.1.0 time_ms=2 duration=2.492519ms size=19 referer=http://172.25.254.50/connections/datasources/edit/alertmanager handler=/api/datasources/proxy/uid/:uid/* status_source=server logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:09:55.576074595Z level=error msg=InternalError error="[plugin.downstreamError] client: failed to call resources: error querying resource: Get \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/label/origin_prometheus/values?match%5B%5D=kube_node_info&start=1748360386&end=1748362186\": dial tcp 10.109.175.244:9090: i/o timeout" remote_addr=10.244.1.0 traceID= logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:09:55.576133342Z level=error msg="Request Completed" method=GET path=/api/datasources/uid/prometheus/resources/api/v1/label/origin_prometheus/values status=500 remote_addr=10.244.1.0 time_ms=10002 duration=10.002799959s size=116 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/datasources/uid/:uid/resources/* status_source=downstream logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:10:05.606246715Z level=error msg=InternalError error="[plugin.downstreamError] client: failed to call resources: error querying resource: Get \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/label/node/values?match%5B%5D=kube_node_info%7Borigin_prometheus%3D~%22%22%7D&start=1748360396&end=1748362196\": dial tcp 10.109.175.244:9090: i/o timeout" remote_addr=10.244.1.0 traceID= logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:10:05.60631776Z level=error msg="Request Completed" method=GET path=/api/datasources/uid/prometheus/resources/api/v1/label/node/values status=500 remote_addr=10.244.1.0 time_ms=10003 duration=10.00376243s size=116 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/datasources/uid/:uid/resources/* status_source=downstream logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:10:05.606327777Z level=error msg=InternalError error="[plugin.downstreamError] client: failed to call resources: error querying resource: Get \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/label/namespace/values?match%5B%5D=kube_namespace_created%7Borigin_prometheus%3D~%22%22%7D&start=1748360396&end=1748362196\": dial tcp 10.109.175.244:9090: i/o timeout" remote_addr=10.244.1.0 traceID= logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:10:05.606375344Z level=error msg="Request Completed" method=GET path=/api/datasources/uid/prometheus/resources/api/v1/label/namespace/values status=500 remote_addr=10.244.1.0 time_ms=10003 duration=10.003738536s size=116 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/datasources/uid/:uid/resources/* status_source=downstream logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:10:15.628274721Z level=error msg=InternalError error="[plugin.downstreamError] client: failed to call resources: error querying resource: Get \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/label/container/values?match%5B%5D=kube_pod_container_info%7Borigin_prometheus%3D~%22%22%2Cnamespace%3D~%22%22%7D&start=1748360406&end=1748362206\": dial tcp 10.109.175.244:9090: i/o timeout" remote_addr=10.244.1.0 traceID= logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:10:15.628347455Z level=error msg="Request Completed" method=GET path=/api/datasources/uid/prometheus/resources/api/v1/label/container/values status=500 remote_addr=10.244.1.0 time_ms=10002 duration=10.002678393s size=116 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/datasources/uid/:uid/resources/* status_source=downstream logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:10:25.643322497Z level=error msg=InternalError error="[plugin.downstreamError] client: failed to call resources: error querying resource: Get \"http://kube-prometheus-stack-prometheus.kube-prometheus-stack:9090/api/v1/label/pod/values?match%5B%5D=kube_pod_container_info%7Borigin_prometheus%3D~%22%22%2Cnamespace%3D~%22%22%2Ccontainer%3D~%22%22%7D&start=1748360416&end=1748362216\": dial tcp 10.109.175.244:9090: i/o timeout" remote_addr=10.244.1.0 traceID= logger=context userId=1 orgId=1 uname=admin t=2025-05-27T16:10:25.643392778Z level=error msg="Request Completed" method=GET path=/api/datasources/uid/prometheus/resources/api/v1/label/pod/values status=500 remote_addr=10.244.1.0 time_ms=10002 duration=10.002621373s size=116 referer="http://172.25.254.50/d/een5u5jx4qzuoa/k8s-dashboard-en-20250125?orgId=1" handler=/api/datasources/uid/:uid/resources/* status_source=downstream
05-29
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值