-->

kubernetes DNS fails

2020-07-24 07:17发布

问题:

I am trying to learn kubernetes, and I have successfully setup a cluster (1-node) on bare-metal, deployed a service and exposed it via ingress.

I tried to implement traefik, to get lets encrypt certificates, but I couldn't get it working and while debugging I noticed, my DNS service wasn't working properly (39 restarts).

I figured I would try to start over, because i been playing around alot, and also to try flannel instead of calico.

I now has a cluster which was created by the following commands:

kubeadm init --pod-network-cidr=192.168.0.0/16
kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml

Testing the DNS with the following commands gives an error:

kubectl create -f https://raw.githubusercontent.com/kubernetes/website/master/content/en/docs/tasks/administer-cluster/busybox.yaml
kubectl exec -ti busybox -- nslookup kubernetes.default

Server:    10.96.0.10
Address 1: 10.96.0.10 kube-dns.kube-system.svc.cluster.local

nslookup: can't resolve 'kubernetes.default'
command terminated with exit code 1

By investigating the status on the pods, i see the dns pod is restarting

kubectl get po --all-namespaces
NAMESPACE     NAME                                        READY     STATUS    RESTARTS   AGE
default       busybox                                     1/1       Running   0          11m
kube-system   etcd-kubernetes-master                      1/1       Running   0          15m
kube-system   kube-apiserver-kubernetes-master            1/1       Running   0          15m
kube-system   kube-controller-manager-kubernetes-master   1/1       Running   0          15m
kube-system   kube-dns-86f4d74b45-8tgff                   3/3       Running   2          25m
kube-system   kube-flannel-ds-6cd9h                       1/1       Running   0          15m
kube-system   kube-flannel-ds-h78ld                       1/1       Running   0          13m
kube-system   kube-proxy-95kkd                            1/1       Running   0          13m
kube-system   kube-proxy-lq7hx                            1/1       Running   0          25m
kube-system   kube-scheduler-kubernetes-master            1/1       Running   0          15m

The DNS pod logs say the following:

kubectl logs kube-dns-86f4d74b45-8tgff dnsmasq -n kube-system
I0621 08:41:51.414587       1 main.go:76] opts: {{/usr/sbin/dnsmasq [-k --cache-size=1000 --no-negcache --log-facility=- --server=/cluster.local/127.0.0.1#10053 --server=/in-addr.arpa/127.0.0.1#10053 --server=/ip6.arpa/127.0.0.1#10053] true} /etc/k8s/dns/dnsmasq-nanny 10000000000}
I0621 08:41:51.414709       1 nanny.go:94] Starting dnsmasq [-k --cache-size=1000 --no-negcache --log-facility=- --server=/cluster.local/127.0.0.1#10053 --server=/in-addr.arpa/127.0.0.1#10053 --server=/ip6.arpa/127.0.0.1#10053]
I0621 08:41:52.255053       1 nanny.go:119]
W0621 08:41:52.255074       1 nanny.go:120] Got EOF from stdout
I0621 08:41:52.256152       1 nanny.go:116] dnsmasq[10]: started, version 2.78 cachesize 1000
I0621 08:41:52.256216       1 nanny.go:116] dnsmasq[10]: compile time options: IPv6 GNU-getopt no-DBus no-i18n no-IDN DHCP DHCPv6 no-Lua TFTP no-conntrack ipset auth no-DNSSEC loop-detect inotify
I0621 08:41:52.256245       1 nanny.go:116] dnsmasq[10]: using nameserver 127.0.0.1#10053 for domain ip6.arpa
I0621 08:41:52.256260       1 nanny.go:116] dnsmasq[10]: using nameserver 127.0.0.1#10053 for domain in-addr.arpa
I0621 08:41:52.256275       1 nanny.go:116] dnsmasq[10]: using nameserver 127.0.0.1#10053 for domain cluster.local
I0621 08:41:52.256320       1 nanny.go:116] dnsmasq[10]: reading /etc/resolv.conf
I0621 08:41:52.256335       1 nanny.go:116] dnsmasq[10]: using nameserver 127.0.0.1#10053 for domain ip6.arpa
I0621 08:41:52.256350       1 nanny.go:116] dnsmasq[10]: using nameserver 127.0.0.1#10053 for domain in-addr.arpa
I0621 08:41:52.256365       1 nanny.go:116] dnsmasq[10]: using nameserver 127.0.0.1#10053 for domain cluster.local
I0621 08:41:52.256379       1 nanny.go:116] dnsmasq[10]: using nameserver 127.0.0.53#53
I0621 08:41:52.256432       1 nanny.go:116] dnsmasq[10]: read /etc/hosts - 7 addresses
I0621 08:50:43.727968       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:50:53.750313       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:51:03.879573       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:51:13.887735       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:51:23.957996       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:51:34.016679       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:51:43.032107       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:51:53.076274       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:52:03.359643       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:52:13.434993       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:52:23.497330       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:52:33.591295       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:52:43.639024       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:52:53.681231       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:53:03.717874       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:53:13.794725       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:53:23.877015       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)
I0621 08:53:33.974114       1 nanny.go:116] dnsmasq[10]: Maximum number of concurrent DNS queries reached (max: 150)

kubectl logs kube-dns-86f4d74b45-8tgff sidecar -n kube-system
I0621 08:41:57.464915       1 main.go:51] Version v1.14.8
I0621 08:41:57.464987       1 server.go:45] Starting server (options {DnsMasqPort:53 DnsMasqAddr:127.0.0.1 DnsMasqPollIntervalMs:5000 Probes:[{Label:kubedns Server:127.0.0.1:10053 Name:kubernetes.default.svc.cluster.local. Interval:5s Type:33} {Label:dnsmasq Server:127.0.0.1:53 Name:kubernetes.default.svc.cluster.local. Interval:5s Type:33}] PrometheusAddr:0.0.0.0 PrometheusPort:10054 PrometheusPath:/metrics PrometheusNamespace:kubedns})
I0621 08:41:57.465029       1 dnsprobe.go:75] Starting dnsProbe {Label:kubedns Server:127.0.0.1:10053 Name:kubernetes.default.svc.cluster.local. Interval:5s Type:33}
I0621 08:41:57.468630       1 dnsprobe.go:75] Starting dnsProbe {Label:dnsmasq Server:127.0.0.1:53 Name:kubernetes.default.svc.cluster.local. Interval:5s Type:33}
W0621 08:50:46.832282       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:37437->127.0.0.1:53: i/o timeout
W0621 08:50:55.772310       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:57714->127.0.0.1:53: i/o timeout
W0621 08:51:02.779876       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:38592->127.0.0.1:53: i/o timeout
W0621 08:51:09.795385       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:39941->127.0.0.1:53: i/o timeout
W0621 08:51:16.798735       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:41457->127.0.0.1:53: i/o timeout
W0621 08:51:23.802617       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:45709->127.0.0.1:53: i/o timeout
W0621 08:51:30.822081       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:37072->127.0.0.1:53: i/o timeout
W0621 08:51:37.826914       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:49924->127.0.0.1:53: i/o timeout
W0621 08:51:51.093275       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:51194->127.0.0.1:53: i/o timeout
W0621 08:51:58.203965       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:35781->127.0.0.1:53: i/o timeout
W0621 08:52:06.423002       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:42763->127.0.0.1:53: i/o timeout
W0621 08:52:16.455821       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:36143->127.0.0.1:53: i/o timeout
W0621 08:52:23.496199       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:44195->127.0.0.1:53: i/o timeout
W0621 08:52:30.500081       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:48733->127.0.0.1:53: i/o timeout
W0621 08:52:37.519339       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:39179->127.0.0.1:53: i/o timeout
W0621 08:52:51.695822       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:51821->127.0.0.1:53: i/o timeout
W0621 08:52:58.739133       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:46358->127.0.0.1:53: i/o timeout
W0621 08:53:06.823714       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:49103->127.0.0.1:53: i/o timeout
W0621 08:53:16.866975       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:52782->127.0.0.1:53: i/o timeout
W0621 08:53:23.869540       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:52495->127.0.0.1:53: i/o timeout
W0621 08:53:30.882626       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:56134->127.0.0.1:53: i/o timeout
W0621 08:53:37.886811       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:55489->127.0.0.1:53: i/o timeout
W0621 08:53:46.023614       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:58523->127.0.0.1:53: i/o timeout
W0621 08:53:53.034985       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:56026->127.0.0.1:53: i/o timeout
W0621 08:54:00.041734       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:42093->127.0.0.1:53: i/o timeout
W0621 08:54:07.050864       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:58731->127.0.0.1:53: i/o timeout
W0621 08:54:14.053858       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:34062->127.0.0.1:53: i/o timeout
W0621 08:54:21.076986       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:44293->127.0.0.1:53: i/o timeout
W0621 08:54:28.080808       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:42738->127.0.0.1:53: i/o timeout
W0621 08:54:41.423864       1 server.go:64] Error getting metrics from dnsmasq: read udp 127.0.0.1:58715->127.0.0.1:53: i/o timeout

It seems the dns gets spammed somehow, but i am at a loss as to how i should proceed.

回答1:

Digging into Javier's answer, I found the solution to the problem here: https://github.com/kubernetes/kubeadm/issues/787

These were the commands needed to work around the problem:

sudo rm /etc/resolv.conf
sudo ln -s /run/systemd/resolve/resolv.conf /etc/resolv.conf


回答2:

According to https://github.com/kubernetes/kubeadm/issues/272#issuecomment-355386674, this could be due to conflicts with other system services like systemd-resolved.service and resolveconf.service. My advice would be:

  • Check if these services are running
  • Try disabling them and restarting the kube-dns proxy.

If that does not work, the next thing would be checking which of the pods is flooding kube-dns with that many requests. In this sense, something like tcpdump could help. Check this guide: https://github.com/coreos/docs/issues/802#issuecomment-215196547