Hostname resolution on nginx/ubuntu container (excerise 7.2, step#11)

furqanbaqai

Hi Guys,

While doing exercise 7.2, step#11, i am getting following error while updating apt repositories:

root@thirdpage:/# apt-get update
Err:1 http://deb.debian.org/debian stretch InRelease Temporary failure resolving 'deb.debian.org'
Err:2 http://security.debian.org/debian-security stretch/updates InRelease Temporary failure resolving 'security.debian.org'

My container's resolv.conf have following content:

root@thirdpage:/# cat /etc/resolv.conf  
nameserver 10.96.0.10  
search default.svc.cluster.local svc.cluster.local cluster.local localdomain  
options ndots:5

NOTE My Host OS /etc/network/interfaces file is configured as:

baqai@lfdtru1804n01:/etc/init.d$ cat /etc/network/interfaces
# This file describes the network interfaces available on your system
# and how to activate them. For more information, see interfaces(5).

source /etc/network/interfaces.d/*

# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interface
auto ens32
iface ens32 inet dhcp

NOTE: I am doing all these exercises on local VMs created on VMWare Workstation.

Appreciate your help..

Thanks.
Furqan Baqai

serewicz

Hello Furqan,

Definitely a DNS issue. We do not test using VMWare though.

Having said that, you did test and found that 10.96.0.10 was the listed DNS server inside the container as I see in the /etc/resolv.conf file. This is most likely the IP address of the kubernetes service, found with kubectl get svc -o wide.

Do apt-get commands work inside the master VM? Same for the worker node? What IP address do they use for DNS?

You may want to use kubectl get pods -o wide and kubectl get svc to find that IP address. Also ensure that the coredns and calico-node pods are properly running.

Regards,

furqanbaqai

Hi Serewicz,

Thank you for your response.

Responding to your questions inline:
1. Does apt-get commands on the host work ?
Yes apt-get is working and ping to a DNS is working as well
2. Output of kubectl get pods -o wide:

baqai@lfdtru1804n01:~$ kubectl get svc -o wide
NAME         TYPE           CLUSTER-IP       EXTERNAL-IP   PORT(S)        AGE   SELECTOR
kubernetes   ClusterIP      10.96.0.1        <none>        443/TCP        21d   <none>
nginx        ClusterIP      10.100.147.5     <none>        443/TCP        16d   io.kompose.service=nginx
registry     ClusterIP      10.102.89.186    <none>        5000/TCP       16d   io.kompose.service=registry
secondapp    LoadBalancer   10.102.242.149   <pending>     80:32000/TCP   34h   example=second
thirdpage    NodePort       10.107.220.47    <none>        80:31325/TCP   24h   example=third

3. Output of kubectl get svc:

baqai@lfdtru1804n01:~$ kubectl get svc
NAME         TYPE           CLUSTER-IP       EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP      10.96.0.1        <none>        443/TCP        21d
nginx        ClusterIP      10.100.147.5     <none>        443/TCP        16d
registry     ClusterIP      10.102.89.186    <none>        5000/TCP       16d
secondapp    LoadBalancer   10.102.242.149   <pending>     80:32000/TCP   34h
thirdpage    NodePort       10.107.220.47    <none>        80:31325/TCP   24h

4. Also ensure that the coredns and calico-node pods are properly running.
   I checked if the pods are running by issuing command kubectl get -n kube-system podsand found out that the pods are seeemd to be working. Not sure if it is something to do with their configuration. Please find the relevant output

baqai@lfdtru1804n01:~$ kubectl get -n kube-system pods
NAME                                    READY   STATUS    RESTARTS   AGE
calico-node-k5tv2                       2/2     Running   48         21d
calico-node-zwm9z                       2/2     Running   56         21d
coredns-86c58d9df4-62h6j                1/1     Running   28         21d
coredns-86c58d9df4-fzhk6                1/1     Running   28         21d
etcd-lfdtru1804n01                      1/1     Running   29         21d
kube-apiserver-lfdtru1804n01            1/1     Running   31         21d
kube-controller-manager-lfdtru1804n01   1/1     Running   29         21d
kube-proxy-8cs5r                        1/1     Running   25         21d
kube-proxy-qfkcc                        1/1     Running   28         21d
kube-scheduler-lfdtru1804n01            1/1     Running   29         21d
traefik-ingress-controller-qhv4s        1/1     Running   1          24h
traefik-ingress-controller-vv8nh        1/1     Running   1          24h

Thanks in advance for your help...

Regards,
Furqan Baqai

furqanbaqai

Hi Guys,

I think i found out the issue. As i am running all labs on VMWare workstation, kubernetes core DNS services are pointing to my vmware gateway for DNS resolution which is evidently failing. Please check below logs:

baqai@lfdtru1804n01:~$ kubectl logs --namespace=kube-system coredns-86c58d9df4-62h6j
.:53
2019-05-20T17:54:20.286Z [INFO] CoreDNS-1.2.6
2019-05-20T17:54:20.286Z [INFO] linux/amd64, go1.11.2, 756749c
CoreDNS-1.2.6
linux/amd64, go1.11.2, 756749c
 [INFO] plugin/reload: Running configuration MD5 = f65c4821c8a9b7b5eb30fa4fbc167769
 [ERROR] plugin/errors: 2 325794039783830212.4647384507963668824. HINFO: unreachable backend: read udp 192.168.0.163:47806->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 325794039783830212.4647384507963668824. HINFO: unreachable backend: read udp 192.168.0.163:49965->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 325794039783830212.4647384507963668824. HINFO: unreachable backend: read udp 192.168.0.163:58770->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 325794039783830212.4647384507963668824. HINFO: unreachable backend: read udp 192.168.0.163:37151->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 325794039783830212.4647384507963668824. HINFO: unreachable backend: read udp 192.168.0.163:42153->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 325794039783830212.4647384507963668824. HINFO: unreachable backend: read udp 192.168.0.163:46402->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 325794039783830212.4647384507963668824. HINFO: unreachable backend: read udp 192.168.0.163:35809->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 325794039783830212.4647384507963668824. HINFO: unreachable backend: read udp 192.168.0.163:40866->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 325794039783830212.4647384507963668824. HINFO: unreachable backend: read udp 192.168.0.163:33442->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 325794039783830212.4647384507963668824. HINFO: unreachable backend: read udp 192.168.0.163:54784->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 www.linuxfoundation.org. AAAA: unreachable backend: read udp 192.168.0.163:47129->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 www.linuxfoundation.org. A: unreachable backend: read udp 192.168.0.163:47656->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 www.linuxfoundation.org. A: unreachable backend: read udp 192.168.0.163:56256->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 www.linuxfoundation.org. AAAA: unreachable backend: read udp 192.168.0.163:42880->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 www.linuxfoundation.org. A: unreachable backend: read udp 192.168.0.163:60223->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 www.linuxfoundation.org. AAAA: unreachable backend: read udp 192.168.0.163:57832->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 www.linuxfoundation.org. A: unreachable backend: read udp 192.168.0.163:56050->192.168.159.2:53: i/o timeout
 [ERROR] plugin/errors: 2 www.linuxfoundation.org. AAAA: unreachable backend: read udp 192.168.0.163:33869->192.168.159.2:53: i/o timeout

Appreciate if somebody help us out to resolve this.

serewicz

Hello,

Debugging DNS can definitely be a task. There are several configurations which can effect resolution, some like routing, may not actually be a DNS issue as much as an overall network issue.

I'm glad you looked at the logs, I tried to post here suggesting that but the message was blocked for some reason. From the output it looks like the UDP requests are going from 192.168.0.163:47806 to 192.168.159.2:53 with a timeout. Either the forwarding in VMWare is not properly configured, it is a routing issue, or perhaps a firewall issue.

I don't know much of anything about how VMWare sets up the network between VMs by default. In the systems we support and use like GCE and AWS this problem is typically a firewall issue. That either we have to open all ports, or as in the case with VirtualBox we need to activate every interface to use promiscuous mode so as to not drop some packets.

Regards,