search

kubernetes / dns

Kubernetes DNS service
Apache License 2.0
898 stars 459 forks source link

NodeLocalDNS not working with custom hosts #592

Open chmodrs opened 1 year ago

chmodrs commented 1 year ago

Hi everyone,

What happened? In EKS 1.22, after installing nodelocaldns, in-cluster DNS and external DNS works normally, but custom hosts stored in coredns configmap, stop working.

What did you expect to happen? I expect that even with nodelocaldns, custom hosts continue to work normally.

How can we reproduce it (as minimally and precisely as possible)?

  1. Configure custom hosts in coredns configmap 
    
# kubectl edit configmap coredns -n kube-system

hosts custom.hosts mydns.rtm { 8.8.4.4 mydns.rtm fallthrough }

2. Ping mydns.rtm

kubectl exec -it nginx2-55764b6d95-dvvdh -- ping mydns.rtm

PING mydns.rtm (8.8.4.4): 56 data bytes 64 bytes from 8.8.4.4: seq=0 ttl=105 time=2.401 ms 64 bytes from 8.8.4.4: seq=1 ttl=105 time=1.158 ms 64 bytes from 8.8.4.4: seq=2 ttl=105 time=1.286 ms

3. Download nodelocaldns manifests and replace values.

!/bin/sh

echo "Downloading nodelocaldns" wget https://raw.githubusercontent.com/kubernetes/kubernetes/v1.22.17/cluster/addons/dns/nodelocaldns/nodelocaldns.yaml -O manifests/nodelocaldns.yaml

echo "replacing values" sed -i '' -e 's/PILLARDNSDOMAIN/cluster.local/g' manifests/nodelocaldns.yaml sed -i '' -e 's/PILLARDNSSERVER/172.20.0.10/g' manifests/nodelocaldns.yaml ### kubectl -n kube-system get service kube-dns sed -i '' -e 's/PILLARLOCALDNS/169.254.20.10/g' manifests/nodelocaldns.yaml ### default nodelocadns ip address

4. Apply manifests.
`# kubectl apply -f manifests/nodelocaldns.yaml`

5. Stop to working

kubectl exec -it nginx2-55764b6d95-dvvdh -- ping mydns.rtm

ping: bad address 'mydns.rtm' command terminated with exit code 1

buccella commented 7 months ago

From this comment, NodeLocal DNS cache will read the kube-dns ConfigMap, but will not read the coredns ConfigMap:

https://github.com/kubernetes/dns/issues/452#issuecomment-865206337

Running CoreDNS with NodeLocal DNS cache seems like a pretty typical use case though, so it would be nice if this was supported.

k8s-triage-robot commented 4 months ago

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

You can:

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

kilsilent commented 4 months ago

Is there any progress on that issue? I'm having the same problem,how did you end up solving the problem @chmodrs

kilsilent commented 4 months ago

/remove-lifecycle stale

kilsilent commented 4 months ago

I fixed the problem, which was caused by a configuration error When these statements are executed, the resulting corefile is like

apiVersion: v1
data:
  Corefile: |
    cluster.local:53 {
    ...
        forward . 172.20.0.10 {
                force_tcp
        }
        prometheus :9253
        health 169.254.20.10:8080
        }
  this cause only cluster.local domain names will be forwarded to coredns, the rest of the domain names will be resolved locally, so this error is reported

If you want to solve this problem, you can configure the domain of the custom domain name into the Corefile, and like that, the problem is solved

apiVersion: v1
data:
  Corefile: |
    rtm cluster.local:53 {
....
        }
k8s-ci-robot commented 4 months ago

@kilsilent: You can't close an active issue/PR unless you authored it or you are a collaborator.

In response to [this](https://github.com/kubernetes/dns/issues/592#issuecomment-1999085513): >/close > Instructions for interacting with me using PR comments are available [here](https://git.k8s.io/community/contributors/guide/pull-requests.md). If you have questions or suggestions related to my behavior, please file an issue against the [kubernetes/test-infra](https://github.com/kubernetes/test-infra/issues/new?title=Prow%20issue:) repository.
k8s-triage-robot commented 1 month ago

The Kubernetes project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

You can:

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

k8s-triage-robot commented 4 days ago

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

You can:

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten