Origin: Nginx container can't resolve internal DNS hostnames

[atoy3731]

When using a variable to proxy to an internal Openshift service within an nginx container's proxy_pass config, NGINX can't resolve the service's DNS due to a required resolver. For instance:

location /api/ {
   set $pass_url http://service.namespace.svc:8080$request_uri;
   proxy_pass  $pass_url;
}

When using standard Kubernetes, I can use kube-dns.kube-system.svc.cluster.local as the resolver:

resolver kube-dns.kube-system.svc.cluster.local;

But Openshift doesn't provide this. I've tried using the IP that is in the container's /etc/resolv.conf, which is just one of the nodes in my cluster that is running the DNS server, but it still can't resolve.

Weirdest part is nslookup service.namespace.svc from inside the container terminal uses the nameserver in /etc/resolv.conf and works fine.

Is there an equivalent to the Kubernetes DNS hostname in Openshift I could use?

Version

openshift v3.11.0+d0c29df-98
kubernetes v1.11.0+d4cacc0

Steps To Reproduce

1. Use an NGINX container.
2. Modify a location block in the NGINX configuration to point to an internal service:

location /api/ {
   set $pass_url http://service.namespace.svc:8080$request_uri;
   proxy_pass  $pass_url;
}

1. Use ps aux | grep nginx to find the master PID and reload NGINX via kill -HUP <pid>
2. Hit the route at /api/
3. Check the NGINX logs.

Current Result

NGINX can't resolve internal hostname.

Expected Result

NGINX proxies to the internal service.

Additional Information

[openshift-bot]

Issues go stale after 90d of inactivity.

Mark the issue as fresh by commenting /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.
Exclude this issue from closing by commenting /lifecycle frozen.

If this issue is safe to close now please do so with /close.

/lifecycle stale

[openshift-bot]

Stale issues rot after 30d of inactivity.

Mark the issue as fresh by commenting /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.
Exclude this issue from closing by commenting /lifecycle frozen.

If this issue is safe to close now please do so with /close.

/lifecycle rotten
/remove-lifecycle stale

[Vandborg]

I am facing the exact same problem. So I'll just watch this issue, if anybody comes up with a solution. :)

[openshift-bot]

Rotten issues close after 30d of inactivity.

Reopen the issue by commenting /reopen.
Mark the issue as fresh by commenting /remove-lifecycle rotten.
Exclude this issue from closing again by commenting /lifecycle frozen.

/close

[openshift-ci-robot]

@openshift-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.

Reopen the issue by commenting /reopen.
Mark the issue as fresh by commenting /remove-lifecycle rotten.
Exclude this issue from closing again by commenting /lifecycle frozen.

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[ainiml]

/reopen

having the same issue, unable to use kube-dns.kube-system.svc.cluster.local as the resolver

[openshift-ci-robot]

@ainiml: You can't reopen an issue/PR unless you authored it or you are a collaborator.

In response to this:

/reopen

having the same issue, unable to use kube-dns.kube-system.svc.cluster.local as the resolver

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[omnibrain]

/reopen

Having the same problem

[openshift-ci-robot]

@omnibrain: You can't reopen an issue/PR unless you authored it or you are a collaborator.

In response to this:

/reopen

Having the same problem

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[robertofabrizi]

Anyone has resolved this? I read to use the DNS at master-ip port 8053, but it still fails to resolve (nginx, doing a manual curl from inside the nginx container does work)

[ainiml]

Anyone has resolved this? I read to use the DNS at master-ip port 8053, but it still fails to resolve (nginx, doing a manual curl from inside the nginx container does work)

I wonder if there's been a change to docker

I used to could ping other containers by their docker container name, but not anymore now