[Container-tools] [aos-devel] Useing the RHEL CDK 2.0

[Subhendu Ghosh]

Yes please 


Sent from Nine

From: James Mills
Sent: Nov 11, 2015 9:18 AM
To: Lalatendu Mohanty
Cc: aos-devel
Subject: Re: [aos-devel] Useing the RHEL CDK 2.0

All,

I've found the problem.

The issue is in fact that /usr/bin/kubectl is a symlink to /usr/bin/openshift (thanks Jason!).  During the atomic run, this happens:

docker run -it --rm  --privileged -v /home/vagrant:/atomicapp -v /run:/run -v /:/host --net=host --name helloapache -e NAME=helloapache -e IMAGE=projectatomic/helloapache projectatomic/helloapache 

and the kubernetes provider portion of the atomicapp looks for /host/usr/bin/kubectl.  Unfortunately at that point, kubectl is a *broken* symbolic link to /usr/bin/openshift.

In order to fix it, you can either remove the symlink and copy openshift directly to kubectl, or you can recreate the symbolic link to be relative instead of absolute:

# cd /usr/bin
# ln -sf openshift kubectl


Should I open a bug against the CDK?

Thanks!
~james


On Tue, Nov 10, 2015 at 9:26 PM, Lalatendu Mohanty <lmohanty at redhat.com> wrote:
>
> On 11/11/2015 03:31 AM, Jason Brooks wrote:
>>
>>
>> ----- Original Message -----
>>>
>>> From: "James Mills" <jamills at redhat.com>
>>> To: "aos-devel" <aos-devel at redhat.com>, "atomic-sme" <atomic-sme at redhat.com>
>>> Sent: Tuesday, November 10, 2015 1:50:46 PM
>>> Subject: [aos-devel] Useing the RHEL CDK 2.0
>>>
>>> All,
>>>
>>> I pulled down the CDK zip and libvirt image from access.redhat.com, and
>>> tried to fire up the basic "helloapache" example.  Unfortunately, I ran
>>> into a few problems:
>>>
>>> The README in the CDK has me clone the adb repo, and the docs from that
>>> repo let me know I can test my setup using the helloapache example [1].
>>> The helloapache [2] non-interactive test fails for me:
>>>
>>> # cat /etc/redhat-release
>>> Red Hat Enterprise Linux Server release 7.2 Beta (Maipo)
>>> # atomic run projectatomic/helloapache
>>> ...
>>> raise ProviderFailedException("No kubectl found in %s" %
>>> ":".join(test_paths))
>>> atomicapp.plugin.ProviderFailedException: No kubectl found in
>>> /usr/bin/kubectl:/usr/local/bin/kubectl
>>>
>>> I went ahead and did the same workflow using only the upstream
>>> atomicapp/dev box, and the example worked as expected:
>>>
>>> #  cat /etc/centos-release
>>> CentOS Linux release 7.1.1503 (Core)
>>> # atomic run projectatomic/helloapache
>>> # kubectl get pods
>>> NAME          READY     STATUS    RESTARTS   AGE
>>> helloapache   1/1       Running   0          1m
>>>
>>> I'm not sure if I'm seeing a bug, or if I'm missing something.  ANy help
>>> would be appreciated.
>>
>> I've seen an issue like this with versions of kube where
>> "kubectl" is actually a symlink to "openshift".
>>
>> Jason
>
>
> In the CDK Vagrant box, we are using RHEL's k8s packages for the k8s setup.
>
> -Lala
>>
>>   
>>>
>>> Thanks!
>>> ~james
>>>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/container-tools/attachments/20151111/23878cbc/attachment.htm>