Red Hat Bugzilla – Bug 1317097
Runtime Error when using registry.access.redhat.com/rhel6 image for oc new-app
Last modified: 2016-05-12 12:32:36 EDT
Description of problem: After pulling, tagging, and pushing the registry.access.redhat.com/rhel6 image to my integrated docker registry I try to run either `oc new-app --image=mytest/rhel6` or `oc new-app --image-stream=rhel6` and I get the following error messages [0]. [0] http://paste-platops.itos.redhat.com/pksei8bxt/rrpnog Version-Release number of selected component (if applicable): openshift version openshift v3.1.1.6 kubernetes v1.1.0-origin-1107-g4c8e6f4 etcd 2.1.2 How reproducible: Everytime Steps to Reproduce: 1. docker pull registry.access.redhat.com/rhel6 2. docker tag registry.access.redhat.com/rhel6 myregistry.apps.example.com/mytest/rhel6 3. docker push myregistry.apps.example.com/mytest/rhel6 4. `oc new-app --image=mytest/rhel6` or `oc new-app --image-stream=rhel6` Actual results: The error message shown in [0]. Expected results: Deploy the pod and create all necessary resources
@Eric I don't meet your error in OSE3.2.0.1 environments with registry.access.redhat.com/rhel6(imageid is 31b925c88737). But can't create container with below error. Events: FirstSeen LastSeen Count From SubobjectPath Type Reason Message --------- -------- ----- ---- ------------- -------- ------ ------- <invalid> <invalid> 1 {default-scheduler } Normal Scheduled Successfully assigned rhel6-1-jb2vp to ip-172-18-0-241.ec2.internal <invalid> <invalid> 4 {kubelet ip-172-18-0-241.ec2.internal} spec.containers{rhel6} Normal Pulling pulling image "registry.access.redhat.com/rhel6:latest" <invalid> <invalid> 4 {kubelet ip-172-18-0-241.ec2.internal} spec.containers{rhel6} Normal Pulled Successfully pulled image "registry.access.redhat.com/rhel6:latest" <invalid> <invalid> 4 {kubelet ip-172-18-0-241.ec2.internal} spec.containers{rhel6} Warning Failed Failed to create docker container with error: API error (500): No command specified <invalid> <invalid> 4 {kubelet ip-172-18-0-241.ec2.internal} Warning FailedSync Error syncing pod, skipping: failed to "StartContainer" for "rhel6" with RunContainerError: "runContainer: API error (500): No command specified\n"
@XiuJuan Can you test it on 3.1.1.6? The customer _and_ I experienced this issue there. Is there a way to determine what changed between the 2 so we can get this fixed? Or backported maybe?
My assumption is that the error in comment #2 is expected. You can't simply run the rhel6 image. It's not an s2i image and you aren't passing any source in. It's not an image that has an entrypoint that is going to do anything. I suspect that's why "No command specified" is probably accurate. I suspect in 3.1.1.6 we simply didn't do a good job of exposing that error message. I think the real question to ask is what is the end goal? I would think you would want to create a build that would add a layer on top of the rhel6 image and actually run a service.
The ultimate goal was to use the rhel6 image as the base for a dockerfile, but they got the same error message that I got when I simply tried to deploy the image. I understand that the image will ultimately fail when you try to deploy _just_ the image, but it shouldn't fail with the error that I got in comment #1 which is the same error that the customer got when they tried to use it properly.
This is what the customer using. a very basic Dockerfile with the following: FROM rhel6:latest EXPOSE 80 COPY base.repo /etc/yum.repos.d/ RUN yum clean all -y && \ yum install -y httpd && \ yum clean all -y CMD /usr/sbin/httpd -DFOREGROUND and they are getting the same error. Thanks & regards, Shanna
@Eric Rreproduce this bug with ose 3.1.1.6 environment and oc v3.1.1.6-21-gcd70c35 client. But works with ose 3.1.1.6 environment and latest oc version. Just my opinion,looks like oc client issue.
This was fixed in origin by https://github.com/openshift/origin/pull/6670
Available in latest 3.2 puddle
This bug has been fixed with $oc version oc v3.2.0.4 kubernetes v1.2.0-origin-41-g91d3e75 No error when create rhel6 image using new-app cmd. $ oc get is NAME DOCKER REPO TAGS UPDATED rhel6 172.31.168.158:5000/xiuwang/rhel6 latest Less than a second ago [wxj@dhcp-129-163 ~]$ oc new-app rhel6 --name=myrhel6 --> Found image 31b925c (2 weeks old) in image stream rhel6 under tag "latest" for "rhel6" * This image will be deployed in deployment config "myrhel6" * The image does not expose any ports - if you want to load balance or send traffic to this component you will need to create a service with 'expose dc/myrhel6 --port=[port]' later --> Creating resources with label app=myrhel6 ... deploymentconfig "myrhel6" created --> Success Run 'oc status' to view your app. new-app should create the app pointing to the destination tag instead of "latest". $ oc get is mysql -n openshift NAME DOCKER REPO TAGS UPDATED mysql 172.31.168.158:5000/openshift/mysql 5.5,5.6,latest 37 hours ago [wxj@dhcp-129-163 ~]$ oc new-app mysql --> Found image 09a304f (2 weeks old) in image stream "mysql" in project "openshift" under tag "5.6" for "mysql"
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2016:1064