This service will be undergoing maintenance at 00:00 UTC, 2017-10-23 It is expected to last about 30 minutes
Bug 1313884 - inconsistent docker api version between docker and atomic
inconsistent docker api version between docker and atomic
Product: Container Development Kit (CDK)
Classification: Red Hat
Component: vms (Show other bugs)
RHEL 7.2.x
Unspecified Unspecified
unspecified Severity high
: beta4
: 2.0
Assigned To: Langdon White
David Kutálek
Vikram Goyal
Depends On:
  Show dependency treegraph
Reported: 2016-03-02 09:27 EST by Ondřej Pták
Modified: 2016-04-19 08:37 EDT (History)
5 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed:
Type: Bug
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)

  None (edit)
Description Ondřej Pták 2016-03-02 09:27:36 EST
Description of problem:
atomic is not able to run atomicapp container and application in openshift

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1. vagrant up
2. vagrant ssh
3. oc login
4. oc new-project testing
5. sudo atomic  run tomaskral/helloflask-atomicapp --provider=openshift --providerconfig=/home/vagrant/.kube/config --namespace=testing

Actual results:
fail on assertion: self.inspect is not None, real problem hiddne by atomic:
400 Client Error: Bad Request ("client is newer than server (client API version: 1.21, server API version: 1.20)")

Expected results:
successful deployment of image

Additional info:
this worked in
Comment 1 Dusty Mabe 2016-03-03 16:14:17 EST
Ok so I hit some issues with Atomic cli when trying to run this:

Status: Image is up to date for

Traceback (most recent call last):
  File "/bin/atomic", line 524, in <module>
  File "/usr/lib/python2.7/site-packages/Atomic/", line 46, in run
    args = self._get_args("RUN")
  File "/usr/lib/python2.7/site-packages/Atomic/", line 361, in _get_args
    labels = self._get_labels()
  File "/usr/lib/python2.7/site-packages/Atomic/", line 279, in _get_labels
    return self._getconfig("Labels", [])
  File "/usr/lib/python2.7/site-packages/Atomic/", line 266, in _getconfig
    assert self.inspect is not None
[vagrant@rhel-cdk ~]$ 

I didn't see the client/server version mismatch though. I ran the atomicapp directly from the command line (cutting out the Atomic CLI middle man and it worked fine using the following command.

sudo docker run -it --rm  --privileged -v $(pwd):/atomicapp -v /run:/run -v /:/host --net=host --name atomicapp tomaskral/helloflask-atomicapp run --provider=openshift --providerconfig=/home/vagrant/.kube/config --namespace=testing

I did notice some other issues with Atomic CLI though [1]. Maybe that version isn't stable.

Comment 2 Ondřej Pták 2016-03-03 18:18:54 EST
Sorry, I didn't mention another related bug, which hides problems with docker api:
Comment 3 Dusty Mabe 2016-03-03 23:32:11 EST
(In reply to Ondřej Pták from comment #2)
> Sorry, I didn't mention another related bug, which hides problems with
> docker api:

Got ya. So I think the issue is between docker and atomic cli; which means atomicapp is not at fault here.

Here are the versions of some components on the cdk box that may help with investigation:

$ rpm -qa | grep 'docker\|atomic' 
Comment 4 David Kutálek 2016-03-04 03:37:22 EST
Actually I think that this issue happened in:

And not happened in:
(this version I tested and it worked)

There was an update of atomic in box #15 I think.
Comment 5 Ondřej Pták 2016-04-19 08:37:56 EDT
now there is docker 1.9.1 in box, with docker api 1.21
version: rhel-cdk-kubernetes-7.2-23.x86_64.vagrant*

Note You need to log in before you can comment on or make changes to this bug.