Bug 1383813

Summary: All pods starting and terminating quickly on 3.4.0.9 with Docker 1.12
Product: OpenShift Container Platform Reporter: Mike Fiedler <mifiedle>
Component: ContainersAssignee: Jhon Honce <jhonce>
Status: CLOSED DUPLICATE QA Contact: DeShuai Ma <dma>
Severity: urgent Docs Contact:
Priority: unspecified    
Version: 3.4.0CC: aos-bugs, jokerman, mmccomas
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2016-10-12 11:41:14 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description Mike Fiedler 2016-10-11 20:27:06 UTC
Description of problem:

Installed 3.4.0.9 on a system updated to Docker 1.12.   The Ansible install was successful (no error messages) but no pods can run on the system.   All attempt to start deployments (router, docker-registry) fail with the following messages in the node logs:

Oct 11 16:16:47 ip-172-31-51-194 atomic-openshift-node: W1011 16:16:47.095749   10935 container.go:352] Failed to create summary reader for "/system.slice/docker-c6a061890c2af7185485dbcca5f884120a824b44d4801a7bf884d87c4d9d9c5e.scope": none of the resources are being tracked.
Oct 11 16:16:47 ip-172-31-51-194 atomic-openshift-node: E1011 16:16:47.141310   10935 docker_manager.go:2127] Failed to setup network for pod "router-1-deploy_default(2ee12e03-8fee-11e6-9927-023dc74be9f5)" using network plugins "redhat/openshift-ovs-multitenant": Error running network setup script: Could not find IP address for container c6a061890c2af7185485dbcca5f884120a824b44d4801a7bf884d87c4d9d9c5e; Skipping pod
Oct 11 16:16:47 ip-172-31-51-194 atomic-openshift-node: I1011 16:16:47.143174   10935 docker_manager.go:1492] Killing container "c6a061890c2af7185485dbcca5f884120a824b44d4801a7bf884d87c4d9d9c5e default/router-1-deploy" with 10 second grace period
Oct 11 16:16:47 ip-172-31-51-194 dockerd-current: time="2016-10-11T16:16:47.143497805-04:00" level=info msg="{Action=stop, LoginUID=4294967295, PID=10935}"
Oct 11 16:16:47 ip-172-31-51-194 dockerd-current: time="2016-10-11T16:16:47.143893617-04:00" level=error msg="Handler for POST /containers/c6a061890c2af7185485dbcca5f884120a824b44d4801a7bf884d87c4d9d9c5e/stop?t=10 returned error: Container c6a061890c2af7185485dbcca5f884120a824b44d4801a7bf884d87c4d9d9c5e is already stopped"
Oct 11 16:16:47 ip-172-31-51-194 dockerd-current: time="2016-10-11T16:16:47.143924202-04:00" level=error msg="Handler for POST /containers/c6a061890c2af7185485dbcca5f884120a824b44d4801a7bf884d87c4d9d9c5e/stop returned error: Container c6a061890c2af7185485dbcca5f884120a824b44d4801a7bf884d87c4d9d9c5e is already stopped"
Oct 11 16:16:47 ip-172-31-51-194 atomic-openshift-node: I1011 16:16:47.144043   10935 docker_manager.go:1531] Container "c6a061890c2af7185485dbcca5f884120a824b44d4801a7bf884d87c4d9d9c5e default/router-1-deploy" exited after 844.597µs
Oct 11 16:16:47 ip-172-31-51-194 atomic-openshift-node: E1011 16:16:47.144095   10935 pod_workers.go:184] Error syncing pod 2ee12e03-8fee-11e6-9927-023dc74be9f5, skipping: failed to "SetupNetwork" for "router-1-deploy_default" with SetupNetworkError: "Failed to setup network for pod \"router-1-deploy_default(2ee12e03-8fee-11e6-9927-023dc74be9f5)\" using network plugins \"redhat/openshift-ovs-multitenant\": Error running network setup script: Could not find IP address for container c6a061890c2af7185485dbcca5f884120a824b44d4801a7bf884d87c4d9d9c5e; Skipping pod"


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

OpenShift 3.4.0.9
Docker 1.12.2-rc2

How reproducible: Always


Steps to Reproduce:
1.  Run an Ansible install of the byo/config.yml playbook on a system with Docker 1.12 installed from 
2.  Deploy the router or the docker-registry DCs


Actual results:

Pods are stuck in ContainerCreating state.   On the nodes you can see the containers briefly start and die.   There are no container logs.    See below for link to node logs

Expected results:

Containers start successfully.
Additional info:

Comment 3 DeShuai Ma 2016-10-12 00:11:28 UTC
This related to https://bugzilla.redhat.com/show_bug.cgi?id=1382997

Comment 4 Mike Fiedler 2016-10-12 11:41:14 UTC
This is a duplicate per comment 3.   Verified pods start with SELinux disabled.

*** This bug has been marked as a duplicate of bug 1382997 ***