Rebase to Kube 0.14

[smarterclayton]

@pweil- please see the commit on how ports have changed.

[smarterclayton]

So much rebase...

[smarterclayton]

[test]

[openshift-bot]

continuous-integration/openshift-jenkins/test SUCCESS (https://ci.openshift.redhat.com/jenkins/job/test_pull_requests_openshift3/1774/)

[smarterclayton]

Squash incoming

[smarterclayton]

[merge]

[smarterclayton]

@mfojtik / @bparees after this change lands someone needs to investigate osc build-logs (it's failing in the end-to-end tests). It might be @derekwaynecarr's changes to require namespace to be specified interacting with the test, but I wasn't able to run it down in a short period of time.

[openshift-bot]

continuous-integration/openshift-jenkins/merge SUCCESS (https://ci.openshift.redhat.com/jenkins/job/test_pull_requests_openshift3/1774/) (Image: devenv-fedora_1252)

[smarterclayton]

[merge]

[smarterclayton]

@pweil- if this doesn't merge tonight I'll need your help looking at TestRouter. I'm seeing:

$ docker logs e0156da40fb3
I0410 05:35:04.312742       1 router.go:64] Commiting router changes
E0410 05:35:04.334364       1 router.go:131] Error reloading router: exit status 127
 Reload output: + config_file=/var/lib/haproxy/conf/haproxy.config
+ pid_file=/var/lib/haproxy/run/haproxy.pid
+ old_pid=
+ path_map_file=/var/lib/haproxy/conf/os_http_be.map
+ sort -r /var/lib/haproxy/conf/os_http_be.map -o /var/lib/haproxy/conf/os_http_be.map
+ '[' -f /var/lib/haproxy/run/haproxy.pid ']'
+ '[' -n '' ']'
+ /usr/sbin/haproxy -f /var/lib/haproxy/conf/haproxy.config -p /var/lib/haproxy/run/haproxy.pid
/var/lib/haproxy/reload-haproxy: line 18: /usr/sbin/haproxy: No such file or directory
F0410 05:35:04.334429       1 router.go:45] exit status 127

in the docker logs when it tries to start haproxy, but don't know why.

[pweil-]

that is strange. I pulled this PR locally, built the images, and ran the container.

[vagrant@openshiftdev origin]$ docker ps
CONTAINER ID        IMAGE                                     COMMAND                CREATED             STATUS              PORTS               NAMES
023ee241b46b        openshift/origin-haproxy-router:ee831e7   "/usr/bin/openshift-   13 seconds ago      Up 12 seconds       80/tcp              sharp_lumiere       
[vagrant@openshiftdev origin]$ docker inspect 023ee | grep Pid
        "PidMode": "",
        "Pid": 7004,
[vagrant@openshiftdev origin]$ sudo nsenter -m -u -n -i -p -t 7004
[root@023ee241b46b /]# ll /usr/sbin/haproxy
-rwxr-xr-x. 1 root root 800656 Mar  6 00:30 /usr/sbin/haproxy
[root@023ee241b46b /]# 

[pweil-]

running hack/test-integration-docker.sh locally did show one issue for me. The v1beta3 ns is not supported by the http master simulator correctly, it was only registering v1beta1 urls. This fixes it for me.

func (s *TestHttpService) startMaster() error {
    masterServer := http.NewServeMux()
    apis := []string{"v1beta1", "v1beta2", "v1beta3"}

    for _, version := range apis {
        masterServer.HandleFunc(fmt.Sprintf("/api/%s/endpoints", version), s.handleEndpointList)
        masterServer.HandleFunc(fmt.Sprintf("/api/%s/watch/endpoints", version), s.handleEndpointWatch)
        masterServer.HandleFunc(fmt.Sprintf("/osapi/%s/routes", version), s.handleRouteList)
        masterServer.HandleFunc(fmt.Sprintf("/osapi/%s/watch/routes", version), s.handleRouteWatch)
    }

    masterServer.HandleFunc("/", s.handleHelloMaster)

    if err := s.startServing(s.MasterHttpAddr, masterServer); err != nil {
        return err
    }

    return nil
}

ok      TestRouter
ok      TestRouterPathSpecificity

[pweil-]

Travis flaked on one run with a bind address already in use but the other two runs passed and jenkins passed as well in my test PR: #1690

[smarterclayton]

Open a pull to my branch

On Apr 10, 2015, at 9:30 AM, Paul [email protected] wrote:

Travis flaked on one run with a bind address already in use but the other two runs passed and jenkins passed as well in my test PR: #1690

—
Reply to this email directly or view it on GitHub.

[pweil-]

submitted

[bparees]

@smarterclayton what's the failure mode/error message for build-logs?

[smarterclayton]

It says build not found (even though it does). It might be a namespace bug, or it could be a redirect bug.

On Apr 10, 2015, at 10:54 AM, Ben Parees [email protected] wrote:

@smarterclayton what's the failure mode/error message for build-logs?

—
Reply to this email directly or view it on GitHub.

[smarterclayton]

[merge]

[openshift-bot]

Evaluated for origin up to 726d211