Build does not proceed, waits for ever

[ht512512]

OS: Ubuntu 16.04
Installed 'oc' and did oc cluster up
Connected to the Web Console
In my-project, created an app ("an-app") using 'https://github.com/openshift/nodejs-ex.git' (actually just clicked try-it button that fills this in).
It started build 1 that did not do anything (just the spinning icon). Canceled after 40 min. Tried again. Same fate for build 2.

Builds
an-app Build #2 is pending created 4 minutes ago
Build #1 was cancelled created an hour ago

Not sure if any other config is required!

[bparees]

all your builds remain in "pending"?

Can you run an "oc describe" on the build pod and the build and report the results?

"oc get pods -o yaml" would also be helpful.

[ht512512]

oc describe builds

• Seems to not like the docker version string !!
  ==============
  Name: an-app-1
  Namespace: my-project
  Created: About a minute ago
  Labels: app=an-app
  buildconfig=an-app
  openshift.io/build-config.name=an-app
  openshift.io/build.start-policy=Serial
  Annotations: openshift.io/build-config.name=an-app
  openshift.io/build.number=1
  openshift.io/build.pod-name=an-app-1-build

Status: Pending
Duration: waiting for 1m29s
Build Config: an-app
Build Pod: an-app-1-build

Strategy: Source
URL: https://github.com/openshift/nodejs-ex.git
Ref: master
From Image: DockerImage centos/nodejs-4-centos7@sha256:a166ee3d5cfd04c6a08f758f35bce9d4fc6cf6b2387d15b2998be6a36ccea63a
Output to: ImageStreamTag an-app:latest
Push Secret: builder-dockercfg-q63gh

Build trigger cause: Build configuration change

Events:
FirstSeen LastSeen Count From SubObjectPath Type Reason Message

---

1m 1m 1 {default-scheduler } Normal Scheduled Successfully assigned an-app-1-build to localhost
1m 8s 7 {kubelet localhost} Warning FailedSync Error syncing pod, skipping: failed to "StartContainer" for "POD" with RunContainerError: "runContainer: docker: failed to parse docker version "17.03.1-ce": illegal zero-prefixed version component "03" in "17.03.1-ce""

[bparees]

paging @csrwng

[bparees]

(fyi this is really an openshift/origin issue, not a nodejs example issue, so next time consider opening the issue there)

[ht512512]

Thanks.

[csrwng]

@ht512512 if you are using cluster up to start your cluster, try with --version=v1.5.0
The other alternative is to downgrade your docker to v1.13.1

[ht512512]

@csrwng Thanks. With --version=v1.5.0 the build fails with unable to clone the app. Seems like origin is not able to network out (Could not resolve host error in the log)

[ht512512]

@csrwng, In monitoring, I see an event with the following error, " kubelet does not have ClusterDNS IP configured and cannot create Pod using "ClusterFirst" policy. Falling back to DNSDefault policy. "

[csrwng]

@ht512512 do you by any chance a firewall running?

[csrwng]

It sounds like this is an issue with pod networking and not necessarily the nodejs example.
Please open in origin if things are not working right after running cluster up (or contact me directly for help)