Name: e2e-runner
Owner: Deis
Description: Docker image for running workflow-e2e
Created: 2016-05-23 04:35:35.0
Updated: 2018-01-01 06:59:16.0
Pushed: 2017-09-06 19:54:31.0
Homepage: null
Size: 107
Language: Shell
GitHub Committers
User | Most Recent Commit | # Commits |
---|
Other Committers
User | Most Recent Commit | # Commits |
---|
Docker image for running workflow-e2e
E2e-runner coordinates the entire e2e run against a Workflow chart, including:
deis
namespace if still exists),tiller
deployment if necessary),See the main run script for the basic outline of actions presented above. It is a good entry point into the finer details of e2e-runner functionality.
To run the tests for a jenkins job you should have a docker run
command that looks like the following:
> /home/jenkins/workspace/${JOB_NAME}/${BUILD_NUMBER}/env.file
er run \
env-file=/home/jenkins/workspace/${JOB_NAME}/${BUILD_NUMBER}/env.file \
jenkins:jenkins \
/home/jenkins/workspace/${JOB_NAME}/${BUILD_NUMBER}:/home/jenkins/logs:rw \
ay.io/deisci/e2e-runner
A few different things are happening here. First we export all the local environment variables to a file so we can pass those into the container at runtime. This allows us to use secret text values for things like $AUTH_TOKEN
. When we call docker run
we make sure we specify the user jenkins:jenkins
. This way the container has write permissions into the volume mount.
cker run -e AUTH_TOKEN=$AUTH_TOKEN quay.io/deisci/e2e-runner
AUTH_TOKEN
- Token needed to talk to k8s claimerCLUSTER_DURATION
- How long to lease the k8s cluster (default: 800 seconds
)GINKGO_NODES
- How many nodes to use when running e2e tests in parallel (default: 30
)