fabric8-pipeline-library
fabric8-pipeline-library copied to clipboard
Fabric8 Pipeline for Jenkins
I've try to use fabric pipeline library with Openshift jenkins, but I got an error > java.util.ServiceConfigurationError: io.fabric8.kubernetes.client.ExtensionAdapter: Provider io.fabric8.kubernetes.client.AppsAPIGroupExtensionAdapter not a subtype when I try to execute mvn --version...
Due to misconfigured `arquillian.xml` and environment variables in the pipeline library integration tests are currently executed in the `user-jenkins` namespace which is not intended for that nor has enough resources...
I find that there are two ways to fetch github token in the pipeline library 1. `withCredentials` to lookup `cd-github` see: https://github.com/fabric8io/fabric8-pipeline-library/blob/0bff0f50a2af9a0dcc4d2c4abb28d1eafcbfdfb9/vars/mavenCI.groovy#L14 2. `Fabric8Command.getGitHubToken` which uses a different mechanism. See:...
Seems to happen when there's low resources or multiple jobs running. This has been seen on OSO and GKE. ``` Executing shell script inside container [maven] of pod [kubernetes-137ebf2065f949d4acac4e019ed07af7-1e96524904d1e] [Pipeline]...
There is roughly 3400 lines of code in about 84 groovy files. I ran some very simple analysis and this is the gist of the analysis. ``` Found 1874 duplicate...
Added a new groovy script that will deploy snapshot and report the route URL on the PR. Similar to [deployOpenShiftSnapshot](https://github.com/fabric8io/fabric8-pipeline-library/blob/master/vars/deployOpenShiftSnapshot.groovy) This is being consumed in https://github.com/fabric8-ui/fabric8-planner/pull/2426
Variables like this in code effectively makes this project difficult to use outside the red hat ecosystem. This needs to be reconsidered in the sprit of FOSS. ``` export FABRIC8_WIT_API_URL="https://api.prod-preview.openshift.io/api/"...
Looks like dead code, [github search]( https://github.com/openshiftio/openshift.io/search?q=bayesianScanner&type=Issues&utf8=%E2%9C%93) returns 0 results. @rupalibehera There seems to be a few more like this. Should we do a review of the old stuff?
Apparently, fabric8-cd bot does not cleanup after automerge, which results with tons of stale branches such as `versionUpdate1f13c367-fffe-426d-915f-658c2ff30ae0`. When the PR is automerged the branch should be deleted if coming...
Add a function to check if the deployment to prod is disabled or enabled. for eg : it can check a gitrepo called 'osio-deploy-prod' and see if there's an indicator...