Julius von Kohout
Julius von Kohout
@andreas-schliebitz you might want to check out https://github.com/kubeflow/manifests/pull/1875 if you are on Openshift. In the end I just wrote my own jupyterlab with code-server and RStudio all included in one...
Definitely for all,because it is not Openshift specific. On a plain Kubernetes with pod security policies you have exactly the same problem.
> So a workaround would be to add > > 1. securitycontext 1000:100 > 2. emptydir volume > 3. volume mount to /var/run/s6 > 4. S6_READ_ONLY_ROOT env > > as...
``` kind: Deployment apiVersion: apps/v1 metadata: annotations: deployment.kubernetes.io/revision: '1' name: test1 namespace: prod-1 spec: replicas: 1 selector: matchLabels: app: test1 template: metadata: labels: app: test1 sidecar.istio.io/inject: 'false' # it also...
> @juliusvonkohout Awesome, thanks for testing that. If you have the time and energy, could you also check if it breaks if you remove the `S6_READ_ONLY_ROOT=1` environment variable? Would like...
> Great, then the change necessary to the notebook-controller is very minimal. I'm not sure how the SCC / PSP can best be handled, as this would probably need to...
> @juliusvonkohout Are you in the Kubeflow Slack? I'd like to discuss and coordinate some things with you regarding this and similar problems when using OpenShift. No, how can i...
i think it is very dangerous because MLMD is not yet separated per namespace https://github.com/kubeflow/pipelines/issues/4790. It will lower the security standards even more if more components break down the namespace...
[buggy_pipeline.py_1.yaml.txt](https://github.com/kubeflow/kfp-tekton/files/5937377/buggy_pipeline.py_1.yaml.txt) [buggy_pipeline.py_2.yaml.txt](https://github.com/kubeflow/kfp-tekton/files/5937378/buggy_pipeline.py_2.yaml.txt) As you can see in the files, the workspace is missing
> Looks like there's a bug with using newlines. This is fixed in the latest kfp code > [kubeflow/pipelines#4993](https://github.com/kubeflow/pipelines/pull/4993) > > For a quick fix, you can run `pip install...