Chapter 2. Known issues
This section lists known issues with Red Hat CodeReady Workspaces 1.2. Where available, workaround suggestions are provided.
2.1. General known issues
2.1.1. Multiple workspaces fail to run at the same time
Provided the following two conditions are met, it will not be possible to run two or more workspaces at the same time:
CodeReady Workspaces uses the
commonPersistent Volume Claim (PVC) strategy
Persistent volumes (PVs) use
ReadWriteOnce(RWO) access mode
To work around this limitation, use one of the following measures:
ReadWriteMany(RWX) access mode for PVs
2.1.2. Undefined error when starting a workspace
When starting workspace a broken websocket connection can cause an Undefined error. The error does not impact the workspace start process, and refreshing the browser page fixes it.
2.1.4. Occasional NullPointerException when opening a workspace from the "Java 1.8" stack with an imported Java project
When opening a workspace (Java 1.8 stack) with an imported Java project, an error (NullPointerException) is encountered. The project does not open. To work around this issue, open the project again.
2.1.5. Delay when launching commands using the command palette
When launching a command using the command palette, a delay of approximately five seconds occurs before the real execution of the command.
To work around this issue, activate the JDT.LS (Java Language Server Protocol) installer in workspace configuration in Dashboard > Workspace details page.
2.1.6. SSH key upload does not work with GitLab or GitHub
An SSH key generated locally and uploaded to the workspace does not work and the
git pull command fails on authorization.
To work around this limitation, use the following command to convert your ssh key:
ssh-keygen -p -f <your_key_file> -m PEM
2.1.7. New terminal windows freeze after a minute of inactivity on macOS X
When a new terminal is opened and the user stops typing, the terminal becomes unresponsive after some time of inactivity. The user must close the terminal and open a new one to continue.
2.1.8. Workspaces fail to start with certain configurations of StorageClass
When the StorageClass resource object in OpenShift is configured with
volumeBindingMode=WaitForFirstConsumer, the workspaces in CodeReady Workspaces fail to start.
To work around the issue, take the following steps:
- In the OpenShift web console, expand Workloads and click Config Maps > custom.
- Click the YAML tab.
In the data section, add the following environment variable and set it to
CHE_INFRA_KUBERNETES_PVC_WAIT__BOUNDenvironment variable is used to disable or enable the wait time for the PVCs to bind.
Run the following commands to manually scale down and then scale up the deployment to ensure that the wait time for PVCs to bind is disabled:
$ oc scale deployment/codeready --replicas=0 $ oc scale deployment/codeready --replicas=1
2.2. Known issues on OpenShift Container Platform 4.1
2.2.1. Logging in using OpenShift OAuth does not work
On OpenShift Container Platform 4.1, authentication with Red Hat SSO that is configured to use OpenShift OAuth does not work. There is no work around for this issue other than not using OpenShift OAuth.
2.2.2. Error marker fails to appear in Java-stack workspaces
On OpenShift Container Platform 4.1, when editing Java source code in a CodeReady Workspaces workspace based on the Java stack, error markers do not appear when expected. At times, an error marker appears when there should be none. Reloading the page solves the issue only temporarily.