* Create view role rather than rely on view clusterRole
* Create view role for k8s. Fix typo
* Typo and Java doc sync
* Add namespaced role to k8s and openshift templates. Role instead ClusterRole for k8s
* Rename view role to workspace-view
* Fix formatting
* Fix javadoc. Revent erroneous changes to postgres template
* Remove hidden Dockerfile
* Revert pg changes
* Introduced two web-socket endpoints for workspace master to split JSON-RPC messages
Based on Dmytro's Kulieshov work https://github.com/eclipse/che/pull/12252
Signed-off-by: Sergii Kabashniuk <skabashniuk@redhat.com>
When using an existing Keycloak instance, it may be necessary to specify a particular realm or client.
This allows them to be speicfied at install time via parameters.
Signed-off-by: David Martin <david.martin@redhat.com>
The property CHE_INFRA_KUBERNETES_PVC_PRECREATE__SUBPATHS is required
for PVC cleanup on current versions of Kubernetes/OpenShift.
Without the property enabled, subpaths are created in PVCs by Kubernetes
when they are needed for pods. However, only the leaf directory created
in this way has write permissions for nonroot users. As a result, the
cleanup pod cannot delete directories created (e.g. the
workspaceX/projects) directory.
For more details, see https://github.com/eclipse/che/issues/12445
Signed-off-by: Angel Misevski <amisevsk@redhat.com>
* Adds prometheus and grafana as to the deployment
moves che's custom charts into "custom-charts" directory so that they don't
interfere with helm which uses the "charts" directory to download and use
all the other remote dependencies.
Signed-off-by: Lukas Krejci <lkrejci@redhat.com>
* Deploy our own ingress for prometheus and grafana so that we can have
templatized names of the same format as the same of the deployment.
Also removed the hardcoded time interval in the sample dashboard so that
it uses the default of "last 6 hrs".
Signed-off-by: Lukas Krejci <lkrejci@redhat.com>
* New lines at the end of the files and whitespace removal.
Remove readiness/liveness probes from Che deployment when debug is
enabled.
This allows to debug Che master for a long period of time without
Che container being killed because of unresponsive probes.
Signed-off-by: Oleksandr Garagatyi <ogaragat@redhat.com>
Signed-off-by: Oleksandr Garagatyi <ogaragat@redhat.com>
Latest versions of minishift ignore Che master docker image that
was built locally inside the VM.
Removeing docker.io from the image name workarounds the issue.
Signed-off-by: Oleksandr Garagatyi <ogaragat@redhat.com>
- Make Resource memory request configurable for k8s and OpenShift.
- Add alias for backwards compatibility to che.workspace.default_memory_limit_mb.
- set memory request to 512Mb.
- document CHE_WORKSPACE_DEFAULT__MEMORY__REQUEST__MB in che.env.
- Configure helm scripts to allow memory attributes on deployment.
Signed-off-by: Sergey Kuperman <sergey.kuperman@sap.com>
Sergii Kabashniuk