-
Bug
-
Resolution: Done
-
Critical
-
3.15.0.GA
-
Team A: Dashboard, devfile and plugin registries, open-vsx, operator + chectl/dsc, devfile-converter, configbump + traefik, image-puller, server, gateway, authentication, try-in-web-ide action, telemetry, Team B: DevWorkspace + Operator, Web Terminal + Operator, editors/IDEs + built-in vscode extensions, Universal Developer Image, machine-exec, dev environment config
-
False
-
None
-
False
-
-
Description of problem:
Modify devspaces-dashboard and devworkspace-controller-manager with new default resource values as the current defaults are not enough to manage the increasing workloads of customers.
Also, devworkspace-controller-manager deployment is managed by CSV and since upgrading the operator replaces the old CSV, the changes gets reverted. So, need to update the values with the right defaults if the current one is not enough - the assumption is that DWO should handle 1000 + concurrent CDEs without any hiccups
In addition, if we manage the operator installation using a subscription(using gitops) and when we reapply the configuration, there is a chance that the resources will get revert
Proposed default values:
For devspaces-dashboard
dashboard:
deployment:
containers:
- resources:
limits:
memory: 5Gi
request:
cpu: 100m
memory: 32Mi
For devworkspace-controller-manager,
spec:
containers:
- resources:
limits:
cpu: '3'
memory: 3Gi
requests:
cpu: 250m
memory: 100Mi
Prerequisites (if any, like setup, operators/versions):
Steps to Reproduce
- Update the default resource values of DWO with a new values
- Upgrade the operator
Actual results:
Since the CSV gets replaces as part of upgrade, the modified value gets reverted to older default value, so the default value should get adjusted with a new value
Expected results:
The default value should get adjusted with a new value
Reproducibility (Always/Intermittent/Only Once):
Always
Acceptance criteria:
Definition of Done:
Build Details:
Additional info (Such as Logs, Screenshots, etc):