Troubleshoot Suite Admin
This section lists some of the issues that you may encounter and suggests workarounds.
Download Log File
See Monitor Modules > Download Logs.
Kibana UI (Log Search)
See Monitor Modules > View Logs in Kibana.
Grafana dashboards Alerts
See Monitor Modules > Configure Grafana Dashboard Alert.
Module Failure during Installation or Upgrade
When installing the CloudCenter Suite for a OpenStack Installation or a VMware vSphere Installation, you have the option to configure NTP server details. If you do not provide the NTP details, workers and nodes may not have their time synchronized with each other. This can potentially cause modules to fail during an installation or upgrade as displayed in the following screenshot.
To workaround this issue, be sure to synchronize the server time for all instances running the CloudCenter Suite.
Log in Failure Due to Session Timeout
If your session has timed out during an idle session, you may sometimes see the message displayed in the following screenshot – even if you have entered the right credentials. Try logging in again.
If you are unable to login due to a wrong password as visible in the following screenshot, contact your suite administrator to reset the password (see Create and Manage Users > User Actions for details).
Blank Screen when Logging into Suite Admin
A user who only belong to one group is abandoned if the group was only specific to one module and if that module was uninstalled. In this case, the abandoned user must follow up with one of the CloudCenter Suite administrators based on their enterprise policies. If a user does not have any active roles, this user may see a blank screen on log in.
SSO Setup Resulting in Login Failure
If you log in as root admin, configure SSO, and was subsequently timed out, you may not be able to log back in. This is because the ADFS user may not have the roles mapped and will not be able to access any modules. This user may not be able to login by using the direct URL (ui/auth/login).
To address this issue, be sure to complete the SSO Setup, add the ADFS user to the Suite Admin Group before any session timeout.
Deployment Repo Uses the Public Repo Even if Kubernetes was Restarted
If you initially use the public repository to install the CloudCenter Suite, the repo for deployments and other activities may continue to point to the public repository. If so, your deployments may continue to reference the public repo even for cases where the Kubernetes nodes were restarted.
After an offline repo is registered with the CloudCenter Suite, users may expect Deployments to automatically and immediately pickup images from the offline repo. This is not the natural behavior.
During product installation/upgrade events, the repository settings are set to Helm (the package manager for Kubernetes). Consequently, offline repository settings are only registered after the next upgrade or new product installation event.
To workaround this natural behavior, you can opt to start a new cluster with the offline repository during the first event to ensure that your environment continues to use this offline repository.
When you uninstall the Workload Manager or Cost Optimizer modules in the Suite Admin, the Kubernetes Persistent Volume Claims (PVCs) are not deleted – they are retained as is for the Suite Administrator to take appropriate steps to backup or manually delete the PVC. The secrets for the Workload Manager and the Cost Optimizer are not deleted when you uninstall the product. To work around this issue, the Suite Administrator must clean up their instances using one of the following suggestions.
- Backup PVC: Take a snapshot of the volume backing up the PVC or just the data contained within. Refer to the Kubernetes Documentation for additional details.
Delete PVC: Manually delete the PVCs by running the following command:
- No labels