Notification Center for failed operations

Your dashboard now has a Notification Center. It lists failed operations on your servers and instances, together with a short explanation of the cause. A badge on the bell icon in the top bar shows the number of unread notifications.

Which failures are reported

• Module operations: failed installs, updates, uploads and removals, including updates started by a Git push.
• Instance operations: instance updates, backup restores and staging refreshes that fail because the server runs out of disk space.
• Webhooks: a Git webhook that could not be set up, for example because the GitHub connection is no longer valid.

Each notification shows the server, the instance and, where applicable, the module. Selecting a notification opens the related server or instance page. For module failures, the error is also shown on the module in the instance page.

Explanation of the cause

Cloudpepper maps common errors to a short explanation. Examples:

• “The selected branch was not found in the repository.”
• “Access to the Git repository was denied. Check the repository credentials and permissions.”
• “The server ran out of disk space during the operation.”
• “A Python dependency failed to build on the server. A required system package may be missing.”

Retention

Notifications are kept for 7 days and then removed automatically. Opening a notification marks it as read. You can also mark all notifications as read. The read status is stored in your browser. Cloudpepper does not send e-mail for these notifications.