This issue tracks the upstream Velero GitHub issue #9193 which is part of the Velero v1.18 milestone.

Description

The backup repository controller connects to the backup repo when the backupRepository CR status is not in normal state.
The backup repository controller is hosted by Velero server.
As a result, similar to #8886, Velero server may be OOM killed once the connecting to backup repo takes lots of memory for large scaled repo.

Upstream Details

• GitHub Issue: https://github.com/vmware-tanzu/velero/issues/9193
• Status: Open
• Assignee: Lyndon-Li
• Labels: Performance, enhancement/backup-repo, repository, area/resilience, scalability
• Created: 2025-08-19T08:17:33Z
• Updated: 2025-09-19T06:20:26Z

This is a performance and scalability issue that could cause the Velero server to be OOM killed when working with large repositories.