Last Modified: Oct 19, 2025
Affected Product(s):
F5OS F5OS-C, Install/Upgrade
Fixed In:
F5OS-C 1.8.2
Opened: May 20, 2025 Severity: 2-Critical
Partition tenants that are configured with slots greater than max-nodes can fail to come back when resetting the controller configuration and restoring it back via save configuration for controller and partition.
Stale tenant pods for partition 1 tenants will still show after restoring the controller configuration and partition configuration, but the impacted tenants may not came back up fully (multinodes case).
Occurs typically when - Partition with ID 1 - Tenant uses virtual slots that do not match the physical slot. - Controller reset-to-default is issued
- Bring partition tenants down if planning to do controller configuration reset-to-default. or - Manually delete default partition 1 after reset-to-default before restoring the save controller configuration. This should take care of removing everything associated to the namespace before the config-restore happens. Partitions with ID different than 1 should clear the namespace automatically, since they don't get recreated as part of reset-to-default.
Startup stale cleanup logic has been adjusted to manage multinode and virtual slots tenants appropriately.