Bug ID 2053301: Upon reaching maximum memory capacity, BIG-IP tenant pods intermittently regenerated after a host reboot

Last Modified: Dec 09, 2025

Affected Product(s):
F5OS F5OS-A(all modules)

Known Affected Versions:
F5OS-A 1.8.3

Opened: Sep 15, 2025

Severity: 2-Critical

Symptoms

BIG-IP Tenant Pods will keep restarting forever. On host(root access)# kubectl get pods Ex: # kubectl get pods NAMESPACE NAME READY STATUS RESTARTS AGE default f5-resource-manager-jlv5q 1/1 Running 0 7h9m default virt-launcher-v171x-1-xnk5s 1/1 Running 0 2m15s ==========> default virt-launcher-v175x-1-24d9w 1/1 Running 0 2m4s ==========> default virt-launcher-new-183-tenant-1-2nz9q 0/1 Running 0 15s ===========> default virt-launcher-v151x-1-bqqr5 0/1 Init:0/1 0 2s The status of the BIG-IP tenants on F5OS user CLI might show as "Started Tenant Instance" but the tenants will not be accessible.

Impact

The BIG-IP tenants will not reachable. No impact to BIG-IP's data.

Conditions

One or more tenants are configured to utilize all available memory and the host undergoes a reboot.

Workaround

With root access to the F5OS host, run the below command on bash shell: # docker restart system_fpga after few(~5) min the BIG-IP tenants will be reachable.

Fix Information

None

Behavior Change

Guides & references

K10134038: F5 Bug Tracker Filter Names and Tips