Bug ID 1712009: Attempting to perform a configuration restore, after downgrading from v1.8.0, makes the system inoperable

Last Modified: Jun 14, 2025

Affected Product(s):
F5OS F5OS, F5OS-A, F5OS-C, Install/Upgrade, Velos(all modules)

Known Affected Versions:
F5OS-A 1.4.0, F5OS-A 1.5.0, F5OS-A 1.5.1, F5OS-A 1.5.2, F5OS-A 1.7.0, F5OS-A 1.8.0, F5OS-C 1.6.0, F5OS-C 1.6.1, F5OS-C 1.6.2

Fixed In:
F5OS-A 1.5.3

Opened: Nov 05, 2024

Severity: 2-Critical

Symptoms

After a downgrading from v1.8.0 and reset-to-default process, ConfD fails to start.

Impact

The system becomes inoperable, with no access to the CLI or UI. Interaction is restricted to a root-level bash login. Following a database reset, access is exclusively available through the serial console.

Conditions

Upgrade system from F5OS-A-1.4.0 or F5OS-C-1.6.0 and later to F5OS 1.8.0. Then downgrade and attempt to perform a config-restore to the prior version configuration.

Workaround

Perform the below steps for a successful configuration restore or reset-to-default operation following a version downgrade from 1.8.0. ===================================================================================== F5 rSeries system's config-restore workaround after downgrading from v1.8.0 ======================================================================== step-1: Log in to the command line interface (CLI) of the system using an account with root access. step-2: Copy the below content to a new file f5_dyncfg_config_restor_fix.xml <!-- File Begin --> <!-- XML file content for fixing the config-restore issue. --> <config xmlns='http://tail-f.com/ns/config/1.0'> <confdConfig xmlns='http://tail-f.com/ns/confd_dyncfg/1.0'> <restconf> <transport> <tcp> <enabled>false</enabled> </tcp> </transport> </restconf> <webui> <enabled>false</enabled> <transport> <tcp> <enabled>true</enabled> </tcp> </transport> </webui> </confdConfig> </config> <!-- End of file --> step-3: Move the file (f5_dyncfg_config_restor_fix.xml) created in step-2 to /var/F5/system/ step-4: Execute the below command. docker exec -it system_manager /confd/bin/confd_load -U -c system -m -l /var/F5/partition/f5_dyncfg_config_restor_fix.xml step-5: delete the file /var/F5/system/f5_dyncfg_config_restor_fix.xml System Controller’s config-restore workaround after downgrading from v1.8.0 =========================================================================== step-1: Log into the command line interface (CLI) of the Active controller using an account with root access. step-2: Copy the below content to file f5_dyncfg_config_restor_fix.xml <!-- File Begin --> <!-- XML file content for fixing the config-restore issue. --> <config xmlns='http://tail-f.com/ns/config/1.0'> <confdConfig xmlns='http://tail-f.com/ns/confd_dyncfg/1.0'> <restconf> <transport> <tcp> <enabled>false</enabled> </tcp> </transport> </restconf> <webui> <enabled>false</enabled> <transport> <tcp> <enabled>true</enabled> </tcp> </transport> </webui> </confdConfig> </config> <!-- End of file --> step-3: Move the file (f5_dyncfg_config_restor_fix.xml) created in step-2 to /var/F5/system/ Step-4: Execute the below command. docker exec -it vcc-confd confd_load -U -c system -m -l /var/F5/system/f5_dyncfg_config_restor_fix.xml step-5: Delete the file /var/F5/system/f5_dyncfg_config_restor_fix.xml Chassis Partition's config-restore workaround after Partition downgrading from 1.8.0 ================================================================================== step-1: Log in to the command line interface (CLI) of the blade using an account with root access. step-2: copy the below content to file f5_dyncfg_config_restor_fix.xml <!-- File Begin --> <!-- XML file content for fixing the config-restore issue. --> <config xmlns='http://tail-f.com/ns/config/1.0'> <confdConfig xmlns='http://tail-f.com/ns/confd_dyncfg/1.0'> <restconf> <transport> <tcp> <enabled>false</enabled> </tcp> </transport> </restconf> <webui> <enabled>false</enabled> <transport> <tcp> <enabled>true</enabled> </tcp> </transport> </webui> </confdConfig> </config> <!-- End of file --> step-3: Move the file (f5_dyncfg_config_restor_fix.xml) created in step-2 to /var/F5/partition<id>/ Step-4: Execute the below command. docker exec -it partition<id>_manager confd_load -U -c system -m -l f5_dyncfg_config_restor_fix.xml step-5: Delete the file /var/F5/system/f5_dyncfg_config_restor_fix.xml Follow the below steps to fix the system after it enters a failed state following a version downgrade v1.8.0 ===================================================================================== To restore functionality, you must access a bash shell using an account with root access (most likely through the system's serial console) and delete the files in the "cdb/" directory and perform a restart. This action will erase all settings, including licensing and the system’s management IP. Next, get a new license, configure the system management IP address, verify or reset the primary key, and initiate a configuration restoration using the previously saved backup. If the system controller is reset using this method, the empty partitions must be recovered from backup and the tenants must then be restored. If a partition experiences this type of failure and is cleared and reset, it must not be deleted or recreated in the system controller. This is because it will result in a mismatch of primary keys and the configuration restoration will not function properly.

Fix Information

None

Behavior Change

Guides & references

K10134038: F5 Bug Tracker Filter Names and Tips