Description
This article describes an alternative way to recover HA Sync Status in a FortiGate Cluster, using a backup configuration file modified.
Scope
FortiGate, Cluster, HA.
Solution
Sometimes, the HA cluster goes into the un-sync mode, due to a non-modifiable configurations, (for example a UUID or system objects).
This method helps to solve.
Prerequisites:
- Physical access to the cluster.
- Serial console cable.
- Identify ports and cables connected (label if needed).
- Super admin account credentials, IP, and port for management.
Preparation of modified file:
- Access to the current primary FortiGate GUI and download a configuration file backup.
- Access to current secondary Fortigate CLI and take note of the following configurations:
show system global <----- Hostname and alias.
show system ha<----- Priority, Management and override status*.
show system interface "interface_name" <----- All management interfaces.
- Open Primary's configuration file with a text editor (notepad++), replace all the above configurations for Secondary unit, and save it as a new file.
Procedure:
- Connect to the secondary FortiGate using a serial console cable.**
- Execute command and confirm command:
execute factoryreset
- While restarting, disconnect network cables from the Fw-Slave (Optional***).
- Configure the laptop ethernet IP with 192.168.1.20/255.255.255.0 no gateway neither DNS is needed.
- Connect the ethernet cable between the laptop and the default management port of FortiGate. (mgmt or port1, may vary depending of the model).
- Access to the secondary FortiGate GUI with default values:
user: admin
password: blank
- Restore the configuration using the modified backup file.
- If possible, keep connected to the wireless network and access cluster GUI to check HA status.
- Once rebooted, change the laptop's NIC address, connect to internal, and validate the hostname, and HA priority. (Optional, only 2.1) has been followed).
- Turn off Fw-Slave again, reconnect cables, and power on. (Optional, only if 2.1) has been followed).
- After a few minutes, the cluster should show up and be in sync.
Important notes:
* If override is enabled be sure priority in secondary is slower than primary.
** Can use SSH but be careful to change between primary and secondary administration.
*** It is always better to recover and validate configurations before rebuilding HA.