This article explains the scenario, where all interfaces show admin down after Juno's upgrade due to duplicate addresses in the FTI interfaces.
LOGS: Error Log: May 10 23:32:22.701 abc.chg dcd[31233]: UI_CONFIGURATION_ERROR: Process: dcd, path: [edit interfaces fti0 unit 42 family inet], statement: address 10.50.3.5/24, Cannot have the same local address on different units of an interface Below are the few units that have duplicate IP addresses: set interfaces fti0 unit 13 family inet address 10.1.1.5/24 set interfaces fti0 unit 14 family inet address 10.1.1.5/24 set interfaces fti0 unit 17 family inet address 10.1.1.5/24 set interfaces fti0 unit 18 family inet address 10.1.1.5/24 set interfaces fti0 unit 24 family inet address 10.50.3.5/24 set interfaces fti0 unit 25 family inet address 10.50.3.5/24 set interfaces fti0 unit 42 family inet address 10.50.3.5/24 set interfaces fti0 unit 62 family inet address 10.50.3.5/24 Please see below DCD is not running : [email protected]> show system processes extensive | match dcd [email protected]> [email protected]# commit full force [edit interfaces fti0 unit 42 family inet] 'address 10.50.3.5/24' Cannot have the same local address on different units of an interface Broadcast Message from [email protected] (no tty) at 5:26 UTC... Force commit synchronized requested on another RE. All configuration sessions on this RE will be terminated re0: configuration check succeeds re1: [edit interfaces fti0 unit 42 family inet] 'address 10.50.3.5/24' Cannot have the same local address on different units of an interface May 11 05:26:31 jlaunchd: craft-control (PID 48760) exited with status=0 Normal Exit May 11 05:26:31 jlaunchd: Registered PID 48775(craft-control): exec_command May 11 05:26:31 jlaunchd: craft-control (PID 48775) started May 11 05:26:31 jlaunchd: Registered PID 48775(craft-control): new process May 11 05:26:31 jlaunchd: del-mstr-ip (PID 48757) exited with status=0 Normal Exit commit complete re0: commit complete [edit] [email protected]# run show system processes extensive | match dcd [edit] [email protected]#
Recovery Steps: - You need to identify all the units with duplicate IP addresses and deactivate/delete those units, You can use a commit check to identify the units. - Post deletion/deactivation do "commit full force", this will initialize DCD which responsible to spawn the interfaces.