Showing results for 
Search instead for 
Did you mean: 

APIC docker0 interface IP address

Trinh Nguyen

Hello community,

All my APICs are having a docker0 interface with the same IP address  This subnet is in the production network VLAN 100 and causes the problem with VLAN 100 to access APIC because route was wrong.

Question: How do I change APICs docker0 ip address? I prefer to change them to a loop-back address like


7 Replies 7

Cisco Employee
Cisco Employee


I will take a look in our lab, but at the moment I am unaware of a method to change the docker0 address.

With that said, what type of service is the APIC trying to reach and what is its path through ACI fabric?

Are you using in-band management and you are expecting the APIC to reply using the VICs/in-band EPG to reach some external Endpoint sourced traffic within the subnet?



I am using OOBMGMT interface.  This service is for management only, nothing wrong in ACI.  I have a TACACS server in that cannot be reach by all APICs

Note that all three APICs are having the same docker0 ip address

external@APIC1:~> ifconfig docker0
docker0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet netmask broadcast

RX packets 128 bytes 8576 (8.3 KiB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 2345321 bytes 98503674 (93.9 MiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

Thanks for looking at this.




I just got confirmation from our development team that it is not configurable at this point. I will file an enhancement to allow that to happen in the future.

As for immediate solutions, I am assuming it is not possible to change the address of the TACACS service that is on the subnet outside of ACI?

Also, could you please send me the output of the following (if acceptable), otherwise please open an SR to send over this info:

>route -rm

>arp -an

>the address of your TACACS server





No, changing the TACACS IP address is not an option.  

These are the outputs of the commands with some IP and MAC mask-out:

IP address of TACACS is 172.17.N.M (it is in the arp table) 

external@APIC1:~> route

Kernel IP routing table

Destination     Gateway         Genmask         Flags Metric Ref    Use Iface

default         172.X.Y.1         UG    16     0        0 oobmgmt     UG    0      0        0 bond0.3961 UH    0      0        0 bond0.3961   U     0      0        0 teplo-1   U     0      0        0 lxcbr0     U     0      0        0 docker0

172.X.Y.0   U     0      0        0 oobmgmt

external@APIC1:~> arp -an

? ( at xxxxxxxxxx [ether] on oobmgmt

? ( at xxxxxxxxxx [ether] on oobmgmt

<output omit>..

? (172.17.N.M) at <incomplete> on docker0

? ( at xxxxxxxxxx [ether] on oobmgmt


Thanks again.


Trinh Nguyen

This issue is now track under bug id CSCve84297.  No workaround

Hello All, 


Did we manage to get a solution for this, I'm facing similar challenge where our resources are in subnet so as the docker subnet in the APIC setup. 

@Trinh Nguyen wrote:

This issue is now track under bug id CSCve84297.  No workaround




Hi All,

Old post but I think its still relevant!


Docker0 interface in APIC continues to hold (last checked in 3.2(3O)), keeping users from accessing their mgmt network in that private segment.


Just a known work around for this is to have the docker0 interface IP changed by TAC.

It requires root access which only TAC has.


Procedure is same as changing interface IP in linux, just users don't have the root access.




***Rate all Helpful Responses***

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: