cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
343
Views
0
Helpful
4
Replies

IPv6 dhcpv6 not getting ip

AKRobert
Level 1
Level 1

Hello. 
We have been trying to migrate to IPv6 and dhcpv6. So far we are having dhcp issue. 
I have a router running iOS-xe and it has the production vlan on it and it’s configured like this for the vlan interface 

interface Vlan1000

 description User VLAN

 no ip address

 ipv6 address <ipv6 ip>1000::1/64

 ipv6 enable

 ipv6 nd prefix default 2592000 604800 no-autoconfig

 ipv6 nd managed-config-flag

 ipv6 nd other-config-flag

 ipv6 dhcp relay destination <dhcpv6 ip>

 ipv6 dhcp relay source-interface Vlan1000

 ipv6 ospf 1 area <area in use, not area 0>

for some reason when the windows 11 client comes online i see a gathering info where it’s talking to dhcp and then a message that says 

sas returned null falling to link local

we have an infoblox appliance as the dhcp server 

thank you

1 Accepted Solution

Accepted Solutions

so i did find the solution to this.

there was a device-tracking policy that blocked UDP and NDP. once i pulled out that policy devices came up and started getting IPs

View solution in original post

4 Replies 4

Jens Albrecht
Level 4
Level 4

Hello @AKRobert,

first of all, the config for your interface Vlan 1000 is correct and should work just fine.

Looks like your DHCPv6 server is either not reachable (routing issue, firewall blocking) or not responding (misconfiguration, service not running). So you need to troubleshoot step by step:

  • Ping from your router to the DCHPv6 server with a source int vlan 1000 to verify IPv6 connectivity
  • Configure your Win11 client with a static IPv6 address and check ping to the DHCPv6 server to verify connectivity
  • Check on your Infoblox appliance whether it receives the DCHPv6 request and sends an offer

Without knowing your network we can only give general advice. If you provice more details about your setup we might be able to give more specific replies.

HTH!

ok. so now that im in front of the computer i can maybe explain this better.

I have a Cisco 3850 acting as my router connected to it i have 2 Layer 2 Cisco Nexus devices (NX-OS)  and connected to those i have several Cisco 3850 Fiber switches acting as Layer 2 distros. the end user buildings (3560, 3850s, 9200 and 9300s) connected to those.

on the L2 3850s i have these 2 commands in the config
ipv6 unicast-routing
ipv6 dhcp-relay trust-all

on the L2 Nexus devices i have these 2 commands in the config
feature dhcp
ipv6 dhcp relay

on the 3850 router i have both the commands  above and for my OSPFv3 i have
router ospfv3 1
router-id 0.0.0.1
!
address-family ipv6 unicast
passive-interface default
area xx nssa

for my vlan interface i have this 
interface Vlan1000
no ip address
ip pim passive
ipv6 address <ipv6 ip for this vlan>::/64 eui-64
ipv6 enable
ipv6 nd prefix default 2592000 604800 no-autoconfig no-onlink
ipv6 nd managed-config-flag
ipv6 nd other-config-flag
ipv6 dhcp relay destination <ipv6 dhcp device ip>
ipv6 dhcp relay source-interface Vlan1000
ospfv3 1 hello-interval 10
ospfv3 1 dead-interval 40
ospfv3 1 ipv6 area xx instance 1

i have extended this vlan all the way down through all the devices to the end user building switch. 
from the end user switch i cant ping my ipv6 gateway

from my router i can ping myself of course and my GW2 ip and from my GW2 i can ping my vlan interface IPv6 IPs

I know for IPv4 i had to have this to advertise my vlans
router ospf 1
area xx nssa no-summary
network xxx.xxx.xxx.0 0.0.0.255 area xx

but these options in ipv6 dont seem to exist



wajidhassan
Level 4
Level 4

Hi, It sounds like your DHCPv6 relay and ND settings are mostly in place, but the client not getting an IP usually points to a routing or relay issue. Make sure the Infoblox DHCP server can receive and reply to the solicit messages, and that the return path is clear. Also verify RA isn't disabled too aggressively—try enabling onlink in the ND prefix to help with routing.

so i did find the solution to this.

there was a device-tracking policy that blocked UDP and NDP. once i pulled out that policy devices came up and started getting IPs