Last few days I was busy with configuring IPV6 DHCP on my Fortigate. On the net I found some examples of IPV6 DHCP configurations but for some reasons it's not working on my FTG. The only way to get it working is to enable autonomous-flag enable. The configuration that I made is as follow:
edit "VLAN10" set vdom "root" set ip 10.x.x.1 255.255.255.0 set allowaccess ping https ssh set vlanforward enable set device-identification enable set fortiheartbeat enable set role lan set snmp-index 6 config ipv6 set ip6-allowaccess ping set ip6-reachable-time 30000 set ip6-retrans-time 3000 set ip6-address 2a02:xx:xxx:100::1/64 set ip6-send-adv enable set ip6-manage-flag enable disable set ip6-other-flag enable end set interface "Trunk" set vlanid 10 next end config system dhcp6 server edit 1 set lease-time 10800 set subnet 2a02:xx:xxx:100::/64 set interface "VLAN10" config ip-range edit 1 set start-ip 2a02:xx:xxx:100::100 set end-ip 2a02:xx:xxx:100::500 next end set dns-server1 ::ffff:208.67.222.222 set dns-server2 ::ffff:208.67.220.220 next
Now I'm just wondering why this is not working. The clients are only getting the IPV6 DNS servers but not the IP
Any idea's what to change in above to et it work?
Solved! Go to Solution.
This is indeed a bug- confirmed by Fortinet support.
I reported it back on the 8th of August after 5.6.1 came out.
The issue is still present in 5.6.2 and apparently will be resolved in 5.6.3.
Ticket reference is 2306843.
As suggested above "set ip6-manage-flag enable disable" makes no sense. It should be either enabled or disabled- not both.
Rolling back to 5.6.0 does resolve the problem.
Hope this is useful.
Kind Regards,
Andy.
This is what I see when I debug and the client connects to the FTG
[debug]server6_recv() called [debug]server6_recv() received information request from fe80::1456:5c07:a61a:6e3e%VLAN10 [debug]server6_recv() dhcp6 solicit: search ifp VLAN10's subnet against interface address=2a02:xx:xxx:100::1 [debug]server6_recv() found service id=1 interface VLAN10 [debug]dhcp6_get_options() get DHCP option client ID, len 14 [debug] DUID: 00:01:00:01:20:c7:12:52:18:65:90:7b:a1:61 [debug]dhcp6_get_options() get DHCP option option request, len 4 [debug] requested option: DNS [debug] requested option: domain search list [debug]dhcp6_get_options() get DHCP option elapsed time, len 2 [debug] elapsed time: 0 [debug]copy_option() set client ID (len 14) [debug]copy_option() set server ID (len 14) [debug]copy_option() set DNS (len 32) [debug]copy_option() set information refresh time (len 4) [debug]server6_send() transmit reply to fe80::1456:5c07:a61a:6e3e%VLAN10 [debug]dhcp6_check_timer() called [debug]dhcp6_check_timer() called [debug]binding_save_timo() called binding_changed=0
hello,
i'm running with the same problem.
it's seems that the managed flag is not enabled even set in the configuration. Windows client don't issue a dhcpv6 request. What's funny is that ubuntu client send dhcprequest and get an IPv6...
Wireshark on the W10 client confirm it. Make a capture on ipv6 and look at router advertisement packet. you will see that enabled or disabled in the configuration file don't make any difference.
the statement in configuration :
[ul]Make me think is a bug (was working previous version). previously it was only set ip6-manage-flag enable.
Maybe i'm missing something. For the moment SLAAC is working fine
Here is my ip6 conf:
config ipv6 set ip6-allowaccess ping set ip6-retrans-time 3000 set dhcp6-information-request enable set ip6-address 2001:41d0:xxxx:xxxx::1/64 set ip6-send-adv enable set ip6-manage-flag enable disable set dhcp6-relay-service enable set dhcp6-relay-ip 2001:41d0:xxxx:xxxx::10
edit 20171023-1608CEST:# downgraded from 5.6.2 to 5.6.0 solved the issue with managed flag.
Regards,
This is indeed a bug- confirmed by Fortinet support.
I reported it back on the 8th of August after 5.6.1 came out.
The issue is still present in 5.6.2 and apparently will be resolved in 5.6.3.
Ticket reference is 2306843.
As suggested above "set ip6-manage-flag enable disable" makes no sense. It should be either enabled or disabled- not both.
Rolling back to 5.6.0 does resolve the problem.
Hope this is useful.
Kind Regards,
Andy.
Well, if the issue might be solved on 5.6.3... I will wait for 5.6.4, just in case !
Thanks for sharing the status Andy.
Just to confirm, this issue appears to be resolved in 5.6.3 and my original ticket has now been closed following my testing.
However, it doesn't seem to be covered in the Release Notes and there is still no "What's new" for 5.6.3 (which I guess could potentially cover it too).
5.6.3 seems to have also fixed LLDP for me (neighboring Cisco switches now showing correct LLDP neighbor info from the Fortigate).
Hope this helps.
ANdy.
Select Forum Responses to become Knowledge Articles!
Select the “Nominate to Knowledge Base” button to recommend a forum post to become a knowledge article.
User | Count |
---|---|
1759 | |
1116 | |
766 | |
447 | |
242 |
The Fortinet Security Fabric brings together the concepts of convergence and consolidation to provide comprehensive cybersecurity protection for all users, devices, and applications and across all network edges.
Copyright 2025 Fortinet, Inc. All Rights Reserved.