|
When a DHCP relay is configured on the FortiGate, there are scenarios where the clients do not get an IP address from the DHCP server.
During the troubleshooting process, the DHCP relay debugs show fthe ollowing type of errors:
diagnose debug application dhcprelay -1
diagnose debug console timestamp enable
diagnose debug enable
spoke # 2025-10-15 08:00:22 (xid:224e9f81) received request message from 0.0.0.0:68 to 255.255.255.255 at port4
2025-10-15 08:00:22 (xid:224e9f81) got a DHCPREQUEST
2025-10-15 08:00:22 (xid:224e9f81) Warning! can't get server id from client message
2025-10-15 08:00:22 Insert option(82), len(7)
2025-10-15 08:00:22 found oif=8/port6 (dst=10.70.70.5, vfid=0, mode=specify, ifname=port6)
2025-10-15 08:00:22 sys_kernel_get_vf_route3_oif() failed (dst=10.70.70.5, vfid=0, vrf=0, oif=8)
2025-10-15 08:00:22 (xid:224e9f81) cannot find outgoing interface.
2025-10-15 08:00:22 (xid:224e9f81) sendmsg() failed or was not complete: No such file or directory
2025-10-15 08:00:25 (xid:224e9f81) received request message from 0.0.0.0:68 to 255.255.255.255 at port4
Looking at the above debug errors, it indicates that the FortiGate is not able to find the outgoing interface correctly. This is mostly because the DHCP server is not reachable from the FortiGate. In this example, the issue is because of the incorrect interface used as the source interface. This can also occur because of a routing issue.
Step 1:
Making sure that a proper route to DHCP server is in the routing table on the firewall by running the below command:
get router info routing-table details 10.70.70.5 <----- DHCP server IP.
spoke # get router info routing-table details 10.70.70.5
Routing table for VRF=0
Routing entry for 10.70.70.0/24
Known via "connected", distance 0, metric 0, best
* is directly connected, port5
In this case the correct route is via port 5 but debug states that its using port 6.
After checking the interface setting on which the DHCP relay is configured, the interface selection is set to 'specify' and a wrong port is specified instead of port 6:
spoke (port4) # show
config system interface
edit "port4"
set vdom "root"
set dhcp-relay-interface-select-method specify
set dhcp-relay-interface "port6"
set dhcp-relay-service enable
set ip 10.30.30.1 255.255.255.0
set allowaccess ping https http
set type physical
set snmp-index 4
set dhcp-relay-ip "10.70.70.5"
set dhcp-relay-source-ip 10.30.30.1
next
end
Valid config:
spoke (port4) # show
config system interface
edit "port4"
set vdom "root"
set dhcp-relay-interface-select-method auto
set dhcp-relay-service enable
set ip 10.30.30.1 255.255.255.0
set allowaccess ping https http
set type physical
set snmp-index 4
set dhcp-relay-ip "10.70.70.5"
set dhcp-relay-source-ip 10.30.30.1
next
end
After making this change, a DHCP request is seen forwarded to the DHCP server with no issues.
spoke # 2025-10-15 08:34:01 (xid:e526a7fb) received request message from 0.0.0.0:68 to 255.255.255.255 at port4
2025-10-15 08:34:01 (xid:e526a7fb) got a DHCPREQUEST
2025-10-15 08:34:01 (xid:e526a7fb) Warning! can't get server id from client message
2025-10-15 08:34:01 Insert option(82), len(7)
2025-10-15 08:34:01 found route to 10.70.70.5 via 10.30.30.1 iif=6 oif=7/port5, mode=auto, ifname=
2025-10-15 08:34:01 (xid:e526a7fb) forwarding dhcp request from 10.30.30.1:67 to 10.70.70.5:67
|
