5130s-ei acts as gateway and dhcp server, some terminals fail to obtain addresses
- 0 Followed
- 0Collected ，53Browsed
Network Topology
5130s as gateway---------multiple pcs
Problem Description
The terminal occasionally fails to obtain an address
Process Analysis
*Apr 21 10:55:24:932 2025 DHCPS/7/PACKET: From 0.0.0.0 port 68, interface Vlan-interface11
Message type: REQUEST (1)
Hardware type: 1, Hardware address length: 6
Hops: 0, Transaction ID: 548480408
Seconds: 3, Broadcast flag: 0
Client IP address: 0.0.0.0 Your IP address: 0.0.0.0
Server IP address: 0.0.0.0 Relay agent IP address: 0.0.0.0
Client hardware address: 02be-6ea1-30b8
Server host name: not configured
Boot file name: not configured
DHCP message type: DHCPDISCOVER (1)
*Apr 21 10:55:24:933 2025 DHCPS/7/EVENT: Receive a DHCPDISCOVER message from Vlan-interface11.
*Apr 21 10:55:24:933 2025 DHCPS/7/ERROR: No free IP in the network 10.10.1.0.
Debugging DHCP all found no available addresses left
<H3C>display dhcp server free-ip pool XXX also showed no available addresses for vlan11
<H3C>display dhcp server expired pool XXX also had no expired addresses for the corresponding subnet
<H3C>display dhcp server conflict found unallocated addresses all marked as conflicting
<sw>display dhcp server conflict
IP address Detect time
10.10.1.1 Apr 21 11:17:23 2025
10.10.1.3 Apr 21 11:17:25 2025
10.10.1.4 Apr 21 11:17:25 2025
10.10.1.8 Apr 21 11:17:27 2025
10.10.1.9 Apr 21 11:17:29 2025
<21SW-01>display dhcp server ip-in-use pool vlan11
IP address Client-identifier/ Lease expiration Type
Hardware address
10.10.1.2 1-1-1 Apr 21 15:32:55 2025 Auto(C)
10.10.1.5 2-2-2 Apr 21 15:17:00 2025 Auto(C)
10.10.1.6 3-3-3 Apr 21 13:53:09 2025 Auto(C)
10.10.1.7 4-4-4 Apr 21 14:17:21 2025 Auto(C)
10.10.1.10 5-5-5 Apr 21 15:23:45 2025 Auto(C)
10.10.1.11 ；；； Apr 21 15:47:32 2025 Auto(C)
10.10.1.12 ；；； Apr 21 16:17:28 2025 Auto(C)
It is speculated that the following mechanism prevents these conflicting addresses from being assigned to endpoints
(5) If no available IP address is found, query the lease-expired or previously conflicted IP addresses in sequence from the current matching address pool. If found, allocate them; otherwise, no action will be taken.
If the client's subnet changes, the server will not assign the previously allocated IP address to the client but will select a new IP address from the address pool matching the new subnet.
When using an IP address that has previously conflicted, only those with a conflict state lasting over one hour can be assigned by the server to a new DHCP client.
However, these conflicting addresses cannot be pinged, and the client did not detect the address conflict.
Packet capture on the terminal PC revealed that some terminals sent abnormal decline messages.
It is speculated that the following mechanism caused the conflict.
1.2.25 display dhcp server conflict
The command display dhcp server conflict is used to show DHCP address conflict information.
Syntax
display dhcp server conflict [ ip ip-address ] [ vpn-instance vpn-instance-name ]
View
any view
Predefined user roles
network-admin
network-operator
Parameters
ip IP address: Displays address conflict information for the specified IP address. If this parameter is not specified, all address conflict information is displayed.
vpn-instance VPN-instance-name: Displays address conflict information within the specified VPN-instance. vpn-instance-name specifies the name of an MPLS L3VPN VPN-instance, which is a case-sensitive character string of 1 to 31 characters. If this parameter is not specified, address conflict information on the public network is displayed.
Usage guidelines
The DHCP server generates address conflict information in the following situations:
The DHCP server detected through ping that another host in the network was already using the IP address before assigning it to the DHCP client.
The DHCP client sent a Decline message to the DHCP server, reporting an IP address conflict with the address assigned by the DHCP server.
The DHCP server detected that the assignable address in the address pool was its own device address.
Solution
After removing the terminals that sent abnormal decline messages, it returned to normal