No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

Interface on server address pool assignment IP address lead to control CPU high

Publication Date:  2014-11-07 Views:  46 Downloads:  0
Issue Description

Version: NE40E & 80E CX600 V600R003C00SPCa00

Patch: V600R003SPH027

Network topo:
 N/A

Problem description:

In a bureau NE40E device as a DHCP Sever , check the device status found in the master CPU reaches 99%, which accounts for DHCP up to 64 percent:

 

[NE40E -diagnose ] disp heal

Slots CPU Usage Memory Usage (Used / Total)

,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

9 MPU (Master)      99%          29%  546MB / 1828MB


DHCP                       64%          0/4c2e5e71             DHCP Dynamic Host Configuration

Handling Process

1. check the equipment configuration:

ip pool fttx-voip-4 server

vpn-instance CTVPN3704-VOIP_UNTR

Gateway 172.42.0.1 255.255.252.0

Section 0 172.42.0.2 172.42.3.254

interface GigabitEthernet1/0/7.3203

description vrrp-for-YYLJ-NE40E-2-ge2/0/4.3203

shutdown

control-vid 3014 dot1q-termination

dot1q termination vid 3014

dot1q termination vid 3353 to 3360

dot1q termination vid 3361 to 3364

dot1q vrrp vid 3014

ip binding vpn-instance CTVPN3704-VOIP_UNTR

ip address 172.42.0.2 255.255.252.0  ---- Current network configuration interface IP address is assigned by address pool

vrrp vrid 106 virtual-ip 172.42.0.1

traffic-policy qos_tpCritical4 inbound

traffic-policy QoSMark8021p6 outbound

arp-proxy enable

arp-proxy inter-sub-vlan-proxy enable

arp broadcast enable

trust upstream SN-QOS-1


2.Enable dhcp server's debug:

<NE40E>debugging dhcps packet

<NE40E> terminal debugging

Info: Current terminal debugging is on.

<NE40E> terminal monitor

Info: Current terminal monitor is on.

 

Jan 22 2014 09:44:07.530.16+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCPS_HandleDiscoverPkt:Giaddr is 2888433666.VrfIndex is 11.User:001e-73c8-f5cc.  --------------------Discover message has been sent continued from the user which ip 172.42.0.2 and mac 001e-73c8-f5cc.

 

 Jan 22 2014 09:44:07.530.17+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter AM_DHCPS_ReqIp to apply ip [ac2a0002] ------- Users with option50 application specified 172.42.0.2 address packets

 

 

Jan 22 2014 09:44:07.530.18+08:00 NE40E DHCPS/7/DHCPS_DBG: Error:

DHCPS_HandleDiscoverPkt 005, ulIpAddr is 2888433666

Jan 22 2014 09:44:07.530.19+08:00 NE40E DHCPS/7/DHCPS_DBG: Error:

DHCPS_HandleDiscoverPkt 003, ulIpAddr is 2888433666

Jan 22 2014 09:44:07.530.20+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

PING PACKET ----- Detect whether the conflict application to send ping packets,  ping must available due to the address is same as the interface's.

 

 

 

Jan 22 2014 09:44:07.530.21+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

DHCP_ADD_PINGHASH success add ulIpAddr is 2888433666

Jan 22 2014 09:44:07.530.22+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

DhcpServer: Send ICMP detect for 2 times

Jan 22 2014 09:44:07.530.23+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCP_FSM_ICMP_SendEcho

Jan 22 2014 09:44:07.530.24+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCP_FSM_ICMP_SendEcho ulIP is 2888433666, ulVrfId is 11

Jan 22 2014 09:44:07.530.25+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

DhcpServer: Sending ICMP ECHO to Target IP: 172.42.0.2

Jan 22 2014 09:44:07.530.26+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

DhcpServer: Send ICMP detect success

Jan 22 2014 09:44:07.530.27+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCP_FSM_ICMP_SendEcho

Jan 22 2014 09:44:07.530.28+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCP_FSM_ICMP_SendEcho ulIP is 2888433666, ulVrfId is 11

Jan 22 2014 09:44:07.530.29+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

DhcpServer: Sending ICMP ECHO to Target IP: 172.42.0.2

Jan 22 2014 09:44:07.530.30+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

DhcpServer: Send ICMP detect success

Jan 22 2014 09:44:07.530.31+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Success const discover end

Jan 22 2014 09:44:07.530.32+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCP_COMM_DealIcmp DHCP_FSM_ICMP_Receive

Jan 22 2014 09:44:07.530.33+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCP_COMM_DealIcmp

Jan 22 2014 09:44:07.530.34+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

pstrIcp->icmp_nsId is 211,g_ulDhcpTaskID is 211

Jan 22 2014 09:44:07.530.35+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCP_COMM_DealIcmp ulFromIp is 2888433666, ulVrfId is 11

Jan 22 2014 09:44:07.530.36+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCP_FSM_ICMP_Receive

Jan 22 2014 09:44:07.530.37+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

DhcpServer: ICMP ECHOREPLY received from Client IP 172.42.0.2

Jan 22 2014 09:44:07.530.38+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCP_FSM_ICMP_Receive ulGiaddr is 2888433666,ulIPAddr is 2888433666,ulVrfIndex is 11

Jan 22 2014 09:44:07.530.39+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCPS_Ping_RecvDelProc

Jan 22 2014 09:44:07.530.40+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

DHCPS_Ping_RecvDelProc ICMP ECHOREPLY received Del timer and hash

Jan 22 2014 09:44:07.530.41+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Success const discover begin

Jan 22 2014 09:44:07.530.42+08:00 NE40E DHCPS/7/DHCPS_DBG: Event:

Enter DHCPS_HandleDiscoverPkt:Giaddr is 2888433666.VrfIndex is 11.User:001e-73c8-f5cc.  ---

The discover process to apply the specified address process will be re-take after assign an conflict address. The process will return to idle if assign an conflict address, thus into circulation leads to high cpu.

 

Jan 22 2014 09:44: 07.530.43 + 08:00 NE40E DHCPS / 7 / DHCPS_DBG: Event:

EnterAM _ DHCPS ReqIptoapply IP [ ac2a0002 ]

 

Information analysis from debug, the user application specified ip address 172.42.0.2, for the interface configuration ip address, since this address will be assigned after the initiative to detect whether a conflict, since the interface is configured with this address, so the inevitable conflict, set address the conflict, and then continue to walk address allocation process, because the distribution is still required to specify the ip address, the current treatment is to address conflicts continue normal distribution, resulting address assignment conflict into conflict allocation allocation cycle, resulting in high DHCP tasks, CPU utilization rate.

 

Root Cause

The reason may be:
1.
Configuration issue;
2
Attack issue;

This case problem is reason one.

Solution

The address of interface is set to gateway address of DHCP server  address pool.

Suggestions

Proposals interface down the address is set to DHCP server The address pool gateway.

END