Dual-line redundancy and load balancing with PBR, SLA, and Eem on Cisco IOS

Background:

With the information office becoming more and more popular, people's office relies on the Internet more and more, it, the network in the enterprise's influence is more and more big. Therefore, the reliability of the enterprise network becomes very important. For an internet company or a company that needs to use a network to work properly, a broken network means that employees cannot work properly and the company has a lot of money to lose. In such a context, large companies and the financial industry have already achieved two-line or multi-line redundancy, and some central enterprises due to the budget and other problems have endured such distress. This article will provide a perfect solution for small and medium-sized enterprises to achieve dual-line redundancy, load balancing and automatic switching through existing equipment or a lower budget.

Let me introduce the basic architecture of double-line internet access for large and medium-sized companies. The general enterprises are through the leasing of 2 different ISP links, such as a unicom link, a telecommunications link, respectively, access to the company network, and then use a dedicated load balancer device for the link load and redundancy, such as foreign F5, radware, domestic deep conviction.

However, like the foreign manufacturers of load balancer a more than 10 200,000, the domestic is also in good tens of thousands of a Taiwan. If the average small and medium-sized company, it budget is limited, such equipment is difficult to buy. But also want to achieve dual-line access, load balancing, dual-line redundancy, the following solutions can be used, only the use of ordinary routers and switches can be achieved.

Blog Content objectives:

1. Basic Overview of PBR

2. Dual-line redundancy with static floating routing

3. Double-line redundancy with SLAs

4. Implement VLAN-based load and redundancy using PBR combined with Sla&eem

Network egress topology diagram:

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M01/6E/51/wKiom1V4_M_xcPGQAAHrLihuhOU460.jpg "title=" 4.png " alt= "Wkiom1v4_m_xcpgqaahrlihuhou460.jpg"/>

The author through the above topological diagram to carry on a comprehensive case explanation. Topology 1 and Topology 2 differ in that topology 1 is more than topology 2 of a router's device redundancy, and the PBR and SLA policies are done on the core switch, while the Topology 2 router assumes the PBR, SLA and other functions.

Topology 1 Explanation: The two routers are connected to the Unicom and the telecom network, the router needs to refer to a default route to the operator, the internal can use static routes or dynamic routing, the router performs NAT function, the intranet IP address is converted into a public network IP address. Use the default route on the layer three switch to point to R2, use PBR to point a portion of the traffic to R1, to achieve load balancing; When the R1 fails, PBR automatically fails, all traffic goes to R2, and the switch uses a floating route to R1, which enables all traffic to be switched to R2 after the R1 failure.

Note: It is generally recommended that link large, stable links use the default route. It is important to note that if the internal server needs to provide services, it is best to put the server on the default route of the link, of course, you can also do NAT on both routers, through DNS to control the external traffic access, when doing PBR also exclude the local server address, if there is VPN traffic, Also pay attention to the control of VPN traffic.

1. Using PBR to implement VLAN-based load Balancing

PBR (policy-based Routing) policy-based routing. We all know that the route is the base destination IP address to forward, when a packet arrives at the router, the router defaults not to view the source IP address of the packet, only the destination IP address, then anza by the table, carried out with the operation, and finally encapsulated a new two-layer header to forward it out. PBR is based on the IP packet's source IP address forwarding, we can manually define which source IP address to which interface to send. For example, the source is the 1.1.x.x IP packet that is emitted from the S0 interface, and the source address is 1.2.x.x packets from the S1 interface. PBR takes precedence over the routing table, and when PBR fails, the matching packet still executes the default route forwarding rule. Cisco, Huawei, Huawei and other enterprise-class routers and three-layer switches generally support the function of PBR, but the specific operation may be different, the principle of datong small meaning.

We implemented VLAN-based load balancing by using PBR, for example, there were 600 people in the company, one VLAN per 100 people, and a total of 6 VLANs. In this way we let the former VLAN 1 2 3 go through the telecom link by default route, let VLAN 4 5 6 go through PBR link.

PBR failure: When PBR fails, all traffic is forwarded through the routing table. PBR through the ARP refresh to determine the failure, 1, when the R1 complete router failure or interface down, SW can immediately perceive, this time PBR failure; If R1 is just the interface IP is removed, SW needs to wait for ARP aging to determine the PBR failure, if it is R1 connected operator link failure, The SW cannot be judged (judging by the SLA that follows)

2. Dual-line redundancy with static floating routing

We know that in general the router exits will write a default route to the operator, when two-wire access we use static floating route to achieve the default route redundancy. We know that the management distance of the static route is 1 by default, we point the default route to R2 on the SW, and write a default route that manages a distance of 5 (as long as it is greater than 1), and when the route of AD 1 is present, the route of AD 5 does not appear on the router, only if the route of AD 1 fails. Ad-5 routing does not work until the routing table is surfaced. In this way, when the R2 fails, the ad for 5 routes works to implement a backup of the default route.

3. Double-line redundancy with SLAs

Although we seem to have implemented redundancy through PBR and floating routing, this is not the case. Routers are less likely to be bad, interfaces are bad or the odds of a bad network are smaller, and more so, there is a problem with the operator's link. In this case, SW is not aware of the problem occurs, or the need for human intervention. To this end, we introduced Cisco SLA technology (other vendors have similar technologies, such as Huawei's Nqa). SLAs can be used to determine the good and bad quality of a link by sending information such as ICMP packets. Static routes can be combined with SLAs, and when an IP address fails, the static route fails. For example, by using an SLA to ping 8.8.8.8 This IP address, when 10 packets are lost, we think that the IP fails, and the static route to track this SLA fails.

4. Implement VLAN-based load and redundancy using PBR combined with Sla&eem

However, after we used the SLA, when the carrier failed, the default route was good to disappear from the router, and the floating route took effect. However, if it is a unicom fault, SLA can perceive, but nothing can be done, PBR is still in force. This time we introduce a technology, Cisco's Eem. EEM joint SLA, the ability to automatically shutdown the SW connection R1 port after the detected IP expires, so that the PBP fails, the traffic is switched to R2.

Important: Two-line redundancy can be implemented on a router through a variety of techniques, and using PBR here is just a starting point. SLA and EEM are introduced to introduce two new technologies. SLAs and Eem can be implemented with a lot of features, super-powerful, if the needs of children's shoes can be studied.

Comprehensive Experiment:

The topology diagram is as follows:

650) this.width=650; "src=" http://s3.51cto.com/wyfs02/M02/6E/4F/wKioL1V5Kljhx-u9AALuJlofKVQ491.jpg "title=" 5.png " alt= "Wkiol1v5kljhx-u9aalujlofkvq491.jpg"/>

We use R1 and R5 to simulate two routers connected to different operators, using R6 to emulate the core switch, R7 and R8 to simulate PC1 and PC2, acting as different VLANs, the switch in the diagram is a fool switch, just play a connection role. Using R2 and R4 to simulate the router devices of Unicom and telecom, the rightmost R3 to simulate a device to be accessed on the Internet. In order to be more realistic, we have NAT configuration in R1 and R5 to convert the private network IP inside the enterprise into public IP.

The basic configuration is as follows:

Carrier Router configuration:

R2 configuration:

R2 (config) #int s1/0

R2 (config-if) #ip add 12.1.1.2 255.255.255.0

R2 (config-if) #no shut

R2 (config-if) #int S1/1

R2 (config-if) #ip add 23.1.1.2 255.255.255.0

R2 (config-if) #no shut

R2 (config-if) #router RIP

R2 (config-router) #ver 2

R2 (config-router) #no Auto

R2 (config-router) #net 23.0.0.0

R2 (config-router) #net 12.0.0.0

Use the dynamic routing protocol to connect networks between R2, R3, and R4.

R3 configuration:

R3 (config) #int s1/0

R3 (config-if) #ip add 23.1.1.3 255.255.255.0

R3 (config-if) #no shut

R3 (config-if) #int S1/1

R3 (config-if) #ip add 34.1.1.3 255.255.255.0

R3 (config) #int Lo0

R3 (config-if) #ip add 3.3.3.3 255.255.255.0

R3 (config-if) #router RIP

R3 (config-router) #ver 2

R3 (config-router) #no Auto

R3 (config-router) #net 3.0.0.0

R3 (config-router) #net 23.0.0.0

R3 (config-router) #net 34.0.0.0

R4 configuration:

R4 (config) #int s1/0

R4 (config-if) #ip add 34.1.1.4 255.255.255.0

R4 (config-if) #no shut

R4 (config-if) #int S1/1

R4 (config-if) #ip add 45.1.1.4 255.255.255.0

R4 (config-if) #no shut

R4 (config-if) #router RIP

R4 (config-router) #ver 2

R4 (config-router) #no Auto

R4 (config-router) #net 34.0.0.0

R4 (config-router) #net 45.0.0.0

Egress Gateway Router configuration:

R1 configuration:

R1 (config) #int S1/1

R1 (config-if) #ip add 12.1.1.1 255.255.255.0

R1 (config-if) #ip Nat outside

R1 (config-if) #no shut

R1 (config-if) #int f0/0

R1 (config-if) #ip add 172.16.1.1 255.255.255.0

R1 (config-if) #ip nat inside

R1 (config-if) #no shut

R1 (config-if) #ip Route 0.0.0.0 0.0.0.0 12.1.1.2

R1 (config-if) #access-list permit IP 172.16.0.0 0.0.255.255 any

R1 (config) #ip NAT inside Source list interface S1/1 overload

NAT is configured on the R1 to convert the intranet IP in the enterprise into a public IP address.

R1 (config) #router rip

R1 (config-router) #ver 2

R1 (config-router) #no Auto

R1 (config-router) #net 172.16.0.0

R1, R5, R6 run dynamic routing protocols, or use static routes on R1 to point to R6.

R5 configuration:

R5 (config) #int s1/0

R5 (config-if) #ip add 45.1.1.5 255.255.255.0

R5 (config-if) #ip Nat outside

R5 (config-if) #no shut

R5 (config-if) #int f0/0

R5 (config-if) #ip add 172.16.1.5 255.255.255.0

R5 (config-if) #ip nat inside

R5 (config-if) #no shut

R5 (config-if) #ip Route 0.0.0.0 0.0.0.0 45.1.1.4

R5 (config) #access-list permit IP 172.16.0.0 0.0.255.255 any

R5 (config) #ip NAT inside Source list interface s1/0 overload

R5 (config) #router rip

R5 (config-router) #ver 2

R5 (config-router) #no Auto

R5 (config-router) #net 172.16.0.0

Analog Switch configuration:

R6 configuration:

R6 (config) #int f0/0

R6 (config-if) #ip ad 172.16.1.6 255.255.255.0

R6 (config-if) #no shut

R6 (config-if) #int f2/0

R6 (config-if) #ip add 172.16.2.1 255.255.255.0

R6 (config-if) #no shut

R6 (config-if) #router RIP

R6 (config-router) #ver 2

R6 (config-router) #no Auto

R6 (config-router) #net 172.16.0.0

PC Configuration:

PC1 (config) #no IP routing

Switch off the routing function on the router, simulate the PC

PC1 (config) #int f2/0

PC1 (config-if) #ip add 172.16.2.2 255.255.255.0

PC1 (config-if) #no shut

PC1 (config-if) #ip Default-gateway 172.16.2.1

PC needs to configure Gateway

PC2 (config) #no IP routing

PC2 (config) #int f2/0

PC2 (config-if) #ip add 172.16.2.3 255.255.255.0

PC2 (config-if) #no shut

PC2 (config-if) #ip Default-gateway 172.16.2.1

Static routing and floating floating static routing configuration

R6 (config) #ip Route 0.0.0.0 0.0.0.0 172.16.1.5

R6 (config) #ip Route 0.0.0.0 0.0.0.0 172.16.1.1 5

This configuration needs to be removed at a later stage to reconfigure, join track, Federated SLA.

Look at the router, only the first one.

R6#show IP route

172.16.0.0/24 is subnetted, 2 subnets

C 172.16.1.0 is directly connected, fastethernet0/0

C 172.16.2.0 is directly connected, fastethernet2/0

s* 0.0.0.0/0 [1/0] via 172.16.1.5

PBR configuration:

R6 (config) #ip access-list extended CNC

R6 (config-ext-nacl) #permit IP host 172.16.2.2 any

Write an ACL first to match the IP that needs to be hit by PBR, which is the host that needs to go through the R1 router.

R6 (config) #route-map PBR-CNC Permit 10

R6 (config-route-map) #match IP address CNC

R6 (config-route-map) #set IP next-hop 172.16.1.1

Write a route-map, match the ACL just written, set the next hop to the R1 IP address.

R6 (config) #int f2/0

R6 (config-if) #ip policy Route-map PBR-CNC

In the f2/0 interface call PBR, note that PBR is called in the incoming interface.

Attention:

If it is a 3560x-e switch, the error message may occur when the PBR is applied to the interface, and the SDM needs to be set

SDM prefer routing//global Use this command

Restarting the switch

Source-based load balancing is now implemented:

Open debug on the R3 test, if the source is 12.1.1.1, the description is to go R1; if the source is 45.1.1.5, the description goes R2

R3#debug IP ICMP

ICMP packet Debugging is on

R7 Ping 3.3.3.3来 to test:

Pc1#ping 3.3.3.3

Type escape sequence to abort.

Sending 5, 100-byte ICMP Echos to 3.3.3.3, timeout is 2 seconds:

!!!!!

Success rate is percent (5/5), round-trip Min/avg/max = 24/40/56 ms

To view the results on R3:

*jun 15:07:00.091:icmp:echo reply sent, src 3.3.3.3, DST 12.1.1.1

*jun 15:07:00.131:icmp:echo reply sent, src 3.3.3.3, DST 12.1.1.1

*jun 15:07:00.155:icmp:echo reply sent, src 3.3.3.3, DST 12.1.1.1

*jun 15:07:00.171:icmp:echo reply sent, src 3.3.3.3, DST 12.1.1.1

*jun 15:07:00.223:icmp:echo reply sent, src 3.3.3.3, DST 12.1.1.1

Use Traceroute to verify:

Pc1#traceroute 3.3.3.3

Type escape sequence to abort.

Tracing the route to 3.3.3.3

1 172.16.2.1 msec msec 4 msec

2 172.16.1.1 4 msec msec msec

3 12.1.1.2 msec msec msec

4 23.1.1.3 msec msec *

R8 Ping 3.3.3.3

Pc2#ping 3.3.3.3

Type escape sequence to abort.

Sending 5, 100-byte ICMP Echos to 3.3.3.3, timeout is 2 seconds:

!!!!!

Success rate is percent (5/5), round-trip Min/avg/max = 20/32/48 ms

View results on R3

*jun 15:07:02.375:icmp:echo reply sent, src 3.3.3.3, DST 45.1.1.5

*jun 15:07:02.403:icmp:echo reply sent, src 3.3.3.3, DST 45.1.1.5

*jun 15:07:02.455:icmp:echo reply sent, src 3.3.3.3, DST 45.1.1.5

*jun 15:07:02.471:icmp:echo reply sent, src 3.3.3.3, DST 45.1.1.5

*jun 15:07:02.527:icmp:echo reply sent, src 3.3.3.3, DST 45.1.1.5

Pc2#traceroute 3.3.3.3

Type escape sequence to abort.

Tracing the route to 3.3.3.3

1 172.16.2.1 8 msec 8 msec 4 msec

2 172.16.1.5 msec 8 msec msec

3 45.1.1.4 msec msec msec

4 34.1.1.3 msec * msec

Configure SLAs and Eem:

R6 (config) #ip SLA 2

R6 (Config-ip-sla) # Icmp-echo 12.1.1.2 source-interface loopback0

R6 (Config-ip-sla-echo) # frequency 10

R6 (Config-ip-sla-echo) #ip SLA Schedule 2 life Forever Start-time Now

Configure an IP SLA named 2, use the loopback interface as the source IP address, 12.1.1.2 this IP address with ICMP detection

R6 (config) #ip SLA 3

R6 (Config-ip-sla) # Icmp-echo 45.1.1.4 source-interface loopback0

R6 (Config-ip-sla-echo) # frequency 10

R6 (Config-ip-sla-echo) #ip SLA Schedule 3 life Forever Start-time Now

Configure an IP SLA named 3, use the loopback interface as the source IP address, 45.1.1.4 this IP address with ICMP detection

R6 (config) #track 2 IP SLA 2 reachability

Configure a track that tracks the accessibility of IP SLA 2

R6 (config) #track 3 IP SLA 3 reachability

R6 (config) #track IP SLA 2

Configure a Taack 200 to track the status of IP SLA 2

R6 (config) #ip Route 0.0.0.0 0.0.0.0 172.16.1.1 5 Track 2

Delete the default and floating routes that you just wrote, add them again, and put a track behind the route. When the Track2 fails, the route fails, and this route plus track is of little significance

R6 (config) #ip Route 0.0.0.0 0.0.0.0 172.16.1.5 Track 3

Add the default route again and track. When the IP SLA3 is not reached, the triggering track3,track3 fails, and the default route expires after the TRACK3 expires. At this point, the ad 5 is configured with a floating routing table, which takes effect and forwards the data. This means that when the carrier link (the telco link) that the default route points to fails, the SLA and the floating route can be used to switch, without the need for EEM.

R6 (config) #ip Route 23.1.1.2 255.255.255.255 172.16.1.1

R6 (config) #ip Route 34.1.1.4 255.255.255.255 172.16.1.5

These two routes are important, since PBR is called on the f2/0 interface, only f2/0 incoming traffic will execute PBR. And we need to monitor 23.1.1.2 (unicom IP) when the packet is thrown to R1, monitoring 34.1.1.4 (telecommunications IP) when the packet is thrown to R2. All default routes point to R2 and do not meet our needs. Therefore, you need to add these two static routes to achieve the above requirements.

R6 (config) #int Lo 0

R6 (config-if) #ip add 172.16.3.1 255.255.255.0

Add the Loopback interface IP address for the newly written SLA, which is used as the source IP address for monitoring

R6 (config) #event Manager applet Sp-down

R6 (Config-applet) # event Track

R6 (Config-applet) # Action 1.1 cli command "en"

R6 (Config-applet) # Action 2.1 CLI command "conf t"

R6 (Config-applet) # Action 3.1 CLI command "int f2/0"

R6 (Config-applet) # Action 4.1 CLI Command "No IP policy route-map pbr-cnc"

R6 (config-applet) # Action 5.1 CLI Command "end"

Write a Eem, when track 200 status is DONW, perform the following action, that is, enter the f2/0 interface, remove the PBR configuration under the interface.

R6 (config) #event Manager applet sp-up

R6 (Config-applet) # event Track

R6 (Config-applet) # Action 1.1 cli command "en"

R6 (Config-applet) # Action 2.1 CLI command "conf t"

R6 (Config-applet) # Action 3.1 CLI command "int f2/0"

R6 (Config-applet) # Action 4.1 CLI command "IP policy route-map pbr-cnc"

R6 (config-applet) # Action 5.1 CLI Command "end"

Write a Eem, when track 200 's status is up, perform the following action, that is, go to the f2/0 interface, add the configuration on the PBR.

All configurations are complete and are tested below.

Test:

Telecom Link Test:

To view the routing table on R6:

R6#show IP route

s* 0.0.0.0/0 [5/0] via 172.16.1.5

23.0.0.0/32 is subnetted, 1 subnets

S 23.1.1.2 [1/0] via 172.16.1.1

34.0.0.0/32 is subnetted, 1 subnets

S 34.1.1.4 [1/0] via 172.16.1.5

Note The default route is pointing to 172.16.1.5

Check the status of the SLA on the R6, OK

Ipslas Latest Operation Summary

Codes: * Active, ^ inactive, ~ Pending

ID Type Destination Stats Return Last (ms) Code Ru N

-----------------------------------------------------------------------

* * Icmp-echo 23.1.1.2 rtt=44 OK 7 seconds ago

Icmp-echo 34.1.1.4 rtt=19 OK 2 seconds ago

ping3.3.3.3 on the PC2.

Pc2#ping 3.3.3.3 Repeat 1000

When simulating a telecom link failure, the R4 s1/0 interface is shutdown

Pc2#ping 3.3.3.3 Repeat 1000

Type escape sequence to abort.

Sending, 100-byte ICMP Echos to 3.3.3.3, timeout is 2 seconds:

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!! U.u.u.u.u.u.u.u.!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

To view SLAs:

Ipslas Latest Operation Summary

Codes: * Active, ^ inactive, ~ Pending

ID Type Destination Stats Return Last

(MS) Code Run

-----------------------------------------------------------------------

* * Icmp-echo 23.1.1.2 rtt=16 OK 7 seconds ago

Icmp-echo 34.1.1.4-timeout 2 seconds ago

You can see 34.1.1.4 is already a timeout.

To view the routing table:

R6#show IP route

s* 0.0.0.0/0 [5/0] via 172.16.1.1

23.0.0.0/32 is subnetted, 1 subnets

S 23.1.1.2 [1/0] via 172.16.1.1

34.0.0.0/32 is subnetted, 1 subnets

S 34.1.1.4 [1/0] via 172.16.1.5

You can see that the floating static is in effect and all traffic has been switched over.

Recovering the s1/0 interface on R4

R4 (config) #int s1/0

R4 (config-if) #no shut

View the routing table again

R6#show IP route

s* 0.0.0.0/0 [1/0] via 172.16.1.5

23.0.0.0/32 is subnetted, 1 subnets

S 23.1.1.2 [1/0] via 172.16.1.1

34.0.0.0/32 is subnetted, 1 subnets

S 34.1.1.4 [1/0] via 172.16.1.5

Can see that it has been restored.

Test Unicom link Failure:

Viewing the configuration of f2/0 in R6

Interface fastethernet2/0

IP address 172.16.2.1 255.255.255.0

IP policy Route-map PBR-CNC

Duplex half

End

On the PC1 long ping3.3.3.3, and then shutdown R2 S1/1 interface, analog Unicom link failure.

Pc1#ping 3.3.3.3 RE

Pc1#ping 3.3.3.3 Repeat 1000

Type escape sequence to abort.

Sending, 100-byte ICMP Echos to 3.3.3.3, timeout is 2 seconds:

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! U.U.U:!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Observation on the R6, you can see the log, Eem has taken effect

r6#

*jun 17:39:48.215:%TRACKING-5-STATE:200 IP SLA 2 State Up->down

*jun 17:39:48.219:%TRACKING-5-STATE:2 IP SLA 2 reachability up->down

*jun 17:39:48.387:%sys-5-config_i:configured from console to on Vty0 (Eem:sp-down)

To view the configuration of the F0/2:

R6#sh Run int f2/0

Current configuration:112 bytes

!

Interface fastethernet2/0

IP address 172.16.2.1 255.255.255.0

Duplex half

End

Restore the interface on R2

R2 (config) #int S1/1

R2 (config-if) #no shut

Observe the logs on the R6:

*jun 17:46:08.223:%TRACKING-5-STATE:200 IP SLA 2 State down->up

*jun 17:46:08.227:%TRACKING-5-STATE:2 IP SLA 2 reachability down->up

*jun 17:46:08.439:%sys-5-config_i:configured from console to on Vty0 (eem:sp-up)

To view the configuration of the F0/2:

R6#sh Run int f2/0

Current configuration:112 bytes

!

Interface fastethernet2/0

IP address 172.16.2.1 255.255.255.0

IP policy Route-map PBR-CNC

Duplex half

End

Hope this article can play a role, readers can combine other technologies to improve their own network.

This article is from "Yang Sen's It Road" blog, please be sure to keep this source http://senyang.blog.51cto.com/3427514/1660961

Dual-line redundancy and load balancing with PBR, SLA, and Eem on Cisco IOS