Traffic Flow Decisions in MPLS Network

-
In this article, we will study how forwarding decisions are made in a MPLS Network. You can check out the configuration of the network at Creating Layer 3 MPLS VPN.

Network Topology:


Download the gns3 topology here:

Traffic Flow Decisions in MPLS Network


In the given Network, We will walk along the packet sent from the inside network of CE1_A for the destination behind CE2_A, i.e. source is 172.16.1.1 and destination is 172.16.2.1. 

Let’s see if the destination is reachable from source.

CE1_A#ping 172.16.2.1 source 172.16.1.1
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 172.16.2.1, timeout is 2 seconds:
Packet sent with a source address of 172.16.1.1
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 68/88/116 ms


Yes, the destination is reachable.


Now let's walk along the packet sent from CE1_A to CE2_A. At this time the packet is at CE1_A and we have executed the command "ping 172.16.2.1 source 172.16.1.1". Below are the tables, that would be referenced to forward the packet to correct path.


To reach the destination IP address 172.16.1.1, CE1_A will reference its CEF table to forward the packets towards exit interface. CE1_A has a default route to reach the destination pointing towards 10.0.0.1, so the packets will be forwarded to 10.0.0.1 via interface GigabitEthernet 0/0.

So, from where the CEF table is built? It is built from the routing table, and we know how the routing table is built. CEF table is built by extracting the required details from various tables on a router, one of which is routing table. CEF takes next hop for all the destinations and do a recursive lookup to find the exit interfaces.


Now, the packet reaches at PE1. Let's check out what tables PE1 will reference to forward the packets further.

To forward the packets destined to 172.16.2.1, PE1 will reference its CEF table to find out the exit interface. After analyzing the CEF table PE1 will find out the exit interface and also the mpls label information, i.e. which label to remove and which to impose. In this case PE1 will  impose the label 208 before forwarding it to the nexthop out of interface GigabitEthernet 0/0.

So, from where the CEF table get the information of Label Forwarding Information Base? It gets this detail from the mpls forwarding-table, which is the LFIB of MPLS. MPLS forwarding-table has an entry for every prefix contained in the routing table with the corresponding local label, outgoing label, outgoing interface and IP next hop.
MPLS forwarding table (LFIB) is built by referencing mpls ldp binding table and routing table. mpls ldp binding table lookup is done to find out the label assigned by the downstream lsr. And routing table lookup is done to find out the IP next hop address, which will be used to find out the layer 2 address of the peer LSR for forwarding the frames.


Now, the packet reaches at P1. Let's check out how P1 will make decision to forward the packets further.


P1 will also forward the packet based on the information collected by CEF table. But look at the output of show mpls ldp bindings 172.16.2.0 24 detail above, P1 has two remote binding for the network 172.16.2.0/24. P1 is receiving labels from both the peer LSRs i.e. PE1 (1.1.1.1) and P2 (3.3.3.3), implying that the prefix 172.16.2.0/24 is reachable via both LSRs. But P1 has only added the label received from P2 to its mpls forwarding-table (LFIB).

So how P1 decided to add only the label received from P2 to mpls forwarding-table? The answer is routing table. When P1 sees two remote bindings for a prefix, it will do a recursive lookup of routing table to find the best path to reach the destination. In this case P1 finds the best path via 10.23.0.2 (3.3.3.3). Below is the mpls ldp neighbor table of P1, which shows the peer LDP Identity along with the peer IP address.


Now the packet reaches P2, It will also follow the same steps as P1 to make the forwarding decision. P2 will reference its CEF table and find out that packets destined to 172.16.2.1 should be forwarded with a label of 408 to the nexthop 10.34.0.2 out of interface GigabitEthernet0/0.



After passing through the P2, packet reaches PE2. PE2 will also reference it's CEF table and find out that the packets destined to 172.16.2.1 should be forwarded to the next hop 10.1.1.2 out of interface GigabitEthernet1/0 without any label, i.e. PE2 will remove the label 408 before forwarding it to CE2_A



Now the packet will reach CE2_A, where CEF table will be referenced for the further path. After referencing the CEF table, CE2_A finds that the destination prefix is directly connected and the IP address is configured locally. So the router will accept the packet and process it.



So, above are the steps that are followed during decision making for forwarding of mpls packets.

Also, it is good to know that for forwarding the frames to next hop, mpls don't have to send arp requests to find the next hop layer 2 address. As all the forwarding has to be done before layer 3 in mpls, mpls stores the next hop layer 2 address in LFIB i.e. mpls forwarding table. Below is an example.



Some basic show commands in mpls environment.

# show ip cef
# show mpls forwarding-table
# show mpls ldp bindings
# show ip route
# show mpls ldp neighbor
# show mpls ip binding
# show mpls interfaces

Summary

The traffic flow decision is made by referencing the FIB. Below are the different tables used in MPLS in a sequence of which table is build referencing which table, e.g. FIB is build using information from LFIB

FIB - Forwarding Information Base - show ip cef
LFIB - Label Forwarding Information Base - show mpld forwarding table
LIB - Label Information Base - show mpls ldp bindings
RIB - Routing Information Base - show ip route




Comments

Post a Comment

Popular posts from this blog

Anyconnect SSL-Client VPN with Self-signed Certificate on Cisco ASA

-
Image
The Cisco AnyConnect Secure Mobility Solution provides a comprehensive, highly secure enterprise mobility solution. the Cisco AnyConnect Secure Mobility Solution continues to lead with next-generation security and encryption, including support for the Suite B set of cryptographic algorithms, and support for IPv6 networks. More importantly, it adapts its tunneling protocol to the most efficient method. In the present scenario, we have to configure Anyconnect SSL remote access VPN for Sales department and Engineering department of a company. Engineering users will have to be provided with access to web server as well as FTP server, while sales users may only have access to the web server.
Read more

Filtering Routes in BGP using Route-maps and Prefix-list

-
Image
Order of preference of attributes in BGP The order of preference varies based on whether the attributes are applied for inbound updates or outbound updates. For inbound updates the order of preference is:     route-map     filter-list     prefix-list, distribute-list For outbound updates the order of preference is:     prefix-list, distribute-list     filter-list     route-map NOTE: The attributes prefix-list and distribute-list are mutually exclusive, and only one command (neighbor distribute-list or neighbor prefix-list) can be applied to each inbound or outbound direction for a particular neighbor. Scenario: We own the AS500 and advertising a network block of 192.0.2.0/24 and 180.179.179.0/16 to two different ISPs.
Read more

Open Shortest Path First (OSPF)

-
Image
The Open Shortest Path First (OSPF) is an Interior Gateway Protocol used to distribute routing information within a single Autonomous System. OSPF is a Link State routing protocol and has following features. Fast convergence VLSM support Scalable to large networks Stub capabilities Authenticaion Efficiency (Sends only hello messages to maintain neighbor relationship) Route tagging capabilities Uses multicast for routing updates. Specifications                              Protocol type            Link-state               Administrative Distance            110                                          Metric            Cost                             Authentication            None, Clear-text, MD5                                Seed metric            20 for all redistributed routes and 1 for BGP routes.      Multicast address used by All             224.0.0.5   Multicast add. used by DR/BDR             225.0.0.6                                            RFC       
Read more

IKEv2 IPsec Site-to-Site VPN configuration on Cisco ASA 8.4(x)

-
Image
Though the crypto IKEv2 proposal command looks similar to the IKEv1 crypto isakmp policy command, there are many differences in how IKEv2 negotiates. Unlike IKEv1, the authentication method and SA lifetime are not negotiable in IKEv2, and they cannot be configured in the IKEv2 proposal. Also the IKEv2 proposal configuration supports specifying multiple options for each transform type and we can configure different pre-shared-key for local and remote authentication. Pseudo-Random Function (PRF) algorithm  is the same as the integrity algorithm, and hence, it is not configured separately. Our goal is to configure site-to-site or l2l IKEv2 IPsec VPN between ASA1 and ASA2 which are running IOS 8.4(2). IP addressing scheme is mentioned in the network diagram.
Read more

IPsec VPN as a Backup for Point-to-Point Link using IP SLA

-
Image
There may be times when you wish to access your remote site even when the point-to-point link (lease line) to that site is down. This is possible if you have also an internet connection to that site. You can configure an IPsec VPN between both the sites. But the problem is that even if you configure IPsec VPN on both the locations, the VPN will never come up when the point-to-point connection goes down. This is because even though that point-to-point connection is down, the interface status of firewall will always remain up due to devices of ISP in between and thus the static route will not let the incoming traffic to match the interesting traffic for IPsec VPN. So the solution to this problem is either to configure dynamic routing protocols or IP SLA. Cisco IP SLA monitor can be used to track the reachability of other location. Here is an example of such scenario using Cisco IP SLA monitor. If you need the gns3 lab for this or have any queries, you can leave your queries and email ID
Read more

Border Gateway Protocol (BGP)

-
Image
Introduction Border Gateway Protocol is an Exterior Gateway Protocol (EGP) used for routing between the autonomous systems. It is the protocol which is used to to make core routing decisions on the internet. BGP has two flavours, eBGP and iBGP. eBGP is routing between the autonomous systems and iBGP is routing within an autonomous systems. BGP is a path-vector routing protocol and has following features: Reliable updaes (TCP -based, port 179) Triggered updates only Rich metric Scalable to massive networks   Specifications              Protocol Type                Path-vector                    eBGP AD                20                     iBGP AD                200               Update Mode                Triggered only                     Transport                TCP/179              Authentication                None, md5                            RFC                RFC 4271 Default Timers                       Holddown time          180 seconds              
Read more

Cisco ASA Active/Active Failover Configuration

-
Image
Active/Active failover is only available to security appliances in multiple context mode. In an Active/Active failover configuration, both security appliances can pass network traffic. In Active/Active failover, you divide the security contexts on the security appliance into failover groups. A failover group is simply a logical group of one or more security contexts. You can create a maximum of two failover groups on the security appliance. The admin context is always a member of failover group 1. Any unassigned security contexts are also members of failover group 1 by default. We have already seen the configuration for Active/Standby failover in the previous article. This article focuses on how to configure an Active/Active Failover configuration on ASA Security Appliance. Network Diagram  (Physical Topology)
Read more

Bypassing Proxy Server in Google Chrome

-
Image
There may be times when you might want to use Google Chrome and do not wish to use the proxy settings for which you are bound to use in Internet Explorer because of your company policies. If you try to change proxy settings in Google Chrome, by default it will open the System Internet Properties dialogue box, and most of the times, change in proxy setting options is restricted by domain group policy. So here is an alternate way to remove or disable proxy settings in Google Chrome.
Read more

Cisco ASA Active/Standby Failover Configuration

-
Image
The security appliance supports two failover configurations: Active/Active Failover and Active/Standby Failover. Each failover configuration has its own method to determine and perform failover. With Active/Active Failover, both units can pass network traffic. This lets you configure load balancing on your network. Active/Active Failover is only available on units that run in multiple context mode. With Active/Standby Failover, only one unit passes traffic while the other unit waits in a standby state. If Active unit fails, secondary will take over and starts forwarding the traffic. When Secondary becomes active, it will also change it's interface IP address and mac address as well.  In Active/Standby configuration, virtually all of the configuration from the active unit is replicated to the secondary unit through a failover cable. This article focuses on how to configure an Active/Standby Failover in ASA Security Appliance. Prerequisites Hardware Requirements The two
Read more