MPLS TE Affinity Attribute

MPLS Traffic Engineering (TE) attribute flags and tunnel affinity allow you to set attributes on interfaces and configure tunnels to either include or exclude specific links based on those attributes. This is also known as link coloring.

Like TE metric or explicit path options, you can use this to decide what path your tunnels should use.

What are attributes? That’s entirely up to you. There are no predefined attributes. The link type (for example, Fiber, DSL, Satellite) could be attributes. Whether a link is encrypted or not, connected to another country. You name it. You can use whatever you think is useful.

An attribute flag is 32-bit, meaning you can use this to define 32 separate properties on a link. It’s written in hexadecimal and, by default, all set to zeroes.

Attribute Flag

Let’s look at an example:

Above, we have an MPLS TE network with links with different attributes:

• Fiber links
• DSL link
• Satellite link
• Encrypted link
• Link to another country

We could use attribute flags like this:

I made up these values myself. I decided that for this network, a Fiber link should have its least significant bit set to 1. For a DSL link, it should be the third least significant bit, etc. You can use hexadecimal values from 0x00000000 up to 0xFFFFFFFF.

A link could have multiple attributes. For example, let’s say a link has these three attributes:

• Fiber (0000 0001)
• Encrypted (0001 0000)
• Connects to another country (0000 1000)

We set the corresponding bits to 1:

Affinity and Mask

Deciding on attribute flags is one part of the story. On our tunnel interfaces, we have to configure two items:

• Affinity
• Mask

These two items determine whether we include or exclude specific links for our tunnel interface.

With affinity, we set the attribute flag value we want to check, and with the mask, we tell the router what bits we care about or don’t care about. This is similar to how a subnet mask works:

• 1 = the attribute bit must match.
• 0 = we don’t

Don’t worry about this too much for now. In the configuration section, I’ll explain four examples of how we use affinity and different masks.

Configuration

Let’s dive into the configuration. This is the topology I’ll use:

We have a small MPLS TE network. Routers PE1, P1, P2, and PE2 run MPLS TE. PE1 is the headend router for four TE tunnels that terminate at PE2. We are going to use these tunnels to test affinity. You can also see that I added attribute flags to each link. These correspond to the examples I showed previously.

hostname CE1
!
ip cef
!
interface Loopback0
 ip address 1.1.1.1 255.255.255.255
 ip router isis 
 isis circuit-type level-2-only
!
interface GigabitEthernet0/0
 ip address 192.168.12.1 255.255.255.0
 ip router isis 
 isis circuit-type level-2-only
!
router isis
 net 49.0001.0001.0001.0001.0001.00
 is-type level-2-only
 metric-style wide
!
end

hostname CE2
!
ip cef
!
interface Loopback0
 ip address 6.6.6.6 255.255.255.255
 ip router isis
 isis circuit-type level-2-only
!
interface GigabitEthernet0/0
 ip address 192.168.56.6 255.255.255.0
 ip router isis 
 isis circuit-type level-2-only
!
router isis
 net 49.0001.0006.0006.0006.0006.00
 is-type level-2-only
 metric-style wide
!
end

hostname P1
!
ip cef
!
mpls traffic-eng tunnels
!
interface Loopback0
 ip address 3.3.3.3 255.255.255.255
 ip router isis 
 isis circuit-type level-2-only
!
interface GigabitEthernet0/0
 ip address 192.168.23.3 255.255.255.0
 ip router isis 
 mpls traffic-eng tunnels
 mpls ip
 isis circuit-type level-2-only
 ip rsvp bandwidth 1000000
!
interface GigabitEthernet0/1
 ip address 192.168.35.3 255.255.255.0
 ip router isis 
 mpls traffic-eng tunnels
 mpls ip
 isis circuit-type level-2-only
 ip rsvp bandwidth 1000000
!
router isis
 mpls traffic-eng router-id Loopback0
 mpls traffic-eng level-2
 net 49.0001.0003.0003.0003.0003.00
 is-type level-2-only
 metric-style wide
!
mpls ldp router-id Loopback0 force
!
end

hostname P2
!
ip cef
!
mpls traffic-eng tunnels
!
interface Loopback0
 ip address 4.4.4.4 255.255.255.255
 ip router isis 
 isis circuit-type level-2-only
!
interface GigabitEthernet0/0
 ip address 192.168.24.4 255.255.255.0
 ip router isis 
 mpls traffic-eng tunnels
 mpls ip
 isis circuit-type level-2-only
 ip rsvp bandwidth 1000000
!
interface GigabitEthernet0/1
 ip address 192.168.45.4 255.255.255.0
 ip router isis 
 mpls traffic-eng tunnels
 mpls ip
 isis circuit-type level-2-only
 ip rsvp bandwidth 1000000
!
router isis
 mpls traffic-eng router-id Loopback0
 mpls traffic-eng level-2
 net 49.0001.0004.0004.0004.0004.00
 is-type level-2-only
 metric-style wide
!
mpls ldp router-id Loopback0 force
!
end

hostname PE1
!
ip cef
!
mpls traffic-eng tunnels
!
interface Loopback0
 ip address 2.2.2.2 255.255.255.255
 ip router isis 
 isis circuit-type level-2-only
!
interface Tunnel1
 ip unnumbered Loopback0
 tunnel mode mpls traffic-eng
 tunnel destination 5.5.5.5
 tunnel mpls traffic-eng priority 7 7
 tunnel mpls traffic-eng bandwidth 750
 tunnel mpls traffic-eng path-option 1 dynamic
 no routing dynamic
!
interface Tunnel2
 ip unnumbered Loopback0
 tunnel mode mpls traffic-eng
 tunnel destination 5.5.5.5
 tunnel mpls traffic-eng priority 7 7
 tunnel mpls traffic-eng bandwidth 750
 tunnel mpls traffic-eng path-option 1 dynamic
 no routing dynamic
!
interface Tunnel3
 ip unnumbered Loopback0
 tunnel mode mpls traffic-eng
 tunnel destination 5.5.5.5
 tunnel mpls traffic-eng priority 7 7
 tunnel mpls traffic-eng bandwidth 750
 tunnel mpls traffic-eng path-option 1 dynamic
 no routing dynamic
!
interface Tunnel4
 ip unnumbered Loopback0
 tunnel mode mpls traffic-eng
 tunnel destination 5.5.5.5
 tunnel mpls traffic-eng priority 7 7
 tunnel mpls traffic-eng bandwidth 750
 tunnel mpls traffic-eng path-option 1 dynamic
 no routing dynamic
interface GigabitEthernet0/0
 ip address 192.168.12.2 255.255.255.0
 ip router isis 
 isis circuit-type level-2-only
!
interface GigabitEthernet0/1
 ip address 192.168.23.2 255.255.255.0
 ip router isis 
 mpls traffic-eng tunnels
 mpls ip
 isis circuit-type level-2-only
 ip rsvp bandwidth 1000000
!
interface GigabitEthernet0/2
 ip address 192.168.24.2 255.255.255.0
 ip router isis 
 mpls traffic-eng tunnels
 mpls ip
 isis circuit-type level-2-only
 ip rsvp bandwidth 1000000
!
interface GigabitEthernet0/3
 ip address 192.168.25.2 255.255.255.0
 ip router isis 
 mpls traffic-eng tunnels
 mpls ip
 isis circuit-type level-2-only
 ip rsvp bandwidth 1000000
!
router isis
 mpls traffic-eng router-id Loopback0
 mpls traffic-eng level-2
 net 49.0001.0002.0002.0002.0002.00
 is-type level-2-only
 metric-style wide
!
mpls ldp router-id Loopback0 force
!
end

hostname PE2
!
ip cef
!
mpls traffic-eng tunnels
!
interface Loopback0
 ip address 5.5.5.5 255.255.255.255
 ip router isis
 isis circuit-type level-2-only
!
interface GigabitEthernet0/0
 ip address 192.168.56.5 255.255.255.0
 ip router isis 
 isis circuit-type level-2-only
!
interface GigabitEthernet0/1
 ip address 192.168.35.5 255.255.255.0
 ip router isis 
 mpls traffic-eng tunnels
 mpls ip
 isis circuit-type level-2-only
 ip rsvp bandwidth 1000000
!
interface GigabitEthernet0/2
 ip address 192.168.45.5 255.255.255.0
 ip router isis 
 mpls traffic-eng tunnels
 mpls ip
 isis circuit-type level-2-only
 ip rsvp bandwidth 1000000
!
interface GigabitEthernet0/3
 ip address 192.168.25.5 255.255.255.0
 ip router isis 
 mpls traffic-eng tunnels
 mpls ip
 isis circuit-type level-2-only
 ip rsvp bandwidth 1000000
!
router isis
 mpls traffic-eng router-id Loopback0
 mpls traffic-eng level-2
 net 49.0001.0005.0005.0005.0005.00
 is-type level-2-only
 metric-style wide
!
mpls ldp router-id Loopback0 force
!
end

I’m using IOSv Version 15.9(3)M4 on all routers.

By default, all interfaces have the attribute flags all set to zero:

PE1#show mpls traffic-eng topology | include Intf | flags        
          frag_id 0, Intf Address:192.168.24.2
          TE metric:10, IGP metric:10, attribute flags:0x0
          frag_id 0, Intf Address:192.168.23.2
          TE metric:10, IGP metric:10, attribute flags:0x0
          frag_id 0, Intf Address:192.168.25.2
          TE metric:10, IGP metric:10, attribute flags:0x0
          frag_id 0, Intf Address:192.168.23.3
          TE metric:10, IGP metric:10, attribute flags:0x0
          frag_id 0, Intf Address:192.168.35.3
          TE metric:10, IGP metric:10, attribute flags:0x0
          frag_id 0, Intf Address:192.168.24.4
          TE metric:10, IGP metric:10, attribute flags:0x0
          frag_id 0, Intf Address:192.168.45.4
          TE metric:10, IGP metric:10, attribute flags:0x0
          frag_id 0, Intf Address:192.168.35.5
          TE metric:10, IGP metric:10, attribute flags:0x0
          frag_id 0, Intf Address:192.168.45.5
          TE metric:10, IGP metric:10, attribute flags:0x0
          frag_id 0, Intf Address:192.168.25.5
          TE metric:10, IGP metric:10, attribute flags:0x0

Tunnels use the following default affinity and mask:

PE1#show mpls traffic-eng tunnels Tunnel 1 | include Affinity
    Bandwidth: 750      kbps (Global)  Priority: 7  7   Affinity: 0x0/0xFFFF

The default affinity is 0x0 with a mask of 0xFFFF. Let me show all 32 bits:

What does this mean?

• The first 16 bits of the mask are set to zeroes, so we don’t care about these.
• The last 16 bits of the mask are set to ones, so we care about these.

The last 16 affinity bits are set to zeroes, so we can use links where the last 16 attribute flag bits are set to zero. Because the attribute flag of a link is set to all zeroes by default, we can use all links.

Link Attributes

Let’s start with the configuration of the link attributes. I’ll use the examples I showed you before:

Let’s configure all interfaces on the routers:

PE1

PE1(config)#interface GigabitEthernet 0/1
PE1(config-if)#mpls traffic-eng attribute-flags 0x19

PE1(config)#interface GigabitEthernet 0/2        
PE1(config-if)#mpls traffic-eng attribute-flags 0x1

PE1(config)#interface GigabitEthernet 0/3
PE1(config-if)#mpls traffic-eng attribute-flags 0x4

P1

P1(config)#interface GigabitEthernet 0/0
P1(config-if)#mpls traffic-eng attribute-flags 0x19

P1(config)#interface GigabitEthernet 0/1
P1(config-if)#mpls traffic-eng attribute-flags 0x1

P2

P2(config)#interface GigabitEthernet 0/0
P2(config-if)#mpls traffic-eng attribute-flags 0x1

P2(config)#interface GigabitEthernet 0/1
P2(config-if)#mpls traffic-eng attribute-flags 0x2

PE2

PE2(config)#interface GigabitEthernet 0/1
PE2(config-if)#mpls traffic-eng attribute-flags 0x1

PE2(config)#interface GigabitEthernet 0/2       
PE2(config-if)#mpls traffic-eng attribute-flags 0x2

PE2(config)#interface GigabitEthernet 0/3       
PE2(config-if)#mpls traffic-eng attribute-flags 0x4

That’s all there is to it.

Affinity and Mask

Now we can configure affinity and a mask for our tunnel interfaces. I’ll give you four examples.

Example 1

Let’s say that our tunnel should only use plain DSL links. No other attributes are allowed. The attribute flag we use for a DSL link looks like this:

We can use our affinity and mask like this:

We set our affinity to the same value as the attribute flag for a DSL link. We set the mask to all ones, which means that all affinity bits have to match. Let’s configure this:

PE1(config)#interface Tunnel 1
PE1(config-if)#tunnel mpls traffic-eng affinity 0x4 mask 0xFFFFFFFF

Verification

Let’s figure out whether this works. Let’s check the tunnel:

PE1#show mpls traffic-eng Tunnels tunnel 1

Name: PE1_t1                              (Tunnel1) Destination: 5.5.5.5
  Status:
    Admin: up         Oper: up     Path: valid       Signalling: connected
    path option 1, type dynamic (Basis for Setup, path weight 10)

  Config Parameters:
    Bandwidth: 750      kbps (Global)  Priority: 7  7   Affinity: 0x4/0xFFFFFFFF
    Metric Type: TE (default)
    AutoRoute:  disabled  LockDown: disabled  Loadshare: 750      bw-based
    auto-bw: disabled
  Active Path Option Parameters:
    State: dynamic path option 1 is active
    BandwidthOverride: disabled  LockDown: disabled  Verbatim: disabled


  InLabel  :  - 
  OutLabel : GigabitEthernet0/3, implicit-null
  RSVP Signalling Info:
       Src 2.2.2.2, Dst 5.5.5.5, Tun_Id 1, Tun_Instance 107
    RSVP Path Info:
      My Address: 192.168.25.2   
      Explicit Route: 192.168.25.5 5.5.5.5 
      Record   Route:   NONE
      Tspec: ave rate=750 kbits, burst=1000 bytes, peak rate=750 kbits
    RSVP Resv Info:
      Record   Route:   NONE
      Fspec: ave rate=750 kbits, burst=1000 bytes, peak rate=750 kbits
  Shortest Unconstrained Path Info:
    Path Weight: 10 (TE)
    Explicit Route: 192.168.25.2 192.168.25.5 5.5.5.5

In the output above, we see our configured affinity and mask. The explicit route we use is the direct DSL link from PE1 to PE2. To find out whether affinity works, we’ll shut the DSL link:

PE1(config)#interface GigabitEthernet 0/3
PE1(config-if)#shutdown

Let’s check the tunnel:

PE1#show mpls traffic-eng tunnels Tunnel 1                             

Name: PE1_t1                              (Tunnel1) Destination: 5.5.5.5
  Status:
    Admin: up         Oper: down   Path: not valid   Signalling: Down
    path option 1, type dynamic

  Config Parameters:
    Bandwidth: 750      kbps (Global)  Priority: 7  7   Affinity: 0x4/0xFFFFFFFF
    Metric Type: TE (default)
    AutoRoute:  disabled  LockDown: disabled  Loadshare: 750      bw-based
    auto-bw: disabled

  Shortest Unconstrained Path Info:
    Path Weight: 20 (TE)
    Explicit Route: 192.168.24.2 192.168.24.4 192.168.45.4 192.168.45.5 
                    5.5.5.5 
  History:
    Tunnel:
      Time since created: 1 days, 47 minutes
      Time since path change: 2 hours, 3 minutes
      Number of LSP IDs (Tun_Instances) used: 469
    Prior LSP:
      ID: path option 1 [215]
      Removal Trigger: path error
      Last Error: PCALC:: No path to destination, 0005.0005.0005.00

As expected, the tunnel is now down because no DSL links remain.

Let’s reactivate the interface before we continue:

PE1(config)#interface GigabitEthernet 0/3 
PE1(config-if)#no shutdown

Example 2

Let’s try something else. Let’s configure a tunnel where we only permit fiber links. We shouldn’t use satellite or DSL links. We don’t care about any other attributes.

Here is the affinity and mask we can use:

Let me explain this:

• The affinity has it’s last bit set to 1 because this matches the fiber attribute flag bit.
• The mask is set to 111 because we care about the last three bits:
  • The second least significant bit indicates a satellite link, so we care that this bit is a 0.
  • The third least significant bit indicates a DSL link, so we care that this bit is a 0.
  • The least significant bit indicates a fiber link, so we care that this bit is a 1.

We don’t care about any of the other attribute flag bits. Let’s configure this: