Answers: MLPPP 2
This lab asks you to do a basic migration: move from using two parallel serial links as independent links to instead treating them as one layer 3 link using MLPPP. This lab requires new configuration as well as analysis of the existing configuration. Do the lab on your own first, and then check here for my answer and explanation.
Answers
Figure 1: MLPPP Topology – Before Migrating to MLPPP
Example 3: Router R1 Config
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 |
interface serial 0/0/0 encapsulation ppp ppp multilink ppp multilink group 2 no ip address ! interface serial 0/0/1 encapsulation ppp ppp multilink ppp multilink group 2 no ip address ! interface multilink 2 encapsulation ppp ppp multilink ppp multilink group 2 ip address 172.20.12.1 255.255.255.0 ! router ospf 1 network 172.20.12.1 0.0.0.0 area 0 no network 172.20.3.1 0.0.0.0 area 0 no network 172.20.4.1 0.0.0.0 area 0 |
Example 4: Router R2 Config
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 |
interface serial 0/0/0 encapsulation ppp ppp multilink ppp multilink group 2 no ip address ! interface serial 0/0/1 encapsulation ppp ppp multilink ppp multilink group 2 no ip address ! interface multilink 2 encapsulation ppp ppp multilink ppp multilink group 2 ip address 172.20.12.2 255.255.255.0 ! router ospf 1 network 172.20.12.2 0.0.0.0 area 0 no network 172.20.3.2 0.0.0.0 area 0 no network 172.20.4.2 0.0.0.0 area 0 |
Commentary
This lab has a rather long configuration answer. This commentary starts by working through the MLPPP configuration, and then discusses some of the migration choices, particularly to do with OSPF and IP addressing.
The requirements asked you to use MLPPP multilink group 2. As a result:
- Both routers create a multilink interface with the interface multilink 2 command
- Both routers use the ppp multilink group 2 subcommand on their physical serial interfaces and on the multilink interface itself
For the layer 3 details, note that the configuration on each router:
- Adds an IPv4 address to each multilink interface, for example, with the ip address 172.20.12.1 255.255.255.0 subcommand under multilink 2 on router R1.
- Both routers remove the IP address configuration from the physical interfaces, because they are no longer needed, with the no ip address interface subcommand.
Next, as normal, the physical serial interfaces plus the multilink interface, on each router, need the same three subcommands. These commands (respectively) enable PPP, enable the MLPPP feature of PPP, and reference the MLPPP multilink group:
encapsulation ppp
ppp multilink
ppp multilink group 2
Finally, note that the OSPF configuration needed to be updated. The initial configuration happened to show OSPF configuration that matched based on all 32 bits of each interface IP address. The MLPPP configuration in this case moved away from using the addresses in subnets 172.20.3.0/24 and 172.20.4.0/24, now using subnet 172.20.12.0/24. The end of Examples 3 and 4 show the addition of one OSPF network command on each router, matching each router’s new IP address in subnet 172.20.12.0/24. It also shows the removal of the two now unneeded network commands.
Thanks Wendell, your posts have been really helpful. However some WAN configurations are not working on PT 7.0. e.g. Int Multilink 2 doesn’t work. Using notepad has been my gateway of last resort
Ha! Made me laugh! 🙂 Good design, always a fallback – notepad. 🙂
What happens with the clock rate command? I dont see it in any side of the link.
Actually, the clock rate command is supplied automatically when the interface has a DCE cable connected, at least by IOS at the later releases. Frankly I ignored it, probably because I didn’t have to code it. I don’t recall what IOS – maybe 15.2M or so?
Anyway, for this lab, there was no implication that the routers were in a lab with a short cable, rather than a real serial link, so it seems ambiguous for the lab whether a clock rate command was required. Either way… if you expect it to be in a lab, with a back-to-back serial cable, make sure there’s a clock rate command on the DCE side.
Thanks for the note…
If using a simulator instead of real hardware try GNS3 it runs real Cisco IOS firmware images in a VM and thus supports things like PPP Multilink.
Wendell
Going through your book now so sorry its like 2 years after your post.
:/
My question is there really a need for the “ppp multilink” command when the very next command is “ppp multilink group 2” ?
In other words, wouldn’t typing the following in the ABOVE section accomplish the same thing as typing what is in the BELOW section since PPP multilink is already in the command “ppp multilink group 2” ?
ABOVE
interface multilink 2
encapsulation ppp
ppp multilink group 2
ip address 172.20.12.2 255.255.255.0
BELOW
interface multilink 2
encapsulation ppp
ppp multilink
ppp multilink group 2
ip address 172.20.12.2 255.255.255.0
As always, love your blog posts!
Hi Jonathan,
Thanks – glad you like the posts!
In answer, without testing, my recollection is “no”. the ppp multilink command (with no other parameters is required. The ppp multilink group 2 does not serve as a dual purpose of both enabling the feature and specifying the group, nor does it cause IOS to generate the ppp multilink command. But I’m running from memory here. I’ll try and test in lab sometime soon.
Wendell