CISCO IOS XR

Jeffrey Fry October 2012 FryGuy.Net

Table of Contents
1. Cisco IOS XR Introduction and Comparison to IOS 2. Cisco IOS XR Prompt and Hostname Differences 3. Basic Configuration Options 4. Configuring an Interface - Basic IPv4 and IPv6 address 5. Bundled Interfaces 6. Software Installation and PIE packages 7. Licensing 8. Aliases 9. Wildcard Masks 10.Processes 11.Remote Access Services Telnet and SSH 12.TACACS Configuration ( default and non-default VRF) 13.Access Lists 14.OSPF 15.EIGRP 16.RIP 17.IS-IS 18.BGP - iBGP and eBGP 19.Route Filtering 20.VRF lite and Dot1Q Trunks 21.Basic MPLS - LDP 22.MPLS VPN 23.L2VPN 24.NHRP (HSRP/VRRP)

Cisco IOS XR Training 2012 Jeffrey Fry

Page 2 FryGuy.Net

There are two topologies that have been used. This is because of what I had access to changed over time. For this lab, here is a topology diagram of what was used: ASR9000 and Cisco 2811
CE1 / R3
S L O T 3 S L O T 2 A S L O T 1 F S S L O L T 0 PVDM1 PVDM2 AIM1 AIM0
R

CE2 / R4
A= ACT S= SPEED FE 0/1 F= FDX L= LINK FE 0/0 A F S L NME0 S L O T 1

Cisco 2811

S L O T 3

S L O T 2 A F S S L O L T 0 PVDM1

A= ACT S= SPEED FE 0/1

F= FDX L= LINK FE 0/0 A F S L

Cisco 2811

NME0

PVDM2

AIM1

AIM0

F0/0

F0/0

Cisco ASR 9000 Series

Cisco ASR 9000 Series

1 1

13

1 5

1 7

19

A9K- 2T- 20GE

A /L 0

A /L 1 0 2 4 6 8 1 0 12 1 4 1 6 18

L L

A /L 0

A /L 1 0 2 4 6 8 1 0 12 1 4 1 6 18

L L

1 1

13

1 5

1 7

19

A9K- 2T- 20GE

A /L 0

A /L 1 0 2 4 6 8 1 0 12 1 4 1 6 18

L L

A /L 0

A /L 1 0 2 4 6 8 1 0 12 1 4 1 6 18

L L

E SB

E SB

A9K- 2T- 20GE

2 0

G0/0/0/11

G0/0/0/11

1 1

13

1 5

1 7

19

PE1 / R1

PE2 / R2

Cisco 12000
R1
Cisco 12000 SERIES

R2
Cisco 12000 SERIES

CSCO

CSC1

7 0 1 2 3 CSCO CSC1 4 5 6 7

LINK LINK
SLOT 0 SLOT 1 SLOT 0 SLOT 1 RESET RESET SLOT 0 SLOT 1 SLOT 0 SLOT 1

LINK 3

G0/3/0/2

G0/5/0/2
AUX CONSOLE

RESET

RESET

LINK 3

LINK 5

LINK 5

LINK

COLL

LINK

COLL

TX RX

TX RX

G0/3/0/3

G0/5/0/3

LINK

COLL

LINK

COLL

TX RX

TX RX

MII

RJ45

MII

RJ45 MII RJ45 MII RJ45

7 7

LINK

8888 8888

8888 8888

8888 8888

LINK

8888 8888

8888 8888

8888 8888

Cisco IOS XR Training 2012 Jeffrey Fry

A9K- 2T- 20GE

G0/1/0/11

G0/1/0/11

1 1

13

1 5

1 7

19

FAIL

FAIL

MAJ

MAJ

CRIT

CRIT

MIN

MIN

MGT LAN 0

MGT LAN 1

CONSOLE

AUX

SYNC 0

SYNC 1

ALARM OUT
ACO CF HDD SYNC

MGT LAN 0

MGT LAN 1

CONSOLE

AUX

SYNC 0

SYNC 1

ALARM OUT
ACO CF HDD SYNC

A9K-RSP-4G
1

A9K-RSP-4G

LAMP TEST

LAMP TEST

ACO 2 0 1

ACO

AUX

AUX CONSOLE ROUTE PROCESSOR

AUX

CONSOLE ROUTE PROCESSOR

CONSOLE

8 FAST ETHERNET FX

ROUTE PROCESSOR

ROUTE PROCESSOR

8 FAST ETHERNET FX

Page 3 FryGuy.Net

1. Cisco IOS XR Introduction and Comparison to IOS

Lets start with the basic difference between Cisco IOS and Cisco IOS XR code, the Operating System. In Cisco IOS, the kernel is monolithic, meaning everything in installed in a single image and all processes share the same address space. There is no memory protection between processes, so if one crashes it can impact all other processes on the box thus forcing or causing a reload of the entire router. The other thing with monolithic code is that it has a run to completion scheduler, so the kernel will not preempt a process that is running; the process must make a kernel call before another process has a chance to execute. In Cisco IOS XR, the kernel is based on an OS called QNX Neutrino that runs some very powerful and reliable systems. QNX runs per their News Release at http://www.qnx.com/news/pr_1329_3.html - things from EKG machines, to Air Traffic Control systems, and among other things automated beer bottle inspection systems. IOS XR offers modularity and memory protection between processes, threads and supports preemptive scheduling as well as the ability to restart a failed process. Protocols like BGP, OSPF, OSPFv3, RIBv4, RIBv6, etc all run in separate spaces if one has a fault, it will not impact the others. Also, an added bonus, if you run multiple routing protocol instances (like OSPF), each process will run in its own memory space this is an important feature of Service Providers any fault with one customer process will not impact another. Another big difference between IOS and IOS XR is the configuration model. IOS is a single stage model meaning that as soon as you make a change, it is applied to the active running config. With IOS XR, you have a running (active) config that you cannot modify directly, all your changes are made in a staging area first before being committed to the running config. After you make your changes, you commit them and promote the staging config to the active config. Before the change is made active, the IOS XR will run a sanity check on it making sure that the commands are correct to a certain degree, if there is a problem it will tell you so that you can correct the error
% Failed to commit one or more configuration items. Please use 'show configuration failed' to view the errors

Cisco IOS XR Training 2012 Jeffrey Fry

Page 4 FryGuy.Net

Here is a table of some of the other significant differences between IOS and IOS XR IOS XR Config changes do not take place immediately Config changes must be COMMITted before taking effect You can check your configuration before applying it Two stage configuration Configuration Rollback IOS Configuration changes take place immediately No commit, changes immediate

No verification, immediate.

Single stage Not easy to do, has to be manually configured and not guaranteed Interface centric

Feature centric

Cisco IOS XR Training 2012 Jeffrey Fry

Page 5 FryGuy.Net

2. Cisco IOS XR Prompt and Hostname Differences

Lets cover the prompt real quick as that is a bit different than what people are used to. Lets look at the standard IOS prompt vs. the IOS XR prompt. IOS: Router# IOS-XR: RP/0/7/CPU0:ios# As you can see the prompt is a bit different. In standard IOS you have the hostname, but in IOS XR you get a bit more information. It breaks down as follows: Prompt Syntax: Type type of interface card (Usually RP for Route Processor) Rack What Rack number this is installed in in a multishelf system, typically 0 if standalone Slot Slot the RP is installed in (7 in this example) Module What execute the user commands or port interface. Usually CPU0 or CPU1 Name Hostname of the router, default here is IOS Ok, now lets change the hostname on typical IOS so you can see the difference. Going forward, BLUE text is prompts and router feedback, RED are commands entered. Router# Router#conf t *Mar 29 16:32:51.507: %SYS-5-CONFIG_I: Configured from console by console Enter configuration commands, one per line. End with CNTL/Z. Router(config)#hostname R1 R1(config)# As you can see, in IOS the hostname changed immediately after hitting Enter. So, lets change the hostname to R1 on IOS XR code: RP/0/7/CPU0:ios# RP/0/7/CPU0:ios#conf t Thu Mar 29 16:00:43.844 UTC RP/0/7/CPU0:ios(config)#hostname R1 RP/0/7/CPU0:ios(config)#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 6 FryGuy.Net

Notice that the hostname did not change? In IOS XR you need to COMMIT your changes in order for them to take effect. But before we commit them, lets do a show config quick RP/0/7/CPU0:ios(config)# RP/0/7/CPU0:ios(config)# sh config Thu Mar 29 16:03:53.060 UTC Building configuration... !! IOS XR Configuration 4.1.1 hostname R1 end RP/0/7/CPU0:ios(config)# Pretty cool, the router will show you the changes you are about to make, this is your staging config changes. Now we can COMMIT the changes RP/0/7/CPU0:ios(config)#commit Thu Mar 29 16:03:04.182 UTC RP/0/7/CPU0:R1(config)# See, once you entered COMMIT, the hostname change from IOS to R1.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 7 FryGuy.Net

3. Basic Configuration Options

Ok, we have seen the basic COMMIT option but what other options do we have for configuration mode? Well, we have a few to choose from. First, what if I am making changes and decide I dont want them? few options. First you could just exit all the way out. You have a

RP/0/7/CPU0:R1(config)#exit Uncommitted changes found, commit them before exiting(yes/no/cancel)? [cancel]: no And once you exit out, all your changes are lost. Ok, that is one option. Another is clear. To demonstrate we will create loopback 666: RP/0/7/CPU0:R1#conf t Sun Apr 1 22:18:52.956 UTC RP/0/7/CPU0:R1(config)#int loop666 RP/0/7/CPU0:R1(config-if)#ip add 6.6.6.6/32 Ok, lets check the candidate configuration: RP/0/7/CPU0:R1(config-if)#show config Sun Apr 1 22:19:03.438 UTC Building configuration... !! IOS XR Configuration 4.1.1 interface Loopback666 ipv4 address 6.6.6.6 255.255.255.255 ! end RP/0/7/CPU0:R1(config-if)# OK, we have it in the candidate configuration now. that so lets clear it. RP/0/7/CPU0:R1(config-if)#clear Now check the candidate configuration again. RP/0/7/CPU0:R1(config)#show config Sun Apr 1 22:19:34.733 UTC Building configuration... !! IOS XR Configuration 4.1.1 end RP/0/7/CPU0:R1(config)# There, all gone! Cisco IOS XR Training 2012 Jeffrey Fry Page 8 FryGuy.Net We changed our mind about

Now, what if we want to make a change but we want to be sure we dont lose connection to the router? Well, we can do a commit confirm, this way if we do lose connection our change will be rolled back! RP/0/7/CPU0:R1#conf t Sun Apr 1 22:23:01.154 UTC RP/0/7/CPU0:R1(config)#int loop 666 RP/0/7/CPU0:R1(config-if)#ip add 6.6.6.6/32 Now, lets look at our commit confirmed options: RP/0/7/CPU0:R1(config-if)#commit confirmed ? <30-65535> Seconds until rollback unless there is a confirming commit minutes Specify the rollback timer in the minutes <cr> Commit the configuration changes to running See, we can have a few seconds or a few minutes. RP/0/7/CPU0:R1(config-if)#commit confirmed 30 Sun Apr 1 22:23:19.344 UTC Pretty cool!

Now, lets see if we have loop666: RP/0/7/CPU0:R1(config-if)#do show int loop666 Sun Apr 1 22:23:34.353 UTC Loopback666 is up, line protocol is up Interface state transitions: 1 Hardware is Loopback interface(s) Internet address is 6.6.6.6/32 MTU 1500 bytes, BW 0 Kbit reliability Unknown, txload Unknown, rxload Unknown Encapsulation Loopback, loopback not set, Last input Unknown, output Unknown Last clearing of "show interface" counters Unknown RP/0/7/CPU0:R1(config-if)# Yup, its there. Now we can wait a few seconds (30 or so) and do the show interface command again. RP/0/7/CPU0:R1(config-if)#do show int loop666 Sun Apr 1 22:25:09.361 UTC Interface not found (Loopback666) RP/0/7/CPU0:R1(config-if)# All gone!

Cisco IOS XR Training 2012 Jeffrey Fry

Page 9 FryGuy.Net

Ok, now lets commit it this time. RP/0/7/CPU0:R1#conf t Sun Apr 1 22:26:20.749 UTC RP/0/7/CPU0:R1(config)#int loop666 RP/0/7/CPU0:R1(config-if)#ip add 6.6.6.6/32 RP/0/7/CPU0:R1(config-if)#commit confirmed 30 Sun Apr 1 22:26:32.913 UTC RP/0/7/CPU0:R1(config-if)# Lets see if the interface is there: RP/0/7/CPU0:R1(config-if)#do show int loop666 Sun Apr 1 22:26:38.421 UTC Loopback666 is up, line protocol is up Interface state transitions: 1 Hardware is Loopback interface(s) Internet address is 6.6.6.6/32 MTU 1500 bytes, BW 0 Kbit reliability Unknown, txload Unknown, rxload Unknown Encapsulation Loopback, loopback not set, Last input Unknown, output Unknown Last clearing of "show interface" counters Unknown Yup, now we can commit it again to make it stay. RP/0/7/CPU0:R1(config-if)#commit Sun Apr 1 22:26:40.299 UTC % Confirming commit for trial session. RP/0/7/CPU0:R1(config-if)# And lets make sure it is still there. RP/0/7/CPU0:R1#sh int loop 666 Sun Apr 1 22:27:09.232 UTC Loopback666 is up, line protocol is up Interface state transitions: 1 Hardware is Loopback interface(s) Internet address is 6.6.6.6/32 MTU 1500 bytes, BW 0 Kbit reliability Unknown, txload Unknown, rxload Unknown Encapsulation Loopback, loopback not set, Last input Unknown, output Unknown Last clearing of "show interface" counters Unknown

RP/0/7/CPU0:R1# Look at that, IOS XR has a commit confirmed just like someone else does as well.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 10 FryGuy.Net

Few other things that is nice to know. You can configure the system in exclusive mode, this way only you can be making changes and nobody else. To do this, just enter configure exclusive RP/0/7/CPU0:R1#configure exclusive You can add comments and notations to your commit that will show up in the rollback. RP/0/7/CPU0:R1#conf t Sun Apr 1 22:32:23.941 UTC RP/0/7/CPU0:R1(config)#int loop 667 RP/0/7/CPU0:R1(config-if)#ip add 6.6.6.7/32 RP/0/7/CPU0:R1(config-if)#exit RP/0/7/CPU0:R1(config)#commit comment Created Loopback 667 For Testing Sun Apr 1 22:33:34.589 UTC RP/0/7/CPU0:R1(config)# Now, if a comment has been added, you can see it via the show configuration history last x detail command RP/0/7/CPU0:R1#sh configuration history last 1 detail Sun Apr 1 22:36:04.053 UTC 1) Event: commit Time: Sun Apr 1 22:33:36 2012 Commit ID: 1000000230 Label: User: user Line: con0_7_CPU0 Client: CLI Comment: Created Loopback 667 For Testing RP/0/7/CPU0:R1#

Ok, lets quickly look at loading a configuration from the disk and overwriting an existing configuration. I have copied a config to disk0a: called newconfig.txt. What I want to do is install this configuration as the running config on the router. 1626 -rwx 204 Wed Oct 17 01:21:30 2012 newconfig.txt

So to start, lets delete the existing configuration RP/0/RSP0/CPU0:R1(config)#commit replace Wed Oct 17 01:21:43.406 UTC This commit will replace or remove the entire running configuration. This operation can be service affecting. Do you wish to proceed? [no]: y RP/0/RSP0/CPU0:ios(config)# RP/0/RSP0/CPU0:ios(config)#exit

Cisco IOS XR Training 2012 Jeffrey Fry

Page 11 FryGuy.Net

Ok, so now we are at an unconfigured device. the disk to the running config.

Now we can load the config on

RP/0/RSP0/CPU0:ios(config)#load disk0a:/newconfig.txt Loading. 204 bytes parsed in 1 sec (203)bytes/sec The configuration is now loaded into the candidate config. is there and then commit it. RP/0/RSP0/CPU0:ios(config)#show confi Wed Oct 17 01:26:17.539 UTC Building configuration... !! IOS XR Configuration 4.1.2 hostname R1 domain name lab.cfg interface Loopback100 ipv4 address 100.100.100.100 255.255.255.255 ! end RP/0/RSP0/CPU0:ios(config)#commit Wed Oct 17 01:26:22.174 UTC RP/0/RSP0/CPU0:R1(config)# There, we have loaded the config and applied the changes. I have loaded another file to the router called ReplaceConfig.txt. This is a new configuration for the router, one that we want to replace the existing config with. RP/0/RSP0/CPU0:R1#conf t Wed Oct 17 01:37:23.638 UTC RP/0/RSP0/CPU0:R1(config)#load disk0a:/ReplaceConfig.txt Loading. 283 bytes parsed in 1 sec (282)bytes/sec RP/0/RSP0/CPU0:R1(config)#show config Wed Oct 17 01:37:38.571 UTC Building configuration... !! IOS XR Configuration 4.1.2 hostname Router1 domain name NewLab.CFG interface Loopback100 ipv4 address 101.101.101.101 255.255.255.255 ! interface TenGigE0/0/0/0 ipv4 address 200.200.200.202 255.255.255.0 ! end Let us check what

Cisco IOS XR Training 2012 Jeffrey Fry

Page 12 FryGuy.Net

RP/0/RSP0/CPU0:R1(config)#commit replace Wed Oct 17 01:37:41.577 UTC This commit will replace or remove the entire running configuration. This operation can be service affecting. Do you wish to proceed? [no]: y RP/0/RSP0/CPU0:Router1(config)# What other options to loaf configuration are there? Well, here is a list: RP/0/RSP0/CPU0:Router1(config)#load ? WORD Load from file bootflash: Load from bootflash: file system commit Load commit changes compactflash: Load from compactflash: file system compactflasha: Load from compactflasha: file system configuration Contents of configuration diff Load from diff file disk0: Load from disk0: file system disk0a: Load from disk0a: file system disk1: Load from disk1: file system disk1a: Load from disk1a: file system ftp: Load from ftp: file system harddisk: Load from harddisk: file system harddiska: Load from harddiska: file system harddiskb: Load from harddiskb: file system lcdisk0: Load from lcdisk0: file system lcdisk0a: Load from lcdisk0a: file system nvram: Load from nvram: file system rcp: Load from rcp: file system rollback Load rollback changes tftp: Load from tftp: file system RP/0/RSP0/CPU0:Router1(config)# You can load from the local disk, RCP, TFTP, FTP, etc if you want.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 13 FryGuy.Net

4. Configuring an interface Basic IPv4 and IPv6 address

First we will take a look at what interfaces we have and review them quickly. We can use the same IOS command we are already familiar with show ip interface brief RP/0/7/CPU0:R1# RP/0/7/CPU0:R1#sh ip int br Thu Mar 29 18:12:04.883 UTC Interface MgmtEth0/7/CPU0/0 MgmtEth0/7/CPU0/1 MgmtEth0/7/CPU0/2 GigabitEthernet0/3/0/0 GigabitEthernet0/3/0/1 GigabitEthernet0/3/0/2 GigabitEthernet0/3/0/3 MgmtEth0/6/CPU0/0 MgmtEth0/6/CPU0/1 MgmtEth0/6/CPU0/2 RP/0/7/CPU0:R1# IP-Address unassigned unassigned unassigned unassigned unassigned unassigned unassigned unassigned unassigned unassigned Status Shutdown Shutdown Shutdown Down Down Up Up Shutdown Shutdown Shutdown Protocol Down Down Down Down Down Up Up Down Down Down

Here you can see that we have an RP in Slot 6 and 7 (Mgmt) and a 4-port Gig card in Slot 3. For this lab, interfaces G0/3/0/2 and G0/3/0/3 are precabled to another router and are currently UP/UP right now. Let configure an IP address on G0/3/0/2 of 150.1.12.1 with a mask of 255.255.255.0 First, lets look at the running config on the interface now: RP/0/7/CPU0:R1# RP/0/7/CPU0:R1#sh run int g0/3/0/2 Thu Mar 29 18:38:29.942 UTC % No such configuration item(s) RP/0/7/CPU0:R1# As you can see, it says No such config, it is telling you that it is unconfigured. RP/0/7/CPU0:R1#conf t Thu Mar 29 18:38:31.891 UTC RP/0/7/CPU0:R1(config)#int g0/3/0/2 RP/0/7/CPU0:R1(config-if)#ip add 150.1.12.1/24 Cisco IOS XR Training 2012 Jeffrey Fry Page 14 FryGuy.Net

Notice, on IOS XR you can use / for the subnet, no more entering 255.255.255.0 : RP/0/7/CPU0:R1(config-if)#show config Thu Mar 29 18:38:44.248 UTC Building configuration... !! IOS XR Configuration 4.1.1 interface GigabitEthernet0/3/0/2 ipv4 address 150.1.12.1 255.255.255.0 ! end RP/0/7/CPU0:R1(config-if)# Another cool thing with IOS-XR is you can find out where you are any time you want just by entering PWD RP/0/7/CPU0:R1(config-if)#pwd Thu Mar 29 19:31:24.666 UTC interface GigabitEthernet0/3/0/2 RP/0/7/CPU0:R1(config-if)# RP/0/7/CPU0:R1(config-if)#comm Thu Mar 29 18:38:46.216 UTC RP/0/7/CPU0:R1(config-if)# Now, lets check the running config on that interface again: RP/0/7/CPU0:R1#sh run int g0/3/0/2 Thu Mar 29 18:42:43.763 UTC interface GigabitEthernet0/3/0/2 ipv4 address 150.1.12.1 255.255.255.0 ! RP/0/7/CPU0:R1# Lets PING our neighbor now 150.1.12.2 RP/0/7/CPU0:R1#ping 150.1.12.2 Thu Mar 29 18:44:39.570 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 150.1.12.2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 3/8/12 ms RP/0/7/CPU0:R1#

Now, lets configure a loopback for R1 of 1.1.1.1/32 RP/0/7/CPU0:R1#conf t Thu Mar 29 19:25:19.486 UTC RP/0/7/CPU0:R1(config)#int l0 RP/0/7/CPU0:R1(config-if)#ip add 1.1.1.1/32 RP/0/7/CPU0:R1(config-if)#exit RP/0/7/CPU0:R1(config)#exit Cisco IOS XR Training 2012 Jeffrey Fry Page 15 FryGuy.Net

Uncommitted changes found, commit them before exiting(yes/no/cancel)? [cancel]:yes RP/0/7/CPU0:R1# Notice this time I did not commit the change, but the system knew I was making changes and asked me if I wanted to commit them. I simply responded with YES and it saved them for me. If I did not want to save them, I could have entered NO and all the changes would have been tossed out. If I would have selected CANCEL, I would go back into edit mode. Now time to configure some IPv6 addresses first 2001:1:1:12::1/64 RP/0/7/CPU0:R1#conf t Thu Mar 29 19:26:21.184 UTC RP/0/7/CPU0:R1(cconfig)#int g0/3/0/2 RP/0/7/CPU0:R1(config-if)#ipv6 address 2001:1:1:12::1/64 RP/0/7/CPU0:R1(config-if)#exit RP/0/7/CPU0:R1(config)#commit Thu Mar 29 19:26:39.769 UTC RP/0/7/CPU0:R1(config)#exit And now we can try to PING our neighbor at 2001:1:1:12::2 RP/0/7/CPU0:R1#ping 2001:1:1:12::2 Thu Mar 29 19:29:11.893 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2001:1:1:12::2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 2/16/68 ms RP/0/7/CPU0:R1# Lets add one under our loopback interface as well well use 2001::1/128 RP/0/7/CPU0:R1#conf t RP/0/7/CPU0:R1(config)#int l0 RP/0/7/CPU0:R1(config-if)#ipv6 add 2001::1/128 RP/0/7/CPU0:R1(config-if)#commit Thu Mar 29 19:30:49.920 UTC RP/0/7/CPU0:R1(config-if)#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 16 FryGuy.Net

5. Interface Bundles
Etherchannels are also different between IOS and IOS XR. they would be configured as such: interface port-channel 1 IP add 10.1.1.1 255.255.255.0 interface FastEthernet0/0 channel-group 1 interface FastEthernet0/1 channel-group 1 In typical IOS,

IOS XE is a little different then IOS as you can choose LACP: interface GigabitEthernet0/0/2 channel-group 12 mode active no shut interface GigabitEthernet0/0/3 channel-group 12 mode active no shut interface Port-channel12 ip address 10.1.1.1 255.255.255.252 And with IOS XR, it is a bit different again. configure Ethernet Bundle 200 First on PE2: RP/0/RSP0/CPU0:PE2#conf t First up though, lets reset the interfaces back to factory by using the no interface command: RP/0/RSP0/CPU0:PE2(config)#no int g0/0/0/11 RP/0/RSP0/CPU0:PE2(config)#commit Instead of a port-channel interface, we do a bundle-ether interface RP/0/RSP0/CPU0:PE2(config)#int bundle-ether 200 RP/0/RSP0/CPU0:PE2(config-if)#ip add 150.1.12.2 255.255.255.0 Now lets look at our bundle options: RP/0/RSP0/CPU0:PE2(config-if)#bundle ? load-balancing Load balancing commands on a bundle maximum-active Set a limit on the number of links that can be active minimum-active Set the minimum criteria for the bundle to be active shutdown Bring all links in the bundle down to Standby state wait-while Set the wait-while timeout for members of this bundle So, for this example we will

Cisco IOS XR Training 2012 Jeffrey Fry

Page 17 FryGuy.Net

Ok, since this is a bundle, we should put restrictions around the max and min links. Normally this is not a problem, but if you had to guarantee bandwidth (say 4G, then you might consider having the min links set to 4, and if you dropped below 4 the interface would go down). RP/0/RSP0/CPU0:PE2(config-if)#bundle maximum-active links 2 RP/0/RSP0/CPU0:PE2(config-if)#bundle minimum-active links 1 Now lets take a quick look at our l oad balancing hash options: RP/0/RSP0/CPU0:PE2(config-if)#bundle load-balancing hash ? dst-ip Use the destination IP as the hash function src-ip Use the source IP as the hash function So, for this example we will use the src-ip RP/0/RSP0/CPU0:PE2(config-if)#bundle load-balancing hash src-ip Now, lets assign the interfaces to the bundle RP/0/RSP0/CPU0:PE2(config-if)#int g0/1/0/11 Just like port-channels, the bundle ID should match the interface number you created. But here we will also look at what bundle options we have: RP/0/RSP0/CPU0:PE2(config-if)#bundle id 200 mode ? active Run LACP in active mode over the port. on Do not run LACP over the port. passive Run LACP in passive mode over the port. There are three ways that Switch 1 Switch 2 Active Active Active Passive On On LACP will link aggregate: Notes This is the recommended configuration. Link will aggregate once negotiation is done Aggregation will happen, but not reccomded

We will use LACP in ACTIVE mode as that is what is recommended by Cisco: RP/0/RSP0/CPU0:PE2(config-if)#bundle id 200 mode active RP/0/RSP0/CPU0:PE2(config-if)#no shut And do the same for G0/0/0/11: RP/0/RSP0/CPU0:PE2(config-if)#int g0/0/0/11 RP/0/RSP0/CPU0:PE2(config-if)#bundle id 200 mode ac RP/0/RSP0/CPU0:PE2(config-if)#no shut

Cisco IOS XR Training 2012 Jeffrey Fry

Page 18 FryGuy.Net

Now lets check our config before we commit: RP/0/RSP0/CPU0:PE2(config-if)#show config Fri Apr 27 01:46:28.451 UTC Building configuration... !! IOS XR Configuration 4.1.2 interface Bundle-Ether200 ipv4 address 150.1.12.2 255.255.255.0 bundle load-balancing hash src-ip bundle maximum-active links 2 bundle minimum-active links 1 ! interface GigabitEthernet0/0/0/11 bundle id 200 mode active no shutdown ! interface GigabitEthernet0/1/0/11 bundle id 200 mode active no shutdown ! end RP/0/RSP0/CPU0:PE2(config)#commit Fri Apr 27 01:46:44.692 UTC Now we can do the other Router, PE1 RP/0/RSP0/CPU0:PE1(config)#no int g0/0/0/11 RP/0/RSP0/CPU0:PE1(config)#commit Fri Apr 27 01:49:05.892 UTC RP/0/RSP0/CPU0:PE1(config)#int bundle-ether 200 RP/0/RSP0/CPU0:PE1(config-if)#ip add 150.1.12.1/24 RP/0/RSP0/CPU0:PE1(config-if)#bundle maximum-active links 2 RP/0/RSP0/CPU0:PE1(config-if)#bundle minimum-active links 1 RP/0/RSP0/CPU0:PE1(config-if)#bundle load-balancing hash src-ip RP/0/RSP0/CPU0:PE1(config-if)#int g0/1/0/11 RP/0/RSP0/CPU0:PE1(config-if)#bundle id 200 mode act RP/0/RSP0/CPU0:PE1(config-if)#no shut RP/0/RSP0/CPU0:PE1(config-if)#int g0/0/0/11 RP/0/RSP0/CPU0:PE1(config-if)#bundle id 200 mode act RP/0/RSP0/CPU0:PE1(config-if)#no shut RP/0/RSP0/CPU0:PE1(config-if)#exit RP/0/RSP0/CPU0:PE1(config)#show config Fri Apr 27 01:50:34.351 UTC Building configuration... !! IOS XR Configuration 4.1.2 interface Bundle-Ether200 ipv4 address 150.1.12.1 255.255.255.0 bundle load-balancing hash src-ip bundle maximum-active links 2 bundle minimum-active links 1 ! Cisco IOS XR Training 2012 Jeffrey Fry Page 19 FryGuy.Net

interface GigabitEthernet0/0/0/11 bundle id 200 mode active no shutdown ! interface GigabitEthernet0/1/0/11 bundle id 200 mode active no shutdown ! end RP/0/RSP0/CPU0:PE1(config)#commit Fri Apr 27 01:50:37.705 UTC RP/0/RSP0/CPU0:PE1(config)#

Now, lets look at our bundle interface: RP/0/RSP0/CPU0:PE1#sh int bundle-eth 200 Fri Apr 27 01:51:04.668 UTC Bundle-Ether200 is up, line protocol is up Interface state transitions: 1 Hardware is Aggregated Ethernet interface(s), address is 6c9c.ed2d.0bab Internet address is 150.1.12.1/24 MTU 1514 bytes, BW 2000000 Kbit (Max: 2000000 Kbit) reliability 255/255, txload 0/255, rxload 0/255 Encapsulation ARPA, Full-duplex, 2000Mb/s loopback not set, ARP type ARPA, ARP timeout 04:00:00 No. of members in this bundle: 2 GigabitEthernet0/0/0/11 Full-duplex 1000Mb/s Active GigabitEthernet0/1/0/11 Full-duplex 1000Mb/s Active Last input 00:00:18, output 00:00:18 Last clearing of "show interface" counters never 5 minute input rate 0 bits/sec, 0 packets/sec 5 minute output rate 0 bits/sec, 0 packets/sec 15 packets input, 1792 bytes, 50 total input drops 0 drops for unrecognized upper-level protocol Received 2 broadcast packets, 13 multicast packets 0 runts, 0 giants, 0 throttles, 0 parity 0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort 12 packets output, 1408 bytes, 0 total output drops As we can see, we are UP and have a full-duplex bandwidth of 2Gs.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 20 FryGuy.Net

So, lets PING! RP/0/RSP0/CPU0:PE1#ping 150.1.12.2 Fri Apr 27 01:51:12.692 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 150.1.12.2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/4/16 ms RP/0/RSP0/CPU0:PE1# Cool! The bundle is working.

Now we can check out some of the details:

RP/0/RSP0/CPU0:PE1#sh bundle bundle-ether 200 Fri Apr 27 02:13:16.767 UTC Bundle-Ether200 Status: Local links <active/standby/configured>: Local bandwidth <effective/available>: MAC address (source): Minimum active links / bandwidth: Maximum active links: Wait while timer: Load balancing: Link order signaling: Hash type: LACP: Flap suppression timer: Cisco extensions: mLACP: IPv4 BFD: Port -------------------Gi0/0/0/11 Link is Active Gi0/1/0/11 Link is Active RP/0/RSP0/CPU0:PE1# Device --------------Local Local

Up 2 / 0 / 2 2000000 (2000000) kbps 6c9c.ed2d.0bab (Chassis pool) 1 / 1 kbps 2 2000 ms Not configured Src-IP Operational Off Disabled Not configured Not configured Port ID -------------0x8000, 0x0002 0x8000, 0x0001 B/W, kbps ---------1000000 1000000

State ----------Active Active

Cisco IOS XR Training 2012 Jeffrey Fry

Page 21 FryGuy.Net

Now we can look at LACP: RP/0/RSP0/CPU0:PE1#show lacp Fri Apr 27 01:52:35.115 UTC State: a - Port is marked as Aggregatable. s - Port is Synchronized with peer. c - Port is marked as Collecting. d - Port is marked as Distributing. A - Device is in Active mode. F - Device requests PDUs from the peer at fast rate. D - Port is using default values for partner information. E - Information about partner has expired.
Bundle-Ether200 Port (rate) -------------------Local Gi0/0/0/11 30s Partner 30s Gi0/1/0/11 30s Partner 30s Port -------------------Local Gi0/0/0/11 Gi0/1/0/11 State Port ID Key System ID -------- ------------- ------ -----------------------ascdA--ascdA--ascdA--ascdA--0x8000,0x0002 0x8000,0x0003 0x8000,0x0001 0x8000,0x0004 0x00c8 0x00c8 0x00c8 0x00c8 0x8000,6c-9c-ed-2d-0b-ac 0x8000,6c-9c-ed-2d-1f-cc 0x8000,6c-9c-ed-2d-0b-ac 0x8000,6c-9c-ed-2d-1f-cc

Receive Period Selection Mux A Churn P Churn ---------- ------ ---------- --------- ------- ------Current Current Slow Slow Selected Selected Distrib Distrib None None None None

RP/0/RSP0/CPU0:PE1#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 22 FryGuy.Net

6. Software installation, PIE packages, and patches

As part of any system, from time to time you need to install updates, patches, and upgrade code. The joys of IOS XR code is that you can actually installed patches that fix bugs, you can perform in-service upgrades and not take down the router (provided you have a dual-supervisor router), as well as add new services to the code. All the necessary PIE packages can be found in the main image, they are not available separately. You can get the main image from CCO Support and Downloads. To navigate to the download, select: Products -> Routers -> Service Provider Edge Routers -> ASR 9000 -> ASR 9006 Then select IOS XR Software for the main images or IOS XR Software Maintenance Upgrades (SMU) for patches for caveats fixes.

Once you select the IOS XR Software, the most recent version of code will be presented on the screen. Select the version that you need and proceed to download it. If you get an error that a contract is required, please open a Cisco TAC case requesting access, they will need the serial number of the chassis in order to prove support. Once you have the image on your computer, we will now need to transfer it. Since the image is over 400 Megs as of 4.1.2, and 4.2.0 is over 700 Megs, TFTP is probably not going to cut it (most TFTP apps do not support files over 32 megs). What you might need to do is find an FTP server program to use - I recommend FileZilla but that is ultimately up to you. Once you have your FTP server setup and ready to go, we now need to get the image copied. For this example, I am using a username of Cisco and a password of cisco

Cisco IOS XR Training 2012 Jeffrey Fry

Page 23 FryGuy.Net

RP/0/RSP0/CPU0:R1# copy ftp://1.1.1.2/ASR9K-iosxr-k9-4.1.2.tar compactflash: Tue Apr 10 02:00:23.038 UTC Source username: [anonymous]?cisco Source password: cisco Destination filename [/compactflash:/ASR9K-iosxr-k9-4.1.2.tar]? (just hit enter) The file copy will now start and will take some time (you will see CCCCCCCCCCCCCCCCCCC) these are large images, so patience is a virtue. Once the file copy is complete, check the compact flash to make sure the images transferred successfully. RP/0/RSP0/CPU0:R2#dir compactflash: Tue Apr 10 02:01:37.766 UTC Directory of compactflash: 131104 6 131264 -rwdrwx -rw9216 Sun Jan 4096 Tue Jan 453611520 Thu Apr 2 08:01:19 2000 4 23:33:44 2000 5 22:14:28 2012 Test LOST.DIR ASR9K-iosxr-k9-4.1.2.tar

1022427136 bytes total (568795136 bytes free) Now that we have the image, we need to extract the tar file. That is done from ADMIN mode. You enter admin mode by typing admin at the command promt. RP/0/RSP0/CPU0:R2#admin Tue Apr 10 02:03:27.052 UTC RP/0/RSP0/CPU0:R2(admin)# Once there, we can install the tar image using the install command: RP/0/RSP0/CPU0:ios(admin)#install add tar compactflash:ASR9K-iosxr-k94.1.2.tar Once you enter that command, the system will start to process the file and show output: Mon Apr 9 21:29:41.420 UTC Install operation 1 '(admin) install add tar /compactflash:ASR9K-iosxr-k9-4.1.2.tar' started by user 'admin' via CLI at 21:29:41 UTC Mon Apr 09 2012. Info: The following files were extracted from the tar file Info: '/compactflash:ASR9K-iosxr-k9-4.1.2.tar' and will be added to the Info: entire router: Info: Info: asr9k-mcast-p.pie-4.1.2 Info: asr9k-mpls-p.pie-4.1.2 Info: asr9k-mini-p.pie-4.1.2 Info: asr9k-mini-p.vm-4.1.2 (skipped - not a pie) Info: asr9k-doc-p.pie-4.1.2 Cisco IOS XR Training 2012 Jeffrey Fry Page 24 FryGuy.Net

Info: asr9k-video-p.pie-4.1.2 Info: asr9k-mgbl-p.pie-4.1.2 Info: asr9k-optic-p.pie-4.1.2 Info: asr9k-upgrade-p.pie-4.1.2 Info: asr9k-k9sec-p.pie-4.1.2 Info: README-ASR9K-k9-4.1.2.txt (skipped - not a pie) Info: The install operation will continue asynchronously. This operation will happen in the background, you will be returned to the command prompt. Once the process is finished, the similar text will appear on the prompt: P/0/RSP0/CPU0:ios(admin)#Info: The following packages are now available to be activated: Info: Info: disk0:asr9k-mcast-p-4.1.2 Info: disk0:asr9k-mpls-p-4.1.2 Info: disk0:asr9k-mini-p-4.1.2 Info: disk0:asr9k-doc-p-4.1.2 Info: disk0:asr9k-video-p-4.1.2 Info: disk0:asr9k-mgbl-p-4.1.2 Info: disk0:asr9k-optic-4.1.2 Info: disk0:asr9k-upgrade-p-4.1.2 Info: disk0:asr9k-k9sec-p-4.1.2 Info: Info: The packages can be activated across the entire router. Info: Install operation 1 completed successfully at 21:38:52 UTC Mon Apr 09 2012. Now that we have the image there, we need to see what inactive PIEs we have to install and activate. The command here is show install inactive summary RP/0/RSP0/CPU0:ios(admin)#sh install inactive summary Mon Apr 9 21:59:10.354 UTC Default Profile: SDRs: Owner Inactive Packages: disk0:asr9k-upgrade-p-4.1.2 disk0:asr9k-optic-4.1.2 disk0:asr9k-doc-p-4.1.2 disk0:asr9k-k9sec-p-4.1.2 disk0:asr9k-video-p-4.1.2 disk0:asr9k-mpls-p-4.1.2 disk0:asr9k-mgbl-p-4.1.2 disk0:asr9k-mcast-p-4.1.2

Cisco IOS XR Training 2012 Jeffrey Fry

Page 25 FryGuy.Net

Now we should be able to activate and install one of the PIE images, here we will activate the MPLS one. RP/0/RSP0/CPU0:ios(admin)#install activate disk0:asr9k-mpls-p-4.1.2 Mon Apr 9 21:59:43.108 UTC Install operation 2 '(admin) install activate disk0:asr9k-mpls-p-4.1.2' started by user 'admin' via CLI at 21:59:43 UTC Mon Apr 09 2012. Error: Cannot proceed with the operation because the upgrade package Error: disk0:asr9k-upgrade-p-4.1.2 is present on boot disk. Error: The disk0:asr9k-upgrade-p-4.1.2 package should only be used when Error: upgrading from software versions prior to 4.0.0. Once the upgrade is Error: complbe immediately doved. No Error: further install operations will be allowed until this is completed. Error: Error: Remove the package disk0:asr9k-upgrade-p-4.1.2 from the entire router Error: by executing the 'install remove disk0:asr9k-upgrade-p-4.1.2' command Error: in admin mode. Error: No further install operations will be allowed until this is Error: completed. Install operation 2 failed at 21:59:44 UTC Mon Apr 09 2012. Ahh, we got an error! The error output tells us that we need to remove the upgrade package from the disk via the install remove command: RP/0/RSP0/CPU0:ios(admin)#install remove disk0:asr9k-upgrade-p-4.1.2 Mon Apr 9 22:00:13.538 UTC Install operation 3 '(admin) install remove disk0:asr9k-upgrade-p-4.1.2' started by user 'admin' via CLI at 22:00:13 UTC Mon Apr 09 2012. Info: This operation will remove the following package: Info: disk0:asr9k-upgrade-p-4.1.2 Now we need to confirm it by just hitting enter: Proceed with removing these packages? [confirm] (just hit enter to confirm) The install operation will continue asynchronoussly.

Now if we do a show install summary, it will tell us that we are in the process of doing something: RP/0/RSP0/CPU0:ios(admin)#sh install summary Mon Apr 9 22:00:22.060 UTC Default Profile: Currently affected by install operation 3 SDRs: Owner Active Packages: No packages.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 26 FryGuy.Net

Once completed, we will be notified on the cli RP/0/RSP0/CPU0:ios(admin)#Install operation 3 completed successfully at 22:00:39 UTC Mon Apr 09 2012. Now, we should be able to install the MPLS PIE RP/0/RSP0/CPU0:ios(admin)#install activate disk0:asr9k-mpls-p-4.1.2 Mon Apr 9 22:03:38.202 UTC Install operation 4 '(admin) install activate disk0:asr9k-mpls-p-4.1.2' started by user 'admin' via CLI at 22:03:38 UTC Mon Apr 09 2012. Info: Install Method: Parallel Process Restart The install operation will continue asynchronously. RP/0/RSP0/CPU0:ios(admin)#RP/0/RSP0/CPU0:Apr 9 22:04:32.428 : insthelper[65]: ISSU: Starting sysdb bulk start session Info: The changes made to software configurations will not be persistent Info: across system reloads. Use the command '(admin) install commit' to Info: make changes persistent. Info: Please verify that the system is consistent following the software Info: change using the following commands: Info: show system verify Info: install verify packages RP/0/RSP0/CPU0:Apr 9 22:04:45.933 : instdir[229]: %INSTALL-INSTMGR-4ACTIVE_SOFTWARE_COMMITTED_INFO : The currently active software is not committed. If the system reboots then the committed software will be used. Use 'install commit' to commit the active software. Install operation 4 completed successfully at 22:04:45 UTC Mon Apr 09 2012. If you want to see the status of the install, you can use the show install request command and it will show you the percentage complete. RP/0/RSP0/CPU0:c20.newthk01(admin)#sh install request Sat May 12 00:43:54.386 UTC Install operation 4 '(admin) install activate disk0:asr9k-mpls-p-4.1.2' started by user 'neteng' via CLI at 00:42:50 UTC Sat May 12 2012. The operation is 85% complete The operation can still be aborted. RP/0/RSP0/CPU0:c20.newthk01(admin)# Once the installation is complete, we need to COMMIT the installation using the install commit command RP/0/RSP0/CPU0:ios(admin)#install commit Mon Apr 9 22:07:17.014 UTC Install operation 5 '(admin) install commit' started by user 'admin' via CLI at 22:07:17 UTC Mon Apr 09 2012. \ 100% complete: The operation can no longer be aborted (ctrl-c for options)RP/0/RSP0/CPU0:Apr 9 22:07:20.238 : instdir[229]: %INSTALL-INSTMGR4-ACTIVE_SOFTWARE_COMMITTED_INFO : The currently active software is now the same as the committed software. Install operation 5 completed successfully at 22:07:20 UTC Mon Apr 09 2012. Cisco IOS XR Training 2012 Jeffrey Fry Page 27 FryGuy.Net

Now if we look at our show install active summary command, we now have the MPLS PIE RP/0/RSP0/CPU0:R2(admin)#sh install active summary Tue Apr 10 02:12:38.009 UTC Default Profile: SDRs: Owner Active Packages: disk0:asr9k-mini-p-4.1.2 disk0:asr9k-k9sec-p-4.1.2 disk0:asr9k-mpls-p-4.1.2 RP/0/RSP0/CPU0:R2(admin)# When it comes to patches, they are rather easy as well. They pretty much follow the same process as packages. Copy the file to flash, install the tar, then activate the patch. For this example, we will copy the CSCtu30994 rn_preorder_key_successor_int function is constantly looping per the readme file. First up, lets copy it from the TFTP server to our CompactFlash card: RP/0/RSP0/CPU0:ASR01#copy tftp: compactflash: Tue May 15 06:12:19.645 UTC Address or name of remote host [192.168.1.1]? (enter) Source filename [/tftp:]?asr9k-p-4.1.2.CSCtu30994.tar Destination filename [/compactflash:/asr9k-p-4.1.2.CSCtu30994.tar]? Accessing tftp://10.100.100.17/asr9k-p-4.1.2.CSCtu30994.tar CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC 911360 bytes copied in 6 sec ( 134936)bytes/sec Once copied, lets switch to ADMIN mode. RP/0/RSP0/CPU0:ASR01#admin Tue May 15 06:15:01.739 UTC

Now we can add the TAR files RP/0/RSP0/CPU0:ASR01(admin)#install add tar compactflash:asr9k-p4.1.2.CSCtu30994.tar Tue May 15 06:15:03.744 UTC /compactflash:asr9k-p-4.1.2.CSCtu30994.tar' started by user 'admin' via CLI at 06:15:04 UTC Tue May 15 2012. Info: The following files were extracted from the tar file Info: '/compactflash:asr9k-p-4.1.2.CSCtu30994.tar' and will be added to the Info: entire router: Info: Info: asr9k-p-4.1.2.CSCtu30994.pie Cisco IOS XR Training 2012 Jeffrey Fry Page 28 FryGuy.Net

Info: asr9k-p-4.1.2.CSCtu30994.txt (skipped - not a pie) Info: The install operation will continue asynchronously. And once the TAR has been added, the following message will appear: Info: The following package is now available to be activated: Info: Info: disk0:asr9k-p-4.1.2.CSCtu30994-1.0.0 Info: Info: The package can be activated across the entire router. Info: Install operation 27 completed successfully at 06:15:39 UTC Tue May 15 2012. Now we can activate this patch: RP/0/RSP0/CPU0:ASR01(admin)#install activate disk0:asr9k-p-4.1.2.CSCtu309941.0.0 Tue May 15 06:15:45.276 UTC Install operation 28 '(admin) install activate disk0:asr9k-p-4.1.2.CSCtu30994-1.0.0' started by user 'admin' via CLI at 06:15:45 UTC Tue May 15 2012. Info: Install Method: Parallel Process Restart The install operation will continue asynchronously. Info: Info: Info: Info: Info: Info: Info: The changes made to software configurations will not be persistent across system reloads. Use the command '(admin) install commit' to make changes persistent. Please verify that the system is consistent following the software change using the following commands: show system verify install verify packages

Once the install is done we need to commit it: RP/0/RSP0/CPU0:ASR01(admin)#install commit Tue May 15 06:17:06.359 UTC Install operation 29 '(admin) install commit' started by user 'admin' via CLI at 06:17:06 UTC Tue May 15 2012. \ 100% complete: The operation can no longer be aborted (ctrl-c for options)RP/0/RSP0/CPU0:May 15 06:17:09.967 : instdir[233]: %INSTALL-INSTMGR4-ACTIVE_SOFTWARE_COMMITTED_INFO : The currently active software is now the same as the committed software. Install operation 29 completed successfully at 06:17:09 UTC Tue May 15 2012. And like that we are patched.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 29 FryGuy.Net

Now, that was not one that required a reload, if you have one of them like CSCtw84381, here you will be prompted that you need to reload. RP/0/RSP0/CPU0:ASR01(admin)#install activate disk0:asr9k-p-4.1.2.CSCtw843811.0.0 Tue May 15 06:30:37.867 UTC Install operation 35 '(admin) install activate disk0:asr9k-p-4.1.2.CSCtw84381-1.0.0' started by user 'admin' via CLI at 06:30:38 UTC Tue May 15 2012. Info: This operation will reload the following nodes in parallel: Info: 0/RSP0/CPU0 (RP) (SDR: Owner) Info: 0/0/CPU0 (LC) (SDR: Owner) Info: 0/1/CPU0 (LC) (SDR: Owner) See, it is asking you to proceed - hit enter for Y Proceed with this install operation (y/n)? [y] (enter) Info: Install Method: Parallel Reload The install operation will continue asynchronously. Once the install is complete, the router will reload and you will need to relogin. Do not forget to do INSTALL COMMIT!!! Note from the Cisco website (http://www.cisco.com/en/US/docs/routers/asr9000/software/asr9k_r3.9/system_m anagement/command/reference/yr39asr9k_chapter14.html) Install operations are activated according to the method encoded in the package being activated. Generally, this method has the least impact for routing and forwarding purposes, but it may not be the fastest method from start to finish and can require user interaction by default. To perform the installation procedure as quickly as possible, you can specify the parallelreload keyword. This action forces the installation to perform a parallel reload, so that all cards on the router reload simultaneously and then come up with the new software. This impacts routing and forwarding, but it ensures that the installation is performed without other issues.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 30 FryGuy.Net

7. Licensing
The ever loving Cisco licensing well, not just Cisco but all vendors have some type of licensing. With the IOS XR in this case, we need a license to run VRF interfaces on our line cards. In order to request a license, you need to have a PAK key that you purchase, once you have that you will need to gather some information to request the license key. From the command promt, enter the admin mode RP/0/RSP0/CPU0:R2#admin Tue Apr 17 01:34:35.939 UTC From there, enter the command show license udi RP/0/RSP0/CPU0:R2(admin)#show license udi Tue Apr 17 01:34:38.950 UTC Local Chassis UDI Information: PID : ASR-9010-AC S/N : FOXXXXXAAAA Operation ID: 1 RP/0/RSP0/CPU0:R2(admin)# This information will be used on the Cisco License site www.cisco.com/go/license (CCO Account required). Once you have submitted the PAK request, [email protected] will send you the license file as an attachment within a few hours. Once you have the file, you will need to copy it to the router via TFTP or some other method. The license file will also include the instructions to add it, I have included them here as well. RP/0/RSP0/CPU0:R1#copy tftp: compactflash Wed Apr 11 05:23:23.259 UTC Address or name of remote host []?1.1.1.2 Source filename [/tftp:]?foo.lic Destination filename [/disk0a:/usr/compactflash]? (enter) Accessing tftp://1.1.1.2/foo.lic C 1199 bytes copied in 0 sec RP/0/RSP0/CPU0:R1#admin

Cisco IOS XR Training 2012 Jeffrey Fry

Page 31 FryGuy.Net

Next we can use the license add command from Admin mode RP/0/1/CPU0:CRS# RP/0/1/CPU0:CRS(admin)#license add compactflash:/foo.lic RP/0/1/CPU0:Mar 16 16:01:37.077 : licmgr[252]: %LICENSE-LICMGR-6LOAD_LICENSE_FILE_OK : All licenses from license file compactflash:/foo.lic added successfully License command "license add compactflash:/foo.lic sdr Owner" completed successfully. RP/0/1/CPU0:CRS(admin)# Now we need to see if is has been added via the show license command RP/0/1/CPU0:CRS(admin)#show license FeatureID: foo (Slot based, Permanent) Available for use 1 Allocated to location 0 Active 0 Pool: Owner Status: Available 1 Operational: Pool: sdr1 Status: Available 0 Operational:

0 0

Once the license has been successfully added, we now need to assign it to a line card slot. Again, this is done from Admin config mode RP/0/RSP0/CPU0:R1(admin)#config To assign the license, the command is license (License) location (LocationID). In our case, we are going to apply A9K-iVRF-LIC. The question mark will show you what location are available for this license. RP/0/RSP0/CPU0:R1(admin-config)#license A9K-iVRF-LIC location ? 0/0/CPU0 Fully qualified location specification 0/1/CPU0 Fully qualified location specification 0/RSP0/CPU0 Fully qualified location specification WORD Fully qualified location specification all all locations Now we can apply the licenses that we have to 0/0 and 0/1: RP/0/RSP0/CPU0:R1(admin-config)#license A9K-iVRF-LIC location 0/0/CPU0 RP/0/RSP0/CPU0:R1(admin-config)#license A9K-iVRF-LIC location 0/1/CPU0 RP/0/RSP0/CPU0:R1(admin-config)#commit Thu Apr 19 03:13:44.883 UTC RP/0/RSP0/CPU0:R1(admin-config)#exit RP/0/RSP0/CPU0:R1(admin)#exit

Cisco IOS XR Training 2012 Jeffrey Fry

Page 32 FryGuy.Net

Once installed, we can check using the show license command. RP/0/RSP0/CPU0:R1#sh license Thu Apr 19 03:13:51.432 UTC FeatureID: A9K-iVRF-LIC (Slot based, Permanent) Total licenses 2 Available for use 0 Allocated to location 0 Active 2 Store name Permanent Store index 1 Pool: Owner Total licenses in pool: 2 Status: Available 0 Operational: 2 Locations with licenses: (Active/Allocated) [SDR] 0/1/CPU0 (1/0) [Owner] 0/0/CPU0 (1/0) [Owner]

RP/0/RSP0/CPU0:R1# There they are, assigned to 0/1 and 0/0 as requested.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 33 FryGuy.Net

8. Aliases
From IOS, Aliases can sometimes make life easier on you and your support staff. In IOS XR, aliases get ramped up a bit, but first lest cover the basics. For this example, we can create an alias to show all the IPV4 interfaces in a brief using a single command, SHV4BR RP/0/7/CPU0:R1#conf t Mon Apr 16 15:05:26.064 UTC RP/0/7/CPU0:R1(config)#alias SHV4BR show ipv4 int brief RP/0/7/CPU0:R1(config)#commit Mon Apr 16 15:05:44.043 UTC Now, lets test the command: RP/0/7/CPU0:R1#shv4br As you can seem the system will re-enter the command from the alias RP/0/7/CPU0:R1#show ipv4 int brief Mon Apr 16 15:05:49.094 UTC Interface Loopback0 Loopback100 Loopback666 Loopback667 MgmtEth0/7/CPU0/0 MgmtEth0/7/CPU0/1 MgmtEth0/7/CPU0/2 GigabitEthernet0/3/0/0 GigabitEthernet0/3/0/1 GigabitEthernet0/3/0/2 GigabitEthernet0/3/0/3 MgmtEth0/6/CPU0/0 MgmtEth0/6/CPU0/1 MgmtEth0/6/CPU0/2 RP/0/7/CPU0:R1# IP-Address 1.1.1.1 100.100.100.100 6.6.6.6 6.6.6.7 unassigned unassigned unassigned unassigned unassigned 150.1.12.1 unassigned unassigned unassigned unassigned Status Up Up Up Up Shutdown Shutdown Shutdown Down Down Up Up Shutdown Shutdown Shutdown Protocol Up Up Up Up Down Down Down Down Down Up Up Down Down Down

Now, onto a cool feature, interface alias! We will create an alias for interface GigabitEthernet0/3/0/2, our connection to R2 on this router. RP/0/7/CPU0:R1#conf t Mon Apr 16 15:23:26.451 UTC RP/0/7/CPU0:R1(config)#alias R2Connection gig0/3/0/2 RP/0/7/CPU0:R1(config)#commit

Cisco IOS XR Training 2012 Jeffrey Fry

Page 34 FryGuy.Net

Now, lets see what happens when we do a show int for that alias RP/0/7/CPU0:R1#sh int r2connection RP/0/7/CPU0:R1#sh int gig0/3/0/2 Mon Apr 16 15:24:00.745 UTC GigabitEthernet0/3/0/2 is up, line protocol is up <--- SNIP - 0 output buffer failures, 0 output buffers swapped out 5 carrier transitions Pretty neat, but it gets better we can actually configure that alias as well! RP/0/7/CPU0:R1#conf t Mon Apr 16 15:24:06.626 UTC RP/0/7/CPU0:R1(config)#int r2connection RP/0/7/CPU0:R1(config)#int gig0/3/0/2 RP/0/7/CPU0:R1(config-if)#exit RP/0/7/CPU0:R1(config)#exit RP/0/7/CPU0:R1# Now, there is another trick with IOS XR, and that is variables! So, what can we do with Variables and Aliases? Well, if there is a command that you use quite often say show interface, why not change it to an alias with a variable. For this example, we will create sint (show interface) and use variable (var1). But first, let us look at what happens when you add a question mark (?) to the end of the command in configuration mode: RP/0/RSP0/CPU0:c21.lab(config)#alias sint ? LINE Alias body with optional parameters e.g,(name) show $name As you can see, it even tells you that you can use variables, might not be obvious, that that is what (name) is. So, let us create our alias: RP/0/RSP0/CPU0:c21.lab(config)#alias sint (var1) show interface $var1 RP/0/RSP0/CPU0:c21.lab(config)#commit RP/0/RSP0/CPU0:c21.lab(config)# Now we can test it on Interface Bundle-Eth 100: RP/0/RSP0/CPU0:c21.lab#sint(Bundle-Eth100) RP/0/RSP0/CPU0:c21.lab#show interface Bundle-Eth100 Bundle-Ether100 is up, line protocol is up Interface state transitions: 3 Hardware is Aggregated Ethernet interface(s), address is 6c9c.ed2d.0bab Internet address is 157.238.206.3/31 MTU 1514 bytes, BW 20000000 Kbit (Max: 20000000 Kbit) Cisco IOS XR Training 2012 Jeffrey Fry Page 35 FryGuy.Net

reliability 255/255, txload 0/255, rxload 0/255 Encapsulation ARPA, Full-duplex, 20000Mb/s loopback not set, ARP type ARPA, ARP timeout 04:00:00 No. of members in this bundle: 2 TenGigE0/0/0/0 Full-duplex 10000Mb/s Active TenGigE0/1/0/0 Full-duplex 10000Mb/s Active Last input 00:00:00, output 00:00:00 Last clearing of "show interface" counters never 5 minute input rate 1000 bits/sec, 1 packets/sec 5 minute output rate 1000 bits/sec, 1 packets/sec 1509709 packets input, 641971670 bytes, 411 total input drops 0 drops for unrecognized upper-level protocol Received 5 broadcast packets, 1298355 multicast packets 0 runts, 0 giants, 0 throttles, 0 parity 0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort 1518092 packets output, 642666596 bytes, 0 total output drops Output 6 broadcast packets, 1300886 multicast packets 0 output errors, 0 underruns, 0 applique, 0 resets 0 output buffer failures, 0 output buffers swapped out 0 carrier transitions RP/0/RSP0/CPU0:c21.lab# Now, another trick we can do is nested aliases! Lets modify the alias sint to show the interface as well as the configuration. RP/0/RSP0/CPU0:c21.lab(config)#alias sint (var1) show interface $var1; show run int $var1 RP/0/RSP0/CPU0:c21.lab(config)#commit RP/0/RSP0/CPU0:c21.lab(config)#

Now we can run that same command [sint(bundle-eth100)]again. RP/0/RSP0/CPU0:c21.lab#sint(bundle-eth100) RP/0/RSP0/CPU0:c21.lab#show interface bundle-eth100 Bundle-Ether100 is up, line protocol is up Interface state transitions: 3 Hardware is Aggregated Ethernet interface(s), address is 6c9c.ed2d.0bab Internet address is 157.238.206.3/31 MTU 1514 bytes, BW 20000000 Kbit (Max: 20000000 Kbit) reliability 255/255, txload 0/255, rxload 0/255 Encapsulation ARPA, Full-duplex, 20000Mb/s loopback not set, ARP type ARPA, ARP timeout 04:00:00 Cisco IOS XR Training 2012 Jeffrey Fry Page 36 FryGuy.Net

No. of members in this bundle: 2 TenGigE0/0/0/0 Full-duplex 10000Mb/s Active TenGigE0/1/0/0 Full-duplex 10000Mb/s Active Last input 00:00:00, output 00:00:00 Last clearing of "show interface" counters never 5 minute input rate 1000 bits/sec, 1 packets/sec 5 minute output rate 1000 bits/sec, 1 packets/sec 1509849 packets input, 642030906 bytes, 411 total input drops 0 drops for unrecognized upper-level protocol Received 5 broadcast packets, 1298474 multicast packets 0 runts, 0 giants, 0 throttles, 0 parity 0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort 1518235 packets output, 642727790 bytes, 0 total output drops Output 6 broadcast packets, 1301008 multicast packets 0 output errors, 0 underruns, 0 applique, 0 resets 0 output buffer failures, 0 output buffers swapped out 0 carrier transitions RP/0/RSP0/CPU0:c21.lab#show run int bundle-eth100 interface Bundle-Ether100 ipv4 address 157.238.206.3 255.255.255.254 bundle load-balancing hash src-ip bundle maximum-active links 2 bundle minimum-active links 1 ! RP/0/RSP0/CPU0:c21.lab# As you can see, it did the show interface bundle-eth100 and show run interface bundle-eth100

Cisco IOS XR Training 2012 Jeffrey Fry

Page 37 FryGuy.Net

9. Wildcard Masks
A really cool thing with IOS XR is interface wildcards. If you want to only see the Loopback interfaces, all of them. Normally you would do something like Show int br | in Loop, but with XR you can use a wildcard (*) RP/0/7/CPU0:R1#sh int l* br Mon Apr 16 17:21:08.088 UTC Intf Intf LineP Encap MTU BW Name State State Type (byte) (Kbps) ---------------------------------------------------------------------Lo0 up up Loopback 1500 0 Lo100 up up Loopback 1500 0 Lo666 up up Loopback 1500 0 Lo667 up up Loopback 1500 0 Lo1000 up up Loopback 1500 0 RP/0/7/CPU0:R1# This works the same if you want to see this in the running config: RP/0/7/CPU0:R1#sh run in l* Mon Apr 16 17:21:53.360 UTC interface Loopback0 ipv4 address 1.1.1.1 255.255.255.255 ipv6 address 2001::1/128 ! interface Loopback100 ipv4 address 100.100.100.100 255.255.255.255 ! interface Loopback666 ipv4 address 6.6.6.6 255.255.255.255 ! interface Loopback667 ipv4 address 6.6.6.7 255.255.255.255 ! interface Loopback1000 vrf LAB ipv4 address 111.111.111.111 255.255.255.255 ! RP/0/7/CPU0:R1#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 38 FryGuy.Net

10. Processes
So, since IOS XR is based on QNX, the SHOW PROCESSES command is a bit different then you would see in IOS. In IOS XR, you can actually query processes and see what is going on. Here is a way to see what BGP is doing : RP/0/RSP0/CPU0:R2#sh processes bgp Tue Apr 24 01:23:06.343 UTC Job Id: 1039 PID: 2941214 Executable path: /disk0/iosxr-routing-4.1.2/bin/bgp Instance #: 1 Version ID: 00.00.0000 Respawn: ON Respawn count: 4 Max. spawns per minute: 12 Last started: Tue Apr 10 05:31:26 2012 Process state: Run Package state: Normal Started on config: ipc/gl/ip-bgp/meta/speaker/default core: MAINMEM Max. core: 0 Placement: Placeable startup_path: /pkg/startup/bgp.startup Ready: 0.338s Available: 25.582s Process cpu time: 63.719 user, 1.074 kernel, 64.793 total JID TID CPU Stack pri state TimeInState HR:MM:SS:MSEC 1039 1 1 312K 10 Receive 0:01:01:0795 0:00:00:0249 1039 2 1 312K 10 Receive 331:51:39:0224 0:00:00:0000 1039 3 0 312K 10 Receive 331:51:39:0223 0:00:00:0001 1039 4 1 312K 10 Sigwaitinfo 331:51:39:0129 0:00:00:0000 1039 5 0 312K 10 Receive 0:00:01:0764 0:00:00:0005 1039 6 1 312K 10 Receive 0:00:01:0760 0:00:00:0016 1039 7 0 312K 10 Receive 0:00:23:0239 0:00:02:0242 1039 8 0 312K 10 Receive 0:00:03:0321 0:00:02:0280 1039 9 1 312K 10 Receive 0:01:01:0796 0:00:00:0005 1039 10 1 312K 10 Receive 0:00:01:0786 0:00:00:0008 1039 11 0 312K 10 Receive 0:00:01:0786 0:00:00:0004 1039 12 0 312K 10 Receive 0:00:01:0786 0:00:00:0001 1039 13 1 312K 10 Receive 0:00:02:0861 0:00:00:0006 1039 14 1 312K 10 Receive 0:00:03:0189 0:00:59:0750 1039 15 0 312K 10 Receive 0:00:01:0816 0:00:00:0223 1039 16 1 312K 10 Receive 331:51:39:0047 0:00:00:0000 1039 17 1 312K 10 Receive 0:00:27:0039 0:00:00:0000 1039 18 1 312K 10 Receive 75:14:02:0621 0:00:00:0002 Cisco IOS XR Training 2012 Jeffrey Fry

NAME bgp bgp bgp bgp bgp bgp bgp bgp bgp bgp bgp bgp bgp bgp bgp bgp bgp bgp Page 39 FryGuy.Net

Now, since IOS XR is based on a flavor of Unix, we have a command similar to TOP called monitor processes. RP/0/RSP0/CPU0:R2#monitor processes Tue Apr 24 01:27:41.959 UTC Computing times... (Screen Clears and the following data is refreshed)

287 processes; 1320 threads; 1086 timers, 6265 channels, 8489 fds CPU states: 99.6% idle, 0.2% user, 0.1% kernel Memory: 4096M total, 2762M avail, page size 4K JID TIDS Chans FDs Tmrs 1 13 291 204 1 65744 1 1 11 0 340 2 16 20 4 60 15 44 20 7 95 22 276 35 4 152 4 16 24 6 71 2 7 11 1 202 9 39 62 14 89 1 6 3 1 355 4 9 15 2 MEM 0 1M 340K 4M 924K 700K 236K 1M 104K 260K HH:MM:SS 998:56:18 0:00:00 0:41:16 0:18:05 0:02:03 0:00:08 0:00:04 0:00:03 0:00:55 0:00:00 CPU 0.12% 0.09% 0.05% 0.04% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% NAME procnto-600-smp-instr ptop sc eth_server sysmgr canb-server mdio_sup ether_ctrl_mgmt serdrvr ssm_process

Cisco IOS XR Training 2012 Jeffrey Fry

Page 40 FryGuy.Net

11. Remote Access Services Telnet and SSH

We need to have a way to remote access this device, and by default SSH and TELNET are not enabled. First up, the easy one telnet. RP/0/RSP0/CPU0:R1(config)#telnet ipv4 server max-servers 10 And like that, we can telnet. Ok, onto SSH but before setting up SSH, we need to generate an RSA key. This is a bit different as you do not do this from config mode. First up, add your domain-name if you do not have one: RP/0/RSP0/CPU0:R1(config)#domain name fryguy.net RP/0/RSP0/CPU0:R1(config)#commit RP/0/RSP0/CPU0:R1#crypto key generate rsa Sat Apr 21 00:36:07.790 UTC The name for the keys will be: the_default Choose the size of the key modulus in the range of 512 to 2048 for your General Purpose Keypair. Choosing a key modulus greater than 512 may take a few minutes. How many bits in the modulus [1024]: 2048 Generating RSA keys ... Done w/ crypto generate keypair [OK] RP/0/RSP0/CPU0:R1# Once we have generated the RSA key, we can now enable the SSH service: RP/0/RSP0/CPU0:R1#conf t Sat Apr 21 00:40:33.845 UTC RP/0/RSP0/CPU0:R1(config)#ssh server v2 RP/0/RSP0/CPU0:R1(config)#commit Sat Apr 21 00:40:39.939 UTC And like that, SSH services are now enabled. Ok, but what if we wanted to limit who has access to the box by IP address, that is where control-plane security comes in. For this example, I will allow 10/8 to access the device. RP/0/RSP0/CPU0:R1(config)#control-plane RP/0/RSP0/CPU0:R1(config-ctrl)#management-plane RP/0/RSP0/CPU0:R1(config-mpp-inband)#int g0/1/0/18 RP/0/RSP0/CPU0:R1(config-mpp-inband-if)#allow SSH peer RP/0/RSP0/CPU0:R1(config-ssh-peer)# address ipv4 10.0.0.0/8 Cisco IOS XR Training 2012 Jeffrey Fry Page 41 FryGuy.Net

RP/0/RSP0/CPU0:R1(config-ssh-peer)# allow Telnet peer RP/0/RSP0/CPU0:R1(config-telnet-peer)#address ipv4 10.0.0.0/8 RP/0/RSP0/CPU0:R1(config-telnet-peer)#exit RP/0/RSP0/CPU0:R1(config-mpp-inband)#comm Sat Apr 21 01:09:45.163 UTC And now to test, from a device on the 10/8 network: user@host [~]> ssh [email protected] [email protected] password: RP/0/RSP0/CPU0:R1# There you go, SSH access from only the 10.0.0.0/8 subnet. And, when it comes close the expiry timer, you will get a message: RP/0/RSP0/CPU0:R1# * * The idle timeout is soon to expire on this line * Received disconnect from 10.2.2.2: 11: user@host [~]>

Cisco IOS XR Training 2012 Jeffrey Fry

Page 42 FryGuy.Net

12. TACACS Configuration (default and non-default VRF)

Ok, so you want to secure your IOS-XR device using TACACS. The first example I will use will be using the default VRF for TACACS authorization and the second will be using a different VRF. For these examples, the tacacs server is at IP 192.168.100.100 and the password is TacacsPassword First up, we need to configure our source interface for TACACS, here we will use loopback0 and the default VRF. RP/0/RSP0/CPU0:PE2(config)#tacacs source-interface Loopback0 vrf default Now we can configure our TACACS server and Password RP/0/RSP0/CPU0:PE2(config)#tacacs-server host 192.168.100.100 RP/0/RSP0/CPU0:PE2(config-tacacs-host)#key 0 TacacsPassword RP/0/RSP0/CPU0:PE2(config-tacacs-host)#exit RP/0/RSP0/CPU0:PE2(config)# Time to create a local console authenticaion method, this way console does not rely on TACACS. You may or may not want to do this, but I am showing it for these examples. RP/0/RSP0/CPU0:PE2(config)#aaa authentication login console local RP/0/RSP0/CPU0:PE2(config)#aaa authorization commands console none Apply the console loging to the line console RP/0/RSP0/CPU0:PE2(config)#line console RP/0/RSP0/CPU0:PE2(config-line)#login authentication console RP/0/RSP0/CPU0:PE2(config-line)#authorization commands console RP/0/RSP0/CPU0:PE2(config-line)#exit RP/0/RSP0/CPU0:PE2(config)# Now we can start to configure our AAA for login, here I am using default
RP/0/RSP0/CPU0:PE2(config)#aaa authentication login default group tacacs+ local

Now for some command authorization, if you want it RP/0/RSP0/CPU0:PE2(config)#aaa authorization commands default group tacacs+ And accounting as well.
RP/0/RSP0/CPU0:PE2(config)#aaa accounting exec default start-stop group tacacs+ RP/0/RSP0/CPU0:PE2(config)#aaa accounting system default start-stop group tacacs+ RP/0/RSP0/CPU0:PE2(config)#aaa accounting commands default start-stop group tacacs+

Since this is IOS XR, I strongly suggest using a commit confirmed here! RP/0/RSP0/CPU0:PE2(config)#commit confirmed minutes 2 Thu Oct 18 03:22:57.487 UTC RP/0/RSP0/CPU0:PE2(config)# Cisco IOS XR Training 2012 Jeffrey Fry Page 43 FryGuy.Net

From another terminal, SSH into the box using a TACACs account, and if successful, commit again. RP/0/RSP0/CPU0:PE2(config)#commit Thu Oct 18 03:23:22.951 UTC % Confirming commit for trial session. RP/0/RSP0/CPU0:PE2(config)#

That is normal TACACS, now time to add in the challenges of a VRF. First up, we need to set our source interface, for this one I will use a different Loopback, Lo100 and use VRF CustA RP/0/RSP0/CPU0:PE2(config)#tacacs source-interface Loopback100 vrf CustA Now we can configure our TACACS server RP/0/RSP0/CPU0:PE2(config)#tacacs-server host 192.168.100.100 RP/0/RSP0/CPU0:PE2(config-tacacs-host)#key 0 TacacsPassword RP/0/RSP0/CPU0:PE2(config-tacacs-host)#exit RP/0/RSP0/CPU0:PE2(config)# Now we need to create a server group for the ACS box. This tells it what VRF the server is in. RP/0/RSP0/CPU0:PE2(config)#aaa group server tacacs+ ACS RP/0/RSP0/CPU0:PE2(config-sg-tacacs)# server 192.168.100.100 RP/0/RSP0/CPU0:PE2(config-sg-tacacs)# vrf CustA Now we can configure our local logins for the console: RP/0/RSP0/CPU0:PE2(config)#aaa authentication login console local RP/0/RSP0/CPU0:PE2(config)#aaa authorization commands console none RP/0/RSP0/CPU0:PE2(config)#line console RP/0/RSP0/CPU0:PE2(config-line)# login authentication console RP/0/RSP0/CPU0:PE2(config-line)# authorization commands console Here I would commit the configs that we have done. RP/0/RSP0/CPU0:PE2(config)#commit And finally configure our AAA for login RP/0/RSP0/CPU0:PE2(config)#aaa authentication login default group ACS local RP/0/RSP0/CPU0:PE2(config)#aaa authorization commands default group ACS none RP/0/RSP0/CPU0:PE2(config)#aaa accounting exec default start-stop group ACS RP/0/RSP0/CPU0:PE2(config)#aaa accounting system default start-stop group ACS RP/0/RSP0/CPU0:PE2(config)#aaa accounting commands default start-stop group ACS RP/0/RSP0/CPU0:PE2(config)#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 44 FryGuy.Net

And finally do the commit confirmed here again RP/0/RSP0/CPU0:PE2(config)#commit confirmed minutes 2 Test remote access via SSH, and if all works - commit it to save RP/0/RSP0/CPU0:PE2(config)#commit % Confirming commit for trial session. And we are done!

Cisco IOS XR Training 2012 Jeffrey Fry

Page 45 FryGuy.Net

13. Access Lists

Access lists - these are the same as IOS Extended access lists. Sorry, not much to say here but you should already be familiar with these. RP/0/RSP0/CPU0:R1(config)#ipv4 access-list RemoteAccess RP/0/RSP0/CPU0:R1(config-ipv4-acl)#permit tcp 216.167.0.0/24 any eq ssh RP/0/RSP0/CPU0:R1(config-ipv4-acl)#commit

Cisco IOS XR Training 2012 Jeffrey Fry

Page 46 FryGuy.Net

14. OSPF
Time for some OSPF configs, these will build off the previous configs we just did. For this lab, the other router, R2, was preconfigured to support the connections. We will place our loopback and out g0/3/0/2 interface into OSPF process LAB and area 0.0.0.0 RP/0/7/CPU0:R1# RP/0/7/CPU0:R1#conf t Thu Mar 29 19:37:52.671 UTC Define our OSPF process name RP/0/7/CPU0:R1(config)#router ospf LAB Now to define our area first RP/0/7/CPU0:R1(config-ospf)#area 0.0.0.0 Now we can place the interfaces into the area, no need to entering subnets RP/0/7/CPU0:R1(config-ospf-ar)#inter loo0 RP/0/7/CPU0:R1(config-ospf-ar-if)#inter g0/3/0/2 RP/0/7/CPU0:R1(config-ospf-ar-if)#exit RP/0/7/CPU0:R1(config-ospf-ar)#exit RP/0/7/CPU0:R1(config-ospf)#exit RP/0/7/CPU0:R1(config)#commit Thu Mar 29 19:38:15.182 UTC RP/0/7/CPU0:R1(config)# Now to look at our IP Protocols running: RP/0/7/CPU0:R1#sh ip proto Thu Mar 29 19:38:24.113 UTC Routing Protocol OSPF LAB Router Id: 1.1.1.1 Distance: 110 Non-Stop Forwarding: Disabled Redistribution: None Area 0.0.0.0 Loopback0 GigabitEthernet0/3/0/2 RP/0/7/CPU0:R1# We can see what we have OSPF LAB running with a RouterID of 1.1.1.1 (our loopback). It tells us what interfaces are in Area 0.0.0.0 as well. Cisco IOS XR Training 2012 Jeffrey Fry Page 47 FryGuy.Net

Now to see if we neighbored up with R2: RP/0/7/CPU0:R1#sh ip ospf nei Thu Mar 29 19:38:33.557 UTC * Indicates MADJ interface Neighbors for OSPF LAB Neighbor ID Pri State Dead Time 2.2.2.2 1 FULL/DR 00:00:37 Neighbor is up for 00:00:12 Total neighbor count: 1 RP/0/7/CPU0:R1# Yup, we have a neighbor of R2 (2.2.2.2) up and in FULL/DR. Time to look at the routing table: RP/0/7/CPU0:R1#sh ip route Thu Mar 29 19:41:06.047 UTC Codes: C - connected, S - static, R - RIP, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP i - ISIS, L1 - IS-IS level-1, L2 - IS-IS level-2 ia - IS-IS inter area, su - IS-IS summary null, * - candidate default U - per-user static route, o - ODR, L - local, G - DAGR A - access/subscriber, (!) - FRR Backup pathc Gateway of last resort is not set L 1.1.1.1/32 is directly connected, 00:15:37, Loopback0 O 2.2.2.2/32 [110/2] via 150.1.12.2, 00:02:43, GigabitEthernet0/3/0/2 C 150.1.12.0/24 is directly connected, 01:02:19, GigabitEthernet0/3/0/2 L 150.1.12.1/32 is directly connected, 01:02:19, GigabitEthernet0/3/0/2 RP/0/7/CPU0:R1# We can see we have a route to R2 loopback interface (2.2.2.2/32), now we should be able to PING it from our Loopback0 interface. RP/0/7/CPU0:R1#ping 2.2.2.2 source lo0 Thu Mar 29 19:41:21.828 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/4 ms RP/0/7/CPU0:R1# Address 150.1.12.2 Interface GigabitEthernet0/3/0/2

Cisco IOS XR Training 2012 Jeffrey Fry

Page 48 FryGuy.Net

For reference, here is a similar IOS config for the same thing: R1(config)#router ospf 1 R1(config-router)#net 1.1.1.1 0.0.0.0 a 0.0.0.0 R1(config-router)#net 150.1.12.0 0.0.0.255 a 0.0.0.0 R1(config-router)#^Z R1# *Mar 29 20:18:29.698: %SYS-5-CONFIG_I: Configured from console by console R1# R1# R1#p 2.2.2.2 so l0 Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds: Packet sent with a source address of 1.1.1.1 !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/1 ms R1# OSPF Advanced Features I just wanted to take a minute and discuss some of the other features available for OSPF. Network Point-to-Point, Point-to-Multipoint, broadcast, non-broadcast RP/0/7/CPU0:R1(config)#router ospf LAB RP/0/7/CPU0:R1(config-ospf)#area 0.0.0.0 RP/0/7/CPU0:R1(config-ospf-ar)#int g0/3/0/2 RP/0/7/CPU0:R1(config-ospf-ar-if)#network ? broadcast Specify OSPF broadcast multi-access network non-broadcast Specify OSPF NBMA network point-to-multipoint Specify OSPF point-to-multipoint network point-to-point Specify OSPF point-to-point network As you can see, all the normal OSPF network interface types are there. You just need to configure them under the OSPF process instead of the interface like in normal IOS.

Authentication IOS XR also supports OSPF authentication, both area and interface. example we will create an MD5 interface authentication. RP/0/7/CPU0:R1(config)#router ospf LAB RP/0/7/CPU0:R1(config-ospf)#area 0.0.0.0 RP/0/7/CPU0:R1(config-ospf-ar)#int g0/3/0/2 Need to enable MD5 authentication RP/0/7/CPU0:R1(config-ospf-ar-if)#authentication message-digest Cisco IOS XR Training 2012 Jeffrey Fry Page 49 FryGuy.Net In this

Then set our MD5 key #1 to Cisco RP/0/7/CPU0:R1(config-ospf-ar-if)#message-digest-key 1 md5 Cisco RP/0/7/CPU0:R1(config-ospf-ar-if)#exit RP/0/7/CPU0:R1(config-ospf-ar)#commit Now, lets look at the interface and make sure we have MD5 authentication enabled. RP/0/7/CPU0:R1#sh ospf LAB int g0/3/0/2 Sun Apr 1 18:31:01.235 UTC GigabitEthernet0/3/0/2 is up, line protocol is up Internet Address 150.1.12.1/24, Area 0.0.0.0 Process ID LAB, Router ID 1.1.1.1, Network Type BROADCAST, Cost: 1 Transmit Delay is 1 sec, State BDR, Priority 1, MTU 1500, MaxPktSz 1500 Designated Router (ID) 2.2.2.2, Interface address 150.1.12.2 Backup Designated router (ID) 1.1.1.1, Interface address 150.1.12.1 Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Hello due in 00:00:01 Index 1/1, flood queue length 0 Next 0(0)/0(0) Last flood scan length is 1, maximum is 1 Last flood scan time is 0 msec, maximum is 0 msec LS Ack List: current length 0, high water mark 3 Neighbor Count is 1, Adjacent neighbor count is 1 Adjacent with neighbor 2.2.2.2 (Designated Router) Suppress hello for 0 neighbor(s) Message digest authentication enabled Youngest key id is 1 Multi-area interface Count is 0 RP/0/7/CPU0:R1# As you can see above, we do. This is all very similar to IOS, so as you can see, the jump to XR is more knowing where to configure something then how to configure something. Now, lets check our neighbor state RP/0/7/CPU0:R1#sh ospf LAB neighbor Sat Mar 31 18:37:07.753 UTC * Indicates MADJ interface Neighbors for OSPF LAB Neighbor ID Pri State 2.2.2.2 1 EXSTART/DR GigabitEthernet0/3/0/2 Neighbor is up for 00:00:31 Total neighbor count: 1 RP/0/7/CPU0:R1# Cisco IOS XR Training 2012 Jeffrey Fry Page 50 FryGuy.Net Dead Time 00:00:36 Address 150.1.12.2 Interface

Then make sure we are getting a route RP/0/7/CPU0:R1#sh route ipv4 ospf Sat Mar 31 18:37:15.279 UTC O 2.2.2.2/32 [110/2] via 150.1.12.2, 00:00:06, GigabitEthernet0/3/0/2 RP/0/7/CPU0:R1# And finallying PINGing R2 loopback from ours RP/0/7/CPU0:R1#ping 2.2.2.2 so l0 Sat Mar 31 18:37:19.151 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/5 ms RP/0/7/CPU0:R1#

Cost Just like normal IOS, we can change the OSPF cost on an interface but same thing here; it is done under the OSPF process RP/0/7/CPU0:R1# RP/0/7/CPU0:R1#conf t Sun Apr 1 18:35:17.061 UTC RP/0/7/CPU0:R1(config)#router ospf LAB RP/0/7/CPU0:R1(config-ospf)#area 0.0.0.0 RP/0/7/CPU0:R1(config-ospf-ar)#int loop0 RP/0/7/CPU0:R1(config-ospf-ar-if)#cost ? <1-65535> Cost

Cisco IOS XR Training 2012 Jeffrey Fry

Page 51 FryGuy.Net

15. EIGRP
First thing we need to do is delete the OSPF process,that is if you still have it. RP/0/7/CPU0:R1#conf t Thu Mar 29 20:07:53.797 UTC RP/0/7/CPU0:R1(config)#no router ospf LAB RP/0/7/CPU0:R1(config)#commit Once that is deleted, we can now continue with EIGRP configuration. Just like IOS, we need to give it a process ID RP/0/7/CPU0:R1(config)#router eigrp 1 Here is where the difference starts, we need to select the Address family first RP/0/7/CPU0:R1(config-eigrp)#address-family ipv4 Enter no auto-summary ( this is habitual to be honest ) RP/0/7/CPU0:R1(config-eigrp-af)#no auto-summary Then assign the interfaces you want in EIGRP RP/0/7/CPU0:R1(config-eigrp-af)#int l0 RP/0/7/CPU0:R1(config-eigrp-af-if)#int g0/3/0/2 RP/0/7/CPU0:R1(config-eigrp-af-if)#exit RP/0/7/CPU0:R1(config-eigrp-af)#exit RP/0/7/CPU0:R1(config-eigrp)#exit RP/0/7/CPU0:R1(config)#commit Thu Mar 29 20:08:59.108 UTC RP/0/7/CPU0:R1(config)#exit Now lets look at our IP Protocols: RP/0/7/CPU0:R1#sh ip protocols Thu Mar 29 20:09:25.058 UTC Routing Protocol: EIGRP, instance 1 Default context AS: 1, Router ID: 1.1.1.1 Address Family: IPv4 Default networks not flagged in outgoing updates Default networks not accepted from incoming updates Distance: internal 90, external 170 Maximum paths: 4 EIGRP metric weight K1=1, K2=0, K3=1, K4=0, K5=0 EIGRP maximum hopcount 100 EIGRP maximum metric variance 1 EIGRP NSF: enabled Cisco IOS XR Training 2012 Jeffrey Fry Page 52 FryGuy.Net

NSF-aware route hold timer is 480s NSF signal timer is 20s NSF converge timer is 300s Time since last restart is 00:00:25 SIA Active timer is 180s Interfaces: Loopback0 GigabitEthernet0/3/0/2 When you issue the same command under IOS, you have Routing for Networks instead of Interfaces: R1#sh ip protocols Routing Protocol is "eigrp 1" Outgoing update filter list for all interfaces is not set Incoming update filter list for all interfaces is not set Default networks flagged in outgoing updates Default networks accepted from incoming updates EIGRP metric weight K1=1, K2=0, K3=1, K4=0, K5=0 EIGRP maximum hopcount 100 EIGRP maximum metric variance 1 Redistributing: eigrp 1 EIGRP NSF-aware route hold timer is 240s Automatic network summarization is not in effect Maximum path: 4 Routing for Networks: 1.1.1.1/32 150.1.12.0/24 Routing Information Sources: Gateway Distance Last Update (this router) 90 00:00:22 150.1.12.2 90 00:00:04 Distance: internal 90 external 170 R1# Now, lets look at our routing table on IOS XR. RP/0/7/CPU0:R1#sh ip route Thu Mar 29 20:09:31.763 UTC Codes: C - connected, S - static, R - RIP, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP i - ISIS, L1 - IS-IS level-1, L2 - IS-IS level-2 ia - IS-IS inter area, su - IS-IS summary null, * - candidate default U - per-user static route, o - ODR, L - local, G - DAGR A - access/subscriber, (!) - FRR Backup path Gateway of last resort is not set L 1.1.1.1/32 is directly connected, 00:44:02, Loopback0 Page 53 FryGuy.Net

Cisco IOS XR Training 2012 Jeffrey Fry

D 2.2.2.2/32 [90/130816] via 150.1.12.2, 00:00:11, GigabitEthernet0/3/0/2 C 150.1.12.0/24 is directly connected, 01:30:45, GigabitEthernet0/3/0/2 L 150.1.12.1/32 is directly connected, 01:30:45, GigabitEthernet0/3/0/2 RP/0/7/CPU0:R1# There, we have a route to R2s loopback. Lets PING it from our loopback to test connectivity. RP/0/7/CPU0:R1#ping 2.2.2.2 so l0 Thu Mar 29 20:09:36.232 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 2/2/5 ms RP/0/7/CPU0:R1# And like that basic EIGRP is done. Now, lets add IPv6 to the EIGRP process. RP/0/7/CPU0:R1#conf t Thu Mar 29 20:27:16.966 UTC RP/0/7/CPU0:R1(config)#router eigrp 1 RP/0/7/CPU0:R1(config-eigrp)#address-family ipv6 RP/0/7/CPU0:R1(config-eigrp-af)#int l0 RP/0/7/CPU0:R1(config-eigrp-af-if)#int g0/3/0/2 RP/0/7/CPU0:R1(config-eigrp-af-if)#commit Thu Mar 29 20:27:28.352 UTC RP/0/7/CPU0:R1(config-eigrp-af-if)# I will be honest here; the correct command to show routes is show route Protocol. Once you add IPv6, you really should to start to use the correct commands. RP/0/7/CPU0:R1#sh route ipv6 Thu Mar 29 20:29:31.952 UTC Codes: C - connected, S - static, R - RIP, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP i - ISIS, L1 - IS-IS level-1, L2 - IS-IS level-2 ia - IS-IS inter area, su - IS-IS summary null, * - candidate default U - per-user static route, o - ODR, L - local, G - DAGR A - access/subscriber, (!) - FRR Backup path Gateway of last resort is not set L D 2001::1/128 is directly connected, 00:58:42, Loopback0 2001::2/128 Page 54 FryGuy.Net

Cisco IOS XR Training 2012 Jeffrey Fry

[90/130816] via fe80::2d0:79ff:fe01:3a78, 00:01:43, GigabitEthernet0/3/0/2 C 2001:1:1:12::/64 is directly connected, 01:02:52, GigabitEthernet0/3/0/2 L 2001:1:1:12::1/128 is directly connected, 01:02:52, GigabitEthernet0/3/0/2 RP/0/7/CPU0:R1# Cool we have an IPv6 route to R2 loopback (2001::2/128) Lets ping that interface from our loopback interface RP/0/7/CPU0:R1#ping 2001::2 sou 2001::1 Thu Mar 29 20:31:56.602 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2001::2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 2/4/10 ms RP/0/7/CPU0:R1# There you go; we have IPv4 and IPv6 connectivity now.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 55 FryGuy.Net

16. RIP (yeah yeah, why would you want to run this? Think PE-CE)
Ok, time for the next routing protocol RIP? Well, if you have a CE device that perfect protocol. Keep in mind that IOS Provider network, so PE-CE relationships RIP. Why would you use IOS XR for only has a few networks, RIP is a XR is code built for a Service are what these routers are about.

Plus it is just good to know different options. RP/0/7/CPU0:R1#conf t Thu Mar 29 20:37:44.801 UTC RP/0/7/CPU0:R1(config)#router rip RP/0/7/CPU0:R1(config-rip)#int l0 RP/0/7/CPU0:R1(config-rip-if)#int g0/3/0/2 RP/0/7/CPU0:R1(config-rip-if)#^Z Uncommitted changes found, commit them before exiting(yes/no/cancel)? [cancel]:yes Notice I did not do a COMMIT, but since the router knows I was making changes it asked me. RP/0/7/CPU0:R1# Lets check our IP protocols: RP/0/7/CPU0:R1#sh ip proto Thu Mar 29 20:39:20.919 UTC Routing Protocol RIP 1 VRFs (including default) configured, 1 active 6 routes, 3 paths have been allocated Current OOM state is "Normal" UDP socket descriptor is 42 VRF Active If-config If-active Routes Updates default Active 2 2 6 30s Now lets look at the RIP process: RP/0/7/CPU0:R1#sh rip Thu Mar 29 20:39:24.892 UTC RIP config: Active: Added to socket: Out-of-memory state: Cisco IOS XR Training 2012 Jeffrey Fry

Paths 3

Yes Yes Normal Page 56 FryGuy.Net

Version: Default metric: Maximum paths: Auto summarize: Broadcast for V2: Packet source validation: NSF: Timers: Update: Invalid: Holddown: Flush: RP/0/7/CPU0:R1#

2 Not set 4 No No Yes Disabled 30 seconds (0 seconds until next update) 180 seconds 180 seconds 240 seconds

Now here is something interesting, the RIP version is 2, yet I did not specify it in the config. This is because IOS XR code only supports v2. RP/0/7/CPU0:R1(config)#router rip RP/0/7/CPU0:R1(config-rip)#ver? ^ % Invalid input detected at '^' marker. Now, lets look at the routing table using the proper command: RP/0/7/CPU0:R1#sh route ipv4 Thu Mar 29 20:40:08.877 UTC Codes: C - connected, S - static, R - RIP, B BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2, E EGP i - ISIS, L1 - IS-IS level-1, L2 - IS-IS level-2 ia - IS-IS inter area, su - IS-IS summary null, * - candidate default U - per-user static route, o - ODR, L - local, G - DAGR A - access/subscriber, (!) - FRR Backup path Gateway of last resort is not set L 1.1.1.1/32 is directly connected, 01:14:39, Loopback0 R 2.2.2.2/32 [120/1] via 150.1.12.2, 00:01:25, GigabitEthernet0/3/0/2 C 150.1.12.0/24 is directly connected, 02:01:22, GigabitEthernet0/3/0/2 L 150.1.12.1/32 is directly connected, 02:01:22, GigabitEthernet0/3/0/2 RP/0/7/CPU0:R1# As you can see, we have a RIP route to R2 L0 2.2.2.2/32 interface. Time for a PING! RP/0/7/CPU0:R1#ping 2.2.2.2 so l0 Thu Mar 29 20:40:15.606 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 2/3/4 ms RP/0/7/CPU0:R1# And like that RIP is configured. Cisco IOS XR Training 2012 Jeffrey Fry Page 57 FryGuy.Net

17. IS-IS
Time for some IS-IS routing! Between IS-IS and OSPF, those are the two most coming SP core routing protocols. RP/0/7/CPU0:R1#conf t Thu Mar 29 22:09:12.786 UTC First we need to name our process RP/0/7/CPU0:R1(config)#router isis LAB Then configure our Network Entity ( Area ) RP/0/7/CPU0:R1(config-isis)#net 49.0000.0000.0001.00 Then we assign the interfaces to the process, RP/0/7/CPU0:R1(config-isis)#int l0 RP/0/7/CPU0:R1(config-isis-if)#address-family RP/0/7/CPU0:R1(config-isis-if-af)#exit RP/0/7/CPU0:R1(config-isis-if)#address-family RP/0/7/CPU0:R1(config-isis-if-af)# exit RP/0/7/CPU0:R1(config-isis-if)#int g0/3/0/2 RP/0/7/CPU0:R1(config-isis-if)#address-family RP/0/7/CPU0:R1(config-isis-if-af)# exit RP/0/7/CPU0:R1(config-isis-if)#address-family RP/0/7/CPU0:R1(config-isis-if-af)# exit RP/0/7/CPU0:R1(config-isis-if)#exit as well as the address family. ipv4 ipv6 ipv4 ipv6

Notice I did not specify an IS-IS Level when I started, but we can set this to Level-2 RP/0/7/CPU0:R1(config-isis)#is-type level-2-only Now, when we show the config, you will notice Level-2 is set to the top of the config when applied, not in the order I entered it. This is the beauty of a staging config, you can enter some things in the wrong order but they will be applied in the correct order. RP/0/7/CPU0:R1(config-isis)#sh config Thu Mar 29 22:10:22.326 UTC Building configuration... !! IOS XR Configuration 4.1.1 router isis LAB is-type level-2-only net 49.0000.0000.0001.00 interface Loopback0 address-family ipv4 unicast ! address-family ipv6 unicast ! ! Cisco IOS XR Training 2012 Jeffrey Fry Page 58 FryGuy.Net

interface GigabitEthernet0/3/0/2 address-family ipv4 unicast ! address-family ipv6 unicast ! ! ! end Now, let us commit our changes. RP/0/7/CPU0:R1(config-isis)#commit RP/0/7/CPU0:R1(config-isis)#exit RP/0/7/CPU0:R1(config)#exit RP/0/7/CPU0:R1# Time to check our IS-IS adjancies. RP/0/7/CPU0:R1#sh isis adjacency Thu Mar 29 22:16:21.989 UTC
IS-IS LAB Level-2 adjacencies: System Id Interface GSR-R2 Gi0/3/0/2 SNPA State Hold Changed 9 NSF IPv4 IPv6 BFD BFD 00:05:52 Yes None None

00d0.7901.3a78 Up

Total adjacency count: 1 RP/0/7/CPU0:R1# We can see we are adjacent with R2 via IPv4 and IPv6. IS-IS routing table and then PING the loopback of R2: RP/0/7/CPU0:R1#sh route ipv4 isis Thu Mar 29 22:17:15.545 UTC Lets look at the IPv4

i L2 2.2.2.2/32 [115/20] via 150.1.12.2, 00:06:36, GigabitEthernet0/3/0/2 RP/0/7/CPU0:R1# RP/0/7/CPU0:R1#ping 2.2.2.2 so l0 Thu Mar 29 22:17:37.226 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/4 ms RP/0/7/CPU0:R1#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 59 FryGuy.Net

Ok, that worked now we can do the same for IPv6. First we should look at the IPv6 IS-IS routes and then ping the loopback of R2. RP/0/7/CPU0:R1#sh route ipv6 isis Thu Mar 29 22:17:43.918 UTC i L2 2001::2/128 [115/20] via fe80::2d0:79ff:fe01:3a78, 00:07:05, GigabitEthernet0/3/0/2 RP/0/7/CPU0:R1#ping 2001::2 so 2001::1 Thu Mar 29 22:17:49.763 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2001::2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 3/9/34 ms RP/0/7/CPU0:R1# Compare that to IOS ISIS config: Configure the process and set the level and NET R1(config)#router isis R1(config-router)#net 49.0000.0000.0001.00 R1(config-router)#is-type level-2 Then change context to the G0/1 interface and enable ISIS for IPv4 and IPv6 R1(config-router)#int g0/1 R1(config-if)#ip router isis R1(config-if)#ipv6 router isis Then change context to the Loop0 interface and enable ISIS for IPv4 and IPv6 R1(config-if)#int l0 R1(config-if)#ip router isis R1(config-if)#ipv6 router isis R1(config-if)#^Z Few more steps, and configuring things under the process make much more sense than under an interface.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 60 FryGuy.Net

18. BGP iBGP and eBGP

BGP, this is where it starts to get different with IOS XR. First up, configuring an iBGP peering with R2s 150.1.12.2 in AS1 and advertise our loopback interface. RP/0/7/CPU0:R1(config)# RP/0/7/CPU0:R1(config)#router bgp 1 Lets define the network we want to advertise, under the address family: RP/0/7/CPU0:R1(config-bgp)#address-family ipv4 unicast RP/0/7/CPU0:R1(config-bgp-af)#net 1.1.1.1/32 RP/0/7/CPU0:R1(config-bgp-af)#exit Now, we can configure the neighbor. Notice all the commands for the neighbor are under the neighbor now not next to the neighbor. RP/0/7/CPU0:R1(config-bgp)#nei 150.1.12.2 RP/0/7/CPU0:R1(config-bgp-nbr)#remote-as 1 RP/0/7/CPU0:R1(config-bgp-nbr)#address-family ipv4 unicast RP/0/7/CPU0:R1(config-bgp-nbr-af)#exit RP/0/7/CPU0:R1(config-bgp-nbr)#comm Thu Mar 29 22:47:05.147 UTC RP/0/7/CPU0:R1(config-bgp)#exit RP/0/7/CPU0:R1(config)#exit Now, time to see if we have a neighbor established: RP/0/7/CPU0:R1#sh bgp nei 150.1.12.2 Thu Mar 29 22:48:13.338 UTC BGP neighbor is 150.1.12.2 Remote AS 1, local AS 1, internal link Remote router ID 2.2.2.2 BGP state = Established, up for 00:00:24 Last read 00:00:24, Last read before reset 00:00:00 Hold time is 180, keepalive interval is 60 seconds Configured hold time: 180, keepalive: 60, min acceptable hold time: 3 Last write 00:00:24, attempted 19, written 19 Second last write 00:00:24, attempted 53, written 53 Last write before reset 00:00:00, attempted 0, written 0 Second last write before reset 00:00:00, attempted 0, written 0 Last write pulse rcvd Mar 29 22:47:49.296 last full not set pulse count 4 Last write pulse rcvd before reset 00:00:00 Socket not armed for io, armed for read, armed for write Last write thread event before reset 00:00:00, second last 00:00:00 Last KA expiry before reset 00:00:00, second last 00:00:00 Last KA error before reset 00:00:00, KA not sent 00:00:00 Last KA start before reset 00:00:00, second last 00:00:00 Cisco IOS XR Training 2012 Jeffrey Fry Page 61 FryGuy.Net

Precedence: internet Neighbor capabilities: Route refresh: advertised and received 4-byte AS: advertised and received Address family IPv4 Unicast: advertised and received Received 2 messages, 0 notifications, 0 in queue Sent 2 messages, 0 notifications, 0 in queue Minimum time between advertisement runs is 0 secs For Address Family: IPv4 Unicast BGP neighbor version 0 Update group: 0.2 Filter-group: 0.1 No Refresh request being processed Route refresh request: received 0, sent 0 0 accepted prefixes, 0 are bestpaths Cumulative no. of prefixes denied: 0. Prefix advertised 0, suppressed 0, withdrawn 0 Maximum prefixes allowed 524288 Threshold for warning message 75%, restart interval 0 min AIGP is enabled An EoR was not received during read-only mode Last ack version 1, Last synced ack version 0 Outstanding version objects: current 0, max 0 Additional-paths operation: None Connections established 1; dropped 0 Local host: 150.1.12.1, Local port: 33432 Foreign host: 150.1.12.2, Foreign port: 179 Last reset 00:00:00 Cool, neighbor is up and active. Now, time to check our BGP summary to see what routes we have. RP/0/7/CPU0:R1#sh ip bgp Thu Mar 29 22:48:51.876 UTC BGP router identifier 1.1.1.1, local AS number 1 BGP generic scan interval 60 secs BGP table state: Active Table ID: 0xe0000000 RD version: 4 BGP main routing table version 4 BGP scan interval 60 secs Status codes: s suppressed, d damped, h history, * valid, > best i - internal, r RIB-failure, S stale Origin codes: i - IGP, e - EGP, ? - incomplete Network Next Hop Metric LocPrf Weight Path *> 1.1.1.1/32 0.0.0.0 0 32768 i *>i2.2.2.2/32 150.1.12.2 0 100 0 i Processed 2 prefixes, 2 paths RP/0/7/CPU0:R1# Cisco IOS XR Training 2012 Jeffrey Fry Page 62 FryGuy.Net

Cool, we have a route to R2 Loopback interface. Lets PING it! RP/0/7/CPU0:R1#ping 2.2.2.2 so l0 Thu Mar 29 22:52:17.899 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/3 ms RP/0/7/CPU0:R1# We have connectivity! Here is the IOS XR Config: RP/0/7/CPU0:R1#sh run | begin bgp Thu Mar 29 22:56:17.937 UTC Building configuration... router bgp 1 address-family ipv4 unicast network 1.1.1.1/32 ! neighbor 150.1.12.2 remote-as 1 address-family ipv4 unicast ! ! ! end RP/0/7/CPU0:R1# Here is the same IOS config. With a single neighbor it is pretty simple. router bgp 1 network 1.1.1.1 mask 255.255.255.255 neighbor 150.1.12.2 remote-as 1 Now for eBGP here is where it starts gets interesting! First we need to configure an IGP so that we can establish Loopback connectivity for this we will use ISIS: RP/0/7/CPU0:R1(config)#router ISIS LAB RP/0/7/CPU0:R1(config-isis)#net 49.0000.0000.0001.00 RP/0/7/CPU0:R1(config-isis)#interface l0 RP/0/7/CPU0:R1(config-isis-if)#address-family ipv4 RP/0/7/CPU0:R1(config-isis-if)#exit RP/0/7/CPU0:R1(config-isis)#interface g0/3/0/2 RP/0/7/CPU0:R1(config-isis-if)#address-family ipv4 RP/0/7/CPU0:R1(config-isis-if)#exit RP/0/7/CPU0:R1(config-isis)#is-type level-2 RP/0/7/CPU0:R1(config-isis)#commit RP/0/7/CPU0:R1(config-isis-if-af)#exit RP/0/7/CPU0:R1(config-isis-if)#exit Cisco IOS XR Training 2012 Jeffrey Fry Page 63 FryGuy.Net

RP/0/7/CPU0:R1(config-isis)#exit Now we need to configure an interface to advertise via BGP here we will create Loop100 with an IP of 100.100.100.100/32 RP/0/7/CPU0:R1(config)#int loop100 RP/0/7/CPU0:R1(config-if)#ip add 100.100.100.100/32 RP/0/7/CPU0:R1(config-if)#comm Thu Mar 29 23:12:31.681 UTC RP/0/7/CPU0:R1(config-if)#exit Now to configured eBGP. We will peer with R2 loopbacks (2.2.2.2) and their remote AS of 2. First we define our BGP processed (ID 1) RP/0/7/CPU0:R1(config)# RP/0/7/CPU0:R1(config-if)#router bgp 1 Define the networks we want to advertise RP/0/7/CPU0:R1(config-bgp)#address-family ipv4 unicast RP/0/7/CPU0:R1(config-bgp-af)#net 100.100.100.100/32 RP/0/7/CPU0:R1(config-bgp-af)#exit Now we can configure our neighbor RP/0/7/CPU0:R1(config-bgp)#nei 2.2.2.2 RP/0/7/CPU0:R1(config-bgp-nbr)#remote-as 2 RP/0/7/CPU0:R1(config-bgp-nbr)#ebgp-multihop RP/0/7/CPU0:R1(config-bgp-nbr)#up loopback 0 RP/0/7/CPU0:R1(config-bgp-nbr)#address-family ipv4 un And finally commit our changes. RP/0/7/CPU0:R1(config-bgp-nbr-af)#comm Thu Mar 29 23:18:06.455 UTC RP/0/7/CPU0:R1(config-bgp-nbr-af)#exit RP/0/7/CPU0:R1(config-bgp-nbr)#exit RP/0/7/CPU0:R1(config-bgp)#exit RP/0/7/CPU0:R1(config)#exit RP/0/7/CPU0:R1# Ok, now that we have that configured time to look at our routing table, we should see a route to 200.200.200.200/32. RP/0/7/CPU0:R1#sh ip route Thu Mar 29 23:24:25.533 UTC Codes: C - connected, S - static, R - RIP, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP i - ISIS, L1 - IS-IS level-1, L2 - IS-IS level-2 ia - IS-IS inter area, su - IS-IS summary null, * - candidate default U - per-user static route, o - ODR, L - local, G - DAGR A - access/subscriber, (!) - FRR Backup path Cisco IOS XR Training 2012 Jeffrey Fry Page 64 FryGuy.Net

Gateway of last resort is not set L 1.1.1.1/32 is directly connected, 02:26:47, Loopback0 i L2 2.2.2.2/32 [115/20] via 150.1.12.2, 00:13:05, GigabitEthernet0/3/0/2 L 100.100.100.100/32 is directly connected, 00:11:53, Loopback100 C 150.1.12.0/24 is directly connected, 02:27:12, GigabitEthernet0/3/0/2 L 150.1.12.1/32 is directly connected, 02:26:47, GigabitEthernet0/3/0/2 RP/0/7/CPU0:R1# Hmm, no route why is that? Is the neighbor up? Lets check: RP/0/7/CPU0:R1#sh ip bgp summ Thu Mar 29 23:25:12.041 UTC BGP router identifier 1.1.1.1, local AS number 1 BGP generic scan interval 60 secs BGP table state: Active Table ID: 0xe0000000 RD version: 7 BGP main routing table version 7 BGP scan interval 60 secs BGP is operating in STANDALONE mode.
Process Speaker RcvTblVer 7 bRIB/RIB 7 LabelVer 7 ImportVer 7 SendTblVer 7 StandbyVer 7

Some configured eBGP neighbors (under default or non-default vrfs) do not have both inbound and outbound policies configured for IPv4 Unicast address family. These neighbors will default to sending and/or receiving no routes and are marked with '!' in the output below. Use the 'show bgp neighbor <nbr_address>' command for details.
Neighbor 2.2.2.2 Spk 0 AS MsgRcvd MsgSent 2 7 6 TblVer 7 InQ OutQ Up/Down 0 0 00:03:09 St/PfxRcd 0!

RP/0/7/CPU0:R1# Yup, we are up for over 3 minutes now but wait, we have an ! mark there no routes received. It says to use the show bgp neighbors address for details. Lets see what that says. RP/0/7/CPU0:R1#sh bgp neighbors 2.2.2.2 Thu Mar 29 23:26:12.572 UTC BGP neighbor is 2.2.2.2 Remote AS 2, local AS 1, external link Remote router ID 2.2.2.2 BGP state = Established, up for 00:04:10 Last read 00:00:05, Last read before reset 00:00:00 Cisco IOS XR Training 2012 Jeffrey Fry Page 65 FryGuy.Net

Hold time is 180, keepalive interval is 60 seconds Configured hold time: 180, keepalive: 60, min acceptable hold time: 3 Last write 00:00:05, attempted 19, written 19 Second last write 00:01:05, attempted 19, written 19 Last write before reset 00:00:00, attempted 0, written 0 Second last write before reset 00:00:00, attempted 0, written 0 Last write pulse rcvd Mar 29 23:26:07.793 last full not set pulse count 14 Last write pulse rcvd before reset 00:00:00 Socket not armed for io, armed for read, armed for write Last write thread event before reset 00:00:00, second last 00:00:00 Last KA expiry before reset 00:00:00, second last 00:00:00 Last KA error before reset 00:00:00, KA not sent 00:00:00 Last KA start before reset 00:00:00, second last 00:00:00 Precedence: internet Enforcing first AS is enabled Neighbor capabilities: Route refresh: advertised and received 4-byte AS: advertised and received Address family IPv4 Unicast: advertised and received Received 8 messages, 0 notifications, 0 in queue Sent 7 messages, 0 notifications, 0 in queue Minimum time between advertisement runs is 30 secs For Address Family: IPv4 Unicast BGP neighbor version 7 Update group: 0.2 Filter-group: 0.1 No Refresh request being processed eBGP neighbor with no inbound or outbound policy; defaults to 'drop' Route refresh request: received 0, sent 0 0 accepted prefixes, 0 are bestpaths Cumulative no. of prefixes denied: 1. No policy: 1, Failed RT match: 0 By ORF policy: 0, By policy: 0 Prefix advertised 0, suppressed 0, withdrawn 0 Maximum prefixes allowed 524288 Threshold for warning message 75%, restart interval 0 min An EoR was not received during read-only mode Last ack version 7, Last synced ack version 0 Outstanding version objects: current 0, max 0 Additional-paths operation: None Connections established 1; dropped 0 Local host: 1.1.1.1, Local port: 58277 Foreign host: 2.2.2.2, Foreign port: 179 Last reset 00:00:00 External BGP neighbor may be up to 255 hops away. RP/0/7/CPU0:R1# Ahh, the neighbor is up but there is a line that says: eBGP neighbor with no inbound or outbound policy; defaults to 'drop' Cisco IOS XR Training 2012 Jeffrey Fry Page 66 FryGuy.Net

Here is the first difference with IOS XR eBGP peers must have a RoutePolicy (route-map) configured to permit routes in and out of them. Instead of a route-map like IOS, IOS XR uses a Route Policy Language (RPL) that is more powerful and easier than IOS. Lets configure a very simple one to pass everything: RP/0/7/CPU0:R1(config)#route-policy PASS RP/0/7/CPU0:R1(config-rpl)#pass RP/0/7/CPU0:R1(config-rpl)#exit RP/0/7/CPU0:R1(config)#commit Thu Mar 29 23:28:08.400 UTC Cool that was easy. Now lets apply that to the eBGP neighbor: RP/0/7/CPU0:R1(config)#router bgp 1 RP/0/7/CPU0:R1(config-bgp)#nei 2.2.2.2 RP/0/7/CPU0:R1(config-bgp-nbr)#address-family ipv4 unicast RP/0/7/CPU0:R1(config-bgp-nbr-af)#route-policy PASS out RP/0/7/CPU0:R1(config-bgp-nbr-af)#route-policy PASS in RP/0/7/CPU0:R1(config-bgp-nbr-af)#commit Thu Mar 29 23:28:32.865 UTC Now, lets look at the routing table for BGP RP/0/7/CPU0:R1#sh route ipv4 bgp Thu Mar 29 23:29:43.865 UTC B 200.200.200.200/32 [20/0] via 2.2.2.2, 00:01:06 RP/0/7/CPU0:R1# Cool, we have a route to R2s Loopback100 interface. PING time! RP/0/7/CPU0:R1#ping 200.200.200.200 sou loop100 Thu Mar 29 23:30:10.013 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 200.200.200.200, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 2/3/5 ms RP/0/7/CPU0:R1# Look at that, we have connectivity! A similar IOS config would look like this: router bgp 1 no synchronization bgp log-neighbor-changes network 100.100.100.100 mask 255.255.255.255 neighbor 2.2.2.2 remote-as 2 neighbor 2.2.2.2 ebgp-multihop 255 neighbor 2.2.2.2 route-map PASS in neighbor 2.2.2.2 route-map PASS out Cisco IOS XR Training 2012 Jeffrey Fry Page 67 FryGuy.Net

ip prefix-list PASS seq 5 permit 0.0.0.0/0 le 32 route-map PASS permit 10 match ip address prefix-list PASS

Cisco IOS XR Training 2012 Jeffrey Fry

Page 68 FryGuy.Net

19. Route Filtering

Ok, now that BGP has been covered, lets talk about filtering routes received from our neighbor. Here I have created some additional Loopbacks on R2 that are being advertised to R1: RP/0/7/CPU0:R1#sh ip route bgp Fri Mar 30 13:13:36.797 UTC B 200.100.200.100/32 B 200.200.200.200/32 B 200.200.200.203/32 B 200.200.200.204/32 B 200.200.200.205/32 B 200.200.200.206/32 B 200.200.200.207/32 B 200.200.200.208/32 B 200.200.200.209/32 B 200.200.200.210/32 RP/0/7/CPU0:R1# [20/0] [20/0] [20/0] [20/0] [20/0] [20/0] [20/0] [20/0] [20/0] [20/0] via via via via via via via via via via 2.2.2.2, 2.2.2.2, 2.2.2.2, 2.2.2.2, 2.2.2.2, 2.2.2.2, 2.2.2.2, 2.2.2.2, 2.2.2.2, 2.2.2.2, 00:00:42 13:45:00 00:00:42 00:00:42 00:00:42 00:00:42 00:00:42 00:00:42 00:00:42 00:00:42

As you can see, we are getting a bunch of 200.200.200.x/32 routes now as well as a 200.100.200.100/32 route. For this exercise, lets filter our all the 200.200.200.x routes we are receiving from our neighbor. Ok, lets create a prefix-set for the loopback we want to permit: RP/0/7/CPU0:R1(config)#conf t RP/0/7/CPU0:R1(config)#prefix-set R2Loopbacks In IOS XR you can add comments via the # RP/0/7/CPU0:R1(config-pfx)## These are the R2 Loopbacks that we will allow RP/0/7/CPU0:R1(config-pfx)#200.100.200.100/32 RP/0/7/CPU0:R1(config-pfx)#end-set Now that we have the prefix-set done we can create the route-policy RP/0/7/CPU0:R1(config)#route-policy R2Loopbacks Notice that IOS XR can use IF statements, you can just imagine how powerful IF and ELSE statements can be when route filtering. RP/0/7/CPU0:R1(config-rpl)#if destination in R2Loopbacks then RP/0/7/CPU0:R1(config-rpl-if)#pass RP/0/7/CPU0:R1(config-rpl-if)#endif RP/0/7/CPU0:R1(config-rpl)#end-policy After we end the policy, we need to commit it RP/0/7/CPU0:R1(config)#commit

Cisco IOS XR Training 2012 Jeffrey Fry

Page 69 FryGuy.Net

Now that we have the policy committed with no errors, we can apply it to the neighbor. We could have waited to commit, but I chose to commit there to make sure all was OK. RP/0/7/CPU0:R1(config)#router bgp 1 RP/0/7/CPU0:R1(config-bgp)#neighbor 2.2.2.2 RP/0/7/CPU0:R1(config-bgp-nbr)#address-family ipv4 un RP/0/7/CPU0:R1(config-bgp-nbr-af)#route-policy R2Loopbacks in RP/0/7/CPU0:R1(config-bgp-nbr-af)#exit RP/0/7/CPU0:R1(config-bgp-nbr)#exit RP/0/7/CPU0:R1(config-bgp)#exi RP/0/7/CPU0:R1(config)#commit Fri Mar 30 13:27:01.945 UTC RP/0/7/CPU0:R1(config)# Now, lets look at our BGP routing table: RP/0/7/CPU0:R1#sh ip route bgp Fri Mar 30 13:27:22.601 UTC B 200.100.200.100/32 [20/0] via 2.2.2.2, 00:14:28 RP/0/7/CPU0:R1# There we go, only getting the 200.100.200.100/32 from R2 now. In IOS this would have looked like: R1(config)#ip prefix-list R2Loopbacks permit 200.100.200.100/32 R1(config)#route-map R2Loopbacks R1(config-route-map)#match ip add prefix-list R2Loopbacks R1(config-route-map)#exit R1(config)#router bgp 1 R1(config-router)#nei 2.2.2.2 route-map R2Loopbacks in R1(config-router)#^Z R1#sh ip route b *Mar 30 14:08:53.048: %SYS-5-CONFIG_I: Configured from console by console (After a few minutes waiting for BGP) R1#sh ip route bgp 200.100.200.0/32 is subnetted, 1 subnets B 200.100.200.100 [20/0] via 2.2.2.2, 00:00:20 R1# While that might not be so bad, the power of RPL grows. This is just a quick intro; future posts will have more and more about RPL. Some other things that we might see are: route-policy check ASPath if as-path passes-through 65500 then drop else pass endif end-policy Cisco IOS XR Training 2012 Jeffrey Fry Page 70 FryGuy.Net

20. VRF lite and Dot1Q Trunks

Ok, time for some VRF lite basics and we can throw in some Dot1Q trunks to go with it. First, lets create our VRF called LAB RP/0/7/CPU0:R1(config)#vrf LAB Now we need to enable the address family for this VRF, there IPv4 Unicast RP/0/7/CPU0:R1(config-vrf)#address-family ipv4 un RP/0/7/CPU0:R1(config-vrf-af)#exit Now we need to enable the IPv6 address family for this VRF RP/0/7/CPU0:R1(config-vrf)#address-family ipv6 unicast Now we can create our Dot1Q trunk to the other router: RP/0/7/CPU0:R1(config-vrf-af)#int g0/3/0/3.100 Little different then IOS, but this actually makes more sense RP/0/7/CPU0:R1(config-subif)#dot1q vlan 100 RP/0/7/CPU0:R1(config-subif)#ip add 150.1.21.1/24 RP/0/7/CPU0:R1(config-subif)#ipv6 add 2001:1:1:21::1/64 RP/0/7/CPU0:R1(config-subif)#vrf LAB Notice that I applied the VRF LAB command after configuring the IP addresses. If this was IOS, I would have lost all that work but since its IOS XR, nothing takes effect until after you COMMIT the changes. Lets look at what will be applied and then commit it. RP/0/7/CPU0:R1(config-subif)#show config Fri Mar 30 14:12:06.649 UTC Building configuration... !! IOS XR Configuration 4.1.1 vrf LAB address-family ipv4 unicast ! address-family ipv6 unicast ! ! interface GigabitEthernet0/3/0/3.100 vrf LAB ipv4 address 150.1.21.1 255.255.255.0 ipv6 address 2001:1:1:21::1/64 dot1q vlan 100 ! end Cisco IOS XR Training 2012 Jeffrey Fry Page 71 FryGuy.Net

RP/0/7/CPU0:R1(config-subif)#comm Fri Mar 30 14:12:12.700 UTC RP/0/7/CPU0:R1(config-subif)# RP/0/7/CPU0:R1# Now we should try to PING over the VRF. Remember, when PINGing over a VRF you need to specify the VRF in the PING command. RP/0/7/CPU0:R1#ping vrf LAB 2001:1:1:21::2 Fri Mar 30 14:17:37.291 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2001:1:1:21::2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 3/13/49 ms RP/0/7/CPU0:R1#ping vrf LAB 150.1.21.2 Fri Mar 30 14:17:40.858 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 150.1.21.2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 2/3/5 ms RP/0/7/CPU0:R1# Full connectivity, there we go! Ok, now we can toss a routing protocol over the link say OSPF PID 100 First though, create a loopback we can advertise over that VRF, RP/0/7/CPU0:R1#conf t Fri Mar 30 14:39:31.441 UTC RP/0/7/CPU0:R1(config)#int loop1000 RP/0/7/CPU0:R1(config-if)#ip add 111.111.111.111/32 RP/0/7/CPU0:R1(config-if)#vrf LAB Now we can configure OSPF First, define the process identifier, here 100 RP/0/7/CPU0:R1(config-if)#router ospf 100 Now to change context to VRF LAB RP/0/7/CPU0:R1(config-ospf)#vrf LAB Configure our Router-ID RP/0/7/CPU0:R1(config-ospf-vrf)#router-id 111.111.111.111 RP/0/7/CPU0:R1(config-ospf-vrf)#area 0.0.0.0 And then assign the interfaces to the area RP/0/7/CPU0:R1(config-ospf-vrf-ar)#int loop1000 RP/0/7/CPU0:R1(config-ospf-vrf-ar)#int g0/3/0/3.100 RP/0/7/CPU0:R1(config-ospf-vrf-ar-if)#exit RP/0/7/CPU0:R1(config-ospf-vrf-ar)#exit

Cisco IOS XR Training 2012 Jeffrey Fry

Page 72 FryGuy.Net

Now we can check our config: RP/0/7/CPU0:R1(config-ospf-vrf)#show configuration Fri Mar 30 14:40:49.074 UTC Building configuration... !! IOS XR Configuration 4.1.1 interface Loopback1000 vrf LAB ipv4 address 111.111.111.111 255.255.255.255 ! router ospf 100 vrf LAB router-id 111.111.111.111 area 0.0.0.0 interface Loopback1000 ! interface GigabitEthernet0/3/0/3.100 ! ! ! ! end And finally commit the changes: RP/0/7/CPU0:R1(config-ospf-vrf)#commit Fri Mar 30 14:40:57.984 UTC Now, lets check our routes: RP/0/7/CPU0:R1#sh route vrf LAB ipv4 Fri Mar 30 14:41:40.746 UTC Codes: C - connected, S - static, R - RIP, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP i - ISIS, L1 - IS-IS level-1, L2 - IS-IS level-2 ia - IS-IS inter area, su - IS-IS summary null, * - candidate default U - per-user static route, o - ODR, L - local, G - DAGR A - access/subscriber, (!) - FRR Backup path Gateway of last resort is not set L 111.111.111.111/32 is directly connected, 00:00:40, Loopback1000 C 150.1.21.0/24 is directly connected, 00:29:26, GigabitEthernet0/3/0/3.100 L 150.1.21.1/32 is directly connected, 00:29:26, GigabitEthernet0/3/0/3.100 O 222.222.222.222/32 [110/2] via 150.1.21.2, 00:00:37, GigabitEthernet0/3/0/3.100 RP/0/7/CPU0:R1# Cisco IOS XR Training 2012 Jeffrey Fry Page 73 FryGuy.Net

Cool, we have a route to R2s Loop1000 of 222.222.222.222/32. time!

Ping test

RP/0/7/CPU0:R1#ping vrf LAB 222.222.222.222 so loop1000 Fri Mar 30 14:41:50.331 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 222.222.222.222, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 2/2/4 ms RP/0/7/CPU0:R1# There you go, we have connectivity!

Cisco IOS XR Training 2012 Jeffrey Fry

Page 74 FryGuy.Net

21. Basic MPLS - LDP

Ok, time for some MPLS! For this lab I will be using the OSPF routing protocol first, then we can move to IS-IS next. All we will be doing here is configuring LDP First up, lets enabled LDP on all OSPF interfaces. Normally you would do this under each interface, but here we will use the MPLS LDP AUTOCONFIG command. This is a good command to use as it ensures that you do not miss configuring LDP on an interface. RP/0/7/CPU0:R1# RP/0/7/CPU0:R1#conf t Sun Apr 1 18:58:04.084 UTC RP/0/7/CPU0:R1(config)#router ospf LAB Under the OSPF LAB process, we need to configure mpls ldp autoconfig and then commit it. RP/0/7/CPU0:R1(config-ospf)#mpls ldp auto RP/0/7/CPU0:R1(config-ospf)#commit Sun Apr 1 18:58:12.277 UTC RP/0/7/CPU0:R1(config-ospf)#exit RP/0/7/CPU0:R1(config)#exit Ok, now lets see what interfaces have LDP on them RP/0/7/CPU0:R1#sh mpls ldp int Sun Apr 1 18:58:18.902 UTC MPLS LDP application must be enabled to use this command Ahh, we configured the command but never enabled MPLS LDP. Remember, if a process is not needed it does not run. So, lets enable the process RP/0/7/CPU0:R1#conf t Sun Apr 1 18:58:22.811 UTC RP/0/7/CPU0:R1(config)#mpls ldp Now, one thing to note on IOS XR, LDP is the only label protocol supported, TDP is not available. RP/0/7/CPU0:R1(config-ldp)#label ? accept Configure inbound label acceptance control advertise Configure outbound label advertisement control allocate Configure label allocation control <cr> RP/0/7/CPU0:R1(config-ldp-lbl)#tag? ^ % Invalid input detected at '^' marker. RP/0/7/CPU0:R1(config-ldp)#comm

Cisco IOS XR Training 2012 Jeffrey Fry

Page 75 FryGuy.Net

Now, lets check our interfaces and check for a neighbor. RP/0/7/CPU0:R1#sh mpls ldp int Sun Apr 1 19:04:23.402 UTC Interface GigabitEthernet0/3/0/2 (0x4000500) Enabled via config: IGP Auto-config Interface GigabitEthernet0/3/0/3.100 (0x4000700) Disabled Yup, we see its enabled on g0/3/0/2 via IGP Auto-config. Now, lets check for a neighbor RP/0/7/CPU0:R1#sh mpls ldp neighbor Sun Apr 1 19:04:27.582 UTC Nice!

Peer LDP Identifier: 2.2.2.2:0 TCP connection: 2.2.2.2:35051 - 1.1.1.1:646 Graceful Restart: No Session Holdtime: 180 sec State: Oper; Msgs sent/rcvd: 13/24; Downstream-Unsolicited Up time: 00:05:57 LDP Discovery Sources: GigabitEthernet0/3/0/2 Addresses bound to this peer: 2.2.2.2 150.1.12.2 200.100.200.100 200.200.200.200 200.200.200.201 200.200.200.202 200.200.200.203 200.200.200.204 200.200.200.205 200.200.200.206 200.200.200.207 200.200.200.208 200.200.200.209 200.200.200.210 RP/0/7/CPU0:R1# There you can see we have LDP neighbor with router-id 2.2.2.2 (R2) on G0/3/0/2. You can also see the ports we are using for this communication. Our local port is 646 and the remote port is 35051. LDP Authentication Ok, now onto neighbor password for LDP (both directed and all) Lets configure a password for our neighbor, 2.2.2.2, of cisco RP/0/7/CPU0:R1#conf t Sun Apr 1 19:39:35.480 UTC This is done under the LDP section RP/0/7/CPU0:R1(config)#mpls ldp RP/0/7/CPU0:R1(config-ldp)#nei 2.2.2.2 password cisco RP/0/7/CPU0:R1(config-ldp)#comm Sun Apr 1 19:40:04.498 UTC

Cisco IOS XR Training 2012 Jeffrey Fry

Page 76 FryGuy.Net

Now here is something different than normal authentication the LDP session resets and you would need to clear the LDP session and RP/0/7/CPU0:Apr 1 19:40:06.205 : tcp[400]: digest from 2.2.2.2:57032 to 1.1.1.1:646

IOS, as soon as you enable enables the password. With IOS, allow it to re-establish. %IP-TCP-3-BADAUTH : Invalid MD5

Ok, I made the change to R2 so the passwords match, now we can look at our neighbor RP/0/7/CPU0:R1#sh mpls ldp neighbor Sun Apr 1 19:40:33.961 UTC Peer LDP Identifier: 2.2.2.2:0 TCP connection: 2.2.2.2:57491 - 1.1.1.1:646; MD5 on Graceful Restart: No Session Holdtime: 180 sec State: Oper; Msgs sent/rcvd: 7/18; Downstream-Unsolicited Up time: 00:00:10 LDP Discovery Sources: GigabitEthernet0/3/0/2 Addresses bound to this peer: 2.2.2.2 150.1.12.2 200.100.200.100 200.200.200.200 200.200.200.201 200.200.200.202 200.200.200.203 200.200.200.204 200.200.200.205 200.200.200.206 200.200.200.207 200.200.200.208 200.200.200.209 200.200.200.210 RP/0/7/CPU0:R1# As you can see, next to the TCP connection, it now says MD5 on. nothing was after the port number. Previously

You can also configure a password for any LDP session, that is done like follows: RP/0/7/CPU0:R1#conf t Sun Apr 1 19:40:45.561 UTC RP/0/7/CPU0:R1(config)#mpls ldp RP/0/7/CPU0:R1(config-ldp)#neighbor ? A.B.C.D IP address of neighbor password Configure password for MD5 authentication for all neighbors RP/0/7/CPU0:R1(config-ldp)#neighbor password cisco RP/0/7/CPU0:R1(config-ldp)#comm Sun Apr 1 19:40:57.167 UTC Now any LDP session must have a password. Now remember this in case you need to do directed LDP session some time down the road.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 77 FryGuy.Net

ISIS Changing Metrics on an interface. To change a metric on an interface in IS-IS, it is pretty simple. Just like before, all configuration are done under the routing protocol section of the config, interface subsection, and address family. RP/0/7/CPU0:R1#conf t Sun Apr 1 22:40:33.251 UTC RP/0/7/CPU0:R1(config)#router ISIS LAB RP/0/7/CPU0:R1(config-isis)#int g0/3/0/2 RP/0/7/CPU0:R1(config-isis-if)#address-family ipv4 un RP/0/7/CPU0:R1(config-isis-if-af)#metric 20 RP/0/7/CPU0:R1(config-isis-if-af)# And to check: RP/0/7/CPU0:R1#sh isis interface g0/3/0/2 Sun Apr 1 22:42:11.124 UTC GigabitEthernet0/3/0/2 Enabled Adjacency Formation: Enabled Prefix Advertisement: Enabled <--SNIP Information removed for brevity --> IPv4 Unicast Topology: Adjacency Formation: Prefix Advertisement: Metric (L1/L2): MPLS LDP Sync (L1/L2): IPv6 Unicast Topology: Adjacency Formation: Prefix Advertisement: Metric (L1/L2): MPLS LDP Sync (L1/L2): IPv4 Address Family: Protocol State: Forwarding Address(es): Global Prefix(es): IPv6 Address Family: Protocol State: Forwarding Address(es): Global Prefix(es): Enabled Running Running 20/20 Disabled/Disabled Enabled Running Running 10/10 Disabled/Disabled Enabled Up 150.1.12.1 150.1.12.0/24 Enabled Up fe80::201:c9ff:fee8:dd7c 2001:1:1:12::/64

LSP transmit timer expires in 0 ms LSP transmission is idle Can send up to 9 back-to-back LSPs in the next 0 ms RP/0/7/CPU0:R1# As you can see, IPv4 now has a metric of 20 whereas IPv6 has the default metric of 10. Cisco IOS XR Training 2012 Jeffrey Fry Page 78 FryGuy.Net

Passive Interfaces Now, typically in ISIS you make the loopback interface passive. To make an interface passive, is very simple. RP/0/7/CPU0:R1#conf t Sun Apr 1 22:45:10.308 UTC RP/0/7/CPU0:R1(config)#router isis LAB Change to the interface uder the protocol RP/0/7/CPU0:R1(config-isis)#int loop0 And set it as passive. RP/0/7/CPU0:R1(config-isis-if)#passive RP/0/7/CPU0:R1(config-isis-if)#commit Authentication Time to configure IS-IS authentication. Again, all this is done under the routing process makes keeping all relevant changes very close together. RP/0/7/CPU0:R1(config)#router ISIS LAB RP/0/7/CPU0:R1(config-isis)#inter g0/3/0/2 Now, to configure authentication we need to set the hello-password. As you can see we have some options listed but for this lab we will use hmac-md5. RP/0/7/CPU0:R1(config-isis-if)#hello-password ? WORD The unencrypted (clear text) hello password accept Use password for incoming authentication only clear Specifies an unencrypted password will follow encrypted Specifies an encrypted password will follow hmac-md5 Use HMAC-MD5 authentication keychain Specifies a Key Chain name will follow text Use cleartext password authentication RP/0/7/CPU0:R1(config-isis-if)#hello-password hmac-md5 cisco Now before we commit, lets look at our neighbors RP/0/7/CPU0:R1(config-isis-if)#do show isis neighbors Sun Apr 1 22:49:07.800 UTC IS-IS LAB neighbors: System Id Interface GSR-R2 Gi0/3/0/2

SNPA State Holdtime Type IETF-NSF 00d0.7901.3a78 Up 7 L2 Capable

Total neighbor count: 1 RP/0/7/CPU0:R1(config-isis-if)#commit Sun Apr 1 22:49:10.443 UTC RP/0/7/CPU0:R1(config-isis-if)# RP/0/7/CPU0:R1#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 79 FryGuy.Net

You may or may not have to clear the process; I did not and was able to catch this in the log with regards to ISIS neighbor authentication failure. RP/0/7/CPU0:Apr 1 22:52:58.265 : isis[1003]: %ROUTING-ISIS-5AUTH_FAILURE_DROP : Dropped L2 LAN IIH from GigabitEthernet0/3/0/2 SNPA 00d0.7901.3a78 due to authentication TLV not found Once I configured the password on the other router, we have neighbors again! RP/0/7/CPU0:R1#sh isis neighbors Sun Apr 1 22:55:55.066 UTC IS-IS LAB neighbors: System Id Interface GSR-R2 Gi0/3/0/2 Total neighbor count: 1 RP/0/7/CPU0:R1# SNPA State Holdtime Type IETF-NSF 00d0.7901.3a78 Up 7 L2 Capable

Cisco IOS XR Training 2012 Jeffrey Fry

Page 80 FryGuy.Net

22. MPLS VPN

Next up is MPLS VPN; actually VPNv4 routes are what these actually are since we are only passing IPv4 traffic in this example. So we have this diagram below CE1 and CE2 are the customer routers and both are running OSPF in Area 0. Then need to talk to each other but do not have a direct connection available, so they have contracted us to provide connectivity via MPLS between them. What we will now do is build a pseudo MPLS network between PE1 and PE2, establish an iBGP peering, create the associated customer VRF and then peer with the customer via OSPF Area 0.

CE1

CE2

Router

Router

OSPF Area 0

OSPF Area 0

ISIS Level 2 MPLS

Router

Router

PE1

PE2

So, first up lets configure CE1 using an IP of 10.3.3.3/32 for the loopback and 10.1.13.3/24 for the link facing PE1. CE1#conf t Enter configuration commands, one per line.

End with CNTL/Z.

First up, Loopback 0. Since this is IOS, you will need to use the full dotted decimal subnet mask CE1(config)#int loop0 CE1(config-if)#ip add 10.3.3.3 255.255.255.255

Cisco IOS XR Training 2012 Jeffrey Fry

Page 81 FryGuy.Net

Now for the interface facing the PE (here f0/0) CE1(config-if)#int f0/0 CE1(config-if)#ip add 10.1.13.3 255.255.255.0 CE1(config-if)#no shut CE1(config-if)#^Z CE1# Ok, now lets get CE2 done since it basically the same but here we will use 10.4.4.4/32 and 10.4.24.4/24 CE2#conf t Enter configuration commands, one per line. End with CNTL/Z. First up, Loopback 0. Since this is IOS, you will need to use the full dotted decimal subnet mask CE2(config)#int loop0 CE2(config-if)#ip add 10.3.4.4 255.255.255.255 Now for the interface facing the PE (here f0/0) CE2(config-if)#int f0/0 CE2(config-if)#ip add 10.1.24.4 255.255.255.0 CE2(config-if)#no shut CE2(config-if)#^Z CE2# Now we can do the OSPF configs for these routers. Since this is a lab, I am just going to put the 10/8 network in Area 0. So, first up - CE1 CE1(config)#router ospf 1 CE1(config-router)#net 10.0.0.0 0.255.255.255 a 0 CE1(config-router)# And now CE2: CE2(config)# router ospf 1 CE2(config-router)#net 10.0.0.0 0.255.255.255 a 0 CE2(config-router)# Easy part done, now to build the PE nework. For the PE network we are going to use ISIS for our internal routing protocol and then use BGP on top of that to connect the routers together to pass the VPNv4 routes. Why ISIS you ask? It is because you can use one process for IPv4 and IPv6 traffic. With OSPF you need to run two processes, OSPF for IPv4 and OSPFv3 for IPv6. A single process makes it easier to support as well as if new protocols come around, ISIS wont really care since it is not IP based (CLNS based).

Cisco IOS XR Training 2012 Jeffrey Fry

Page 82 FryGuy.Net

Now it is time to get some IP addresses on PE1. We will use G0/1/0/11 for connection to PE2 and also create Loopback0. The IPs for the connection to PE2 will be 150.1.12.0/24 and the Loopback will be 150.1.1.1/32. RP/0/RSP0/CPU0:R1#conf t Fri Apr 20 00:34:18.971 UTC RP/0/RSP0/CPU0:R1(config)#int g0/1/0/11 RP/0/RSP0/CPU0:R1(config-if)#ip add 150.1.12.1/24 RP/0/RSP0/CPU0:R1(config-if)#no shut RP/0/RSP0/CPU0:R1(config-if)#commit Fri Apr 20 00:34:25.555 UTC RP/0/RSP0/CPU0:R1(config-if)# RP/0/RSP0/CPU0:R1(config-if)#int loop0 RP/0/RSP0/CPU0:R1(config-if)#ip add 150.1.1.1/32 RP/0/RSP0/CPU0:R1(config-if)#commit Fri Apr 20 00:34:39.839 UTC RP/0/RSP0/CPU0:R1(config-if)# Ok, lets get PE2 done now and test the interface connectivity. confirm that, we can do ISIS. RP/0/RSP0/CPU0:R2#conf t Fri Apr 20 00:35:39.031 UTC RP/0/RSP0/CPU0:R2(config)#int g0/1/0/11 RP/0/RSP0/CPU0:R2(config-if)#ip add 150.1.12.2/24 RP/0/RSP0/CPU0:R2(config-if)#no shut RP/0/RSP0/CPU0:R2(config-if)#int loop0 RP/0/RSP0/CPU0:R2(config-if)#ip add 150.2.2.2/32 RP/0/RSP0/CPU0:R2(config-if)#comm Fri Apr 20 00:35:54.565 UTC RP/0/RSP0/CPU0:R2(config-if)# After we

Ok, now lets do a PING test to see if we have connectivity: RP/0/RSP0/CPU0:R2#ping 150.1.12.1 Fri Apr 20 00:36:09.946 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 150.1.12.1, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/1 ms RP/0/RSP0/CPU0:R2# Good, now onto ISIS. For this we will use ISIS area 49.0000.0000.000X.00 where X = Router number and Level-2 only area. PE1: Lets define the routing process - Core RP/0/RSP0/CPU0:R1(config)#router isis Core Set are Network ID RP/0/RSP0/CPU0:R1(config-isis)#net 49.0000.0000.0001.00 Cisco IOS XR Training 2012 Jeffrey Fry Page 83 FryGuy.Net

And our IS Type RP/0/RSP0/CPU0:R1(config-isis)#is-type level-2 Set the loopback interface into ISIS and place it in PASSIVE mode RP/0/RSP0/CPU0:R1(config-isis)#int loop0 RP/0/RSP0/CPU0:R1(config-isis-if)#passive RP/0/RSP0/CPU0:R1(config-isis-if)#address-family ipv4 un RP/0/RSP0/CPU0:R1(config-isis-if-af)#exit Now for g0/1/0/11 RP/0/RSP0/CPU0:R1(config-isis-if)#int g0/1/0/11 RP/0/RSP0/CPU0:R1(config-isis-if)#address-family ipv4 unicast RP/0/RSP0/CPU0:R1(config-isis-if-af)#exit And finally, commit our changes. RP/0/RSP0/CPU0:R1(config-isis-if)#commit Fri Apr 20 01:02:31.714 UTC Now, lets get PE2 setup the same way: RP/0/RSP0/CPU0:R2(config)#router isis Core RP/0/RSP0/CPU0:R2(config-isis)#net 49.0000.0000.0002.00 RP/0/RSP0/CPU0:R2(config-isis)#is-type level-2 RP/0/RSP0/CPU0:R2(config-isis)#int loop0 RP/0/RSP0/CPU0:R2(config-isis-if)#passive RP/0/RSP0/CPU0:R2(config-isis-if)#address-family ipv4 u RP/0/RSP0/CPU0:R2(config-isis-if-af)#exit RP/0/RSP0/CPU0:R2(config-isis-if)#exit RP/0/RSP0/CPU0:R2(config-isis)#int g0/1/0/11 RP/0/RSP0/CPU0:R2(config-isis-if)#address-family ipv4 un RP/0/RSP0/CPU0:R2(config-isis-if-af)#exit RP/0/RSP0/CPU0:R2(config-isis-if)#commit Ok, lets check our ISIS neighbors RP/0/RSP0/CPU0:R2#sh isis neighbors Fri Apr 20 01:10:31.813 UTC IS-IS Core neighbors: System Id Interface R1 Gi0/1/0/11 Total neighbor count: 1 RP/0/RSP0/CPU0:R2# Yup, all neighbored up. Time to check the routes: RP/0/RSP0/CPU0:R2#sh ip route isis Fri Apr 20 01:10:54.269 UTC i L2 150.1.1.1/32 [115/10] via 150.1.12.1, 00:07:06, GigabitEthernet0/1/0/11 RP/0/RSP0/CPU0:R2# Cisco IOS XR Training 2012 Jeffrey Fry Page 84 FryGuy.Net

SNPA State Holdtime Type IETF-NSF 6c9c.ed26.ab91 Up 22 L2 Capable

Cool, we have a Level2 route to 150.1.1.1 via R1. Now, lets PING to make sure. RP/0/RSP0/CPU0:R2#ping 150.1.1.1 so l0 Fri Apr 20 01:11:26.132 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 150.1.1.1, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/1 ms RP/0/RSP0/CPU0:R2# Connectivity is working, cool! Next up, LDP. First up PE1: RP/0/RSP0/CPU0:R1(config)#mpls ldp Like all other IOS XR commands, you assign the interfaces under the protocol. RP/0/RSP0/CPU0:R1(config-ldp)#int g0/1/0/11 RP/0/RSP0/CPU0:R1(config-ldp-if)#comm Fri Apr 20 01:18:00.216 UTC RP/0/RSP0/CPU0:R1(config-ldp-if)# Ok, PE2 RP/0/RSP0/CPU0:R2(config)#mpls ldp RP/0/RSP0/CPU0:R2(config-ldp)#int g0/1/0/11 RP/0/RSP0/CPU0:R2(config-ldp-if)#comm Fri Apr 20 01:18:08.116 UTC Now lets check our LDP neighbors: RP/0/RSP0/CPU0:R2#sh mpls ldp neighbor Fri Apr 20 01:21:21.957 UTC Peer LDP Identifier: 150.1.1.1:0 TCP connection: 150.1.1.1:646 - 150.2.2.2:43857 Graceful Restart: No Session Holdtime: 180 sec State: Oper; Msgs sent/rcvd: 12/10; Downstream-Unsolicited Up time: 00:00:26 LDP Discovery Sources: GigabitEthernet0/1/0/11 Addresses bound to this peer: 150.1.1.1 150.1.12.1 RP/0/RSP0/CPU0:R2# Cool, we have a LDP session with PE1 and we can see the IPs bound to the peer. We are getting there, we still have BGP, VRF, and the OSPF configuration to do yet. We will save the BGP part until last so for now, VRF time. Cisco IOS XR Training 2012 Jeffrey Fry Page 85 FryGuy.Net

For this example, we will call our VRF R3R4 since we are connecting R3 (CE1) and R4 (CE2). PE1 up first: RP/0/RSP0/CPU0:R1#conf t Fri Apr 20 01:38:35.869 UTC Lets define the name of our VRF RP/0/RSP0/CPU0:R1(config)#vrf R3R4 Now we need to configure the appropriate address family, ipv4 unicast. RP/0/RSP0/CPU0:R1(config-vrf)# address-family ipv4 unicast Now we need to define our route-targets that we are going to import, and export. What is a route-target? Quickly it is a 64-bit BGP community that is used for tagging prefixes, making every prefix unique and also allows the remote PE routers to know what routes belong to what VRF (import). For this example, we will use 100:100 for both. RP/0/RSP0/CPU0:R1(config-vrf-af)# import route-target RP/0/RSP0/CPU0:R1(config-vrf-import-rt)# 100:100 RP/0/RSP0/CPU0:R1(config-vrf-import-rt)# export route-target RP/0/RSP0/CPU0:R1(config-vrf-export-rt)# 100:100 And commit the changes. RP/0/RSP0/CPU0:R1(config-vrf-export-rt)#commit Fri Apr 20 01:38:39.866 UTC Now we can create the same VRF with the same route-targets: RP/0/RSP0/CPU0:R2(config)#vrf R3R4 RP/0/RSP0/CPU0:R2(config-vrf)# address-family ipv4 unicast RP/0/RSP0/CPU0:R2(config-vrf-af)# import route-target RP/0/RSP0/CPU0:R2(config-vrf-export-rt)# 100:100 RP/0/RSP0/CPU0:R2(config-vrf-export-rt)# export route-target RP/0/RSP0/CPU0:R2(config-vrf-export-rt)# 100:100 RP/0/RSP0/CPU0:R2(config-vrf-export-rt)#comm Fri Apr 20 01:45:18.380 UTC Ok, time to check to see if the VRF is there: RP/0/RSP0/CPU0:R2#sh vrf R3R4 Fri Apr 20 01:45:52.204 UTC VRF RD RT R3R4 100:100 import 100:100 export 100:100 RP/0/RSP0/CPU0:R2#

AFI SAFI IPV4 Unicast IPV4 Unicast

Yup, we have a VRF. Now we can assign the interfaces facing the CE routers to the appropriate VRF, configure the IP addresses, and then do a PING test across the interface. Cisco IOS XR Training 2012 Jeffrey Fry Page 86 FryGuy.Net

PE1: RP/0/RSP0/CPU0:R1#conf t Fri Apr 20 01:48:48.712 UTC Lets get to our interface, G0/1/0/19 RP/0/RSP0/CPU0:R1(config)#interface GigabitEthernet0/1/0/19 Now we can assign the VRF of R3R4 RP/0/RSP0/CPU0:R1(config-if)# vrf R3R4 Configure our IP RP/0/RSP0/CPU0:R1(config-if)# ipv4 address 10.1.13.1 255.255.255.0 Since this is a 100M link, we will need to hard code it for the GBICs sake. RP/0/RSP0/CPU0:R1(config-if)# speed 100 And Commit our changes RP/0/RSP0/CPU0:R1(config-if)#comm Fri Apr 20 01:48:51.120 UTC RP/0/RSP0/CPU0:R1(config-if)# Once that is done, let do PE2 the same way. PE2: RP/0/RSP0/CPU0:R2#conf t Fri Apr 20 01:48:45.677 UTC RP/0/RSP0/CPU0:R2(config)#interface GigabitEthernet0/1/0/19 RP/0/RSP0/CPU0:R2(config-if)# vrf R3R4 RP/0/RSP0/CPU0:R2(config-if)# ipv4 address 10.1.24.2 255.255.255.0 RP/0/RSP0/CPU0:R2(config-if)# speed 100 RP/0/RSP0/CPU0:R2(config-if)#commit Fri Apr 20 01:48:51.059 UTC RP/0/RSP0/CPU0:R2(config-if)# Now we can test a ping from PE1 to CE1 and PE2 to CE2. RP/0/RSP0/CPU0:R1#ping 10.1.13.3 Fri Apr 20 01:57:01.969 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 10.1.13.3, timeout is 2 seconds: UUUUU Success rate is 0 percent (0/5) RP/0/RSP0/CPU0:R1# Hmm, that failed why? Well, when an interface lives in a VRF, you need to PING from that VRF. Lets try that again using VRF R3R4 RP/0/RSP0/CPU0:R1#ping vrf R3R4 10.1.13.3 Fri Apr 20 01:57:11.522 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 10.1.13.3, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/2 ms Cisco IOS XR Training 2012 Jeffrey Fry Page 87 FryGuy.Net

There, that worked. Lets check R2 RP/0/RSP0/CPU0:R2#ping vrf R3R4 10.1.24.4 Fri Apr 20 01:56:49.742 UTC Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 10.1.24.4, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/1 ms RP/0/RSP0/CPU0:R2# Ok, we have connectivity. CEs. Now we can get OSPF working between the PE and the

First up, PE1 We need to specify what we want to call our OSPF process, here I just used R3R4 RP/0/RSP0/CPU0:R1(config)#router ospf R3R4 Now we need to configure OSPF for the VRF RP/0/RSP0/CPU0:R1(config-ospf)# vrf R3R4 Now for the area RP/0/RSP0/CPU0:R1(config-ospf-vrf)# area 0

And then place the interfaces that we want in area 0 RP/0/RSP0/CPU0:R1(config-ospf-vrf-ar)# interface GigabitEthernet0/1/0/19 And commit our changes ( I just hit CTRL-Z) Uncommitted changes found, commit them before exiting(yes/no/cancel)? [cancel]:yes RP/0/RSP0/CPU0:R1# Ok, that is PE1 now for PE2 RP/0/RSP0/CPU0:R2#conf t Fri Apr 20 02:13:03.521 UTC RP/0/RSP0/CPU0:R2(config)#router ospf R3R4 RP/0/RSP0/CPU0:R2(config-ospf)# vrf R3R4 RP/0/RSP0/CPU0:R2(config-ospf-vrf)# area 0 RP/0/RSP0/CPU0:R2(config-ospf-vrf-ar)# interface GigabitEthernet0/1/0/19 RP/0/RSP0/CPU0:R2(config-ospf-vrf-ar-if)#exit RP/0/RSP0/CPU0:R2(config-ospf-vrf-ar)# exit RP/0/RSP0/CPU0:R2(config-ospf-vrf)# exit RP/0/RSP0/CPU0:R2(config-ospf)#com Fri Apr 20 02:13:14.843 UTC RP/0/RSP0/CPU0:R2(config-ospf)#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 88 FryGuy.Net

Ok, PE2 done. Now we can check for OSPF neighbor in that VRF. To do that, we need to use the following command: show ospf (OSPF Process) vrf (VRF Name) neighbor RP/0/RSP0/CPU0:R1#sh ospf R3R4 vrf R3R4 neighbor Fri Apr 20 02:18:16.826 UTC * Indicates MADJ interface Neighbors for OSPF R3R4, VRF R3R4 Neighbor ID Pri State 10.3.3.3 1 FULL/DR GigabitEthernet0/1/0/19 Neighbor is up for 00:00:05 Total neighbor count: 1 RP/0/RSP0/CPU0:R1# Ok, lets check PE2: RP/0/RSP0/CPU0:R2#sh ospf R3R4 vrf R3R4 neighbor Fri Apr 20 02:19:19.588 UTC * Indicates MADJ interface Neighbors for OSPF R3R4, VRF R3R4 Neighbor ID Pri State 10.4.4.4 1 FULL/DR GigabitEthernet0/1/0/19 Neighbor is up for 00:00:03 Total neighbor count: 1 RP/0/RSP0/CPU0:R2# Ok, both PE routers are neighbored up with the CE routers. Now, if we look at CE1s routing table what will we see? CE1#sh ip route Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2 i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2 ia - IS-IS inter area, * - candidate default, U - per-user static route o - ODR, P - periodic downloaded static route Gateway of last resort is not set 10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks Cisco IOS XR Training 2012 Jeffrey Fry Page 89 FryGuy.Net Dead Time 00:00:39 Address 10.1.24.4 Interface Dead Time 00:00:39 Address 10.1.13.3 Interface

C C CE1#

10.1.13.0/24 is directly connected, FastEthernet0/0 10.3.3.3/32 is directly connected, Loopback0

We only see our local routes, nothing from CE2 yet. This is because we have not built the VPNv4 session between PE1 and PE2 yet. We need to configure BGP VPNv4 in order to get the two PE routers to pass the tagged routes to each other. So, onto BGP we go! For this we will peer with PE2 loopback (150.2.2.2) using AS 1. PE1 first: RP/0/RSP0/CPU0:R1#conf t Fri Apr 20 02:21:32.174 UTC First we define our BGP process and AS number RP/0/RSP0/CPU0:R1(config)#router bgp 1 Enable vpv4 address family RP/0/RSP0/CPU0:R1(config)#address-family vpnv4 unicast Now we can configure our neighbor and all the info. RP/0/RSP0/CPU0:R1(config-bgp-af)# neighbor 150.2.2.2 RP/0/RSP0/CPU0:R1(config-bgp-nbr)# remote-as 1 Remember to specify the loopback as the update-source RP/0/RSP0/CPU0:R1(config-bgp-nbr)# update-source Loopback0 Now we can enable VPNv4 address family with that neighbor RP/0/RSP0/CPU0:R1(config-bgp-nbr)# address-family vpnv4 unicast Now we can configure the VRF parameters that BGP needs to know First define our VRF RP/0/RSP0/CPU0:R1(config-bgp-nbr-af)# vrf R3R4 Assign our Route Distinguisher (RD) RP/0/RSP0/CPU0:R1(config-bgp-vrf)# rd 100:100 Enable IPv4 Unicse for this VRF RP/0/RSP0/CPU0:R1(config-bgp-vrf)#

address-family ipv4 unicast

And finally redistribute our OSPF learned routes into BGP VRF R3R4 RP/0/RSP0/CPU0:R1(config-bgp-vrf-af)# redistribute ospf R3R4 match internal external RP/0/RSP0/CPU0:R1(config-bgp-vrf-af)# ^Z Uncommitted changes found, commit them before exiting(yes/no/cancel)? [cancel]:yes RP/0/RSP0/CPU0:R1#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 90 FryGuy.Net

Ok, now that that is done we need to do the same thing on PE2 RP/0/RSP0/CPU0:R2(config)#router bgp 1 RP/0/RSP0/CPU0:R2(config-bgp)# address-family ipv4 unicast RP/0/RSP0/CPU0:R2(config-bgp-af)# address-family vpnv4 unicast RP/0/RSP0/CPU0:R2(config-bgp-af)# neighbor 150.1.1.1 RP/0/RSP0/CPU0:R2(config-bgp-nbr)# remote-as 1 RP/0/RSP0/CPU0:R2(config-bgp-nbr)# update-source Loopback0 RP/0/RSP0/CPU0:R2(config-bgp-nbr)# address-family vpnv4 unicast RP/0/RSP0/CPU0:R2(config-bgp-nbr-af)# vrf R3R4 RP/0/RSP0/CPU0:R2(config-bgp-vrf)# rd 100:100 RP/0/RSP0/CPU0:R2(config-bgp-vrf)# address-family ipv4 unicast RP/0/RSP0/CPU0:R2(config-bgp-vrf-af)# redistribute ospf R3R4 match internal external RP/0/RSP0/CPU0:R2(config-bgp-vrf-af)#exit RP/0/RSP0/CPU0:R2(config-bgp-vrf)#exit RP/0/RSP0/CPU0:R2(config-bgp-vrf)#comm Fri Apr 20 02:27:10.491 UTC Ok, since this is a VPNv4 neighbor we need to check to see if we are neighbored up: RP/0/RSP0/CPU0:R2#sh bgp vpnv4 unicast summary Fri Apr 20 02:28:05.467 UTC BGP router identifier 150.2.2.2, local AS number 1 BGP generic scan interval 60 secs BGP table state: Active Table ID: 0x0 RD version: 3889240856 BGP main routing table version 25 BGP scan interval 60 secs BGP is operating in STANDALONE mode. Process StandbyVer Speaker 25 Neighbor St/PfxRcd 150.1.1.1 2 RcvTblVer 25 bRIB/RIB 25 LabelVer 25 ImportVer 25 SendTblVer 25

Spk 0

AS MsgRcvd MsgSent 1 14168 14173

TblVer 25

InQ OutQ 0

Up/Down

0 00:00:48

RP/0/RSP0/CPU0:R2# Yup, we are up and we can see what we are receiving 2 prefixes as well! Wonder what they are? To find out, use the show bgp vpnv4 unicast command RP/0/RSP0/CPU0:R2#sh bgp vpnv4 unicast Fri Apr 20 02:29:01.202 UTC BGP router identifier 150.2.2.2, local AS number 1 BGP generic scan interval 60 secs Cisco IOS XR Training 2012 Jeffrey Fry Page 91 FryGuy.Net

BGP table state: Active Table ID: 0x0 RD version: 3889240856 BGP main routing table version 25 BGP scan interval 60 secs Status codes: s suppressed, d damped, h history, * valid, > best i - internal, r RIB-failure, S stale Origin codes: i - IGP, e - EGP, ? - incomplete Network Next Hop Metric LocPrf Weight Path Route Distinguisher: 100:100 (default for vrf R3R4) *>i10.1.13.0/24 150.1.1.1 0 100 0 ? *> 10.1.24.0/24 0.0.0.0 0 32768 ? *>i10.3.3.3/32 150.1.1.1 2 100 0 ? *> 10.4.4.4/32 10.1.24.4 2 32768 ? Processed 4 prefixes, 4 paths RP/0/RSP0/CPU0:R2# Nice, we can see we have routes from CE1 and CE2. Now, lets see if CE1 has routes to CE2 CE1#sh ip route Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2 i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2 ia - IS-IS inter area, * - candidate default, U - per-user static route o - ODR, P - periodic downloaded static route Gateway of last resort is not set C C CE1# 10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks 10.1.13.0/24 is directly connected, FastEthernet0/0 10.3.3.3/32 is directly connected, Loopback0

Nope, hmm. What did we forget? I know, we redistributed OSPF into BGP, but we did not redistribute BGP into OSPF. Lets get that fixed. PE1: RP/0/RSP0/CPU0:R1(config)#router ospf R3R4 RP/0/RSP0/CPU0:R1(config-ospf)#vrf R3R4 RP/0/RSP0/CPU0:R1(config-ospf-vrf)# redistribute bgp 1 RP/0/RSP0/CPU0:R1(config-ospf-vrf)#comm Fri Apr 20 02:31:44.637 UTC RP/0/RSP0/CPU0:R1(config-ospf-vrf)#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 92 FryGuy.Net

And on PE2: RP/0/RSP0/CPU0:R2(config)#router ospf R3R4 RP/0/RSP0/CPU0:R2(config-ospf)#vrf R3R4 RP/0/RSP0/CPU0:R2(config-ospf-vrf)# redistribute bgp 1 RP/0/RSP0/CPU0:R2(config-ospf-vrf)# ^Z Uncommitted changes found, commit them before exiting(yes/no/cancel)? [cancel]:yes RP/0/RSP0/CPU0:R2# Ok, lets check CE1 for routes to CE2 now CE1#sh ip route Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2 i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2 ia - IS-IS inter area, * - candidate default, U - per-user static route o - ODR, P - periodic downloaded static route Gateway of last resort is not set C C O IA O IA CE1# 10.0.0.0/8 is variably subnetted, 4 subnets, 2 masks 10.1.13.0/24 is directly connected, FastEthernet0/0 10.3.3.3/32 is directly connected, Loopback0 10.4.4.4/32 [110/12] via 10.1.13.1, 00:00:51, FastEthernet0/0 10.1.24.0/24 [110/11] via 10.1.13.1, 00:00:51, FastEthernet0/0

There they are, lets do a PING CE1#ping 10.4.4.4 so l0 Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 10.4.4.4, timeout is 2 seconds: Packet sent with a source address of 10.3.3.3 !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/4 ms CE1# Nice, we can PING! Now, one other thing that you should notice with the CE1 routing table, routes to CE2 are seen as O IA, OSPF InterArea routes. This is what is expected when you run the same CE OSPF process ID over a MPLS network the BGP will carry the extra attributes creating what is called a Super Backbone. When we decode the BGP route information using the show bgp vpnv4 unicast vrf R3R4 10.1.13.0/24 command, we get the following output - notice the extended community information, this is where the extra information is carried. We will actually pull up both 10.1.13.0 and 10.1.14.0 so you can see. Cisco IOS XR Training 2012 Jeffrey Fry Page 93 FryGuy.Net

RP/0/RSP0/CPU0:R2#sh bgp vpnv4 unicast vrf R3R4 10.1.13.0 Fri Apr 20 02:45:47.177 UTC BGP routing table entry for 10.1.13.0/24, Route Distinguisher: 100:100 Versions: Process bRIB/RIB SendTblVer Speaker 24 24 Last Modified: Apr 20 02:27:22.347 for 00:18:24 Paths: (1 available, best #1) Not advertised to any peer Path #1: Received by speaker 0 Not advertised to any peer Local 150.1.1.1 (metric 10) from 150.1.1.1 (150.1.1.1) Received Label 16001 Origin incomplete, metric 0, localpref 100, valid, internal, best, group-best, import-candidate, imported Received Path ID 0, Local Path ID 1, version 24 Extended community: RT:100:100 OSPF route-type:0:2:0x0 OSPF routerid:150.1.1.1 RP/0/RSP0/CPU0:R2#sh bgp vpnv4 unicast vrf R3R4 10.1.24.0 Fri Apr 20 03:23:37.990 UTC BGP routing table entry for 10.1.24.0/24, Route Distinguisher: 100:100 Versions: Process bRIB/RIB SendTblVer Speaker 20 20 Local Label: 16001 Last Modified: Apr 20 02:19:16.347 for 01:04:21 Paths: (1 available, best #1) Advertised to peers (in unique update groups): 150.1.1.1 Path #1: Received by speaker 0 Advertised to peers (in unique update groups): 150.1.1.1 Local 0.0.0.0 from 0.0.0.0 (150.2.2.2) Origin incomplete, metric 0, localpref 100, weight 32768, valid, redistributed, best, group-best, import-candidate Received Path ID 0, Local Path ID 1, version 20 Extended community: RT:100:100 OSPF route-type:0:2:0x0 OSPF routerid:150.2.2.2 RP/0/RSP0/CPU0:R2# There is a way to prevent this from happening and that is to create a DomainID for the OSPF process on one of the PE routers. RP/0/RSP0/CPU0:R2#conf t Fri Apr 20 03:30:11.463 UTC

Navigate to the OSPF VRF process Cisco IOS XR Training 2012 Jeffrey Fry Page 94 FryGuy.Net

RP/0/RSP0/CPU0:R2(config)#router ospf R3R4 RP/0/RSP0/CPU0:R2(config-ospf)#vrf R3R4 Now, lets see what Domain-id types we have See RFC 4577 for more info on these. RP/0/RSP0/CPU0:R2(config-ospf-vrf)#domain-id type ? 0005 Type 0x0005 0105 Type 0x0105 0205 Type 0x0205 8005 Type 0x8005 RP/0/RSP0/CPU0:R2(config-ospf-vrf)#domain-id type 0005 value ? WORD OSPF domain ID ext. community value in Hex (6 octets) Now lets set it to a value RP/0/RSP0/CPU0:R2(config-ospf-vrf)#domain-id type 0105 value AABBCCDDEEFF When you do this, the routes on CE are now E2 routes: CE1#sh ip route Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2 i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2 ia - IS-IS inter area, * - candidate default, U - per-user static route o - ODR, P - periodic downloaded static route Gateway of last resort is not set 10.0.0.0/8 is variably subnetted, 4 subnets, 2 masks C 10.1.13.0/24 is directly connected, FastEthernet0/0 C 10.3.3.3/32 is directly connected, Loopback0 O E2 10.4.4.4/32 [110/2] via 10.1.13.1, 01:04:24, FastEthernet0/0 O E2 10.1.24.0/24 [110/1] via 10.1.13.1, 01:04:24, FastEthernet0/0 CE1# This can also work in reverse, if you want to create a SuperBackbone but the OSPF processes are different, you can set the domain-id to be the same. RP/0/RSP0/CPU0:R1(config-ospf-vrf)#domain-id type 0105 value AABBCCDDEEFF CE1#sh ip route Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2 E1 - OSPF external type 1, E2 - OSPF external type 2 i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS ia - IS-IS inter area, * - candidate default, U - per-user o - ODR, P - periodic downloaded static route Gateway of last resort is not set Cisco IOS XR Training 2012 Jeffrey Fry Page 95 FryGuy.Net

10.0.0.0/8 is variably subnetted, 4 subnets, 2 masks C 10.1.13.0/24 is directly connected, FastEthernet0/0 C 10.3.3.3/32 is directly connected, Loopback0 O IA 10.4.4.4/32 [110/12] via 10.1.13.1, 00:00:04, FastEthernet0/0 O IA 10.1.24.0/24 [110/11] via 10.1.13.1, 00:00:04, FastEthernet0/0 CE1# There, back to IA routes again. There is much more to domain-id, but I will save that for another day.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 96 FryGuy.Net

23. L2VPN
Ok, now it is time for some L2VPN. Here we will use the same diagram as before, but instead of providing MPLS VPN between CE1 and CE2, we are going to create a L2VPN so that CE1 and CE2 think that they are directly connected to each other.

First up, CE1 CE1(config-if)#int f0/0 CE1(config-if)#ip add 10.1.34.3 255.255.255.0 CE1(config-if)#int l0 CE1(config-if)#ip add 10.1.3.3 255.255.255.255 CE1(config-if)#router ospf 1 CE1(config-router)#net 10.0.0.0 0.255.255.255 a 0 CE1(config-router)# Now CE2 Cisco IOS XR Training 2012 Jeffrey Fry Page 97 FryGuy.Net

CE2(config-if)#ip add 10.1.34.4 255.255.255.0 CE2(config-if)#int l0 CE2(config-if)#ip add 10.4.4.4 255.255.255.255 CE2(config-if)#router ospf 1 CE2(config-router)#net 10.0.0.0 0.255.255.255 a 0 CE2(config-router)# Ok, now time for the PE routers. PE1: First, lets reset the interface to all its defaults: RP/0/RSP0/CPU0:PE1(config)#default interface g0/1/0/19 RP/0/RSP0/CPU0:PE1(config)#commit Now PE2: RP/0/RSP0/CPU0:PE2(config)#default interface g0/1/0/19 RP/0/RSP0/CPU0:PE2(config)#commit Now, lets kill our OSPF sessions. On both routers: PE1: RP/0/RSP0/CPU0:PE1(config)#no router ospf 100 RP/0/RSP0/CPU0:PE1(config)#commit PE2: RP/0/RSP0/CPU0:PE2(config)#no router ospf 100 RP/0/RSP0/CPU0:PE2(config)#commit OK, now we can build out L2VPN cross-connects. Fist up, we need to get to the L2VPN configuration RP/0/RSP0/CPU0:PE1(config)#l2vpn Now to configure our X-Connect group RP/0/RSP0/CPU0:PE1(config-l2vpn)#xconnect group R3R4 And our Point-to-Point settings RP/0/RSP0/CPU0:PE1(config-l2vpn-xc)#p2p R3_to_R4 Place the interface in the P2P group R3_to_R4 RP/0/RSP0/CPU0:PE1(config-l2vpn-xc-p2p)#interface g0/1/0/19 Specify our Neighbor for this with a pseudowire ID (think of it as a circuit ID) and then commit our changes RP/0/RSP0/CPU0:PE1(config-l2vpn-xc-p2p)#neighbor 150.2.2.2 pw-id 304 RP/0/RSP0/CPU0:PE1(config-l2vpn-xc-p2p-pw)#comm Now, this is unique to our CE devices, we need to specify the speed in order to get the interfaces up as the CE routers here do not support Gigabit Ethernet RP/0/RSP0/CPU0:PE2(config-if)#int g0/1/0/19 RP/0/RSP0/CPU0:PE2(config-if)#spee 100 RP/0/RSP0/CPU0:PE2(config-if)#comm Now for PE2: Cisco IOS XR Training 2012 Jeffrey Fry Page 98 FryGuy.Net

RP/0/RSP0/CPU0:PE2(config)#l2vpn RP/0/RSP0/CPU0:PE2(config-l2vpn)#xconnect group R3R4 RP/0/RSP0/CPU0:PE2(config-l2vpn-xc)#p2p R3_to_R4 RP/0/RSP0/CPU0:PE2(config-l2vpn-xc-p2p)#interface g0/1/0/19 The Pseudo-wire ID must match. RP/0/RSP0/CPU0:PE2(config-l2vpn-xc-p2p)#neighbor 150.1.1.1 pw-id 304 RP/0/RSP0/CPU0:PE2(config-l2vpn-xc-p2p-pw)#comm RP/0/RSP0/CPU0:PE2(config-if)#int g0/1/0/19 RP/0/RSP0/CPU0:PE2(config-if)#spee 100 RP/0/RSP0/CPU0:PE2(config-if)#comm

Now we can look at our L2VPN Cross-connects RP/0/RSP0/CPU0:PE1#sh l2vpn xconnect Tue Apr 24 03:40:36.619 UTC Legend: ST = State, UP = Up, DN = Down, AD = Admin Down, UR = Unresolved, SB = Standby, SR = Standby Ready XConnect Segment 1 Segment 2 Group Name ST Description ST Description ST -------------------------------------------------------------R3R4 R3_to_R4 UP Gi0/1/0/19 UP 150.2.2.2 304 UP ---------------------------------------------------------------------RP/0/RSP0/CPU0:PE1# There you go, that looks good. CE1#p 10.1.34.4 Now, can we PING between CE1 and CE2?

Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 10.1.34.4, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/4 ms Yup, PING works. Hmm, wonder what CDP looks like? CE1#sh cdp nei Capability Codes: R - Router, T - Trans Bridge, B - Source Route Bridge S - Switch, H - Host, I - IGMP, r - Repeater Device ID CE2 CE1# Local Intrfce Fas 0/0 Holdtme 160 Capability R S I Platform 2811 Port ID Fas 0/0

Now if we look at OSPF: CE1#sh ip ospf neighbor Neighbor ID Pri State Dead Time Address Interface Page 99 FryGuy.Net

Cisco IOS XR Training 2012 Jeffrey Fry

10.4.4.4

FULL/DR

00:00:33

10.1.34.4

FastEthernet0/0

All neighbored up! That means we should be able to PING between loopback interfaces: CE1#p 10.4.4.4 so l0 Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 10.4.4.4, timeout is 2 seconds: Packet sent with a source address of 10.1.3.3 !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/4 ms CE1#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 100 FryGuy.Net

24. NHRP ( HSRP & VRRP)

Next Hop Resolution Protocol comes in two fashions on the IOS XR. The first is Cisco proprietary and called Hot-Standby Router Protocol or HSRP and the other is the industry standard called Virtual Router Redundancy protocol or VRRP. This is something that many customers use in order to maintain the availability of a default gateway on the network. If your customer uses a static route to a next hop, you might be using this as well. Like everything else with IOS XR, NHRP is handled a little differently. With IOS, you configure your standby commands under the interfaces; in IOS XR you use ROUTER HSRP or Router VRRP. For this lab we will use interface Te0/1/0/0 and a subnet of 150.1.12.0/24. First up R1: RP/0/RSP0/CPU0:R1(config)#int tenGigE 0/1/0/0 RP/0/RSP0/CPU0:R1(config-if)#ip add 150.1.12.1/24 RP/0/RSP0/CPU0:R1(config-if)#no shut RP/0/RSP0/CPU0:R1(config-if)#commit RP/0/RSP0/CPU0:R1(config-if)# Now R2: RP/0/RSP0/CPU0:R2(config)#int tenGigE 0/1/0/0 RP/0/RSP0/CPU0:R2(config-if)#ip add 150.1.12.2/24 RP/0/RSP0/CPU0:R2(config-if)#no shut RP/0/RSP0/CPU0:R2(config-if)#commit Now we should test PING from R1 to R2: RP/0/RSP0/CPU0:R1#ping 150.1.12.2 Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 150.1.12.2, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/2 ms RP/0/RSP0/CPU0:R1# Good, we have connectivity.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 101 FryGuy.Net

HSRP is up first! First thing, from config mode, enter router hsrp RP/0/RSP0/CPU0:R1(config)#router hsrp Now we tell it what interface we are configuring HSRP under RP/0/RSP0/CPU0:R1(config-hsrp)#interface tenGigE 0/1/0/0 And then we configure our HSRP ID and associated information. Instead of using the standby command, we are using the HSRP command. RP/0/RSP0/CPU0:R1(config-hsrp-if)#hsrp 100 ipv4 150.1.12.100 RP/0/RSP0/CPU0:R1(config-hsrp-if)#hsrp 100 priority 150 RP/0/RSP0/CPU0:R1(config-hsrp-if)#hsrp 100 preempt RP/0/RSP0/CPU0:R1(config-hsrp-if)#hsrp 100 authentication cisco RP/0/RSP0/CPU0:R1(config-hsrp-if)#exit RP/0/RSP0/CPU0:R1(config-hsrp)#exit RP/0/RSP0/CPU0:R1(config)#exit And save your changes! Uncommitted changes found, commit them before exiting(yes/no/cancel)? [cancel]:yes Ok, now to configure R2 RP/0/RSP0/CPU0:R2(config)#router hsrp RP/0/RSP0/CPU0:R2(config-hsrp)#interface tenGigE 0/1/0/0 RP/0/RSP0/CPU0:R2(config-hsrp-if)#hsrp 100 ipv4 150.1.12.100 RP/0/RSP0/CPU0:R2(config-hsrp-if)#hsrp preempt RP/0/RSP0/CPU0:R2(config-hsrp-if)#hsrp authentication cisco RP/0/RSP0/CPU0:R2(config-hsrp-if)#comm Back to R1 to validate HSRP RP/0/RSP0/CPU0:R1#sh hsrp
Interface Te0/1/0/0 P indicates configured to preempt. | Grp Pri P State Active addr Standby addr 100 150 P Active local 150.1.12.2 Group addr 150.1.12.100

RP/0/RSP0/CPU0:R1# There we go, we now have HSRP on R1 active with R2 as standby.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 102 FryGuy.Net

Time for some VRRP One thing cool about VRRP, you dont have to burn an IP address just for the virtual. You can use an actual physical IP address of a router. If that router goes off-line, then the other router will just assume the IP address. R1 up first, and we will use the R1 Te0/1/0/0 IP address for the virtual. First though, we need to remove HSRP and save the changes. RP/0/RSP0/CPU0:R1(config)#no router hsrp RP/0/RSP0/CPU0:R1(config)#commit Ok, to configure VRRP the command is router vrrp RP/0/RSP0/CPU0:R1(config)#router vrrp Again, then you tell it what interface RP/0/RSP0/CPU0:R1(config-vrrp)#interface tenGigE 0/1/0/0 Since VRRP likes IPV6, we need to use the address-family command RP/0/RSP0/CPU0:R1(config-vrrp-if)#address-family ipv4 Then configure our VRRP ID RP/0/RSP0/CPU0:R1(config-vrrp-address-family)#vrrp 1 Assign the IP, here I am using the same IP as our physical interface RP/0/RSP0/CPU0:R1(config-vrrp-virtual-router)#address 150.1.12.1 RP/0/RSP0/CPU0:R1(config-vrrp-virtual-router)#text-authentication cisco RP/0/RSP0/CPU0:R1(config-vrrp-virtual-router)#commit RP/0/RSP0/CPU0:R2(config-vrrp-virtual-router)# Now for R2, but this time we will decrease the priority so that R1 is the active router RP/0/RSP0/CPU0:R2(config)#no router hsrp RP/0/RSP0/CPU0:R2(config)#commit RP/0/RSP0/CPU0:R2(config)#router vrrp RP/0/RSP0/CPU0:R2(config-vrrp)#interface tenGigE 0/1/0/0 RP/0/RSP0/CPU0:R2(config-vrrp-if)#address-family ipv4 RP/0/RSP0/CPU0:R2(config-vrrp-address-family)#vrrp 1 RP/0/RSP0/CPU0:R2(config-vrrp-virtual-router)#address 150.1.12.1 RP/0/RSP0/CPU0:R2(config-vrrp-virtual-router)#text-authentication cisco RP/0/RSP0/CPU0:R2(config-vrrp-virtual-router)#priority 50 RP/0/RSP0/CPU0:R2(config-vrrp-virtual-router)#commit RP/0/RSP0/CPU0:R2(config-vrrp-virtual-router)# RP/0/RSP0/CPU0:R2#

Cisco IOS XR Training 2012 Jeffrey Fry

Page 103 FryGuy.Net

Now back to R1 to see the VRRP status: RP/0/RSP0/CPU0:R1#show vrrp IPv4 Virtual Routers: A indicates IP address owner | P indicates configured to preempt | | Interface vrID Prio A P State Master addr VRouter addr Te0/1/0/0 1 255 A P Master local 150.1.12.1 IPv6 Virtual Routers: A indicates IP address owner | P indicates configured to preempt | | Interface vrID Prio A P State Master addr VRouter addr RP/0/RSP0/CPU0:R1# Nice, R1 is the MASTER. Wonder what R2 says? RP/0/RSP0/CPU0:R2#sh vrrp IPv4 Virtual Routers: A indicates IP address owner | P indicates configured to preempt | | Interface vrID Prio A P State Master addr VRouter addr Te0/1/0/0 1 50 P Backup 150.1.12.1 150.1.12.1 IPv6 Virtual Routers: A indicates IP address owner | P indicates configured to preempt | | Interface vrID Prio A P State Master addr VRouter addr RP/0/RSP0/CPU0:R2# It says its the backup! Nice.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 104 FryGuy.Net

This page intentionally left black for notes.

Cisco IOS XR Training 2012 Jeffrey Fry

Page 105 FryGuy.Net