BGP Routing Protocol: BY: Gebretinsae G

BGP Routing Protocol
BY: Gebretinsae G.
Network Div | 02 April 2012 | P001

January 2017
BGP Contents
o BGP overview
o BGP terminology
o BGP packet types and connection status
o BGP route announcing methods
o BGP route announcing principles
o BGP Basic Configurations
o IGW links in ET network
o BGP Troubleshooting Commands
o ET IGW Traffic Troubleshooting
P_002
BGP
Internet development
 Till Dec,1969 there was only 4 sites

 Now it has developed to a global network
 The whole network is divided into several ASs
 Easy to manage
P_003
BGP
How big is the Internet?
 A router running BGP in Internet has:
 The size of the routing table is bigger than 70 Mb
 There are more than 500,000 routes
 More than 65,000 AS numbers
BL-GW-1#show ip forwarding route summary

The total active and backup routes of global and all VPNs:
Route Source Count
static: 34
direct: 17
ospf: 0
bgp: 3096
P_004
BGP
Autonomous Systems
 AS(autonomous system ) is a big network that is under the same
technical management.
 IGPs work in the same AS （ autonomous system ）
 EGPs connect different AS （ autonomous system ）
P_005
BGP
AS （ Autonomous Systems ）
 Every AS network has its own AS number
 Routing policy in each AS is independent
 AS number range ： 1 ~ 65535--
X-GW-Routers#show running-config bgp

router bgp 24757
no synchronization
P_006
BGP
BGP （ Border Gateway Protocol ）
 BGP is a kind of EGP
 BGP is a commonly used standard protocol
 Version that we used nowadays ： BGPv4
 BGP use AS number to avoid route loop
 Inner routing protocol: RIP/IGRP/OSPF/EIGRP/ISIS
P_007
BGP
BGP is used among ASs
 BGP is used among ASs to ensure that there is no loop in the

network
P_008
BGP
Comparison of Routing Protocols
Routing Interior or Distance vector Hierarchy Metric
Protocol Exterior or link state Required
OSPF Interior Link-state Yes Cost

ISIS Interior Link-state Yes Metric(Cost)
EIGRP Interior Hybrid(advanced No Composite
distance vector)
RIP Interior Distance vector No Hope count

(Distance)
BGP Exterior Path vector No Path Vectors or

Attributes
P_009
BGP
Huawei
P_0010
BGP
P_0011
BGP
BGP routing, foundation concepts
 Border Gateway Protocol is a standardized exterior gateway protocol
designed to exchange routing and reach ability information between
autonomous systems on the Internet.
 BGP is often classified as a path vector protocol, but is sometimes also
classed as a distance vector routing protocol
 BGP is more scalable
 The place of BGP in your routed world
EGP-BGP for the internet
To make your ISP typically redundant
To Connect your enterprise’s network to ISP
 internal BGP in your ISP or enterprise
P_0012
BGP
The facts about BGP
• BGP runs on top of TCP(port 179)
• TCP used for reliability
• Updates are incremental and triggered
• Metric is the biggest you have ever seen( so many attributes)
• Slowest routing protocol on the planet to converge
• One of the most tunable protocol
• Without tuning, BGP behaves just like RIP
Styles of BGP implementations

• To advertise your network to internet, you need BGP
• To receive partial updates from your ISP
• To receive a full update
P_0013
BGP
When to use BGP?
 BGP is suitable in these occasions:

 One AS allow data stream to pass through it to
reach another AS (such as ISP)
 One AS has several exits to other ASs
 One AS wants to control the data stream within
itself
 Of course, you need to be familiar with the operation
and function of BGP to make good use of it
P_0014
BGP
When we don’t use BGP?
 BGP is not effective for all occasions. In the following

occasions we may not use BGP :
 There is only one exit to Internet or other AS
 Within your AS ,you don’t care about route policy
and route selection
 If the router is not powerful enough to handle too
many changes and updates in route
 Limited understanding for route filtering and route
selection
 Bandwidth between ASs is limited
 At this time we can use static route
P_0015
BGP
Static Route Review
P_0016
BGP
BGP Facts
 BGP is a kind of enhanced distance vector

routing protocol
 Transmission protocol ： TCP ， port number ：
179
 Support CIDR （ classless inter domain routing ）
 Route updates only send added route or
withdrawn route
 Rich in route filtering and route policy
configuration
P_0017
BGP
BGP in IP packet
P_0018
BGP
Peers = Neighbors
 Two routers establish TCP connection between
each other to exchange BGP route information after TCP
connection is established.
 The relationship between them is peer or neighbor.
 BGP neighbors is divided into two types: IBGP and EBGP.
P_0019
BGP
External BGP
 Neighbors are in different ASs （ autonomous systems ）
 Neighbors are usually directly connected
P_0020
BGP
Internal BGP
 Neighbors are in the same AS
 Neighbors do not need to be directly connected
P_0021
BGP
Internal BGP
 There can be several non-BGP routers between peers
 BGP information can be passed through non-BGP topology
P_0022
BGP
BGP message types
BGP has four kinds of messages:

 OPEN – used to establish BGP connection
 KEEPALIVE – used to keep BGP connection
 UPDATE – used to update or withdraw BGP
route
 NOTIFICATION – BGP error notification
P_0023
BGP
OPEN Messages
 OPEN message is used to establish BGP connection.
 It includes following contents:
 Version number
 AS number
 Keep time
 BGP identification
 Optional parameters
P_0024
BGP
BGP KEEPALIVE Message
 By default keep alive message are sent every 60
seconds(between peers)
 Hold time interval is 180 seconds
 Keep alive message will reset Hold timer to 0 If Hold
timer expires, the peer will be regarded as dead
 Keep alive and Hold timer can be changed When
establishing BGP, the two parts will use the less value of
hold time
 Keep alive message is 19 bytes long
P_0025
BGP
BGP UPDATE Messages
 The route with same attributes can be sent out in one
update message
 Update message can also be used to withdraw those
unreachable routes
 If the route is stable there is no update message
 Update can only be aimed at the attributes of the routes
 Update packets have the same function as keep alive
packets to reset hold timer
P_0026
BGP
BGP NOTIFICATION Messages
 When error is detected, Notification message is sent out
 Notification message will close BGP session
 Possible error information ： certification failure, route
loop etc.
P_0027
BGP
BGP connection status
 Idle ： This is the first status when BGP initiates
 Connect ： the status when BGP is waiting for the success of TCP
connection
 Active ： the status when BGP restart to establish TCP connection
 Open sent ： when TCP establishing succeeds ， BGP begin to send
OPEN message and wait for OPEN message from counterpart
 Open confirm ： after receiving OPEN message from neighbor,

BGP waits for Keep-alive message or Notification message
 Established ： this is the last stage or stable status for the

neighbors to negotiate ， BGP begin to exchange Update packets
between each other
P_0028
BGP
P_0029
BGP
BGP route announcing method
1—Network command
 RTB(config)#router bgp 200
 RTB(config-router-bgp)#network 18.0.0.0 mask
255.0.0.0
P_0030
BGP
BGP route announcing method
2—route redistribution
 RTB (config) # router bgp 200

 RTB (config-router-bgp) # redistribute ospf
P_0031
BGP
P_0032
BGP
P_0033
BGP
BGP route announcing principles
 When there are multiple paths, BGP Speaker will use the
best one for itself;
 BGP Speaker will only announce the routes that it is
using to their neighbors;
 BGP Speaker will announce the routes that it gets from EBGP
neighbor to all the BGP neighbors, including EBGP and IBGP
neighbors;
 BGP Speaker will not announce the routes that it gets from IBGP
neighbor to all the other IBGP neighbors;
 The BGP Speaker will announce the routes that it learns from
IBGP neighbor to its EBGP neighbor
 Once the connection is established, BGP Speaker will announce all
the BGP routes to the neighbors
P_0034
BGP
P_0035
BGP
P_0036
BGP
P_0037
BGP
P_0038
BGP
P_0039
BGP
P_0040
BGP
BGP routes take effect condition
 BGP routes take effect condition ： next hop of

route is reachable P_0041
BGP
BGP synchronization
 Because IBGP neighbors may not be directly connected, so these non-BGP routers
will take part in the process of packets forwarding based on BGP routes it is required
that IGP routers also have related BGP routes
 Routes learned from IBGP neighbor that is also learned from IGP can be announced
to EBGP neighbor, this is called BGP and IGP synchronization. This function is turned
on by default, it needs to be turned off manually
P_0042
BGP
IBGP full mesh
 Inside AS, all the routers which BGP routes pass through should
establish IBGP neighbor between each other, to ensure that
packets still have route guide after disabling BGP
synchronization
P_0043
BGP
How BGP finds the best path in Cisco/ZTE
0. Ignore routes with inaccessible next hope address
1. Prefer the path with the highest weight
2. Prefer the path with the highest Local-preference
3. Prefer the path that was locally originated via a network command
4. Prefer the path with shortest AS-path
5. Prefer the path with lowest origin type(where IGP is lower than EGP,
and EGP is lower than Incomplete).
6. Prefer the path with the lowest multi-exit discriminator (MED)
7. Prefer eBGP over iBGP paths
8. Prefer the path with the lowest IGP metric to the BGP next hope
(prefer the path through the closest IGP neighbor).
9. Determine if multiple paths require installation in the routing table for
BGP multipath
10. When both paths are external, prefer the path that was received first
(the oldest one) Network Div | 02 April 2012 | P001
P_0044
BGP
11. Prefer the route that comes from with the lowest router ID
12. If the originator or router ID is the same for multiple paths, prefer the
path with the minimum cluster list length
13. Prefer the path that comes from the lowest neighbor address
Here's a brief outline of the "order of preference" for filling the IP

routing table.
 Connected routes (IP addresses and routes of router interfaces)
first; then
Static routes (routes configured in router configurations with 'ip
route' statements); then
Routes learned via an IGP (RIP, RIPv2, OSPF, IS-IS, ...); then
Routes learned via BGP and other EGPs.
P_0045
BGP
BGP Extended Applications
P_0046
BGP
P_0047
BGP
P_0048
BGP
P_0049
BGP
BGP Tables:
Neighbor table: The connected BGP friends

A list of all configured BGP neighbors. To view it, use
show ip bgp summary command
BGP Table: A list of all BGP routes (can be ping).
A list of networks known by BGP, along with their paths and attributes. To view
it, use show ip bgp command
Routing Table: A list of the best routes
A list of paths to each network used by the router, and the next hope for each
network. To view it, use show ip route command
P_0050
BGP
BGP configuration steps
 Enable BGP routing process, <as-number> is the AS system ID of router,
1~65535.
 IPCORE Private AS# 65108
 Addis Ababa IPRAN Private AS# 65109
 IPCORE Public AS# 24757
Configure a BGP neighbor and AS number of a peer
Configure BGP Route Advertisement
P_0051
BGP
Establishing EBGP Neighbor
 EBGP ： physical connected directly
--Use IP addresses of interconnected interfaces to create BGP connection
--Use IP addresses of loopback interfaces to create BGP connection
P_0052
BGP
Configuration EBGP Neighbor
 Using Direct Connected Address
P_0053
BGP
Verifying EBGP Neighbor
 Using --Show IP BGP neighbor
P_0054
BGP
Establishing EBGP Neighbor
 Using --Loopback address
P_0055
BGP
Verifying EBGP Neighbor
P_0056
BGP
Establishing IBGP Neighbor
 IBGP ： physical connected directly or not connected directly
-- Use IP addresses of loopback interfaces to create BGP connection
-- valuable next hop （ have reachable route ）
-- Synchronization
-- Full mesh(RR)
P_0057
BGP
Establishing IBGP Neighbor
 Using --Loopback address
P_0058
BGP
Verifying IBGP Neighbor
P_0059
BGP
Configuring BGP Route Advertisement
 1. Use network command to advertise route
 Advertise the physical network routes
 2. Use redistribute command to advertise route
 Redistribute all the routes learned from IGP
protocol(s) ( static ， RIP 、 OSPF 、 ISIS etc )
 3. BGP route aggregation
 Aggregate the routes to one summary route and
advertise to other BGP neighbors, therefore can reduce
the size of IP routing table significantly.
 Make sure the mask of all the routes are equal while
configuring route aggregation on routers, in order to
avoid black hole route Network Div | 02 April 2012 | P001
P_0060
BGP
Using network to Advertise Route
P_0061
BGP
Verifying Advertise Route
 Using -- Show IP BGP route
P_0062
BGP
Using redistribute to Advertise Route
 R2 runs OSPF as IGP protocol.
 Now it needs to redistribute all
OSPF routes information on R2 into BGP.
P_0063
BGP
Verifying redistribute to Advertise Route
 Using - Show IP BGP Route
P_0064
BGP
Configuring Route Aggregation
P_0065
BGP
Verifying Route Aggregation
P_0066
BGP
Internet Service Design
Internet Service general connection
 As shown in the following figure, all internet Broad bands are aggregated at
the
Core Switches (T160G/T64G) which are connected to the PE router, and a GER
router is connected to the PE to provide the NAT service.
P_0067
BGP
IGW ( International Gateway )
 All the traffics go to the internet through our international
Gateways.
 In ET Network, we have four international Gateway sites.
Addis Ababa(Bole)
Bahirdar
Diredawa
Shashemene.
 We use prefix filters in our gateway to control the incoming

traffics from internet
P_0068
BGP
IGW ( International Gateway )…
 Internet Service Traffic Flow
 Broad-Band User Traffic Flow Network Topology
P_0069
BGP
Leased line Traffic Flow Network Topology
P_0070
BGP
Public IP User Traffic Flow Network topology
P_0071
BGP
International Customer VPN Traffic Flow Network topology
P_0072
BGP
BGP Troubleshooting Commands
5 show ip bgp neighbor out 197.254.194.21 This displays all Prefix List
advertised through this neighbor
P_0073
BGP
International GW Troubleshooting Commands
BL-GW-1#show running-config bgp
! <BGP>
router bgp 24757
no synchronization
maximum-paths 6
redistribute ospf-ext 2
aggregate-address 213.55.64.0 255.255.192.0 count 0
aggregate-address 213.55.64.0 255.255.224.0
aggregate-address 213.55.89.0 255.255.255.0 count 0 summary-only
.
. Network Div | 02 April 2012 | P001
P_0074
. BGP
neighbor 105.16.0.137 remote-as 37100
neighbor 105.16.0.137 activate
neighbor 105.16.0.137 ebgp-multihop ttl 2
neighbor 105.16.0.137 password SEAC7/IPTR/99802
neighbor 105.16.0.137 prefix-list ONLY_DEFAULT in
neighbor 105.16.0.137 prefix-list CAT-4xSTM4-SEACOM-1 out
neighbor 105.16.0.137 update-source loopback13
neighbor 105.22.120.5 password SEAC7-IPTR-99803
neighbor 105.22.120.5 prefix-list CAT-WEREDANET-1 out
neighbor 41.189.225.25 ebgp-multihop ttl 2
neighbor 41.189.225.25 password 20#DtEtc13
neighbor 41.189.225.25 prefix-list CAT-DJibouti_2xSTM4 out
neighbor 41.189.225.25 route-map Set_Default_Nexthop_DjiboutTel in
P_0075
. Network Div | 02 April 2012 | P001
BGP
.
BL-GW-1#show ip bgp summary
Neighbor Ver As MsgRcvd MsgSend Up/Down State/Pfx
Rcd
10.1.0.2 4 65108 3049585 2408277 39w5d 319
3
10.1.0.90 4 24757 1306832 1283222 46w3d 207
3
41.189.225.25 4 30990 63891 58607 2w6d 0
41.189.225.51 4 30990 105902 797 00:13:16 1
105.16.0.137 4 37100 2877574 58609 2w6d 1
105.22.120.5 4 37100 2879986 58600 2w6d 1

.
.
.
BL-GW-1#show bgp vpnv4 unicast summary
Rcd Network Div | 02 April 2012 | P001
P_0076
BGP
BL-GW-1#show ip bgp neighbor out 41.189.225.25

Routes Sent to This neighbor:
Dest NextHop Metric LocPrf Path

197.156.115.0/24 213.55.97.2 i
196.188.0.0/18 213.55.97.2 i
197.156.64.0/18 213.55.97.2 i
197.156.74.0/24 213.55.97.2 i
197.156.78.0/24 213.55.97.2 i
197.156.100.0/24 213.55.97.2 i
213.55.64.0/18 213.55.97.2 i
213.55.96.0/24 213.55.97.2 i
213.55.101.0/24 213.55.97.2 i
213.55.107.0/24 213.55.97.2 i
P_0077
BGP
KK-ER-DATA-A#show ip bgp summary
Rcd
10.153.192.44 4 64520 232468 237545 3w4d 1
10.153.192.45 4 64520 142502 146420 2w1d 1
10.153.193.188 4 64521 7254 7054 19:06:23 50
10.153.198.220 4 64521 253802 252244 3w6d 50

.
.
.
KK-ER-DATA-A#show bgp vpnv4 unicast summary
Rcd
10.0.2.114 4 65108 870378 239972 3w6d 373
46
10.0.2.115 4 65108 771520 239972 3w6d 373
42 P_0078
BGP
KK-ER-DATA-A#show running-config bgp

! <BGP>
router bgp 65108
timers bgp 10 30
no synchronization
redistribute connected
redistribute static
no neighbor 10.0.2.113 activate
neighbor 10.0.2.113 password encrypt xxx
neighbor 10.0.2.115 update-source loopback1 P_0079
. BGP
.
MW-RR#show ip bgp summary

Rcd
MW-RR#show bgp vpnv4 unicast summary

Rcd
10.0.2.9 4 65108 2708700 9422521 39w5d 16
10.0.2.10 4 65108 2040708 6318218 29w6d 7
10.0.2.13 4 65108 773182 6122217 33w6d 5

.
.
.
P_0080
BGP
MW-RR#show running-config bgp

! <BGP>
router bgp 65108
no bgp default ipv4-unicast
no bgp default route-target filter
no synchronization
bgp dampening
neighbor PE-node peer-group
neighbor PE-node remote-as 65108
no neighbor PE-node activate
neighbor PE-node password encrypt xxx
neighbor PE-node update-source loopback1
neighbor 10.0.2.9 peer-group PE-node
.
.
.
P_0081
BGP
MW-RR#show running-config bgp

! <BGP>
.
.
.
address-family vpnv4
neighbor PE-node activate
neighbor PE-node route-reflector-client
neighbor PE-node send-community
neighbor 10.0.13.1 route-reflector-client
neighbor 10.0.13.1 send-community
neighbor 10.0.13.2 route-reflector-client
neighbor 10.0.13.2 send-community
P_0082
BGP
show bgp vpnv4 unicast labels
show ip bgp summary
show ip bgp neighbor
show ip bgp route
show ip bgp peer-group PE-node
show ip bgp route network x
P_0083
BGP
OA-ER-DATA-B#show ip forwarding route summary
The total routes of global and all VPNs:
Route Source Count
static: 579
direct: 640
martian: 0
address: 641
ospf: 0
bgp: 119878
rip: 0
isis-level1: 0
isis-level2: 2204
icmp: 0
snmp: 0
nat: 0
pat: 0
vrrp: 236
ppp: 0
asbrvpn: 0
rsvpte: 0
user-ipaddr: 0
user-network: 0
static-vrf: 2
ipsec: 0
perVrf-label: 0
ps-busi: 0
ps-user: 0
ldp-area: 0
user-special: 0
dhcp-dft: 0
dhcp-static: 0
nat64-sl: 0
P_0084
ves: 0 Network Div | 02 April 2012 | P001
Total: 124180 BGP
BGP
Configuration
bgp neighbor commands
Route distinguisher
Route target
Import/export route target
Address family ipv4
Troubleshooting
bgp neighbor commands
Route distinguisher
Route target
Import/export route target
Address family ipv4
P_0085
BGP
BGP Summary
P_0086
BGP
THANK YOU!

BGP Routing Protocol: BY: Gebretinsae G

Uploaded by

Copyright:

Available Formats

BGP Routing Protocol: BY: Gebretinsae G

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BGP Routing Protocol: BY: Gebretinsae G

Uploaded by

Copyright:

Available Formats

BGP Routing Protocol

Network Div | 02 April 2012 | P001

 Till Dec,1969 there was only 4 sites

BL-GW-1#show ip forwarding route summary

X-GW-Routers#show running-config bgp

 BGP is used among ASs to ensure that there is no loop in the

OSPF Interior Link-state Yes Cost

RIP Interior Distance vector No Hope count

BGP Exterior Path vector No Path Vectors or

Styles of BGP implementations

 BGP is suitable in these occasions:

 BGP is not effective for all occasions. In the following

 BGP is a kind of enhanced distance vector

BGP has four kinds of messages:

 Open confirm ： after receiving OPEN message from neighbor,

 Established ： this is the last stage or stable status for the

 RTB (config) # router bgp 200

 BGP routes take effect condition ： next hop of

Here's a brief outline of the "order of preference" for filling the IP

Neighbor table: The connected BGP friends

Configure a BGP neighbor and AS number of a peer

Configure BGP Route Advertisement

 We use prefix filters in our gateway to control the incoming

41.189.225.51 4 30990 105902 797 00:13:16 1

105.16.0.137 4 37100 2877574 58609 2w6d 1

105.22.120.5 4 37100 2879986 58600 2w6d 1

BL-GW-1#show ip bgp neighbor out 41.189.225.25

Dest NextHop Metric LocPrf Path

10.153.192.45 4 64520 142502 146420 2w1d 1

10.153.193.188 4 64521 7254 7054 19:06:23 50

10.153.198.220 4 64521 253802 252244 3w6d 50

KK-ER-DATA-A#show running-config bgp

MW-RR#show ip bgp summary

MW-RR#show bgp vpnv4 unicast summary

10.0.2.10 4 65108 2040708 6318218 29w6d 7

10.0.2.13 4 65108 773182 6122217 33w6d 5

MW-RR#show running-config bgp

MW-RR#show running-config bgp

show bgp vpnv4 unicast labels

show ip bgp summary

show ip bgp neighbor

show ip bgp route

show ip bgp peer-group PE-node

show ip bgp route network x

Network Div | 02 April 2012 | P001

You might also like