Viewing 1 - 11 of 11
BGP_MP_NLRI.cap (2.9 KB)
|Packets: 24||Duration: 60s||Downloads: 10836|
IPv6 routes are carried as a separate address family inside MP_REACH_NLRI attributes.
4-byte_AS_numbers_Mixed_Scenario.cap (414 bytes)
|Packets: 4||Duration: 60s||Downloads: 6357|
Router "B" (AS 2) at 172.16.3.2 does not support 4-byte AS numbers, while router "A" (AS 10.1 / 655361) at 172.16.3.1 does.
Router "A" receives an UPDATE for the 18.104.22.168/8 subnet from an external router ("D") in the AS 40.1 / 2621441 (not shown), and it forwards it to "B" (pkt n. 2): AS_PATH contains "23456 23456" (the first stands for AS 10.1, the second for the originating AS 40.1), but NEW_AS_PATH contains the real 4-byte AS numbers.
At pkt n. 3 "B" receives the same subnet directly from "D" and sends it to "A", including the original NEW_AS_PATH attribute previously appended by "D".
4-byte_AS_numbers_Full_Support.cap (1.2 KB)
|Packets: 9||Duration: 56s||Downloads: 5831|
Router at 172.16.1.2 (hostname "D", AS 40.1 / 2621441) clears a previous established peering with 172.16.1.1 (hostname "A", AS 10.1 / 655361); They both support 32-bit ASN.
While opening the new session, they negotiate the "Four-octet AS Number Capability" (pkts n. 2 and 3).
Then, both "A" and "D" send some UPDATEs containing 4-octect encoded AS_PATH attributes (pkts n. 6 and 9). Please note: WireShark may show wrong paths unless you force 4-byte encoding in the Preferences / Protocols / BGP options.
BGP_MD5.cap (1.7 KB)
|Packets: 16||Duration: 61s||Downloads: 6999|
An EBGP with TCP MD5 authentication enabled
BGP_redist.cap (378 bytes)
|Packets: 2||Duration: n/a||Downloads: 6765|
The OSPF metric is preserved and propagated within the MPLS cloud by the MP-BGP MED attribute.
IBGP_adjacency.cap (2.3 KB)
|Packets: 17||Duration: 63s||Downloads: 4342|
Routers 3 and 4 form an internal BGP relationship. This is evidenced by the OPEN messages in packets #4 and #5, which show both routers belong to the same AS (65300). Also note that IBGP packets are not subject to a limited TTL as are EBGP packets.
EBGP_adjacency.cap (2.7 KB)
|Packets: 24||Duration: 182s||Downloads: 4413|
The external BGP adjacency between routers 1 and 2 is brought online and routes are exchanged. Keepalives are then exchanged every 60 seconds. Note that the IP TTL (normally 1) has been increased to 2 with ebgp-multihop to facilitate communication between the routers' loopback interfaces.
BGP_soft_reset.cap (2.0 KB)
|Packets: 17||Duration: 180s||Downloads: 4000|
R1 performs a soft bidirectional reset (clear ip bgp soft) on its adjacency with R2. The ROUTE-REFRESH message is visible in packet #7. Note that the TCP connection remains uninterrupted, and neither router views the reset as disruptive.
BGP_notification.cap (764 bytes)
|Packets: 9||Duration: n/a||Downloads: 3922|
R1 has been misconfigured to expect R2 to reside in AS 65100. R2 attempts to peer with R1 advertising itself correctly in AS 65200. R1 issues a NOTIFICATION in packet #5 citing a "bad peer AS" error and terminates the TCP connection.
BGP_hard_reset.cap (3.2 KB)
|Packets: 32||Duration: 208s||Downloads: 3815|
A hard reset (clear ip bgp) is performed on R1 for its adjacency with R2. Packet #7 shows R1 sending a packet with the TCP FIN flag set, indicating the connection is to be torn down. The TCP connection is then reestablished and UPDATEs are retransmitted.
Viewing 1 - 11 of 11