As Considering that F5 is a full proxy, How to capture the server-side traffic, Some notes giving
Source IP must be client IP, where actual request coming ,in other notes VIP is source Ip for Server-side traffic? Any one give clear idea on this?
without SNAT: None(no Automap)
Client IP :172.16.16.100
VS : 172.16.16.1
Pool_HTTP with 1 member : 1. WEB_A : 10.10.10.11
SNAT: None(no Automap)
Something I always do is capture some more information on the BIG-IP and use Wireshark to filter it out.
This command is what F5 Support usually use:
tcpdump -nn -s0 -i 0.0:nnnp -w /shared/tmp/Cxxxx_tcpdump_$(date +%d_%b_%H_%M_%S)_$HOSTNAME.pcap host x.x.x.x or host y.y.y.y or host z.z.z.z
x.x.x.x = IP address of client connecting to the virtual server. Or if this IP address is unknown, the IP address of the virtual server.
y.y.y.y = IP address of first pool member
z.z.z.z = IP address of second pool member.
The -i 0.0:nnnp flag will capture traffic on all VLANs and it will also enable the F5 Ethernet Trailer. The F5 Ethernet trailer will gather F5 specific information which can be analyzed in Wireshark.
Then, using the F5 Ethernet Trailer, I filter out the sessions based on the Flow Session ID, captured by the F5 Ethernet Trailer. Jason describes it in detail in the same article.
That way, the tcpdump syntax is easy to remember, it does not miss any information and it is quick to work with in Wireshark.
