Skip to main content

VPN and Amazon AWS

Recently I have worked on few VPN projects where my customers requested VPN connections to their Amazon VPCs (Virtual Private Cloud). As you probably know, today’s companies are very complex and in the same building there can be many organizations, contractors, etc. In my project I had to set up VPN tunnels for three different organizations located behind the same VPN concentrator. One of them had their VPC in US, the second one in Europe. The project was completed without any issues. The problems appeared when third organization, located behind the same VPN concentrator. They requested a new tunnel to their VPC located in Europe. Let’s sum up all requirements:
  • Customer “A” has a tunnel to the VPC “A” in US DC
  • Customer “B” has a tunnel to the VPC “B” in Europe DC 
       Above ones already configured and both are working fine.
  • Customer “C” requires a new tunnel to the VPC “C” in Europe DC.
 
 <---customer site--->|         |<-----Amazon sites----->

                /----\   
               |Cust-A|
                \----/                      /----\   
                   |         --[US DC]-----| VPC-A|
                   |       /                \----/
                   |      /
                   |     /
  /----\         -----  /                   /----\ 
 |Cust-B|-------| ASA |--------[EMEA DC]---| VPC-B| 
  \----/         -----                 \    \----/ 
                   |                    \ 
                   |                     \        
                   |                      \  /----\   
                   |                        | VPC-C|
                   |                         \----/
                /----\   
               |Cust-C|
                \----/  

When I installed ASA I configured one outside interface with one public IP. Then when I started to work on first two VPNs I received two files (one per each customer) with configuration generated by VPCs. The config files contained all settings together with peers IPs (public). As I mentioned before the both tunnels were set up with no issues. When I asked for the third tunnel settings (Customer “C” -> VPC-C) I found the peer IP was the same as for the customer “B” tunnel. It meant Amazon didn’t allocate the public IP for each my customers VPCs. Amazon has only one public IP per region and in my case I couldn’t set up the tunnel for the customer “C” with the current configuration. The easiest way is to have only one tunnel for VPC-B and VPC-C but from security reasons the customer “B” can’t have access to “VPC-C” and vice versa. One of the proposed solutions was a new design of ASA and implementation of sub-interfaces on the outside interface with more public IPs. Of course for such solution you need to have more public IPs what can be expensive depends on the number of IP addresses. The second option is using a BGP protocol, accepted by Amazon, but the ASA doesn’t support it.

                /----\   
               |Cust-A|
                \----/                             /----\   
                   |                --[US DC]-----| VPC-A|
                   |              /                \----/
                   |             /
                   |            /                  /----\     
  /----\         ----- gig0/0.1 ------ [EMEA DC]--| VPC-B|  
 |Cust-B|-------| ASA |gig0/0.2 -------[EMEA DC]   \----/
  \----/         ----- gig0/0.3                 \           
                   |                             \ 
                   |                              \        
                   |                            /----\   
                   |                           | VPC-C|
                   |                            \----/
                /----\   
               |Cust-C|
                \----/  

If you planning to work with Amazon products like VPC , take into account this one huge limitation because if your environment requires 99.999% availability, every big change in your design can be painful.

Comments

  1. HI, really interesting one, overall now for me! i will have later a deeper read on this one for sure

    ReplyDelete
  2. working for Amazon you can even suggest them to be more flexible :)

    ReplyDelete

Post a Comment

Popular posts from this blog

What should you know about HA 'override enabled' setting on Fortigate?

High availability is mandatory in most of today's network designs. Only very small companies or branches can run their business without redundancy. When you have Fortigate firewall in your network you have many options to increase network availability. You can use Fortigate Clustering Protocol ( FGCP ) or Virtual Router Redundancy Protocol ( VRRP ). FGCP has two modes: 'override' disabled (default) and 'override' enabled . I'm not going to explain how to set up HA as you can find many resources on Fortinet websites: https://cookbook.fortinet.com/high-availability-two-fortigates-56/ https://cookbook.fortinet.com/high-availability-with-fgcp-56/ Let's recap what is the main difference between them. The default HA setting is 'override' disabled and this is an order of selection an active unit: 1) number of monitored interfaces - when both units have the same number of working (up) interfaces check next parameter 2) HA uptime - an

MAC Authentication Bypass

One of the method to control your network is using MAB feature. It is helpful in case you have devices without dot1x functionality. Today I will try to implement basic configuration and analyze log messages. There is only one switch SW1 and one device attached to port Fa1/0/2.   ! aaa new - model aaa authentication dot1x default group radius ! ! int Fas1 / 0 / 2 authentication host - mode single - host authentication port - control auto mab ! I haven’t configured ACS yet but let’s see what error message I receive:   SW1 ( config - if ) # mab - ev ( Fa1 / 0 / 2 ): Received MAB context create from AuthMgr mab - ev ( Fa1 / 0 / 2 ): Created MAB client context 0x1100000F mab : initial state mab_initialize has enter mab - ev ( Fa1 / 0 / 2 ): Sending create new context event to EAP from MAB for 0x1100000F ( 0000.0000 . 0000 ) mab - sm ( Fa1 / 0 / 2 ): Received event 'MAB_START' on handle 0x1100000F mab : during state mab_initia

Inpection of asymmetric sessions on FortiGate

There is one feature available on FortiGate, and I think you should know it, as it modifies a bit what we know about stateful firewalls. In past every packet was treated individually and you had to create policies in both directions. With stateful firewalls we can track connections, and by checking couple of attributes, we can treat them as part of the same session. For example when you initiate connection from a host1 to host2, the returning connection from host2 to host1 will be treated as part of the same connection (session). They have to have the same source/destination and destination/source IPs, port numbers and interfaces.There is an exception from this rule and FortiGate in some specific cases can accept connections on port which was not used in the initial connection. Let me explain how it works on the below example:      The host1 has a default gateway on R1 (10.0.1.2), but you may notice that it is not the optimal path to host2 subnet. When we analyze the packet flo