Thanks everybody for the feedback.  I understand what's happening now.

On Thu, Jun 20, 2013 at 12:33 AM, Samarth Chidanand <[email protected]>wrote:

> Yeah…You got that right Mike… Best to have specific Auto NAT’s to avoid
> asymmetric matches/RPF failures. ****
>
> ** **
>
> ** **
>
> Samarth Chidanand****
>
> Vice President of Technical Training – IPexpert India Inc****
>
> CCIE #18535 (R&S, Security)****
>
> CCSI #34585****
>
> ** **
>
> ** **
>
> ** **
>
> *From:* [email protected] [mailto:
> [email protected]] *On Behalf Of *Mike Rojas
> *Sent:* Thursday, June 20, 2013 8:59 AM
> *To:* Joe Astorino; [email protected]
>
> *Subject:* Re: [OSL | CCIE_Security] 8.4 VPN Hairpin****
>
> ** **
>
> Joe;
>
> (Stupid Outlook sorry for the previous e-mail)
>
> object network obj_any
>  nat (any,outside) dynamic interface
>
> Lets say that the VPN client goes out being Natted to the interface IP,
> everything is good, BUT, the reply packet from the source on the internet,
> will ALSO try to hit the same NAT because the object contains any on it.
>
> With this object****
>
> object network obj-192.168.50.0****
>
>  nat (outside,outside) dynamic interface
>
> You are saying ONLY if the VPN subnet is comming on the outside NAT the
> host to the interface, the reply from the internet wont hit this because
> you are defining the VPN subnet on the object, hence wont be Natted and No
> RPF check should occur.
>
> Is the same error as 8.2 remember "Asymmetric NAT rules match for reverse
> and forward flows" because you will try to reach an IP and then, the reply,
> since the first object contains any,  will try to NAT the internet IP as
> well returning the RPF.
>
> Hope it helps.
>
> Mike ****
> ------------------------------
>
> From: [email protected]
> Date: Wed, 19 Jun 2013 21:31:17 -0400
> To: [email protected]
> Subject: Re: [OSL | CCIE_Security] 8.4 VPN Hairpin****
>
> Anybody? Really interested to know the answer. I have read everything I
> can find on the topic.****
>
>
> Sent from my iPhone****
>
>
> On Jun 19, 2013, at 9:42 AM, Joe Astorino <[email protected]>
> wrote:****
>
> So another NAT question with 8.4 code.  Say you have RA VPN configured
> such that the VPN pool is 192.168.50.0/24.  You wish to configure
> hairpinning so that VPN users can access the internet off the ASA.  ****
>
> I figured hey...I need to PAT for inside --> outside, DMZ --> outside as
> well as VPN outside --> outside so why not use "any" and be super
> efficient!??
> ****
>
>
> object network obj_any
>  nat (any,outside) dynamic interface****
>
> When connected to VPN and attempting to ping an internet address this
> fails due to rpf-check on the packet-tracer output.  Packet tracer shows
> these steps
>
> Phase: 8
> Type: NAT
> Subtype:
> Result: ALLOW
> Config:
> object network obj_any
>  nat (any,outside) dynamic interface
> Additional Information:
> Dynamic translate 192.168.50.50/0 to 50.198.34.193/52258
>
> Phase: 9
> Type: NAT
> Subtype: rpf-check
> Result: DROP
> Config:
> object network obj_any
>  nat (any,outside) dynamic interface
> Additional Information:
>
>
> I suppose what this is saying is that the nat (any,outside) bit would
> probably get hit again for the return traffic, thus trying to PAT the
> return traffic to the outside interface.
>
> If I do this it works:
>
> object network obj_any
>  nat (inside,outside) dynamic interface
> !****
>
> object network obj-192.168.50.0****
>
>  nat (outside,outside) dynamic interface
>
> ****
>
> I know this works, but I don't understand why the first option doesn't
> work.  With the first configuration I expect:****
>
> - packet comes in from the VPN on the outside interface and will be
> hairpin routed back out the same interface****
>
> - The nat (any,outside) rule matches as the ingress interface "any" should
> match the outside****
>
> - dynamic PAT is done such that the source is translated to the outside
> interface IP****
>
> - xlate is created****
>
> - When the return traffic comes back, it should see the xlate table and
> translate things back.  ****
>
> So I guess it comes down to the return traffic.  Why would the return
> traffic hit the nat (any,outside) rule and not just be unnatted going back
> to where it came from based on the xlate table.****
>
> Thanks for any help****
>
> ** **
>
> ** **
>
>
> --
> Regards,
>
> Joe Astorino
> CCIE #24347
> http://astorinonetworks.com
>
> "He not busy being born is busy dying" - Dylan****
>
>
> _______________________________________________ For more information
> regarding industry leading CCIE Lab training, please visit
> www.ipexpert.com Are you a CCNP or CCIE and looking for a job? Check out
> www.PlatinumPlacement.com****
>



-- 
Regards,

Joe Astorino
CCIE #24347
http://astorinonetworks.com

"He not busy being born is busy dying" - Dylan
_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit 
www.ipexpert.com

Are you a CCNP or CCIE and looking for a job? Check out 
www.PlatinumPlacement.com

Reply via email to