Support Forum
The Forums are a place to find answers on a range of Fortinet products from peers and product experts.
MotorOil
New Contributor

Progress IPsec Phase 2 Failure

We are running a 800C and are getting a IPsec Phase 2 error every hour (Key Lifetime set to 3600 on both ends).  Build v5.2.1,build618

 

Our connection is in-house to Amazon's data center (VPC connection for those familiar).  We have been able to duplicate the error on a test connection and have tried several settings to make it go away.  Turning off DPD, extending key lifetime (not adjustable on Amazon's end). 

10 REPLIES 10
emnoc
Esteemed Contributor III

That's probably okay, is the VPN staying up? or during the rekey are you having to do anything manual?

PCNSE 

NSE 

StrongSwan  

PCNSE NSE StrongSwan
MotorOil
New Contributor

drops for 15 to 30 seconds and reconnects on its own.

MotorOil

Phase 1 is fine, only the phase 2 is failing every hour.  Causes a disconnect of our key systems and staff needing to reconnect every hour....

emnoc
Esteemed Contributor III

That's not good. So is this a phase2 re-key or a phase1 re-key? ( monitor you phase1  SA SPI at the 1 hr mark )

 

PCNSE 

NSE 

StrongSwan  

PCNSE NSE StrongSwan
emnoc
Esteemed Contributor III

So how do you have the phase2 set in the cfg and mainly for the below items?

 

 

set auto-negotiate disable set keepalive enable set keylife-type seconds

set keylifeseconds 3600

 

 

 

This sounds like a phase2-keylife not matched or being honored.

 

 

PCNSE 

NSE 

StrongSwan  

PCNSE NSE StrongSwan
Alex_K
New Contributor

Hi!

 

Sorry for resurrecting this old thread but it looks like I'm having similar symptoms between Fortigate 100D and Amazon VPC.

 

In Log & Report->VPN Events every now and then I see negotiate failure messages "progress IPsec phase 2", Direction=inbound, Role=responder, RemotePort=500. It looks like the tunnel is always up and I have no problems pinging hosts from both ends, but since this new setup is not rolled out to users yet, I can't really say if it will be stable.

What I also noticed was that I could RDP into servers in Amazon without any issues. However, when I try to RDP from Amazon back to premises, it works, but every 2-3-4th time, meaning that in first several attempts I get a standard RDP connection error. During RDP failures, I can ping hosts fine.

 

 

linfante
New Contributor II

Bump.

 

Was there ever a solution to this? We are having the exact same issue. Pretty sure it's key lifetime as the VPN error messages happen every hour at the same time. FG800 and FG100 IPSec VPN to AWS as well.

Okabe
New Contributor

double BUMP!

 

This thread is 4 years old and does anyone already come out with a solution? 

mahmoud_nasef

Triple BUMP!

Top Kudoed Authors