cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
2279
Views
0
Helpful
7
Replies

Intermittent lost to VPN tunnelling (Gateway to Gateway) to AWS

VAEGROUP01
Level 1
Level 1

Hi,

I've recently installed and configured a Cisco RV325 Small Bussiness router in our office. I configured Gateway to Gateway VPN tunnelling to AWS and I was successfully connected to AWS and can ping to the cloud computer. However this tunnel dropped intermittently for 2-3 mins and reconnect again. I am sure what else to check or enabled on the Cisco router. Any suggestions or advise is greatly appreciated.

 

Thanks  

7 Replies 7

stonefury
Level 1
Level 1

Do you know that you must keep traffic coming into the AWS VPN?

You can create an EC2 instance and establish and ssh connection.

 

while :; do ssh -v -i blah.pem ec2-user@x.x.x.x; sleep 5; date >> /tmp/ssh-aws.log; $(date | mail -s "NOTICE: $HOSTNAME SSH Re-Connect" "youremail"); done

Require this to keep the connection alive.

$ cd .ssh

$ cat config
Host *
Protocol 2
KeepAlive yes
ServerAliveInterval 5

 

I am very interested in VPN tunnel stability in general with AWS. I am finding our tunnel drops randomly, not every day, not predictable.  Not using BGP, which I've heard greatly improves stability, using BGP and the secondary tunnel.

 

But my problem might not even be router related.

Thanks Stonefury for your comments but I am not quiet sure what you're saying, Can you please phrase it in a simple terms? Thanks and I appreciate your help.

 

Do you know that you must keep traffic coming into the AWS VPN?

You can create an EC2 instance and establish and ssh connection.

- See more at: https://supportforums.cisco.com/discussion/12534086/intermittent-lost-vpn-tunnelling-gateway-gateway-aws#sthash.rDLdM4DW.dpuf

o you know that you must keep traffic coming into the AWS VPN?

You can create an EC2 instance and establish and ssh connection.

- See more at: https://supportforums.cisco.com/discussion/12534086/intermittent-lost-vpn-tunnelling-gateway-gateway-aws#sthash.rDLdM4DW.dpuf

o you know that you must keep traffic coming into the AWS VPN?

You can create an EC2 instance and establish and ssh connection.

- See more at: https://supportforums.cisco.com/discussion/12534086/intermittent-lost-vpn-tunnelling-gateway-gateway-aws#sthash.rDLdM4DW.dpuf

Not sure how familiar you are with the site-to-site tunnels, but the tunnel to the AWS VPN will tear down if no traffic is coming from behind your cisco router.

 

One option is to ping from a machine at your business location to an EC2 instance once the VPN is established.

When I create an EC2 instance, I get a public and a private (local) IP address for that instance. Let's just say 54.x.x.x and the local IP is 10.0.x.x.

 

From a machine behind the cisco router, just do:

ping 10.0.x.x

 

Assuming your tunnel is up, the ping will work, correct?  Leave the ping running and see if your tunnel stays up permanently. 

 

 

Thank you for your quick respond.

Yes I've ping the machine on AWS cloud continuously but yet the tunnel is still dropping off.

I noticed that my Remote gateway is 54.240.204.91 but for some odd reason after an hour the tunnel drop off and it is trying to communicate with 54.240.204.92 .

 

Please see my attachment for more information. I thank you for your help.

Hello VAEGROUP01,

That is interesting that it drops the connection by itself in an hour. It would be interesting to get the answers for the questions below:

Is there a connection timeout limit set somewhere on any of the ends?

Is there a way you can see the logs at the time when its dropped?

Also, when you say that it is connected to xxx.91 and after dropping it tries to connect to x.x.x.92, is there a load balancer somewhere?

And does it ever successfully connect to x.x.x.92?

Regards,

Saji Thomas

Hi Saji,

I can not find connection timeout set on the Cisco router, however I've ticked " Keep Alive".

In about an hour it dropped because it is trying to connect to x.x.x.92. Come to think of it there is a WAN load balance, WAN1 set as primary.

It has never successfully connect to x.x.x.92 , it normally reconnect itself after 10 sec to 2 mins.

Someone suggested that I should look for AWS Fully Qualified name for dynamic remote gateway IP address of x.x.x.91 and x.x.x.92. What do you think?

Thanks for your help.

As Stonefury said, "the tunnel to the AWS VPN will tear down if no traffic is coming from behind your cisco router." Did you check with AWS support if that is correct?

Also, check if there is any AWS related FQDN that points to the x.x.x.91/92 IP address (like a load balancer or something). And I am sure that their is something related to the 1 hour time mystery as it always disconnects .91 after one hour and never connects to .92.

Not sure if Amazon will let you do it but is it possible that you can create something like a static connection to that .91 IP. Also continuously ping (-t) the .91 IP so that something trying to keep it alive.   

Thanks! Saji

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: