AWS VPN service outage due to DNS not resolve

Question

Hi,

We have three AWS accounts that have AWS VPN Client services configured. Two of them are not working. The issue started happening suddenly on January 14 2022 (with no known changes to our configurations).

I have sanity checked our OVPN configurations. Booting up my Wireshark on my laptop, it seems that the reason why I could not connect is likely due to 
failure to resolve DNS A records related AWS VPN related services.

I saw the following DNS names got queried (all of which the DNS servers responded with "Standard query response, No such name"):

For the first account, DNS A record queries that failed:

1. ed-.cvpn-endpoint-0a..prod.clientvpn.us-east-1.amazonaws.com 
2. 00-.cvpn-endpoint-0a..prod.clientvpn.us-east-1.amazonaws.com
3. .cv--0a..prod.clientvpn.us-east-1.amazonaws.com
4. a4-.cvpn-endpoint-0a..prod.clientvpn.us-east-1.amazonaws.com
5. 82-.cvpn-endpoint-0a..prod.clientvpn.us-east-1.amazonaws.com

For the second account, DNS A record queries that failed:

1. 12-.cvpn-endpoint-0c-.prod.clientvpn.us-east-1.amazonaws.com

DNS servers (mine and some others on the Internet) could not resolve those. So, it's unlikely just my DNS servers as I checked it against multiple DNS servers.

Is AWS Client VPN currently experiencing outage? Is it just more widespread than issues in my two accounts? Or is my configuration wrong somewhere?

Thanks,
Rob

Accepted Answer

We managed to get them "resolved"; at least whatever seemingly irrelevant things I did managed not to trigger this bug.

1. I went to the Client VPN endpoint settings and found that there were some lingering security group associations that show up in GuI but the underlying SG is already gone.
I deleted those SGs.
2. Turn logging settings off and then back on again

Those actions do not seem to be directly related to AWS not being able to resolve VPN endpoint DNS -- but somehow correlated with the fixes.

AWS VPN service outage due to DNS not resolve

相關內容