aws-vpn-client: Linux: transient authentication failures

Hi, I got this working on Linux!

However, I’m running into a problem where authentication will fail what seems like somewhere between 30 and 50% of the time, with no changes at all between retries:

AUTH: Received control message: AUTH_FAILED,Invalid username or password

I’m not familiar enough with the internals of the authentication scheme to know what would cause this. Given the transient nature though, I wonder if maybe the message is being truncated or improperly formatted at some point in the process.

I realize that this issue description isn’t much to go on, so if you have any suggestions of other information I could provide, or techniques I could use to narrow down the source of the problem, let me know.

About this issue

  • Original URL
  • State: closed
  • Created 4 years ago
  • Comments: 24 (8 by maintainers)

Most upvoted comments

Hi, i am 90% sure that cause is very simple - native client keeps tcp connection open and my implementation not. I will try to fix that but no timelines, very busy atm

finally got some time working on that. will try to do some updates this week

Hi, i think i will try to find some time today or tomorrow to fix that

I had exactly the same problem. I often got “AUTH_FAILED” and it took many tries to establish a connection. But I think I solved at least my case.

The problem seemed to be that the first and second openvpn commands use different remote IP addresses. It seems that --remote flag doesn’t work when the vpn.conf includes remote line. So basically what I did:

  1. Removed remote and remote-random-hostname from vpn.conf.
  2. In aws-connect.sh changed port from 1194 to 443. Also make sure VPN_HOST is correct.

So far I have not seen any AUTH_FAILED messages, but I’ve only used it for a day.

Hopefully will find some time next week. Should be easy to fix but will need some rework