Launcher doesn't open. (can anyone access Frontiers store website?)

[Joe Rowland]

Failed to synchronise time with server

This will most likely be due to a temporary network or server issue. Please check your connection to the internet. If you have not run your launcher in a long time, check to see if this one has not been superseded.

[Joe Rowland]

I was playing the game yesterday just fine.

[Finn]

do you use ipv6? In the last 24 hours have had to disable ipv6 on my adaptor to get into the game (i'm dual stack), auth.frontierstore.net seems to currently be timing out on ipv6 but works fine on ipv4.

[Joe Rowland]

Finn = Legend

[Finn]

I dont know how to report this issue to frontier, I had a look at their support site and got kinda lost.

[Joe Rowland]

Yeah at least for now I can load the game and site though. Before that I was powerless. Thanks man.

[drkazimoto]

Originally posted by Finn:

do you use ipv6? In the last 24 hours have had to disable ipv6 on my adaptor to get into the game (i'm dual stack), auth.frontierstore.net seems to currently be timing out on ipv6 but works fine on ipv4.

Thank you Cmdr, you are a hero

[ToysRsus]

Originally posted by Finn:

do you use ipv6? In the last 24 hours have had to disable ipv6 on my adaptor to get into the game (i'm dual stack), auth.frontierstore.net seems to currently be timing out on ipv6 but works fine on ipv4.

Same here. Thanks for the hint!

I am surprised that not more people are affected. Maybe not all IPv6 networks are affected? I'm going through an IPv6 tunnel from Hurrican Electric.

[edenist]

Can confirm, their HTTPS is failing on IPv6.

$ curl -6 https://auth.frontierstore.net -v
*   Trying 2a05:d018:d09:dd02:b87:e06a:188:8e4f:443...
* TCP_NODELAY set
* Connected to auth.frontierstore.net (2a05:d018:d09:dd02:b87:e06a:188:8e4f) port 443 (#0)
* ALPN, offering h2
* ALPN, offering http/1.1
* Cipher selection: ALL:!EXPORT:!EXPORT40:!EXPORT56:!aNULL:!LOW:!RC4:@STRENGTH
* successfully set certificate verify locations:
*   CAfile: /etc/ssl/certs/ca-certificates.crt
  CApath: none
* TLSv1.2 (OUT), TLS header, Certificate Status (22):
* TLSv1.2 (OUT), TLS handshake, Client hello (1):
* Operation timed out after 300283 milliseconds with 0 out of 0 bytes received
* Closing connection 0
curl: (28) Operation timed out after 300283 milliseconds with 0 out of 0 bytes received

Hopefully they get it fixed up soon. Disabling IPv6 to play is a bit annoying ;-)

[ToysRsus]

Originally posted by edenist:

Hopefully they get it fixed up soon. Disabling IPv6 to play is a bit annoying ;-)

What's really annoying is that whensoever I forget to turn off IPv6 before the launch, it not only runs into the error, but also "updates" the whole game afterwards, which then costs 20 GB and an hour of download and installation time.

[edenist]

Originally posted by 1xok:

What's really annoying is that whensoever I forget to turn off IPv6 before the launch, it not only runs into the error, but also "updates" the whole game afterwards, which then costs 20 GB and an hour of download and installation time.

That is really annoying :/ Can't say that has happened to me fortunately...

I have, however, developed a workaround for those on linux, which lets you get past the broken IPv6 API server without disabling IPv6 on your machine. [Presumably this should work on windows too with the relevant changes to file locations].

I sniffed the packets and the client is attempting to connect to api.zaonce.net.

Add the following line to your /etc/hosts file, where <ipv6 addr> is the address of your interface.

 <ipv6 addr> api.zaonce.net 

Then, run the following command [as root/sudo!] which will forward traffic from an IPv6 socket through to the IPv4 server for api.zaonce.net [resolves to 54.77.204.156]

 ncat -l -p 443 -c "nc 54.77.204.156 443" -k 

*note: ncat is included with the 'nmap' package on most distros, and essentially provides IPv6 support to netcat.

The game will now run as normal with IPv6 enabled :-)

[ToysRsus]

Originally posted by edenist:

 ncat -l -p 443 -c "nc 54.77.204.156 443" -k 

But this will redirect all HTTPS traffic to 54.77.204.156. It's a way to analyze the problem, but nothing I'm going to do. Frontier needs to restore IPv6 connectivity for authentication service.

I still wonder why so few people complain. Half of Germany primarily uses IPv6 because of DS light.

[edenist]

Originally posted by 1xok:

Originally posted by edenist:

 ncat -l -p 443 -c "nc 54.77.204.156 443" -k 

But this will redirect all HTTPS traffic to 54.77.204.156. It's a way to analyze the problem, but nothing I'm going to do. Frontier needs to restore IPv6 connectivity for authentication service.

I still wonder why so few people complain. Half of Germany primarily uses IPv6 because of DS light.

No, not all HTTPS traffic. It opens a socket on port 443 for your IPv6 interface. That is, it is listening at localhost:443. It will only forward traffic going to localhost:443. The override added to /etc/hosts changes api.zaonce.net to go to localhost, but all other port 443 traffic will carry on as usual.

[ToysRsus]

It's working again, though! Some Frontier admin seems to have noticed the problem.

Originally posted by edenist:

No, not all HTTPS traffic. It opens a socket on port 443 for your IPv6 interface.

Yes, just the incoming traffic. But is in conflict with a local running Apache server on my computer. I also don't want to add IP addresses to /etc/hosts. You could do something with a Nginx, but honestly I don't think this are good solutions. If IPv6 doesn't work, you have to turn it off, best isolated for Elite: Dangerous. This is still the cleanest solution in this case if you have a dual stack.

If this happens more often, I would create an IPv4-only gai.conf for Elite which I map into the namespace of Elite via the unshare command and a bind mount. A colleague gave me this idea today. This could work. It would only affect Elite and nothing else.

[edenist]

Originally posted by 1xok:

But is in conflict with a local running Apache server on my computer.

Oh, well then yes of course it will cause a conflict then :p

If you're hosting things on the same box then it probably isn't a solution. But for me it's quick and easy and avoids disabling IPv6.

Glad it's back and working again, it's great to see Frontier supporting IPv6.... I wish more devs would do so!