Pretty sure I see the problem…

On Sun, Jun 3, 2018 at 1:30 PM, Stephen Davies <sdav...@sdc.com.au> wrote:
> Here is the debug output:[root@se5 ~]# openconnect -vvvv 
> --config=/etc/openconnect.conf --useragent="Cisco AnyConnect VPN Agent for 
> Windows 4.2.01035" remotehost
> POST https://remotehost/vendor

The "/vendor" value here must be the *same value* as the placeholder
you called 'ggggg' in the config file.

> Attempting to connect to server 1.2.3.4:443
> SSL negotiation with remotehost
> Connected to HTTPS on remotehost
> Got HTTP response: HTTP/1.1 404 Not Found
> Cache-Control: no-cache
> Pragma: no-cache
> Connection: Close
> Date: Sun, 03 Jun 2018 10:22:24 GMT
> X-Frame-Options: SAMEORIGIN
> HTTP body http 1.0 (-1)
> SSL socket closed uncleanly
> Unexpected 404 result from server
> GET https://remotehost/vendor
> Attempting to connect to server 1.2.3.4:443
> SSL negotiation with remotehost
> Connected to HTTPS on remotehost
> Got HTTP response: HTTP/1.0 302 Temporary moved
> Set-Cookie: tg=string; path=/; secure
> Content-Length: 0
> Cache-Control: no-cache
> Pragma: no-cache
> Connection: Close
> Date: Sun, 03 Jun 2018 10:22:24 GMT
> X-Frame-Options: SAMEORIGIN
> Location: /+webvpn+/index.html
> HTTP body length:  (0)
> GET https://remotehost/+webvpn+/index.html
> SSL negotiation with remotehost
> Connected to HTTPS on remotehost
> Got HTTP response: HTTP/1.1 301 Moved Permanently
> X-Transcend-Version: 1
> Location: https://remotehost/+CSCOU+/anyconnect_unsupported_version.html
> Content-Type: text/html
> Content-Length: 0
> HTTP body length:  (0)
> GET https://remotehost/+CSCOU+/anyconnect_unsupported_version.html
> Got HTTP response: HTTP/1.1 200 OK
> Content-Type: text/xml
> Transfer-Encoding: chunked
> Cache-Control: no-cache
> Pragma: no-cache
> Connection: Keep-Alive
> Date: Sun, 03 Jun 2018 10:22:24 GMT
> X-Frame-Options: SAMEORIGIN
> HTTP body chunked (-2)
> Please upgrade your AnyConnect Client
> Failed to obtain WebVPN cookie
>
> Here is the config file:
>
> [root@se5 ~]# cat /etc/openconnect.conf
> usergroup=gggg
> user=xxxx
> interface=tun0

You are using the "usergroup" setting. Remove it.

Usergroup doesn't do what you expect: it changes the URL that the
client submits a POST to… which leads to a 404 error… which leads to
the client retrying with GET… which is effectively the behavior of
very old clients… which leads to a redirect to a page for old clients.
I think that probably you used to have an old Cisco server which
actually used the URL-as-usergroup mechanism, and it got replaced with
a newer one.

Assuming it works without --usergroup, and you get a prompt asking you
which realm/group to join… the option that you want to use to set this
automatically is --authgroup, NOT --usergroup.

  GROUP: [Employees|Contractors|Pandas|Leprechauns]: Leprechauns
  Please enter your username and password.
  Username: johnsmith
  Password: ********

  => openconnect --authgroup=Leprechauns --user=johnsmith

If I'm right about this… the manual for the --usergroup option should
be clarified to tell people that they probably want the --authgroup
option. :-D

Dan

_______________________________________________
openconnect-devel mailing list
openconnect-devel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/openconnect-devel

Reply via email to