Captive Portal (Local Directory) stops working after Directory Connector trial expire

[wektek]

Here's a screenshot taken from the "test" Slackware Linux box.

The previous screenshot was taken from a FC14 Linux box that was connected over HTTP to the same Untangle server (192.168.2.254) via VPN tunnel (unfortunately, each time I need to take a screenshot using the "test" Slackware Linux box, I have to jump into the car and drive over to the site ... I thought I'd save some time ... and fuel).

As you can see, I'm able to connect over HTTP to the Untangle server admin page using the "test" Slackware Linux box. I'm running 4 other Untangle installs, but have never encountered this issue before. The reason that I (mistakenly) thought it was as a result of the Directory Connector was because the problem started after the trial expired.

[dmorris]

Can you post a screenshot of your settings on the "Captive Page" in captive portal?

Yes, you've said that like six times, but I don't see what this has to do with Directory Connector.
If you want to believe thats what the issue is I can recommended that you just buy Directory Connector. However, I think you'll be sadly disappointed when that does not fix the issue.

[wektek]

Here you go. I've taken 2 screenshots - one of the upper portion of the page and one of the lower portion.

Re my Directory Connector reference in previous posting, I did say that I mistakenly thought it was the issue (my bad). You have said quite plainly that it's not the issue/cause and I fully believe you - no problem. If I've said something to annoy you, I apologise.

[dbunyard]

I suppose I'm a little tardy to the party but what does your "User Authentication" page look like?

[dmorris]

Hrm. I'm truly stumped.

I guess what I would try is to get on the slackware test machine, and telnet to cnn.com or google.com or somewhere (actually first check that it resolves to the correct IP - it should because Untangle will provide DNS to hosts that are still captive)

then type "GET /index.html HTTP/1.1"

it should respond with some sort of redirect to a certain URL.
You can also achieve the same effect using wget or curl, but they may try to follow the redirect.

The redirect should point to the captive portal login page.
192.168.2.254/something.
See if you can figure out what that URL is.
You should be able to reach that page since its hosted in the same place as the administration which you can already reach from the slackware machine.

edit: no offense taken. I just doubt it has anything to do with Directory Connector and I'd rather check first things first and grasp at straws second. If you saw something in the logs in /var/log/uvm or you see some evidence that its Directory Connector i'm all ears, but I don't have any theory why that would effect CP.

[wektek]

Hello. Here you go:

- user_auth_main.png
- user_auth_localusers.png
- captive_hosts.png (in case you need to see these settings too)

[dbunyard]

I have to say, I'm with Dirk on this one. It makes no sense why it doesn't work. I would follow his suggestion and see where it gets you. That's about the best I can think of. Maybe watch the sessions on the UT box also and see if there is anything odd.

[wektek]

Hi gents.

Tks for your help in the matter - very much appreciated. It's 04:00 hrs now here in Ireland and I haven't slept in over 20 hrs, so I'm going to get some shut-eye. I'll head back onsite later this morning and locally try out Dirk's suggestions and revert back with updates.

If all else fails could a reinstall of Captive Portal resolve the issue? I know I'm asking how long is a piece of string, but ...

Is there a way of backing up the Local Directory? I assume that a full system config backup would also backup whatever the issue is with CP, so that may not work as a method of restoring users (or is my brain melting at this stage)?