Can't Turn Off VPN Client

[Preskitt.man]

I have 5 VPN Clients defined on my AC86U router running Merlin 386.5_2. Client #4 will not turn off. If in the VPN Director, I click on Stop Client - goes through a process, but client remains on. If I go to the client, it switch it from On to Off, and then click Apply, it is back on after the Apply finishes. The Connect after reboot flag is set to No. Yet, interface always shows this client connected to the VPN server. I have clients defined on all the other slots as well, but they turn on/off as expected. Mostly I leave them off. Frankly, I'd be happy enough to just delete this client, but don't see any way to do that.

 

[Preskitt.man]

One small update - clicked the "Default" button at bottom of page - which brings up a warning this brings client back to factory default settings. Told it to proceed, and and when all was said and done, the client is still there and connected.

 

[Preskitt.man]

Can't believe not a single suggestion/comment on this. Appears to be an obvious bug to firmware. Certainly not a design feature.

Ideas, anybody?

 

[L&LD]

How long has the router been 'up'? Does a reboot help?

Have you tried using a different computer and/or browser to navigate the GUI?

Are clients currently connected through VPNs? Does turning them off/rebooting those client devices change the outcome?

 

[Viktor Jaep]

Can't believe not a single suggestion/comment on this. Appears to be an obvious bug to firmware. Certainly not a design feature.

Ideas, anybody?

Have you tried wiping out all the fields/settings in your VPN client #4 slot... hitting apply?

You can also try this from a different angle... open up an SSH window and type the command:

service stop_vpnclient4

Then run:

echo $(nvram get vpn_client4_state)

If it comes back with "0", then its disconnected. If it's "2", then it's still connected.

 

[Preskitt.man]

OK - tried the SSH approach first - value came back "2" - still connected.
I clicked on "Default" at bottom - accepted the warning that this would put the client back to it's default state - and then clicked apply. Client stayed "On" with a connection to the vpn server.
I entered an invalid Username and Password for Authentication - click apply - still comes back as On, with a connection to the vpn server.
Erased the Server address and Port number. Clicked apply - still comes back connected to a valid address.
This client has taken on a life of it's own.
I have definitions set for all 5 VPN clients - each client points to a different vpn server. Each server has a different geo-location. Except for client 4, they all work correctly. Each vpn server is from PIA and is an openvpn server. How I could get reconnected with a bad username/password really, really baffles me.

Here are the Syslog entries for my various endeavors
Jun 6 21:57:25 rc_service: service 9616:notify_rc stop_vpnclient4
Jun 6 21:57:25 rc_service: waitting "update_sta_binding" via cfg_server ...
Jun 6 21:57:40 rc_service: skip the event: stop_vpnclient4.
Jun 6 21:59:58 rc_service: httpd 2046:notify_rc stop_vpnclient4;clearvpnclient4
Jun 6 21:59:58 rc_service: waitting "update_sta_binding" via cfg_server ...
Jun 6 22:00:13 rc_service: skip the event: stop_vpnclient4;clearvpnclient4.
Jun 6 22:04:36 rc_service: httpd 2046:notify_rc restart_vpnclient4
Jun 6 22:04:36 rc_service: waitting "update_sta_binding" via cfg_server ...
Jun 6 22:04:50 rc_service: skip the event: restart_vpnclient4.
Jun 6 22:05:23 rc_service: httpd 2046:notify_rc stop_vpnclient4
Jun 6 22:05:23 rc_service: waitting "update_sta_binding" via cfg_server ...
Jun 6 22:05:38 rc_service: skip the event: stop_vpnclient4.
Jun 6 22:05:39 rc_service: httpd 2046:notify_rc restart_vpnclient4
Jun 6 22:05:39 rc_service: waitting "update_sta_binding" via cfg_server ...
Jun 6 22:05:53 rc_service: skip the event: restart_vpnclient4.
Jun 6 22:11:38 wlceventd: wlceventd_proc_event(491): eth5: Deauth_ind 48:44:F7:53:4AC, status: 0, reason: Disassociated due to inactivity (4), rssi:0
Jun 6 22:11:38 wlceventd: wlceventd_proc_event(508): eth5: Disassoc 48:44:F7:53:4AC, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0
Jun 6 22:12:31 rc_service: httpd 2046:notify_rc restart_vpnclient4
Jun 6 22:12:31 rc_service: waitting "update_sta_binding" via cfg_server ...
Jun 6 22:12:46 rc_service: skip the event: restart_vpnclient4.

 

[Preskitt.man]

How long has the router been 'up'? Does a reboot help?

Have you tried using a different computer and/or browser to navigate the GUI?

Are clients currently connected through VPNs? Does turning them off/rebooting those client devices change the outcome?

This issue has been going on for at least several weeks - and router is rebooted every Tuesday morning at 3am.
I have navigated to the router from two different computers - one running Windows 11, the other Windows 10.
No clients are connected to the VPN. The server for this vpn client is an openvpn server from PIA and located in Denmark.

 

[Preskitt.man]

Ok - wouldn't say problem is solved, but it is fixed.
After futzing around with various settings, but the client kept coming back as if nothing was wrong, I then rebooted, logged back in and navigated to the client gave and it had error on connection.
Turned off client - and this time it went off.
Re-entered correct values and turned client on - it connected properly.
AND most importantly, I now turned the client off, and it went off.
Very weird.

 

[criminala]

Facing the same issue currently .
Found this thread but no real solution apparently .

The sad thing is that my outgoing connections for the NAS were going through this particular VPN instance and currently is getting denied any outgoing connections . Even after removing the client from this VPN instance rule .
That's why I thought to disable the VPN client connection as a solution (which didnt work) , after that I just wanted to delete and get rid of the VPN client (click on default button) but that also didn't work !

Rebooted the router several times but no help .

VPN comes back and stays on and keeps denying outgoing connection for NAS .


RT-AX86U router is on 388.2_2 .

Tried the SSH commands but as you can see , vpn stays on .

@RT-AX86U:/tmp/home/root# service stop_vpnclient2

Done.
@RT-AX86U:/tmp/home/root# echo $(nvram get vpn_client2_state)
2

Also none of my other VPN slots on the router were able to connect anymore .


In the end what did it for me was :

Set all bogus values for the corrupted VPN . It will still connect .
Autoconnect on boot off . save .
HARD unplug the router (no powerbutton usage)
wait 10 seconds
HARD plug the router back in (no powerbutton usage)
Now I could disable the vpn connection , and also delete it with the default button .
At this point I could also connect other vpn slots again .

O and say an Hail Mary after each of the above steps .