You are here

Tunnel server port 5525 not open

18 posts / 0 new
Last post
N8JJ
Tunnel server port 5525 not open
I can't get the tunnel server to work.

When I probe port 5525, it is closed but port 80 is open and I can connect to the node via it's wan port.
So I know I have a connection.

This used to work.

Any ideas?
AE6XE
AE6XE's picture
Was the IP address of your
Was the IP address of your node reserved on your home network?  If the home router, cable mode, or node was power cycled, maybe there is a new IP address and the previous port 5524 forward is now going to a different/old address?

Joe AE6XE
N8JJ
tunnel server port
5525 right, not 5524?  I did port scan them all and only 80 showed up.
Yes it's on the correct IP address, because I can connect to the port 80 on that address.
I think I will try to connect a laptop directly to the VLAN 1 port from the switch to make sure my network isn't blocking it somehow.
AE6XE
AE6XE's picture
Yea, typo.  5525.    test

Yea, typo.  5525.    test access with "telnet <ip address> 5525".    If it connects and returns a vtun server info, then the network path is good.   If it does not connect then ether the vtun server is not running or there is a network config issue.

Install of the vtun packages needs to be done via the UI buttons and the node should have rebooted in the process.  

N8JJ
vtund not running
This is a NSM9 and the tunnel software was loaded and the tunnel UI seems to work.
however, vtund is NOT running.
It's been rebooted many times.
Is there an issue with tunnels and the NSM9?
 
AE6XE
AE6XE's picture
Let's take a look at the
Let's take a look at the details. Can you attach the support data? Button at bottom of Administration page in setup. Joe AE6XE
K5DLQ
K5DLQ's picture
do you have an ENABLED tunnel
do you have an ENABLED tunnel client or server connection in the UI?
N8JJ
support data

support data

Support File Attachments: 
AE6XE
AE6XE's picture
We are zeroing in...   I do
We are zeroing in...   I do not see the vtund service/process running.  Let's see if we can find out why.    Do you have a way to telnet or ssh into the node and run a command?   Need to see the output when the vtund service is restarted.  Can you run the command "/etc/init.d/vtund restart"?     Need the output from this command to see what is happening.

As a fallback, reboot the node.  Then within a few minutes capture another support download.  We're looking for messages in 'logread' trying to start up the service, should look something like this:
 
Fri Sep 21 13:47:19 2018 daemon.info vtund[1482]: VTUN server ver 3.X 09/20/2018 (stand)

Joe AE6XE

 
N8JJ
Tunnel
Of course.  It works on a NSM2 but not on the NSM9.  However, the client worked on the M9.
2G devices are not very usefull around here, the 900 is much better over the air.
N8JJ
It said "you need to specify
It said "you need to specify whom to kill" when I run that command
Does that mean there is not enough memory to run it?
I have nothing else loaded on the node.
Here is the new dump
 
Support File Attachments: 
AE6XE
AE6XE's picture
N8JJ,  looks like you only
N8JJ,  looks like you only need to update how the data is entered in setup.  Here's the error:

Fri Sep 21 20:11:31 2018 user.emerg syslog: vtund[1286]: Invalid clause 'W8BI-DARA-OMNI.LOCAL.MESH-172-31-189-32' line 12
Fri Sep 21 20:11:31 2018 user.emerg syslog:
Fri Sep 21 20:11:31 2018 daemon.err vtund[1286]: Invalid clause 'W8BI-DARA-OMNI.LOCAL.MESH-172-31-189-32' line 12
Fri Sep 21 20:11:31 2018 user.emerg syslog: vtund[1286]: No hosts defined
Fri Sep 21 20:11:31 2018 daemon.err vtund[1286]: No hosts defined

I suspect the problem will be solved by removing ".LOCAL.MESH" in the name.  The code may not be checking for all invalid possibilities.

Replace "V8BI-DARA-OMNI.LOCAL.MESH" with "V8BI-DARA-OMNI".

Joe AE6XE
 
K5DLQ
K5DLQ's picture
yes.  that probably makes the
yes.  that probably makes the vtun client name WAY too long.
 
N8JJ
Success Thanks
That seems to have fixed that problem.  Is there a name length limit?  What is it?
Another symptom is that when you hit the save button, the node goes out to lunch for awhile instead of coming back with the setting saved message.
I have to go back to the status screen.
It seems to do this about half the time and I haven't been able to identify the conditions.  Maybe it is overwriting something.
It doesn't seem to reboot because the uptime does not reset but it is dead for a minute or so.
K5DLQ
K5DLQ's picture
If memory serves me correctly
If memory serves me correctly, the Vtun client name can only be 40 characters (which is comprised on the NODE name & "-" & tunnel client IP address.)
AE6XE
AE6XE's picture
Also, doesn't the client side
Also, doesn't the client side determine its hostname, without .local.mesh?   The Client connects in providing the name, password?   Would have to match.

Joe AE6XE
K5DLQ
K5DLQ's picture
absolutely.  The field should
absolutely.  The field should contain only the NODENAME (not fully qualified domain name)
 
k1ky
k1ky's picture
NodeName length for Tunnel success
I believe we have found that node names less than 31 characters seem to work best for tunnel clients.   Anything longer "may" yield mixed results.  Best rule of thumb: If it doesn't work, try a shorter name.  Basically if it fits in the display area of the box, you should be good to go.

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer