Well, I finally have everything working.
Thanks Mike, commenting out the eth0 lines brought the dongle up.
What I found is if I put the wwan0 ahead of the eth0 in the interfaces file, it will try for the wwan0 and if it’s not there, move on to the eth0.
The node (6370) is now showing up in the IRLP active node list. – That’s a win.
However, when on the 4G dongle, the node VK3RSR-R wasn’t showing up in EchoLINK.
Swap back to eth0 and VK3RSR-R comes straight up in EchoLINK.
I wouldn’t have thought that how the Pi is getting internet would have mattered.
While connected on the 4G dongle, I ran the Troubleshooting script in the node and it passed everything except Test 1 – DNS and Test 10 which it’s not expected to pass with EchoIRLP enabled.
I swapped back to eth0 and re-ran the Troubleshoot, it passed Test 1.
This would suggest there is something stopping the DNS service from running when wwan0 is selected.
This is confirmed by NOT being able to Ping any domain name but being able to Ping the IP address for the domain when connected to wwan0.
I’m able to ping both Domain name and IP address when connected to eth0.
During bootup, I see a message “A start job is running for Raise network interfaces”
Then I get a “Failed to start Raise network interfaces” when booting up with wwan0.
I tried adding dns-nameservers and the IP addresses of my carriers DNS servers to the /etc/network/interfaces file after the WWAN0 interface, but that didn’t make any difference.
I also enabled static DNS addresses in the dongle with my carriers DNS servers to no avail.
I then tried to Ping my carriers DNS servers and found I could not do so.
I changed the static DNS servers in my dongle to 18.104.22.168 and 22.214.171.124 (Google’s public DNS servers) and hey presto, My node comes up in EchoLINK and works fine.
Doing this doesn’t stop the “Failed to start” message, however, I noticed the Google DNS addresses have been inserted into the resolve.conf file from the 4G dongle’s static DNS or elsewhere.
I then went back and added the Google DNS servers into the /etc/network/interfaces file (dns-nameservers 126.96.36.199 188.8.131.52) for the wwan0 interface and tested it. All worked fine.
So, there must be something in Debian or IRLP scripts that causes the Telstra DNS addresses to be blocked.
So, in summary, to get it working all I had to do was add a wwan0 interface into the /etc/network/interfaces file and include a line dns-nameservers 184.108.40.206 220.127.116.11 for the wwan0 interface and place the wwan0 interface ahead of the eth0 interface.
Either way, I’ve resolved my problem.
Next, just because I like putting myself through mental pain, I’ll try the same with some other model 4G dongles I have laying around and see if they have the same symptoms.
Thanks again to all those that offered tips and hints that led me down the learning path of Unix/Linux/Debian/Raspbian/etc
Yours gratefully and 73’s for now,
From: IRLP@irlp.groups.io <IRLP@irlp.groups.io> On Behalf Of Mike Alsop
Sent: Friday, 4 October 2019 11:02
Subject: Re: [IRLP] WWAN configuration and 4G Dongles on node 6370
I am pretty sure if you comment out any entries in /etc/network/interfaces relating to eth0 and then reboot, you should be on the dongle.
On Thu, 2019-10-03 at 09:00 -0500, k9dc wrote:
IRLP will use whatever interface is chosen by the Linux routing table. What does the command ‘ip route show' reveal?
On Oct 3, 2019, at 07:04, Peter Chaplin <
Thanks to everyone who has provided tips for this.
From these tips, I have been able to get wwan0 showing up when i run ifconfig
This has led to getting the 4G dongle working and showing up an IP address in the netconfig application, but IRLP want's to use eth0 instead of wwan0 as it's interface, which of course eth0 is unplugged.
How do I get the IRLP node to use wwan0 instead of eth0.
I've tried changing the adaptor in /etc/network/interfaces but that doesn't seem to make a difference.