You are here

new node, no name resolution

20 posts / 0 new
Last post
new node, no name resolution

I just picked up a second GL750 (creta) device and loaded aredn FW. Got it setup and configured, just like I did the first one (working great, love it!)

But this one doesn't seem to resolve names!

  • It is setup stand alone: no ethernet connections, Wifi access, and 2G mesh to my hAP, which is sharing WAN and working great. I can see all the meshed nodes on Mesh Status and I verified it is configured exactly like my other Gl750 (which is working perfectly). No special configuration, mostly defaults.
  • While on the GL750 wifi, I can only access the node via 'localnode.local.mesh', not by it's name 'KG7KMV-GL750-2.local.mesh'. I cannot access any other nodes it is connected to by their names or IPs.
  • I can access the node from the mesh.

Any thoughts here? I've tried a few things, but I'm not a "power user". Should I try the Reset to Firstboot button and start over?

nc8q's picture
new node, no name resolution

Maybe another set of eyes might find a difference.
Please, instead of using words, show us your configuration
73, Chuck


K5DLQ's picture
Do you have a custom DNS

Do you have a custom DNS configuration on your computer, perhaps?  

no custom DNS that I know of.

no custom DNS that I know of. Same issue on PC, phone, and tablet, only on this node.

now some screenshots!
I had installed the blockkonwnencryption package, but removed it many reboots ago
(pasted screenshots didn't stick when I saved.. trying again)

Image Attachments: 
nc8q's picture
kg7kmv-gl750-2 -> kg7kmv-router 100% 60%

I am concerned with the 60% NLQ and tunneled neighbors sharing WAN.
You should strive to achieve 100% LQ and NLQ.
"Location Not Available" !
Are both GL750s and kg8kmv-router in the same building or on the same property?

Trouble shooting:
From a computer on the LAN of your kg7kmv-gl750-2,
I would like to see the results of a PING with a destination node.
gelmce@nc8q-mesh:~$ ping -c 1 kf8mz-ar150
PING kf8mz-ar150.local.mesh ( 56(84) bytes of data.
64 bytes from KF8MZ-AR150.local.mesh ( icmp_seq=1 ttl=61 time=21.9 ms

--- kf8mz-ar150.local.mesh ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 21.913/21.913/21.913/0.000 ms

(Windows OS has a slightly different syntax)

If the PING is successful, I would like to see a traceroute.
gelmce@nc8q-mesh:~$ traceroute kf8mz-ar150
traceroute to kf8mz-ar150 (, 30 hops max, 60 byte packets
 1  localnode.local.mesh (  0.551 ms  0.536 ms  0.845 ms
 2  dtdlink.NC8Q-hAP.local.mesh (  1.947 ms  1.934 ms  1.922 ms
 3  mid2.KF8MZ-MTIK5G (  19.434 ms  19.880 ms  34.906 ms
 4  KF8MZ-AR150.local.mesh (  34.894 ms  35.134 ms  35.122 ms

(Windows OS command is 'tracert')

- Why are kg7kmv-router and KI0E-HAP-4 sharing their internet connection with the entire local AREDN network?
- I am concerned that tunneled nodes may have multiple tunnels/routes and there may be some 'looping'.
There was a bug-fix in the nightly to ensure that tunneled nodes do not link via exposed WAN sharing.

- I see some 'not so unique' hostnames on your network.
(example: raspberrypi, which is the default hostname of all Raspberry Pis.)
The hostnames of computers connected to the mesh at large must be unique.
Typically you should prefix your amateur radio callsign to the computer’s hostname in order to have the best chance of it being unique on the mesh network.

Expand to: The hostnames of computers, devices, cameras, IP phones, PBXs,... connected to the mesh at large must be unique.

73, Chuck

Thanks for all the

Thanks for all the recommendations!

- The link between kg7kmv-gl750-2 and kg7kmv-router is across my house, in a temporary testing situation while I sit at my work bench to get it setup.
- I didn't set the location on the 750 because I don't know where it will end up. I loan them out to other hams interested in trying it before buying. 
- We have a very small network over here with 3 hams, very early stages of experimenting and learning AREDN. I just turned off my WAN sharing, but it's helpful sometimes getting new nodes connected which otherwise have no WAN connection.
- I can address the hostnames with the other operator. As of now there are only two of us with any services on our network.

On the issues with my gl750:
- I did a reset to firstboot. Then had a heck of a time getting it to take the new node name. when I clicked save settings the page would 404 error. Long story short, I eventually got it to take the name/password. Still not doing hostnames for me when connected to the 5G wifi.

ping and traceroute tests (from PC connected to kg7kmv-router meshed to kg7kmv-gl750-2):

root@KG7KMV-GL750-2:~# ping kg7kmv-router.local.mesh
PING kg7kmv-router.local.mesh ( 56 data bytes
64 bytes from seq=0 ttl=64 time=12.384 ms
64 bytes from seq=1 ttl=64 time=4.959 ms
64 bytes from seq=2 ttl=64 time=1.635 ms
64 bytes from seq=3 ttl=64 time=1.645 ms
64 bytes from seq=4 ttl=64 time=1.946 ms
64 bytes from seq=5 ttl=64 time=3.317 ms
--- kg7kmv-router.local.mesh ping statistics ---
6 packets transmitted, 6 packets received, 0% packet loss
round-trip min/avg/max = 1.635/4.314/12.384 ms
root@KG7KMV-GL750-2:~# traceroute kg7kmv-router.local.mesh
traceroute to kg7kmv-router.local.mesh (, 30 hops max, 38 byte packets
 1  KG7KMV-router.local.mesh (  2.447 ms  1.966 ms  3.911 ms

pretty much the same result with a PC connected to the LAN on kg7kmv-gl750-2.

I discovered that while my PC is on the LAN of kg7kmv-gl750-2 I have name resolution. It's only when connected to its WiFi that names don't work for the connected client.

nc8q's picture
It's only when connected to its WiFi that names don't work for t

"It's only when connected to its WiFi that names don't work for the connected client."


Then, perhaps, the same test 'when connected to its WiFi'.

"when I clicked save settings the page would 404 error."
This is likely because, when the node rebooted, your PC dropped its DHCP client address.
'404 error' was likely generated by your web browser.
Wait 1 to 3 minutes and do a  'page refresh'.
Sometimes I do a
ping -i 9 <nodename>
and watch for a response. "-i 9" wait 9 seconds between PINGs
(Windows OS argument syntax may be different)

"No special configuration, mostly defaults."
er, uh, in Advanced Configuration, what is not default?

I see in your images that you have 2 outdoor rated radios and
everyone else is a tunnel.

Which 'names' don't work for the connected client.
Can you PING them 'when connected via AR750's WiFi'?

73, Chuck

Sadly, we are mostly tunnel

Sadly, we are mostly tunnel connections right now. Our network is young and we don't have a good hilltop node yet nor tall towers to get over all the trees. The 3 active ops are loaning out devices for other hams to try out, who are not tech-savvy.

"Mostly defaults" means the only things I changed were:
- node name
- root password
- enable 5G wifi


here are some screenshots

here are some screenshots from my PC connected via cable to the gl750 LAN.
the gl750 is connected to my home network via cable on the WAN port.
the gl750 has a (poor) mesh connection to kg7kmv-router.
I expect that my PC should have full mesh access AND full internet access like this, and I get that with other nodes in this same setup.

Image Attachments: 
nc8q's picture
gl750 is connected to my home network via cable on the WAN - not

Your image in #4 indicates that gl-ar750-2 does not have a WAN connection.
This may need to be fixed.

The bottom image in #9 indicates that 'router' is not offering DNS for internet names.
This may need to be fixed.

I advise against sharing internet access across the network --- full time.

If you want internet access, it may be best to enable it on one of your home devices
(hAP, GL-AR750-1, GL AR750-2) then connect to that device via its 5 GHz Wi-Fi.

Can you use your mouse and cut-n-paste the text from your windows cli into your post?

73, Chuck


I've changed up the setup

I've changed up the setup several times. I started fresh here.
GL750 is plugged in WAN <-> home LAN.
Screenshot shows it has a WAN IP and Default gateway on my LAN.
I SSH to the GL750 and ran a few commands. The output is attached, including the node ifconfig report.
- traceroute to another node is successful
- traceroute to is successful
- Node has mesh and internet access!

on my PC, which is attached to the GL750 wifi, it can reach mesh nodes but has no internet access.
Here is the ipconfig output:

Wireless LAN adapter Wi-Fi:
   Connection-specific DNS Suffix  . : local.mesh
   Link-local IPv6 Address . . . . . : fe80::b047:15db:645f:3717%7
   IPv4 Address. . . . . . . . . . . :
   Subnet Mask . . . . . . . . . . . :
   Default Gateway . . . . . . . . . :

No other connections. I noticed that the default gateway is blank. That seems like the issue why my PC can't get on the node WAN.

Can a hardware issue cause this? I got the node open box on ebay.

debug file attached as well.
Image Attachments: 
File Attachment: 
nc8q's picture
Windows user please help.
Wireless LAN adapter Wi-Fi:
   Connection-specific DNS Suffix  . : local.mesh
   Link-local IPv6 Address . . . . . : fe80::b047:15db:645f:3717%7
   IPv4 Address. . . . . . . . . . . :
   Subnet Mask . . . . . . . . . . . :

   Default Gateway . . . . . . . . . :


All looks good to me regarding your kg7kmv-ar750-2.
The output from your unknown Windows CLI command does not seem correct to me.
I cannot advise as I don't do Windows.

73, Chuck


KK6BML's picture
Updated hAP - same name resolution problem.

I have 2(or now 3) hAP lite' - one sits fixed in my house attached toan nm2, a nm5 and a rPi - backup server for Laguna Niguel ACS services. this hAP was updated to 10/16/22. I have a second (and now 3rd thinking 2nd might be damaged - plus "one is noe, two is one" redundancy adage) that I deply mobile with a switch, NM5 & NM2 node and ip cameras (Summple &a Foscam) for Emergency and other events. I upgraded the second one 11/20/22 or there about. basic Setup, Port forwading page, advanced configuration look the same. however, the second node is displaying all the symptoms described in this thread. I reserve an ip address for a camera,and publish it as alwasy - appears on mesh status correctly yet clicking on link give a response of not connected to the internet (isn't resolving the name), can't ping by name, but can ping by direct ip address and can connect to the camera (or other services) by keying ip address directly into browser. This is across the board - any Mac os or ios device and my windows machine. doesn't matter if connected by ethernet or wifi or remotely via mesh rf.  Further - get the same result trying to click through to ANY name (node or service) on the Mesh Status page. And yes, I've emptied my cache (and cleaned my cookies and restarted the browser and tried several different browsers and rebooted my devices) - to no avail.
Main thing I note in common in the thread and it sparked interest - when I check DHCP configuration (ipconfig in windows terminal or network adapter setting on MAC) the connection is configured with a proper ip address and subnet mask from the DHCP setup on the hAP but NO Primary Gateway(windows speak) / Router(Mac speak) address in the DHCP assignment. All other devices I have that can be connected to as DHCP server all have this value (it is always the LAN address of the DHP Server device).

It is driving me crazy trying to prepare for our major annual event in 2 weeks. I'm working with KK6LZB in Laguna Niguel to see that the nodes & cameras are all configured and accessable by our main control module (ZoneMinder on the LNACS mesh server) for the Laguna Niguel christmas parade. BTW - this node worked completely correctly under ther 3.21 build

Bob Evans KK6BML

PS - I've been at this for 4 hrs tonight on top of about 12 the last week so don't have energy or time to create and upload screen shots or device log files at this point

I did end up getting mine

I did end up getting mine working by enabling "aredn.@wan[0].lan_dhcp_defaultroute". 

Just very strange that one device needed that and the other one does not.

Same issue

Not sure if reviving an older thread is frowned upon or not; if so, sorry.

I'm having a DNS resolution issue like described here. I have a TP-LINK CPE510 outside, wired into a MikroTik hAP ac lite through port 5 (dtd). CPE510 is on 5GHz while hAP is on 2.4GHz. I am accessing via a laptop plugged into port 2 of the hAP. DHCP is enabled on hAP and disabled on CPE510 (maybe this should be flipped?)

I can access the hAP through my localnode.local.mesh and it's hostname, kd2vuj-hap.local.mesh. However, I can only access the CPE510 (kd2vuj-cpe510.local.mesh) through it's direct 10.x.x.x IP address. Ping from my laptop works for all hAP addresses but fails for CPE510.

I've tried resetting devices and enabling the "aredn.@wan[0].lan_dhcp_defaultroute" option on both nodes but it didn't seem to fix anything. I tried setting the DNS specifically to the hAP IP but no change either.

KM4TBQ's picture
Last fall I began learning

Last fall I began learning AREDN from the ground up, and flashed at least 80 nodes. When I started, we were on FW v3.22.8. Of course, once I had flashed 90% of the nodes, FW v3.22.12 was released, so I had had to go back and upgrade all those nodes that I had just configured. Never once did this problem you are describing arise before the FW update, and I immediately noticed that "localnode...." (which I saved as a bookmark) would no longer work. I also had to manually type in the "10.X.X.X" as you described, and everything from there would work fine. I don't know enough to know the exact cause, but it definitely showed up with FW release version 3.22.12. I was flashing nodes day and night for months, and specifically remember noticing it directly after the release. 

K6CCC's picture
How long have you waited

How long have you waited after reboot?  With some of the newer builds, it can take a while for names to propagate.

Well it's been about 5.5

Well it's been about 5.5 hours now, still not resolving the CPE510 hostname.

ng1p's picture
Same issue with a MicroTik Router Board, Rocket and Nanostation

I am setting up for a demo tomorrow at a local hamfest and this same issue was driving me crazy. This was my demo setup from last year and it worked great running code. I thought it would be good to upgrade to to show off some of the new things but the DNS issues was killing me. I first tried just downgrading to on the Router Board. This seems to help but testing between the Rocket and Nanostation running sometimes they could resolve each other but most of the time they could not. My next step I went back to (know last good code) and all is working correctly with DNS now. I dont have any more time today to test but that is what I have found so far. This is a strange one.

ng1p's picture
More testing MicroTik Router Board, Rocket and Nanostatation

I did more testing just upgrading my MicroTik Router Board to 3.22.6 and then to 3.22.8 from 3.22.1 and DNS worked fine while resolving the DtD connected Nanostation and RF connected Rocket (Nanostation to Rocket) I left the Nanostation and Rocket at 3.22.1 and only was upgrading the Router Board.

Once again when I upgrade the Router Board to 3.22.12 the DNS issue started again, I could resolve the rocket or the Nanostation but never both. Seems like rebooting would flip things some times. Upgrading the the latest Nightly Build I see the same effect (DNS doesn't work right as with 3.22.12). Downgrading back to 3.22.8 all is fine again and no issues.  So for sure some sort of bug with 3.22.12 and above.

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer