I'm having an issue where my VirtualHere CloudHub (rpi) is showing up as an ipv6 DNS server on my network. For most clients this isn't an issue, but on our Android phones for example, they seem to get stuck looking at the ipv6 DNS and it completely breaks DNS (for everything, these are NOT VirtualHere clients). I've done a lot of testing and if I disable ipv6 on VirtualHere CloudHub the other clients work perfectly. The problem is, it seems there's a daemon that keeps restarting ipv6 on the CloudHub, so this only works temporarily. Is there any way to get the CloudHub to stop using ipv6? Or better yet, to get the CloudHub to stop the DNS server completely? I already disabled AVAHI (not using auto-discover) and I set the NetworkInterface to the ipv4 address in the config.ini.
.
i think this is provided by dnsmasq, config is here :https://lede-project.org/docs/user-guide/dns_configuration but i cant see how to turn off ipv6...
Thanks, I merely deleted the
Thanks, I merely deleted the contents of that file and all is working perfectly for me. Don't need DNS at all as I'm not using auto-discover.
.
Ok in 3.5.2 of the cloudhub firmware this issue is fixed, if the cloudhub is participating in a network the dhcp and dns forwarding servers are turned off
Awesome! What's the proper
Awesome! What's the proper method for updating the RPi3 running CloudHub? Do I just have to reload the SD with Etcher or can I initiate an update from SSH?
.
Unfortunately you need to re-write the sdcard, there is no way to update it via the client (The other other flash based CloudHub devices can be updated via the client directly)
.
Okay, thanks - got it updated. The RPi was still showing up as a DNS server until I cleared the /etc/config/dhcp file again. Not a big deal because everything's working again, but if you want help troubleshooting let me know.
.
OK, to confirm it should work like this
1. You write the firmware to the pi3 (or any cloudhub)
2. The cloudhub device has dhcp server enabled
3. You connect via wifi to the cloudhub
4. You switch from "Standalone" to "Participate" mode and the cloudhub reboots
5. The DHCP/DNS Forwarder server(s) are not running anymore. To find if it is running ssh to the cloudhub and type
ps | grep dns
Can you double check that there is no dnsmasq running on your cloudhub when in participate mode. I tested all the cloudhub images and none show this running when in partipate mode (which fixed the issue you previously reported).
.
So, when I run
ps | grep dns
in "Standalone" mode I get:root@CloudHub_a52a94:~# ps | grep dns
787 dnsmasq 1100 S /usr/sbin/dnsmasq -C /var/etc/dnsmasq.conf.cfg01411c
857 root 1012 R grep dns
and in "Participate" mode I get:
root@CloudHub_a52a94:~# ps | grep dns
697 root 1012 R grep dns
which should indicate that DNS is not running, as you mentioned. However, even after multiple reboots I am still seeing the RPi advertise as an IPv6 DNS server until I clear /etc/config/dhcp. Interesting.
.
OK i think i see the issue, can you write the new 3.5.3 of the cloudhub firmware
.
That seems to have worked, thanks!