[SOLVED] 2x OpenWrt devices: one router and one client

[inc0gnit0]

Hi,

I have two devices which both run the latest stable release 18.06.5. Unfortunately, my Netgear EX6150v2 rage extender cant connect to internet. It is not about configuring as relaydaemon yet, I first want to make it connect to the internet in order to update LuCI with latest updates and download the necessary software.

I can connect to both of them with LuCI and SSH. I gave the first device (TP-Link router) 192.168.1.1 and the Netgear device the address 192.168.1.2 and told it to use 192.168.1.1 as gateway and checked Disable DHCP for this interface in Network > Interfaces > LAN > Edit. Although these steps are not unfamiliar to me, I checked the quick start troubleshooting guide

Network diagnostics shows me:

Ping: "ping: "bad address 'google.com'" but if I go for 8.8.8.8 or 192.168.1.1 everything is fine!
traceroute: "traceroute: bad address 'google.com'" plus same as above
NSlookup: ";; connection timed out; no servers could be reached" - doesnt matter if I search for 8.8.8.8 or google

Similar behaviour when I SSL into the box:

root@OpenWrt:~# opkg update

 * opkg_download: Failed to download http://downloads.openwrt.org/releases/18.06.5/packages/arm_cortex-a7_neon-vfpv4/telephony/Packages.gz, wget returned 4.
 * opkg_download: Check your network settings and connectivity.

Any ideas what I could check next?

Thx for any input!

inc0gnit0

[siga]

have a look on /etc/resolv.conf

[psherman]

@inc0gnit0

update LuCI with latest updates

Whatever you do, do not use opkg upgrade on any openwrt device unless a) you know what you are doing and upgrade only specific/known safe packages, or b) are willing to deal with the fact that you will almost certainly have problems that will require a reset to defaults situation.

I say this because you do not need to, and should never upgrade any of the core openwrt packages with this method (and you mentioned doing just this in your op). It will break things. Period.

Using opkg update followed by opkg install is, of course, totally fine to install additional packages.

[hnyman]

@inc0gnit0

Network diagnostics shows me:

Ping: "ping: "bad address 'google.com'" but if I go for 8.8.8.8 or 192.168.1.1 everything is fine!
traceroute: "traceroute: bad address 'google.com'" plus same as above
NSlookup: ";; connection timed out; no servers could be reached" - doesnt matter if I search for 8.8.8.8 or google

@inc0gnit0

and the Netgear device the address 192.168.1.2 and told it to use 192.168.1.1 as gateway

You need to also tell it to use 192.168.1.1 as the DNS server...

Basic "dumb AP" advice in wiki:

Wireless Access Point (aka "Dumb" Access Point)

My old post about it:

Not accessible after disabling DHCP

Just to complement the previous advice: remember to also set the gateway and DNS address on the new dumb AP. Otherwise the router itself will be unable to access internet (although clients connected via it will be, as they use DHCP to get address from the main router.) Old advice written by me: If you connect the two routers using a fixed connection between LAN ports, the whole thing is ridiculously easy, as you can dumb down the APs. (The key to making things really easy is to NOT use the WAN…

[inc0gnit0]

root@bridge:~# cat /etc/resolv.conf 
search lan
nameserver 127.0.0.1

Gateway and DNS address have been both set earlier, but that didnt do the trick. :-/ Thx for all your help!

[lleachii]

@inc0gnit0

root@bridge:~# cat /etc/resolv.conf

Can you look at the actual setting from /etc/config/network under LAN. If there is no DNS, add it:

option dns '192.168.1.1'

I'm not sure why @siga asked you to show /etc/resolv.conf

[inc0gnit0]

hnyman! Your link did it! I also had to disable firewall, dnsmasq and odhcpd! Thank you ALL so much for being so helpful!