I have an OpenWRT wireless repeater which was working ok. It was wirelessly connected to my main OpenWRT wireless router, where all the configurations are (networks, DHCP, firewall, etc).
Now I have physically changed my main OpenWRT router but I have restored the same configuration, so there should be no difference. However, the wireless repeater does not work anymore, it does not seem to connect to the main router anymore. Why?
Not sure, but I think this is not the first time it happens to me.
You realize that you literally don't provide any information at all to go by, other than '$unspecified_stuff no longer works after I replaced $other_unspecified_stuff with an unknown configuration'?
Well, we could guess - but I don't like guessing, as it prolongs the debugging unnecessarily.
And it might, if that has been done correctly (and provided there are no MAC filters in place), but again, way too much guesswork. Covering all potential aspects would involve writing an open-ended book series, compared to 5 sentences based on facts and covering the specific case.
But what I'm mostly 'worried' about, is:
because this screams problems, depending on how literally "restored" is supposed to be taken, given that raw configuration files are not compatible between different models, nor strictly speaking even different individual devices of the same model (when it comes to MAC addresses and similar).
…but this is already a prologue to the book series I didn't want to start writing.
I was under the impression that OpenWrt was smarter than that.
You know, to avoid spoofing the access point, especially when used as a travel router and public APs with passwords just written on a sheet of paper.
But, I could be wrong; I always check use 'bssid too'.
Page 2, device paths (which do depend on the hardware design) embedded into /etc/config/wireless.
Page 3, detection order of 2.4- and 5 GHz radios (who's going to become designated radio0 and who's going to become radio1), this does differ between different device models.
Page 4, OpenWrt has 'recently' changed the default interface naming for wireless devices (e.g. phy-ap0 and similar), so if the replacing involved upgrading the OpenWrt version… another can of worms.
Lots of tangents and potential pitfalls we could take, but that's wasted time and efforts without knowing what the OP is dealing with.
what is "OpenWrt", many users (multiple times a day) misrepresent random OEM firmwares as "but that is OpenWrt based", which has serious consequences on the question if 4addr is possible, apart from the more fundamental question of a vastly different syntax of the configuration files
what is the hardware (e.g. allowed interface combinations, or phrased differently, brcmfmac := no AP-STA modes for you).
I'm really not exaggerating by calling this an open ended book series, without any details being presented up front.
Sorry for my little info. It was exactly the same model and version and used WDS. The scanning made it connect (I did not know that was necessary despite having the same configuration) but then I found other problems and I finally connected it through cable with a much easier configuration.
When you click the "Join this network" button it creates a connection that is locked to the MAC address of the AP (which is the bssid parameter). If you replace the AP with different hardware, or sometimes even if you only upgrade OpenWrt to a different major version, the MAC will change and you need to reconfigure the client for the new MAC. Or leave the BSSID box blank and it will connect to any AP simply by matching the SSID.
