Need Help Setting Up Repeater - And Some Teaching

[Garghoul]

I am trying to set up a Buffalo WZR-HP-G300NH as a WiFi repeater to connect to a Netgear C6300 main base station with the IP-address 192.168.0.1.

Using https://www.gargoyle-router.com/wiki/li ... bridge.jpg as a template and consulting https://openwrt.org/docs/guide-user/net ... figuration as well, I am still unsure about a few things.

1.) My WZR only seems to be able to connect to the Netgear base station, when I opt for 192.168.0.1 as the Bridge-IP setting, which is the only change I made.
- a) Isn't scanning for and choosing a "SSID to join" alone take care of the WiFi connection settings between the WZR and the Netgear?
- b) Is that the only IP-address choice here?
- c) What exactly does Bridge here actually mean? Where from does it bridge to?
- d) Does Bridge here now become something like the gateway address of the base station I want to connect to? If not, then how is it possible, to have 2 devices with the same 192.168.0.1 address (my WZR and the Netgear base station) in the same subnet?

2.) After the restart following the change of settings, where is Gargoyle's configuration web server on the WZR now logically located (with its Bridge (which side?), its LAN ports and WiFi AP/Gateway or somewhere else) and under which IP address can it be reached from which ports: LAN, WAN, WiFi?

3.) Since a subsequent brief mains voltage drop, Gargoyle now doesn't transmit any WiFi signal (Netgear is still up and unaltered), nor can I reach its webserver, neither via LAN-ports nor WiFi, with neither 192.168.1.1, nor *.0.1, nor *.0.32 (which once seemed to have been the bridge?-address assigned to the WZR by the Netgear).
- a) How do I get back in there?
- b) Is the only remaining way now to access Gargoyle's settings to do a hard-reset and start all over?

And just for completeness - where and how does the alternative setting to "Connect Via Client Bridge" change things?

<Sigh> So many open questions

[Lantis]

As a preface, if you’re not on version 1.10.0 or later, this conversation takes a completely different path. So please confirm that in your next reply.

You should examine this diagram for details.
https://www.gargoyle-router.com/wiki/li ... relayd.png

The basic idea is that the “bridge ip” is kind of a “dummy address” at which we establish the relay.

1.
- Yes, but you must choose a bridge address outside the subnet of the upstream AP. Everything else will be automatic.
- Yes.
- It “bridges” (pseudobridge technically) the two networks. When we say that this is a pseudobridge, we mean that is is not a real layer 2 bridge, but an emulated one using relayd at a layer 3 level. The result is the same in most cases.
- You won’t ever end up with two devices with the same IP address, that’s not how networks operate. But they will end up on the same subnet. This occurs because the upstream AP will hand out a DHCP address to the router itself which of course will be in the subnet of the upstream AP. We call this the “relay IP”.

2. It is available at:
WAN: Relay IP
LAN: Relay IP OR Bridge IP (if the client uses a static IP in the right subnet)
After changing the setting, the client in use may need to disconnect/reconnect to the router to get a new IP in the right subnet before being able to use the network correctly. Some devices do this automatically, some need a kick.

3.
Based on your questions, I’d say you should reset this time, as it is hard to gauge what has been done correctly and what hasn’t.
In future, if it cannot establish a link to the upstream AP, you can connect to it with a static IP on its bridge IP on the LAN side.
The repeated wifi signal going down if it cannot connect to the upstream AP is expected (but not nice) behaviour. Essentially the wifi driver goes into a permanent search mode, preventing the transmission of another signal.

I hope this clarifies a bit more, please post back if anything remains unclear.

[ispyisail]

The repeated wifi signal going down if it cannot connect to the upstream AP is expected (but not nice) behaviour. Essentially the wifi driver goes into a permanent search mode, preventing the transmission of another signal.

I found this out the hard way

You need a good strong wifi signal

For this reason i'm lukewarm with relayd mode

[Lantis]

There's a way to fix it, but I need to build it into Gargoyle.

Openwrt has the "Travelmate" package which operates under the same premise. Essentially if the upstream AP goes down for a defined period of time, swap to a config which does not rely on it so that the repeated AP comes back online, allowing you to login and check the issue. It also then scans in the backgrouind to try and reconnect.

It's another todo list item lol

[Garghoul]

Lantis, thank you so much for your swift reply.

Yes, I am on 10.0.X 20180324. I had a gut feeling, that I still forgot to add something important. Sorry about that.

https://www.gargoyle-router.com/wiki/li ... relayd.png makes me wonder: the clients on the secondary router are only connected via LAN ports. Coincidence or is client access via WiFi there no longer possible?

1.)
- a) Interesting. On my first try I used an arbitrary IP address, different from the upstream Netgear's *68.0.1 and Gargoyle's default *.1.1, but I only succeeded with a connection, once I changed it to Netgear's *.0.1.
I am curious to see what happens this time, after the reset, with a unique subnet address.

- b)

Is that the only IP-address choice here?

I probably should have worded this a bit better: "Is that the only IP-address option here?", which was meant in reference to the Netgear base station's 192.168.0.1.

-d) Was it then just coincidence that it was working with my chosen Bridge IP being the Netgear base station's 192.168.0.1? And it was working, because my, as you explained it, "pseudobridge" setting didn't really collide with the real upstream subnet gateway address?

3.) your answer here pretty much confirms, what I had pieced together for this part, but without really knowing, if I am making the right assumptions.

It is here now way past midnight. I'll still initiate the router's reset before going to sleep and start playing with it later, once I am up again.
I'll then report back how things turned out.
Thanks a bunch. This, so far, helped quite a bit.

[tapper]

Hi Garghoul I will tell you about my setup to se if it helps. Pleas ignore my crappy spelling.
I have a wrt3200acm as my mane router running OpenWrt and a wrt1900ac running Gargoyle as a repeter bridge. My mane router is at 192.168.1.1 and I set my bridge to be on 192.168.2.1 outside the mane routers subnet. You have to make shure that the bridge has the same SSID, Password and channel settings as your mane router. So the Package that does the bridging on the wrt1900ac which is my bridge askes my mane router for a IP using DHCP and gets 192.168.1.06 From my PC that is connected to my mane routers lan port I can log on the userinterface of my bridge on the IP handed out by DHCP. If the bridge cant contact my mane router for a IP then I have to connect to it with a device on one of it's lan ports. Then set a static IP of 192.168.2.1x and then I can then get in to it's userinterface to make changes. This can be a pane to get working but is rock solid for me for long times. Good luck! Edited post to fix IP addresses!

[Lantis]

Every IP address you’ve listed Tapper is on the same subnet...
That isn’t a valid config.

[tapper]

Crap! I should not post when I am sleepy. Mane router is 192.168.1.1 bridge is 192.168.2.1 Static IP for the PC would be 192.168.2.2 Thanks for pointing that out mate.

[ektus]

I'm struggling with this after updating the router from 1.7.x to 1.15.x. This router acts as a bridge to a remote wifi access point. The network topology is as follows:

[main router 192.168.0.180] ---LAN--- [bridge router 192.168.0.174, LAN port] ---WiFi bridge--- [remote router/ap 192.168.0.173 in AP mode] ---LAN--- [wired client 192.168.0.8]

The wireless bridge is used to wirelessly link together two segments of the same subnet. With the old firmware, this simply worked for the last 8 years. With the relayd setup, I'm struggling. Either I can reach the bridge router (Bridge IP 192.168.0.174 inside subnet 192.168.0.x and AP/Gateway IP 192.168.2.174), but nothing behind it, or nothing at all (bridge IP 192.168.2.174 and AP/Gateway 192.168.0.180).

Terminology on the connection page is not very clear to me.
https://www.gargoyle-router.com/wiki/li ... bridge.jpg
If there is a "dummy" management ip, why is it there at all and not hidden? All I need is one IP to reach the bridge router, and that has to be in main subnet, as everything else goes to the WAN side of the main router. My PC has a wired connection to the main router.
There is also a static assignment of the bridge routers IP in the main router DHCP table.

Is this setup possible at all, and what should I enter in connection page?

Help?

Additional information: After setting an additional route to 192.168.2.0 on the main router, I can reach the bridge router on its 192.168.2.174 address. Yet if I SSH in, the routing table of the bridge router is virtually empty and it can't reach anything.

Code: Select all

login as: root
root@192.168.2.174's password:


BusyBox v1.35.0 (2023-02-09 08:37:35 UTC) built-in shell (ash)

------------------------------------------------------------------
|            _____                             _                 |
|           |  __ \                           | |                |
|           | |  \/ __ _ _ __ __ _  ___  _   _| | ___            |
|           | | __ / _` | '__/ _` |/ _ \| | | | |/ _ \           |
|           | |_\ \ (_| | | | (_| | (_) | |_| | |  __/           |
|            \____/\__,_|_|  \__, |\___/ \__, |_|\___|           |
|                             __/ |       __/ |                  |
|                            |___/       |___/                   |
|                                                                |
|----------------------------------------------------------------|
| Gargoyle version 1.15.X   | OpenWrt 22.03 branch               |
| Gargoyle revision d78aeafe| OpenWrt commit f7541ae             |
| Built February 26, 2023   | Target  ath79/default              |
------------------------------------------------------------------
root@stall0:~# route
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
192.168.2.0     *               255.255.255.0   U     0      0        0 br-lan
root@stall0:~# ping 192.168.0.173
PING 192.168.0.173 (192.168.0.173): 56 data bytes
ping: sendto: Network unreachable
root@stall0:~# ping 192.168.0.180
PING 192.168.0.180 (192.168.0.180): 56 data bytes
ping: sendto: Network unreachable

From the status page of the bridge router:

Code: Select all

Bridge

    Relay IP Address:
    Bridge IP Address: 192.168.2.174
    LAN Gateway IP: 192.168.0.180
    LAN Netmask: 255.255.255.0
    Bridge MAC Address 14:CC:20:98::
    Connected via: Client Bridge
    Bridge SSID: stall2

The Relay IP Address is missing. The remote AP doesn't have DHCP enabled.

[Lantis]

Could you post your settings for the remote AP for clarity please?