Same subnet is fine, but after I installed OpenWrt and connected my R1 LAN to R2 WAN then it automatically act as DHCP client.
Your intermediate ap may also be causing issues for you. But what you want for your openwrt system is this:
How is it physically connected to the upstream network? Is it connected by the WAN port or the LAN port?
R1.1 is a cable modem/router that has 2 LAN port
R1 LAN => R1.1 LAN
R1.1 LAN => R2 WAN
Ok... so hopefully that R1.1 device is not impacting anything here. What IP does it have? How did you set it?
That's the problem, I don't see it on my R1 connected devices, lol.. I just reset that router to default and set the DHCP to disabled.
Usually you need to manually configure the IP of a dumb AP type device to be something on your main subnet. You probably didn't do that, thus why you cannot reach it.
Yes, I did manually set the IP of that R1.1 but it was long ago, but I'm sure it's 192.168.254.2 or .3 but I can access it. I'll try to reconfigure that AP and see that will fix R1 workstation to access R2 workstations.
You probably would be best served setting your R2 device as a dumb AP. I'd recommend resetting that device and then following the instructions (key steps: set the LAN IP to an address in your existing subnet, turn off the DHCP server, and connect the upstream network to a LAN port on your R2 device).
I guess so, then I can't use the WAN port correct? the reason I want to use the WAN port is because I have 4 workstations and R2 has 4 LAN ports only.
You can use your WAN port, although it is possible it won't give you full speed results because it would need to be part of a bridge (which is a software equivalent to a switch). You can try it, of course, and it may be totally fine. But if the performance isn't good, a simple ethernet switch may be a better option.
You'll be creating a bridge for your lan that includes eth0 and eth1.1
Alright, thank you for all the help. But I will try to reconfigure 1st the R1.1 and see if that's blocking R1 workstations, also do I need to change the route setup in R1 if R2 is connected to R1.1?
If R1.1 is really a dumb AP, it shouldn't really matter where it is located. However, it if is routing in any way, it would affect how things communicate.
Maybe you could draw a diagram that includes information about the physical topology, the IP addresses of each device, and also the model numbers of each of your infrastructure parts and the OS it is running.
Just got home, I did reset my AP and rebooted everything.
So here's my topology.
ISP Fiber
Router 1 Huawei HG8145V5
also the DHCP server
192.168.254.254
from Router 1 LAN 1 port going to my Router 1.1 AP LAN 1 port
Router 1.1 AP Skyworth CM5100.g2
DHCP disabled
192.168.254.10
from Router 1.1 LAN 2 port going to my Router 2 WAN port
Router 2 TP-Link TL-WR941N/ND v5 (OpenWrt)
DHCP client
192.168.254.143 (192.168.1.1)
from Router 2 LAN ports are my thinclients, IP is 192.168.1.X
all machine running windows 10
But wouldn't routing (if the WAN port is left as WAN) probably have even bigger performance impact?
1 Like
On many devices, all 5 ports are connected to the switch, and then split in to LAN/WAN by means of VLANs (sometimes with a single CPU connection (eth0), sometimes with two connections (eth0 and eth1).
However, the WAN port on the device in question is not connected to the hardware switch (unless I am mistaken about it). In other words, the architecture is 4 LAN ports on a hardware switch that connect to eth1 on the CPU and the WAN port is a single port that connects to eth0 on the CPU.
In this type of 4+1 architecture, the only way to make the WAN port behave as part of the switch is to use a bridge interface -- i.e. via the CPU since that is the only path available to connect eth0 with eth1/switch0. Typically, the CPU is not particularly well optimized for this purpose and it may not have the bandwidth to bridge at the full line rate (unlike the 4-port switch which has hardware specifically designed for that function). The result is that bandwidth between eth0 and eth1 may potentially be much slower than between all other ports on the switch.
Now, if you leave the WAN port as a "WAN" (or really any distinct network), you'd be in a situation where you would necessarily need to have a different subnet for the "WAN" vs the "LAN" (even if the firewall is disabled or bypassed) -- this would require routing. The routing performance may or may not be more efficient than the bridge, depending on the design of the router (i.e. does it have a hardware accelerated routing path).
Or, you could just leave the WAN port unused.
Just as a reference, I have run a test with an ER-X (running EdgeOS). The ER-X is capable of routing at 1Gbps (total bandwidth through the routing engine) using hardware acceleration, but if HW NAT is disabled, the bandwidth is much lower (~250Mbps). Inside, the unit has 5 ports connected to a hardware switch which will operate at full line rate (1Gbps) whenever it is purely switching. However, if I logically remove 2 of the ports from the switch (it does this with creative use of VLANs behind the scenes) and then assign them to a bridge interface, the bandwidth gets reduced to ~250Mbps. I tested this with iperf to illustrate what I'm talking about... [side note: I am not entirely certain why I'm not getting consistent 940Mbps max transfers, but it is something with my 11 year old Mac Pro running some hacks to keep it alive for a little longer]. You will see a major difference in performance as shown below.
This first one is using the ER-X as a proper switch:
Connecting to host 10.0.1.52, port 5201
Reverse mode, remote host 10.0.1.52 is sending
[ 4] local 10.0.1.219 port 55383 connected to 10.0.1.52 port 5201
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.00 sec 75.3 MBytes 631 Mbits/sec
[ 4] 1.00-2.00 sec 90.4 MBytes 758 Mbits/sec
[ 4] 2.00-3.00 sec 111 MBytes 930 Mbits/sec
[ 4] 3.00-4.00 sec 89.7 MBytes 753 Mbits/sec
[ 4] 4.00-5.00 sec 95.1 MBytes 798 Mbits/sec
[ 4] 5.00-6.00 sec 97.3 MBytes 816 Mbits/sec
[ 4] 6.00-7.00 sec 105 MBytes 881 Mbits/sec
[ 4] 7.00-8.00 sec 112 MBytes 940 Mbits/sec
[ 4] 8.00-9.00 sec 112 MBytes 940 Mbits/sec
[ 4] 9.00-10.00 sec 112 MBytes 941 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-10.00 sec 1002 MBytes 841 Mbits/sec sender
[ 4] 0.00-10.00 sec 1000 MBytes 839 Mbits/sec receiver
iperf Done.
And the next one uses the exact same physical ports but configured as a bridge:
Connecting to host 10.0.1.52, port 5201
Reverse mode, remote host 10.0.1.52 is sending
[ 4] local 10.0.1.219 port 55420 connected to 10.0.1.52 port 5201
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.00 sec 29.9 MBytes 251 Mbits/sec
[ 4] 1.00-2.00 sec 30.3 MBytes 254 Mbits/sec
[ 4] 2.00-3.00 sec 28.7 MBytes 241 Mbits/sec
[ 4] 3.00-4.00 sec 30.4 MBytes 255 Mbits/sec
[ 4] 4.00-5.00 sec 30.3 MBytes 254 Mbits/sec
[ 4] 5.00-6.00 sec 30.3 MBytes 255 Mbits/sec
[ 4] 6.00-7.00 sec 30.3 MBytes 254 Mbits/sec
[ 4] 7.00-8.00 sec 30.1 MBytes 252 Mbits/sec
[ 4] 8.00-9.00 sec 30.2 MBytes 253 Mbits/sec
[ 4] 9.00-10.00 sec 30.4 MBytes 255 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-10.00 sec 303 MBytes 255 Mbits/sec sender
[ 4] 0.00-10.00 sec 301 MBytes 253 Mbits/sec receiver
iperf Done.
1 Like