Beta Asuswrt-Merlin 3006.102.4 Beta is now available

[dave14305]

Somewhere within the multiple levels of Makefiles - no idea where specifically. The processing is done by a tool written by Asus if I remember.

Do you agree that something during build is modifying client_function.js to remove the times symbol used in the client dropdown to delete offline clients? It looks like neither the line of code in client_function.js nor the tools in release/src/router/tools have changed in a long time.

 

[jksmurf]

I had to remove YazDHCP in the Alpha versions as it wasn't working and interfered with manually trying to assign IPs. Haven't tried with the Beta.

I’d be interested in your feedback if you do install YazDHCP again.

With many thanks to @Jeffrey Young and @bennor I installed it (for the first time) yesterday on the Beta, along with adding manual assignments to the VLANs utilizing the dnsmasq-x.config.add files. Early days but both see, to be working OK. I’m not adding additional manual assignments to the primary WAN though so have not stress tested it as you might have.

 

[RMerlin]

Do you agree that something during build is modifying client_function.js to remove the times symbol used in the client dropdown to delete offline clients? It looks like neither the line of code in client_function.js nor the tools in release/src/router/tools have changed in a long time.

No idea, I haven`t looked. All I know is that pages do get processed at build time, both to inject language strings and to remove whitespaces and comments.

 

[outlaw78]

It's just a switch to enable it at compile time. But what this switch does is that some of the regular pages are replaced by different pages during build. These alternative pages need to be developped and maintained on top of the regular ones.

Could the switch be "flipped" and a disclaimer added that the ROG Gui isn't supported or maintained and use at your own risk like you did with AI-Mesh back in the day before enabled that in your builds? I realize this could introduce breaking bugs. It was just a thought.

 

[RMerlin]

Could the switch be "flipped" and a disclaimer added that the ROG Gui isn't supported or maintained and use at your own risk like you did with AI-Mesh back in the day before enabled that in your builds? I realize this could introduce breaking bugs. It was just a thought.

People don't read, I would still constantly get asked "Can you fix this? It's broken."

When I started work on 3006, there were a number of issues with the ROG version of the new Guest Network Pro pages, and I did not have the patience to track down and fix them all. The UI is already the most annoyng part that I have to deal with whenever merging new GPLs because a number of pages have received significant changes on my side, making a lot of merging requiring to be done manually. The new Network page is a nightmare to deal with, it's written in a completely different coding style from any Javascript I have ever worked with.

The ROG UI is gone, and that's final.

 

[jksmurf]

The ROG UI is gone, and that's final.

I would suggest that the ROG statement needs to be in bold at the top of every 3006 FW release and a sticky on this forum but...

People don't read

 

[HeadBanger]

Having done some more digging on my node 5Ghz issue this morning, it seems that some sort of smart connect is being used for just the nodes as I can connect to 5GHz it’s just that the SSID is displayed as my 2.4GHz SSID name. Very odd.

HB

 

[jksmurf]

Having done some more digging on my node 5Ghz issue this morning, it seems that some sort of smart connect is being used for just the nodes as I can connect to 5GHz it’s just that the SSID is displayed as my 2.4GHz SSID name. Very odd.

HB

This might be a shot in the dark, but I see you're using 3006 Main with 3004 (but not 3006) capable nodes, like me.

Propagation of bands to nodes differs depending on if you use nodes with Pro (3006 branch FW) or non-Pro (3004 branch FW) nodes (see rest of thread too). When @visortgw started looking at this he concluded that:

the limit of two devices per band per node (plus primary router) total has been lifted when all nodes are running 3006. I can push 2.4 and 5 MHz IoT and guest networks to all AiMesh nodes.

So if you can only run 3004 on the nodes, there is a limit to how many bands you can propagate per interface. Here's someone with a similar limitation. If you're only seeing one band (you base it on what you can connect to?), maybe it is related to this; or maybe not.

 

[HeadBanger]

This might be a shot in the dark, but I see you're using 3006 Main with 3004 (but not 3006) capable nodes, like me.

Propagation of bands to nodes differs depending on if you use nodes with Pro (3006 branch FW) or non-Pro (3004 branch FW) nodes (see rest of thread too). When @visortgw started looking at this he concluded that:



So if you can only run 3004 on the nodes, there is a limit to how many bands you can propagate per interface. Here's someone with a similar limitation. If you're only seeing one band (you base it on what you can connect to?), maybe it is related to this; or maybe not.

Many thanks for replying.

Yes, I’m using a 3006 (AXE-16000) as the main and various AX 3004 nodes (shown in my signature). I’m not running a guest network and I’ve not started tinkering with VLAN yet.

This is just normal WiFi. The nodes all broadcast 5Ghz using the 2.4Ghz SSID name. Only the main AXE-16000 correctly broadcasts the separate band SSIDs.

HB

 

[alan6854321]

This is not a valid script name. It must be dnsmasq-sdn.postconf (literally).

Does this apply even if I don't have any guest networks?

I have a /jffs/scripts/dnsmasq.postconf which just loads my hostnames.
Do I need to rename it dnsmasq-sdn.postconf after upgrading to 3006?

 

[bennor]

@alan6854321, no it should not apply. You would still use dnsmasq.postconf for your main LAN hostnames.

If you are not using Guest Network Pro (or renamed to Network), VLAN, SDN feature on the Asus-Merlin 3006 firmware then you shouldn't likely have a need to use a dnsmasq-sdn.postconf file.

In the previously discussed examples in this thread, some people were using dnsmasq-sdn.postconf to script manual IP reservations for their Guest Network Pro clients.

 

[chris.at]

AX86U-Pro: In Guest Network Pro, if you configure a guest or iot network and select "use same subnet as main network" the guest/iot clients can reach everyhting inside the main network. In the previous 3004 release this was blocked. They used the same ip subnet but could only reach the internet. Is this a bug, or a new "feature"? If it's a feature, it makes gues/iot network useless if you select "same subnet", so be careful. Also, < and > are no longer allowed within a guest networks wifi key. Both are no show stoppers, but to be mentioned.

 

[Piotrasd]

there is expected FW 3006.102 for GT-AXE11000 ??

 

[rung]

Dirty upgrade. System is main of RT-AX86U Pro (was 3004.388.8_4) and wireless AiMesh node of RT-AX58U (3004.388.9, upgraded earlier this week without issue). I did NOT remove and reattach node yet. All clients successfully reconnected without issue including two main 2.4GHz/5GHz networks (separate SSIDs), GN1 @ 2.4GHz across node and GN1 @ 5GHz just at main. Only change they saw was the new subnets from 192.168.101.x and 192.168.102.x to 192.168.52.x and 192.168.53.x, respectively. ssh works fine (key length is RSA 3024). amtm works and shows no updates. Tailmon successully monitoring tailscale - so entware looks good too. Let me know if there are any areas you would like me to check before I try out any of the new functionality of 3006 (and accidentally screw it all up lol). Thank you for all your hard work!

 

[MDM]

AX86U-Pro: In Guest Network Pro, if you configure a guest or iot network and select "use same subnet as main network" the guest/iot clients can reach everyhting inside the main network. In the previous 3004 release this was blocked. They used the same ip subnet but could only reach the internet. Is this a bug, or a new "feature"? If it's a feature, it makes gues/iot network useless if you select "same subnet", so be careful. Also, < and > are no longer allowed within a guest networks wifi key. Both are no show stoppers, but to be mentioned.

Did you notice this button?
And there is also a Set AP Isolated switch to prevent communications between connected Guest devices...

 

[Ev0]

Flashed my BE92U from Stock 3.0.0.6_102_37526 to this Beta1 and had no issues.

I didn't need to do anything special, no dirty fix needed, just uploaded the fw and it updated.

I didn't factory reset either, my wireguard client was retained (although set to wgc5 rather than 1, but not an issue).

Things I have noticed:

I seem to have around 10% more free ram than I had with the stock fw.

The clients I had set to use vpn (wireguard) or wan, were not retained, but easy enough to re-add them.

Ethernet ports 2 and 4 are swapped around (on stock ethernet 2 shows 2.5gbps connection and eth 4 was 1gbps, with Merlin, Ethernet 2 shows 1gbps and eth 4shows 2.5gbps), not really an issue, just an observation.

So far it seems stable.

 

[MDM]

there is expected FW 3006.102 for GT-AXE11000 ??

No, see here.
And read the first post carefully, since you did not the one in this thread.

 

[rung]

Dirty upgrade. System is main of RT-AX86U Pro (was 3004.388.8_4) and wireless AiMesh node of RT-AX58U (3004.388.9, upgraded earlier this week without issue). I did NOT remove and reattach node yet. All clients successfully reconnected without issue including two main 2.4GHz/5GHz networks (separate SSIDs), GN1 @ 2.4GHz across node and GN1 @ 5GHz just at main. Only change they saw was the new subnets from 192.168.101.x and 192.168.102.x to 192.168.52.x and 192.168.53.x, respectively. ssh works fine (key length is RSA 3024). amtm works and shows no updates. Tailmon successully monitoring tailscale - so entware looks good too. Let me know if there are any areas you would like me to check before I try out any of the new functionality of 3006 (and accidentally screw it all up lol). Thank you for all your hard work!

Okay, even though the router showed all clients connected, I found a couple of clients on the GN claiming there was no internet. So, in a very unscientific way, I changed two things at once. I turned off DNS director and added custom DNS servers to the GN (1.1.1.2, 1.0.0.2). The clients seemed to be happy after those two changes. I'll keep monitoring them...

 

[visortgw]

Okay, even though the router showed all clients connected, I found a couple of clients on the GN claiming there was no internet. So, in a very unscientific way, I changed two things at once. I turned off DNS director and added custom DNS servers to the GN (1.1.1.2, 1.0.0.2). The clients seemed to be happy after those two changes. I'll keep monitoring them...

Yes, when DNS Director is set to Router for a guest network, DNS does not resolve.

UPDATE: I have confirmed that this has been resolved in this beta (see posts #124 through #132 below).

 

[bennor]

AX86U-Pro: In Guest Network Pro, if you configure a guest or iot network and select "use same subnet as main network" the guest/iot clients can reach everyhting inside the main network. In the previous 3004 release this was blocked. They used the same ip subnet but could only reach the internet. Is this a bug, or a new "feature"? If it's a feature, it makes gues/iot network useless if you select "same subnet", so be careful. Also, < and > are no longer allowed within a guest networks wifi key. Both are no show stoppers, but to be mentioned.

Its the same problem, when Use same subnet as main LAN is enabled, that has been repeatedly complained about in the stock Asus 3006 firmware. Made several lengthy posts about it in the past pointing this out in the stock firmware last year when Asus released the 3006 firmware for the RT-AX86U Pro. See here and here. The Guest Network IoT preset default settings simply doesn't work like one would expect, clients isolated from main LAN. I stopped using the IoT preset and use the Customized Network preset in Guest Network Pro, and have Use same subnet as main LAN disabled.