MerlinAU MerlinAU v1.4.2 - The Ultimate Firmware Auto-Updater (GNUTON SUPPORT!)

[jksmurf]

my reported issues in Alpha and Beta almost instantly showed their head.

Sorry to hear about the FW update giving you issues, hopefully the period of instability in the firmware for your nodes is short lived.

A cynic might say well at least MerliAU worked flawlessly to update it …

 

[ExtremeFiretop]

Sorry to hear about the FW update giving you issues, hopefully the period of instability in the firmware for your nodes is short lived.

A cynic might say well at least MerliAU worked flawlessly to update it …

I hope so too. I think this is the first firmware update that has flat out failed me since I started this project... Stuff happens. Reports mostly seem positive for devices running as the primary router.

But seeing multiple reports about the devices in AiMesh similar to my experience is enough to know I may just have to skip this one and try again on the next release.

"AT LEAST MERLINAU WORKED FLAWLESSLY TO UPDATE IT"

 

[aex.perez]

I'm happy the update worked out for you.
388.9 flashed successfully for me over night using MerlinAU on my node, and my reported issues in Alpha and Beta almost instantly showed their head.

Tried to reboot the entire network and the node is gone, offline and unreachable now. Similar to reports from @aex.perez

Downgrade in my immediate future today, unfortunately. But the MerlinAU script worked perfectly and updated it without a hitch with version 1.4.1 so I think we are over the hump now of the development changes with the WebUI.

Martinski and I still have a few small things we would like to work on for 1.4.2, but for now it's mostly listen to reports and validate reports.

Well I updated the wait period to give me time as I expected this might happen then downloaded the FW in the hopes I would tackle this today, and my pains with the Alpha 2 and Beta would be a memory based on the testing @RMerlin did. That he went out of his way to try, even though AiMesh is closed source and firmly out of his control I greatly appreciate. Since it worked for him, limited but worked, I'm encouraged and going to try something (or somethings depending on my success or lack of it).

WPS reset, router and node (since I've had so much practice lately) apply 388.9, WPS reset again to be sure. Then setup everything from scratch. Router, creating and adding nodes to AiMesh then recovering from backup (BACKUPMONto the rescue). Then, if it still fails, do the process again and configure everything, manually without the backup.

Either method will get me to 388.9, on the router and nodes and I suspect get everything connecting over WiFi like before, just the level of effort. Should both fail, then its back to 388.8-4 I'll go for a bit.

Anybody have suggestions on what data to capture if it fails, to try and see why, I did the basics but in the heat of the moment I'm sure I could've done more. Since I'm taking a little time to plan this, thought I'd put it out there (but the basics I did the last time and didn't find anything)

Then I'l try to figure out why only the router sent a FW update notification, while the nodes (also on 1.4.1 and confiured identically) didn't though an email tests (AMTM and MerlinAU) were successful on all of them, but I got a lot to do first.

 

[nlurker]

The email I got this time appears to be in plain text. It says Click here to review the changelog: but there is nothing to click on. There was a link last time. I haven't changed any settings.

 

[ExtremeFiretop]

Well I updated the wait period to give me time as I expected this might happen then downloaded the FW in the hopes I would tackle this today, and my pains with the Alpha 2 and Beta would be a memory based on the testing @RMerlin did. That he went out of his way to try, even though AiMesh is closed source and firmly out of his control I greatly appreciate. Since it worked for him, limited but worked, I'm encouraged and going to try something (or somethings depending on my success or lack of it).

WPS reset, router and node (since I've had so much practice lately) apply 388.9, WPS reset again to be sure. Then setup everything from scratch. Router, creating and adding nodes to AiMesh then recovering from backup (BACKUPMONto the rescue). Then, if it still fails, do the process again and configure everything, manually without the backup.

If it works let me know. And I'll do a similar process... I know AiMesh is closed source so all I can do is observe and report. Try to increase awareness around the issue.

But if a true factory reset of the parent resolves it then I'll have to bite the bullet and try that. So far now I've only been trying on the node, and when it failed, factory reset the nodes without factory resetting the parent. I just usually wrap it up by doing a backupmon restore of the primary to restore all the node specific configurations like bound devices, etc.

Anybody have suggestions on what data to capture if it fails, to try and see why, I did the basics but in the heat of the moment I'm sure I could've done more. Since I'm taking a little time to plan this, thought I'd put it out there (but the basics I did the last time and didn't find anything)

Similar for me, I'd like to know, I did also try poking at some stuff in alpha and beta but didn't find anything conclusive so I ended up just downgrading due to time restraints.

Then I'l try to figure out why only the router sent a FW update notification, while the nodes (also on 1.4.1 and confiured identically) didn't though an email tests (AMTM and MerlinAU) were successful on all of them, but I got a lot to do first.

Only the parent will ever send email notifications for updates available of the nodes.

Mine did so as seen attached.

The nodes only send their own emails for things likes actually starting the update process and successful updates post install.

 

[ExtremeFiretop]

The email I got this time appears to be in plain text. It says Click here to review the changelog: but there is nothing to click on. There was a link last time. I haven't changed any settings.

More information will be required. As you can see by the screenshot in my post above, it appears to work in my instance so if you could please fill in some details to this report it would be appreciated.

What model device? What firmware to what firmware? What version of MerlinAU?

Did you have any logs of the update process (main menu --> log options (lo) --> Update Logs (lg) )

Or screenshots of the emails or the update process? Maybe your configuration file as well?

Thanks.

 

[nlurker]

More information will be required. As you can see by the screenshot in my post above, it appears to work in my instance so if you could please fill in some details to this report it would be appreciated.

What model device? What firmware to what firmware? What version of MerlinAU?

Did you have any logs of the update process (main menu --> log options (lo) --> Update Logs (lg) )

Or screenshots of the emails or the update process? Maybe your configuration file as well?

Thanks.

Here is the email:

New Firmware Update for ASUS Router​

A new F/W Update version 3004.388.9.0 is available for the RT-AX58U router.

The F/W version that is currently installed:
3004.388.8.4

Number of days to postpone flashing the new F/W Update version: 15

Please click here to review the changelog:


The firmware update is expected to occur on: 2025-04-25 03:03:00

Sent by the "MerlinAU" utility.
From the "Asus" router.

2025-Apr-10 Thu 03:03:06 AM DST

Here is the log:

2025-04-10 03:03:04 MerlinAU v1.4.1
2025-04-10 03:03:04 Running the update task now... Checking for F/W updates...
2025-04-10 03:03:05 NVRAM cfg_device_list is NOT populated. No Mesh Nodes were found.
2025-04-10 03:03:11 The email notification was sent successfully [NEW_FW_UPDATE_STATUS].
2025-04-10 03:03:11 Using release information for Merlin Firmware.
2025-04-10 03:03:13 The firmware update to 3004.388.9.0 version is currently postponed for 15 day(s).
2025-04-10 03:03:13 The firmware update is expected to occur on 2025-04-25 03:03:00.

Where do I find my configuration file? Thanks.

 

[ExtremeFiretop]

Here is the email:

New Firmware Update for ASUS Router​

A new F/W Update version 3004.388.9.0 is available for the RT-AX58U router.

The F/W version that is currently installed:
3004.388.8.4

Number of days to postpone flashing the new F/W Update version: 15

Please click here to review the changelog:


The firmware update is expected to occur on: 2025-04-25 03:03:00

Sent by the "MerlinAU" utility.
From the "Asus" router.

2025-Apr-10 Thu 03:03:06 AM DST

Here is the log:

2025-04-10 03:03:04 MerlinAU v1.4.1
2025-04-10 03:03:04 Running the update task now... Checking for F/W updates...
2025-04-10 03:03:05 NVRAM cfg_device_list is NOT populated. No Mesh Nodes were found.
2025-04-10 03:03:11 The email notification was sent successfully [NEW_FW_UPDATE_STATUS].
2025-04-10 03:03:11 Using release information for Merlin Firmware.
2025-04-10 03:03:13 The firmware update to 3004.388.9.0 version is currently postponed for 15 day(s).
2025-04-10 03:03:13 The firmware update is expected to occur on 2025-04-25 03:03:00.

Where do I find my configuration file? Thanks.

Thanks for that; you can use:

cat /jffs/addons/MerlinAU.d/custom_settings.txt

Just make sure to blank out/remove the password from the "credentials" field.

 

[ExtremeFiretop]

Nvm; I got enough from the last post to find the problem in the code;
Will submit a PR to version 1.4.2 in dev shortly.

 

[ExtremeFiretop]

Well I updated the wait period to give me time as I expected this might happen then downloaded the FW in the hopes I would tackle this today, and my pains with the Alpha 2 and Beta would be a memory based on the testing @RMerlin did. That he went out of his way to try, even though AiMesh is closed source and firmly out of his control I greatly appreciate. Since it worked for him, limited but worked, I'm encouraged and going to try something (or somethings depending on my success or lack of it).

WPS reset, router and node (since I've had so much practice lately) apply 388.9, WPS reset again to be sure. Then setup everything from scratch. Router, creating and adding nodes to AiMesh then recovering from backup (BACKUPMONto the rescue). Then, if it still fails, do the process again and configure everything, manually without the backup.

Either method will get me to 388.9, on the router and nodes and I suspect get everything connecting over WiFi like before, just the level of effort. Should both fail, then its back to 388.8-4 I'll go for a bit.

Anybody have suggestions on what data to capture if it fails, to try and see why, I did the basics but in the heat of the moment I'm sure I could've done more. Since I'm taking a little time to plan this, thought I'd put it out there (but the basics I did the last time and didn't find anything)

Then I'l try to figure out why only the router sent a FW update notification, while the nodes (also on 1.4.1 and confiured identically) didn't though an email tests (AMTM and MerlinAU) were successful on all of them, but I got a lot to do first.

I just finished a complete factory reset of both the parent and the node; reconnected them successfully after the upgrade completely took the node offline.
And unfortunately still the same issue trying to bind specific devices to the node. (As my original report) https://www.snbforums.com/threads/3...-gt-ax11000-and-gt-axe11000.93693/post-946239

In @RMerlin 's test in the beta thread, you can clearly tell he is using a Wired backhaul instead of Wireless:

Beta - Asuswrt-Merlin 3004.388.9 Beta is now available

Any further details/reports regarding the empty Networkmap? I still ccannot reproduce the issue here, so I will need the error messages from someone's browser console to be able to investigate. The only thing I noticed was that, over https, that frame can take 2-3 seconds before it loads, but...

www.snbforums.com

So with that in mind I tried again with a Wired backhaul and still the same issue:

The results are the same... But this time since the factory reset I do have a single 5GHz devices connecting... it's my work device (previously bound to the primary router before the reset) so I checked it out compared to my old config; and it's using WP3 for the 5GHz network on my 6E node.
All my previous devices are configured to use WPA2 for the node...?

I thought that was weird since I know for a fact the default setting is WPA2/WPA3-Personal; so I went to the network tab and forced it to WPA2-Personal only for the 5GHz network and rebooted and would you look at that:

Seems like the WPA2/WPA3-Personal setting is really acting as a WPA3 personal for some reason, and not allowing devices with WPA2-personal to connect to the nodes. (Unless explicitly set as WPA2-Personal)
Now time to restore and seeing if changing this setting works.

 

[aex.perez]

I just finished a complete factory reset of both the parent and the node; reconnected them successfully after the upgrade completely took the node offline.
And unfortunately still the same issue trying to bind specific devices to the node. (As my original report) https://www.snbforums.com/threads/3...-gt-ax11000-and-gt-axe11000.93693/post-946239

View attachment 64928

In @RMerlin 's test in the beta thread, you can clearly tell he is using a Wired backhaul instead of Wireless:

Beta - Asuswrt-Merlin 3004.388.9 Beta is now available

Any further details/reports regarding the empty Networkmap? I still ccannot reproduce the issue here, so I will need the error messages from someone's browser console to be able to investigate. The only thing I noticed was that, over https, that frame can take 2-3 seconds before it loads, but...

www.snbforums.com

So with that in mind I tried again with a Wired backhaul and still the same issue:

View attachment 64925View attachment 64929

The results are the same... But this time since the factory reset I do have a single 5GHz devices connecting... it's my work device so I checked it out compared to my old config; and it's using WP3 for the 5GHz network on my 6E node.
All my previous devices are configured to use WPA2 for the node...?

I thought that was weird since I know for a fact the default setting is WPA2/WPA3-Personal; so I went to the network tab and forced it to WPA2-Personal only for the 5GHz network and rebooted and would you look at that:

View attachment 64930View attachment 64931

Seems like the WPA2/WPA3-Personal setting is really acting as a WPA3 personal for some reason, and not allowing devices with WPA2-personal to connect to the nodes.
Now time to restore and seeing if changing this setting works.

doing that now, as I suspected that might be the issue, but roaming assistant was throwing error in the log about WiFi signal strength for a node and a device that are a foot apart so I got distracted chasing that down, trying this now!

And guess what! Me too! Got a few stragglers but working on it!

 

[visortgw]

doing that now, as I suspected that might be the issue, but roaming assistant was throwing error in the log about WiFi signal strength for a node and a device that are a foot apart so I got distracted chasing that down, trying this now!

Yes, the closest node is not always the best connection.

 

[aex.perez]

Yes, the closest node is not always the best connection.

I know but goto 388.9 goes bonkers. 388.8-4 everything in tos place. At @ExtremeFiretop suggestion/post/discovery I too went back to WPA2-Personal from WPA2/WPA3-Personal and bunch more stuff connected. Now I gotta back out some other changes I made and see what else gets fixed.

 

[aex.perez]

I know but goto 388.9 goes bonkers. 388.8-4 everything in tos place. At @ExtremeFiretop suggestion/post/discovery I too went back to WPA2-Personal from WPA2/WPA3-Personal and bunch more stuff connected. Now I gotta back out some other changes I made and see what else gets fixed.

Everything is connecting back to the respective nearest nodes again. Without changing anything I did to try and fix it based on the poor signal strength in the log.

Tempted to leave it alone, but now that I know what to lookout for, I'm going to back out a few changes I made and see if it all still works as it is now.

 

[ExtremeFiretop]

Everything is connecting back to the respective nearest nodes again. Without changing anying I did to try and fix it based on the poor signal strength in the log.

Tempted to leave it alone, but now that I know what to lookout for, I'm going to back out a few changes I made and see if it all still works as it is now.

Seems we identified the root cause at least for us... Devices that can make the jump to WPA3 won't have any issues (like my work device) but devices that need the transitional setting of WPA2 will get stuck.

Maybe worth posting in the other firmware releases thread for @RMerlin to investigate and push up the chain. Or maybe a release note. Thanks for coming to my TED talk.

 

[aex.perez]

Seems we identified the root cause at least for us... Devices that can make the jump to WPA3 won't have any issues (like my work device) but devices that need the transitional setting of WPA2 will get stuck.

Maybe worth posting in the other firmware releases thread for @RMerlin to investigate and push up the chain. Thanks for coming to my TED talk.

I think it's definately related to that, and more as I have a few devices that wouldn't connect, like relatively new Amazon Echo's and a Nest (now Google) that can do both and didn't connect until the others that couldn't were able to connect one I made the change to WPA2, and after that to seal the deal I backed out the changes I made to Roaming Assistant, and the SmartConnect rule to support the three SSIDs I have, one for 2.4 ghz devices, one for 5Ghz devices and the other for both.

After a power cycle of them, not connecting to anything now

 

[jksmurf]

Seems like the WPA2/WPA3-Personal setting is really acting as a WPA3 personal for some reason, and not allowing devices with WPA2-personal to connect to the nodes. (Unless explicitly set as WPA2-Personal)

Interesting, you could be on to something there.

I’ll have to check, but I think I had to change my IoT network (makes sense anyway) to just WPA2-Personal whereas before the IoT devices were all happy with the WPA2/WPA3-Personal setting (presumably the older ones just connect under the WPA2 part).

 

[aex.perez]

Interesting, you could be on to something there.

I’ll have to check, but I think I had to change my IoT network (makes sense anyway) to just WPA2-Personal whereas before the IoT devices were all happy with the WPA2/WPA3-Personal setting (presumably the older ones just connect under the WPA2 part).

A few of the devices that didn't connect (releatively new and can do WPA2/WPA3), I've had to go in and set them up, exactly the same as before and then they connected after I made the WPA2 change. The older one that should have a problem with that connected to the router but not to a node until I made the change to WPA2-Personal at the router for all the SSID's.

Just have one or two devices left, that even after a power cycle didn't conect that I anticipate will as soon as I reset them up, using the exact same settings as before

 

[jksmurf]

A few of the devices that didn't connect (releatively new and can do WPA2/WPA3), I've had to go in and set them up, exactly the same as before and then they connected after I made the WPA2 change. The older one that should have a problem with that connected to the router but not to a node until I made the change to WPA2-Personal at the router for all the SSID's.

Just have one or two devices left, that even after a power cycle didn't conect that I anticipate will as soon as I reset them up, using the exact same settings as before

You definitely need to comment in the FW release thread(s), if only to see if there’s a lightbulb moment amongst other users.

 

[jksmurf]

You definitely need to comment in the FW release thread(s), if only to see if there’s a lightbulb moment amongst other users.

I didn’t (don’t) want to comment in the FW threads as the Network Client list has been an (ASUS) disaster for about a billion years, but I just noticed when I put my IoT network back to WPA2-personal, that whilst most (not all) IoT devices nicely attached themselves to closer nodes, the same devices (on the nodes) appear in the Network Map as being under the Primary LAN SSID. Does my head in.