[Release 384/NG] Asuswrt-Merlin 384.4 is now available

[alienx2]

yes, thats the way asus things are working or not ...
maybe no solution except downgrade
maybe factory reset, format drive a.s.o. could sometimes be helpful.
My USB stick made problems as well, but now on USB2.0 i got it working finaly.

What type of drive is it and which port is it attached to (USB3 or USB2)?

The 'Reducing USB 3.0 interference' set to ON causes the USB3 port to try and run as USB2. Some real USB3 drives don't like this, so it should be set to OFF if you are having problems.

I think this firmware version have bug? or not? i checked system log here:

Apr 13 11:54:22 kernel: __ntfs_error: 5 callbacks suppressed
Apr 13 11:54:22 kernel: tntfs error (device sda2, pid 4421): ntfs_mft_data_extend_allocation_nolock(): Not enough space in this mft record to accommodate extended mft data attribute extent.  Cannot handle this yet.
Apr 13 11:54:22 kernel: tntfs error (device sda2, pid 4421): ntfs_mft_record_alloc(): Failed to extend mft data allocation.
Apr 13 11:54:22 kernel: tntfs error (device sda2, pid 4421): ntfs_create_helper(): Failed to allocate a new on-disk inode (error 95).
Apr 13 11:54:22 kernel: tntfs error (device sda2, pid 4421): ntfs_create_helper(): Failed (error 95).
Apr 13 11:54:22 kernel: tntfs error (device sda2, pid 4422): ntfs_mft_data_extend_allocation_nolock(): Not enough space in this mft record to accommodate extended mft data attribute extent.  Cannot handle this yet.
Apr 13 11:54:22 kernel: tntfs error (device sda2, pid 4422): ntfs_mft_record_alloc(): Failed to extend mft data allocation.
Apr 13 11:54:22 kernel: tntfs error (device sda2, pid 4422): ntfs_create_helper(): Failed to allocate a new on-disk inode (error 95).
Apr 13 11:54:22 kernel: tntfs error (device sda2, pid 4422): ntfs_create_helper(): Failed (error 95).
Apr 13 11:54:22 kernel: tntfs error (device sda2, pid 4420): ntfs_mft_data_extend_allocation_nolock(): Not enough space in this mft record to accommodate extended mft data attribute extent.  Cannot handle this yet.
Apr 13 11:54:22 kernel: tntfs error (device sda2, pid 4420): ntfs_mft_record_alloc(): Failed to extend mft data allocation.

I tried restart router.. then write access was working then 20 minutes later stopped write.

 

[bidi]

i find that i am unable to generate share links in AiCloud. frankly i don't know for how long this problem has been present, since i haven't used this feature in a long time. is it working for anyone else with this firmware version (and i just need to reboot or something) or is it a genuine bug? i strongly suspect this feature is closed source and RMerlin has no control over it.

 

[telUK]

No, disable implicit on both. Only keep explicit/802.11ac beamforming enabled, on the 5 GHz band.

I will have to check, Im pretty sure I only have 802.11ac beamforming enabled in 5ghz which I thought was the new name for explicit beamforming that Asus changed.

Thanks for the advise

 

[scjr]

No, disable implicit on both. Only keep explicit/802.11ac beamforming enabled, on the 5 GHz band.

I have 802.11ac beamforming enabled on the 5 GHz band and implicit off.

On 2.4 GHz I set both to off. Should I keep explicit on for the 2.4 GHz band?

Thanks.

 

[sfortis]

guys i need some help (at lease if someone has the same issue)

Im using 384.4_2 on AC56 and ive a very strange problem. When my PC (on ethernet, 1gbps) is downloading at full speed (my vdsl is 100/10), most of the wifi devices are disconnecting(!).

i've tried factory defaults, turning off QOS, switching channels, etc etc. Issue remains.

I think that this is a problem of 384.x firmware series, since i dont remember something similar with older 374.x

any help is greatly appreciated!

 

[uncle.f]

Running 384.4_2 on an AC-68u. Upgraded fine, initialized to stock, then reapplied all my settings from screenshots.

Everything functions, but the network map listing shows many of my wireless clients as connected via ethernet. I believe they are on my guest network, all IOT devices I try to keep away from my intranet. Not a big deal, but does anyone else have this problem?

I reported the same for 384.4_2 on AC-68U back here.
[Release 384/NG] Asuswrt-Merlin 384.4 is now available

Checked it again with version 380.70 and it is fine, so the issue is new in 384.

I actually went back to 380.70 after a week of using 384.4_2 (and I never gone back in the past!)
I did factory reset and applied all settings new after the upgrade to 384 but after some days my devices reported no Internet connection (even if WiFi was available).
I logged in to the router via SSH and tried to execute a simple command: "ip a".
The shell command was hanging forever. Even after reboot I could not execute "ip a" or "ps" (some other commands like "ip r" worked fine.
I could not accept such behavior and went back to 380.70.
So far I am very happy with it. I guess new firmware needs more work.

 

[johhnyu]

OK, so known bug then, thanks.

Hopefully I don't see lockups, that's much more serious than a mislabeled network map!

 

[test011]

384.4, RT-AC68U
httpd is always running even if only HTTPS gui is enabled.
Result in port 80 is always in use. Anyway to kill it?

I am running nginx and custom Let's encrypt (a certificate for multiple domains) script. httpd is taking port 80 so, it is no longer possible.

Anyway if I am not using it why should httpd be running all the time. How can I take back port 80? I only use HTTPS for GUI.

 

[RMerlin]

On 2.4 GHz I set both to off. Should I keep explicit on for the 2.4 GHz band?

I would keep it disabled, since devices using the 2.4 GHz band are often older or lower-end devices that might be more picky with compatibility.

httpd is always running even if only HTTPS gui is enabled.
Result in port 80 is always in use. Anyway to kill it?

Not at this time, Asus forces it to run on certain models due to features related to their AiTunnel technology which requires it. Best you can do is force it to a different port (which will probably break some of these rarely-used Asus technologies such as AiHome.).

https://github.com/RMerl/asuswrt-merlin.ng/issues/79

Asus has recently been adding the ability to change that to a different port. This functionality should appear on the webui in a future version.

 

[test011]

I would keep it disabled, since devices using the 2.4 GHz band are often older or lower-end devices that might be more picky with compatibility.



Not at this time, Asus forces it to run on certain models due to features related to their AiTunnel technology which requires it. Best you can do is force it to a different port (which will probably break some of these rarely-used Asus technologies such as AiHome.).

https://github.com/RMerl/asuswrt-merlin.ng/issues/79

Asus has recently been adding the ability to change that to a different port. This functionality should appear on the webui in a future version.

Thank you for the link. changing the port with nvram command works.
But since I am not using AiTunnel, the description sounds like a backdoor or a security hole. There should be a way to disable it.

 

[RMerlin]

Thank you for the link. changing the port with nvram command works.
But since I am not using AiTunnel, the description sounds like a backdoor or a security hole. There should be a way to disable it.

That's more or less what the setting I added under Tools -> Other Settings does, but since I can't tell for sure the FULL impact of disabling httpd when Asus doesn't allow so, I decided to play it safe and not mess with it.

 

[drunki]

Have been running 384.4_2 for quite some time now on my AC68U - rock solid for me (basic config). No issues with the network map as everything is presented correctly - both wireless and ethernet devices.

Thank you for the great work RMerlin !

 

[AsusRouterUser]

I have an RT-AC5300 that I am currently running the 380.70 firmware on and I want to switch to the 384.4_2 firmware, but I have a question? In the notes on the first page on this thread it says..."If coming from 380.xx, it's strongly recommended to do a factory default after flashing 384, and manually reconfiguring." My question on that is how do you reset to factory defaults? I also have a RT-N66U and on that it is you hold down the WPS button for 10 seconds while you power it on, is the RT-AC5300 done the same way?

Thanks, AsusRouterUser

 

[Grisu]

yes, or on admin page with initialize.

 

[uncle.f]

No issues with the network map as everything is presented correctly - both wireless and ethernet devices.

Enable guest WiFi, join some device to it and then check if the network map is still correct for that device.

 

[drunki]

Just tried it and can confirm connecting a device via wireless is not accurately presented in my case.

 

[jpclarke]

Hi,

I have started to use ubiquiti access points for WiFi coverage in my house and have disabled the WiFi on my router, one issue I am experiencing is the cpu is stuck permanently at 50% active (showing as 100% on one core in the ui).

The top command shows the process is /usr/sbin/acsd.

The router is rt-ac3200 running Merlin FW 384.4_2

Anyone else see this behaviour if all WiFi is disabled on their router?

Any pointers to stop the constant cpu utilisation with WiFi disabled?

Thanks

 

[AsusRouterUser]

yes, or on admin page with initialize.

Thanks.

 

[Vexira]

A

Hi,

I have started to use ubiquiti access points for WiFi coverage in my house and have disabled the WiFi on my router, one issue I am experiencing is the cpu is stuck permanently at 50% active (showing as 100% on one core in the ui).

The top command shows the process is /usr/sbin/acsd.

The router is rt-ac3200 running Merlin FW 384.4_2

Anyone else see this behaviour if all WiFi is disabled on their router?

Any pointers to stop the constant cpu utilisation with WiFi disabled?

Thanks

Reboot it should help, I mean turn it off for 30 second and let it rest then turn it back on.

 

[Grisu]

Reboot it should help, I mean turn it off for 30 second and let it rest then turn it back on.

this devices have NVRAM, NV=non volatile, that means they wont forget, not in 1 second nor in 1 year, so it really makes no sense to turn of 30s or similar precedures 30-30-30 or whatever is wroten anywhere.

And there is NO battery to hold regional supply power for something else which could be cleared by removing or shorten it.

let it turned on an press reset button or turn off (as long as you're comftable) and hold WPS while turning on.
It can be a difference how long you press one or the other button, but not how long to turn it off!