Release Asuswrt-Merlin 388.2 is now available for select models

[John Fitzgerald]

G'Day from Australia. I upgraded my AX58U router without resetting it and now the 5Ghz band is not working properly. Websites and ping tests fail for any device that connects to 5Ghz. I wonder if this issue affects people in countries that have more channel options because of different rules and regulations?

View attachment 49563
View attachment 49564

Safe your "config" file and "jffs" file.
Then do a GUI Factory reset, and check the box next to it, and reboot.
When you get to the welcome screen go to "advanced" (last option) and restore your "config" file.

If you have scripts, load them fresh.

 

[houmi]

Since upgrading to 388.2 , both my AX86U seem to be losing wifi access every so often and long pings.

64 bytes from 192.168.50.1: icmp_seq=16 ttl=64 time=2.019 ms
Request timeout for icmp_seq 17
Request timeout for icmp_seq 18
Request timeout for icmp_seq 19
Request timeout for icmp_seq 20
64 bytes from 192.168.50.1: icmp_seq=21 ttl=64 time=2.255 ms
64 bytes from 192.168.50.1: icmp_seq=22 ttl=64 time=2.302 ms

 

[sweetlyham]

Safe your "config" file and "jffs" file.
Then do a GUI Factory reset, and check the box next to it, and reboot.
When you get to the welcome screen go to "advanced" (last option) and restore your "config" file.

If you have scripts, load them fresh.

Came back here to provide an update and saw your message, which is funny and I'll explain why.

I decided to do a WPS factory reset followed by factory default restore with the option to the right ticked, then reloaded the same firmware (Basically I followed this - https://www.snbforums.com/threads/ax88-packet-loss.62891/#post-563326). Unfortunately, I was still having issues on the 5Ghz band, at one point it became unusable on 5Ghz and had to resort to 2Ghz.

I have gone back to using 388.1 for now & can confirm 5Ghz works again, unfortunately I don't have the time to debug it so hopefully someone else can

 

[houmi]

Came back here to provide an update and saw your message, which is funny and I'll explain why.

I decided to do a WPS factory reset followed by factory default restore with the option to the right ticked, then reloaded the same firmware.

Still having issues on 5Ghz band, it actually become unconnectable at one point.

Since have reloaded 388.1 back on and can now use 5Ghz, unfortunately I don't have the time to debug it so hopefully someone else can

Same issue... I disabled 160Mhz as a test and it's making 388.2 slightly better, but I still get ping request timeouts every so often.

 

[sweetlyham]

Since upgrading to 388.2 , both my AX86U seem to be losing wifi access every so often and long pings.

64 bytes from 192.168.50.1: icmp_seq=16 ttl=64 time=2.019 ms
Request timeout for icmp_seq 17
Request timeout for icmp_seq 18
Request timeout for icmp_seq 19
Request timeout for icmp_seq 20
64 bytes from 192.168.50.1: icmp_seq=21 ttl=64 time=2.255 ms
64 bytes from 192.168.50.1: icmp_seq=22 ttl=64 time=2.302 ms

Glad it isn't just me that was having the same issue on 388.2, hopefully someone can debug it. I have gone back to 388.1. I couldn't even do a ping when on 388.2 as it would just timeout - back on 388.1 and can confirm I can ping again.

 

[houmi]

Glad it isn't just me that was having the same issue on 388.2, hopefully someone can debug it. I have gone back to 388.1. I couldn't even do a ping when on 388.2 as it would just timeout - back on 388.1 and can confirm I can ping again.

I just bought a new Sony TV and I realized it was losing wifi, I thought it is the Sony and moved it to ethernet. Then I realized my Tivo Stream 4Ks were losing wifi as well and my iPhone was on 5G more than on wifi, then realized it's the router when I started doing pings on wifi... definitely something going on.

I also disabled Diversion thinking it might be it but the problem continued.

But the problem seems to be the 160Mhz setting...

 

[sweetlyham]

I just bought a new Sony TV and I realized it was losing wifi, I thought it is the Sony and moved it to ethernet. Then I realized my Tivo Stream 4Ks were losing wifi as well and my iPhone was on 5G more than on wifi, then realized it's the router when I started doing pings on wifi... definitely something going on.

I also disabled Diversion thinking it might be it but the problem continued.

But the problem seems to be the 160Mhz setting...

I already had 160mhz off by default so unsure what's happening!

 

[IKNOWNOTHING]

Thanks. Will try this solution later for ipvanish. Using Merlin 388.1 and vpnmon-2. After upgrading to 388.2 VPN was broke, so reverted to 388.1 for now.

Loaded the most recent ipvanish ovpn configs (23-4) And reverted back to 388.2 again. Didn't work either. Compression is at lzo adaptive. So now back on 388.1 again, which is (and always was) working flawless. Does anyone has an idea what to try next?

 

[IKNOWNOTHING]

with this firmware I am not able to load ExpressVPN - even after removing keysize 256 entry the configuration loads but never gets an IP like so
Connected (Local: 10.159.0.46 - Internet not redirected)
There is redirection on VPN director to move ipad traffic using this VPN client and still nothing.
All this worked in previous version and while I wait for EVPN to make changes to their config (highly doubt that they will or even understand as the questions I am getting makes me feel that even the best and most expesinve are same as other)

Any help will be appreicated
RT_AX86U here

Edit:
I worked over chat with ExpressVPN for close to 2 hours and after their new configuration that removes ‘keysize’ and adds ‘verify-x509-name Server name-prefix’ it seems there is no dice in successfully getting an IP from EVPN
EVPN mentioned that I have to reload previous software that worked and that they do not support any routers that they don’t sell or are not on their supported list, it so happens that 86U is not. Just stating this in case others have or are facing same issues with ExpressVPN

We seem to have the same problem. On my side it is Ipvanish that is not loading. Where you able to solve the problem yet?

 

[John Fitzgerald]

Came back here to provide an update and saw your message, which is funny and I'll explain why.

I decided to do a WPS factory reset followed by factory default restore with the option to the right ticked, then reloaded the same firmware (Basically I followed this - https://www.snbforums.com/threads/ax88-packet-loss.62891/#post-563326). Unfortunately, I was still having issues on the 5Ghz band, at one point it became unusable on 5Ghz and had to resort to 2Ghz.

I have gone back to using 388.1 for now & can confirm 5Ghz works again, unfortunately I don't have the time to debug it so hopefully someone else can

Back in the Beta thread I posted, with screenshot, that my green wifi symbol was only half light.
I had the router unplugged overnight and it still came back the same way.
Then with no intervention on my part it corrected itself after about 24 hours.

I got no indication from others as to what the issue was but recently I had a script that would not load without freezing, corrupting the whole system.
So my post above is what cleared the issue, with help from the author. (problem not on their end)
From alpha to beta to final release, some sort of corruption occurred, also with Entware having a major release between beta2 and Release.

It's a pain, but let it run for a while and stick your clients on the other band for a day or two and keep an eye on it.
This release thread is loaded with odd problems but those that have done a complete reset seem to be fairing best.
Stuck radio? Did you unplug, do an "electrical reset"?(one minute, then plug back in)

 

[houmi]

Mea Culpa... I tested the wireless ping test on my main router and it was fine, the AIMesh Node was the one acting up, so I had to hard reset it and now it's working ok. They are connected via wired backhaul (1Gbps)

Is hard reset a requirement now between Merlin releases ? I was going from 388.1 -> 388.2 on the two AX86U's....

 

[L&LD]

Not a requirement every time, but when issues arise, it is usually the quickest way to resolve them.

 

[Damit1]

Yeah, its me here again.

12 days a row, best release for a lot of time.
Ax56u one network

And AX88U*2+ AX56U as AiMesh system as well.

 

[Kyjiep]

To the attention of all those whose static entry in the ARP is dying after the update to the 388.2 release. Today I conducted another experiment - I reset the router to factory settings before updating from 388.2alpha1 to 388.2release, after the update I made only the initial configuration and again reset to factory settings with a checkmark in the checkbox, after that I once again performed only the initial configuration of the router and did another reset to factory settings, but without a checkmark in the checkbox. After that, I made the initial configuration of the router and created a static entry in the ARP, and this entry has been working for 1 hour. After my previous repeated attempts to update, the static entry in the ARP disappeared 3-8 minutes after it was created. I don't know what exactly brought success this time, maybe it's just a matter of chance, maybe not quite a chance, I think you should try it.

P.S.:
And still the bug remained, I just noticed that the static entry in ARP does not disappear if it is created for an IP that is not on the same subnet as the router. For example: the address of the router is 192.168.50.1. If a static entry is created for ip 192.168.1.254, then it lives for an arbitrarily long time, until the router is rebooted. If you create a static entry for ip 192.168.50.254, then in most cases it dies after 5-10 minutes, but occasionally it lives much longer, until the router is rebooted. I will look and try different options further.

P.P.S.:
It seems I have found a crutch with which the desired static entry survives.
So: the ip of the router is 192.168.50.1, it is necessary that the static entry (192.168.50.254) at ff:ff:ff:ff:ff:ff does not die.
Create a services-start script with the following content:

#!/bin/sh
arp -i br0 -s 192.168.1.254 ff:ff:ff:ff:ff:ff
arp -i br0 -s 192.168.50.254 ff:ff:ff:ff:ff:ff

Entries in the script should only be in that order.
We reboot the router.
The presence of an entry (192.168.1.254) at ff:ff:ff:ff:ff:ff prevents the death of an entry (192.168.50.254) at ff:ff:ff:ff:ff:ff.
Checked 5 times in a row - the static entry (192.168.50.254) at ff:ff:ff:ff:ff:ff does not die.

P.P.P.S.:
Today I discovered that this method does not always work (((.

 

[aex.perez]

Took awhile but I think I've captured it, whatever it is. Got back online after a weekend away, because if it was going to go it was going to happen while I was away (Murphy's Law in action).

It did, noticed the speedtests were off, went to the date and time of the first drop and found these in the external log withing the 30 minute window (12/42). Performance never/hasn't recovered. Asymmetrical results Router / WiFi / Wired, enough to notice on WiFi / Wired but not enough that it impacts the day to day of things. As we get back up to speed in terms of our use, I expect I'll be rebooting the AX88u later this evening. Spanning tree still off, Memory at 75%, nvram 56%, no QOS, swap 0Kb, jffs 7%, modem speedtest in the 980's. Log level set to debug and all messages logged... Still digging through the log (24hr x 7 for a few days) for any other anomalies, if any...

kernel: Init chrdev /dev/idp with major 190
kernel: tdts: tcp_conn_max = 8000
kernel: tdts: tcp_conn_timeout = 300 sec
kernel: SHN Release Version: 2.0.2 36f59aa
kernel: UDB Core Version: 0.2.20
kernel: Init chrdev /dev/idpfw with major 191
kernel: IDPfw: flush fc
kernel: IDPfw: IDPfw is ready
kernel: sizeof forward pkt param = 280
BWDPI: fun bitmap = 3
rc_service: cfg_server 22412:notify_rc update_sta_binding;
custom_script: Running /jffs/scripts/service-event (args: update_sta_binding)

Haven't done any research other than to correlate the log to the graph but all the entries before and after this snippet are what I typically observe, so far. But if anyone wants to see the entire log for that day, the days before or after, I have it them.

 

[joseterl]

Thanks for your work Merlin! pity the problems with XT12 and GLP, hopefully they will be solved soon, I always buy the router model that has problems

 

[octopus]

Thanks for your work Merlin! pity the problems with XT12 and GLP, hopefully they will be solved soon, I always buy the router model that has problems

Microsoft OneDrive

onedrive.live.com

 

[RAH-66]

@RMerlin
version 388.2_1 will be only for XT12, or for everyone AX else too?

 

[RMerlin]

@RMerlin
version 388.2_1 will be only for XT12, or for everyone AX else too?

It will be for all Wifi 6 models, as it contains an SDK fix that may affect certain hardware revisions of other models, and it will be less confusing than releasing it just for the 6-7 models that may be affected. So far there will be very little changes in that release however.

 

[Wishmaster1965]

Factory Reset and seeing issues connecting to the admin page, get the login page then spinning circle for a very long time.