[Beta] Asuswrt-Merlin 384.15 Beta and 384.13_3 are now available

[DarkKnight75]

I'm on 384.15 beta 1 and they are there since Asus has not removed them. I use Scribe / syslog-ng, so scrape them out of the webgui syslog to their own log - net result, they are gone for me unless I go looking for them.

I have never had devices dropping from 2.4 or 5 ghz on my network. I see reports of those who do. The experts here make recommendations on wireless settings that will resolve them, but many want the "marketing magic promises" more than stability with slight compromises and do not make those changes.

Interesting.see my reply to Maverikcdn...My note 9 dropped on 384.14 and -2 several times before i went back to 384.13.

 

[Maverickcdn]

Ya unfortunately tis the nature of wireless connectivity

Delicate balance between performance and stability. Especially so on the 5Ghz Band

Id download inSSIDer and see if it can help tell you anything about the interference levels in your area and might help to select something that gives you reasonable performance without any issues.

Im also going to hold out for Stable 384.15 before doing anymore testing with my network.

Ill just also mention I only noticed drop outs on the 5Ghz network, I dont run 2.4

 

[Butterfly Bones]

Interesting.see my reply to Maverikcdn...My note 9 dropped on 384.14 and -2 several times before i went back to 384.13.

The mobile devices on my network are an iPhone Xs max, iPad 9", and Google Pixel 2 XL. No drops that I know or have seen. All my internal devices have static IPs, so I can identify any oddities in my logs.

 

[DarkKnight75]

Ya unfortunately tis the nature of wireless connectivity

Delicate balance between performance and stability. Especially so on the 5Ghz Band

Id download inSSIDer and see if it can help tell you anything about the interference levels in your area and might help to select something that gives you reasonable performance without any issues.

Im also going to hold out for Stable 384.15 before doing anymore testing with my network.

Ill just also mention I only noticed drop outs on the 5Ghz network, I dont run 2.4

Yeah, that's the nature of it. I do use an Analyer in my home, most networks in the 2.4ghz are using 1, 6 , 11 so i stuck with channel 2 (20 ) for 2.4ghz and 157 20/40/80 for 5hz. No drops on 384.13 (very stable as I was running it last year for almost 2 months and had no issues, so i'm back

 

[DarkKnight75]

The mobile devices on my network are an iPhone Xs max, iPad 9", and Google Pixel 2 XL. No drops that I know or have seen. All my internal devices have static IPs, so I can identify any oddities in my logs.

Gotcha....Interesting...so far running solid back on 384.13

 

[Wadadli]

Running 15 Beta1 on my AC5300. I have it running as a secondary router to my main one that is connected to the net. The 5300 is set to reboot every day at 1:30 AM. This morning it did so and was given an IP by my main router for 1440 minutes (24 hrs). This was confirmed by logs on both routers. I just noticed however that at 1:30 PM, the 5300 requested a new IP from the main router. I'm just wondering why it would need a new IP half way through it's lease.

 

[Deleted member 22229]

Thats how lease renew works it will always request at half life.

 

[Wadadli]

Thats how lease renew works it will always request at half life.

Not correct ! (I should have done some reading first)
According to RFC2131, The timer for the DHCPREQUEST by the client is a configurable value. It just defaults to 0.5 of the lease time.
FYI: https://tools.ietf.org/html/rfc2131#section-4.4.5

 

[dlandiss]

most networks in the 2.4ghz are using 1, 6 , 11 so i stuck with channel 2

It has been suggested by those with knowledge of the internal protocols of Wi-Fi that by using a non-standard (ie, not 1, 6, 0r 11) channel, you are subjecting both yourself and others to more interference, not less. The protocols used are said to be more tolerant of same-channel interference than of adjacent-channel.

 

[DarkKnight75]

It has been suggested by those with knowledge of the internal protocols of Wi-Fi that by using a non-standard (ie, not 1, 6, 0r 11) channel, you are subjecting both yourself and others to more interference, not less. The protocols used are said to be more tolerant of same-channel interference than of adjacent-channel.

In theory yes, reality no. My AC86u seems to perform best on Ch 2 20mhz 2.4ghz and auto 20/40/80 channel 157 for 5hz running on 384.13.

 

[Mutzli]

In theory yes, reality no. My AC86u seems to perform best on Ch 2 20mhz 2.4ghz and auto 20/40/80 channel 157 for 5hz running on 384.13.

I can confirm that, channel 2 is always the best for my environment. For the 5Ghz it's any channel above 149.

 

[_robodoc_]

Interesting.see my reply to Maverikcdn...My note 9 dropped on 384.14 and -2 several times before i went back to 384.13.

I can confirm your findings: went back to 384.13 for same reason.
(RT-AC86U on Merlin)

 

[sanke1]

Beta 1 is rock solid on my RT-AX88U provided I keep 2.4 Ghz and 5 Ghz channels separate. 2.4 Ghz network is on channel No. 11. The 5 Ghz channel has 160 Mhz width with Channel No. 40 set manually. OFDMA is enabled and all other settings are at their default. No wifi or wan drops or any weird behavior since last 5 days. Huge thumbs UP.
Just don't forget to set DNS servers manually to Google or Cloudflare to get rid of Disconnected error message.

 

[Rik Stigter]

On my AX88 running 384.15beta pixelserver won't start, not even after removing/re-installing it (i.e. Diversion standard). From the system log:

Feb 7 09:35:08 Diversion: created br0ixelserv-tls 192.168.1.2, from /opt/etc/init.d/S80pixelserv-tls
Feb 7 09:35:08 pixelserv-tls[309]: pixelserv-tls 2.3.1 (compiled: Jan 31 2020 13:27:14 flags: tfo tls1_3) options: 192.168.1.2
Feb 7 09:35:08 : Failed to start pixelserv-tls (Diversion) from /opt/bin/diversion (which is an executable)

Restarting pixelserv-tls doesn't change the situation. I'm running Samba (from the GUI), Diversion v4.1.8, Skynet v7.0.9 (with the GUI-plugin) and uiDivStats v1.3.0

I've checked and pixelserver-tls is available in /opt/bin with flags -rwxr-xr-x. Anybody else seen this problem and (more important perhaps) does anyone know how to fix it?

 

[Butterfly Bones]

On my AX88 running 384.15beta pixelserver won't start, not even after removing/re-installing it (i.e. Diversion standard). From the system log:

Feb 7 09:35:08 Diversion: created br0ixelserv-tls 192.168.1.2, from /opt/etc/init.d/S80pixelserv-tls
Feb 7 09:35:08 pixelserv-tls[309]: pixelserv-tls 2.3.1 (compiled: Jan 31 2020 13:27:14 flags: tfo tls1_3) options: 192.168.1.2
Feb 7 09:35:08 : Failed to start pixelserv-tls (Diversion) from /opt/bin/diversion (which is an executable)

Restarting pixelserv-tls doesn't change the situation. I'm running Samba (from the GUI), Diversion v4.1.8, Skynet v7.0.9 (with the GUI-plugin) and uiDivStats v1.3.0

I've checked and pixelserver-tls is available in /opt/bin with flags -rwxr-xr-x. Anybody else seen this problem and (more important perhaps) does anyone know how to fix it?

Since RMerlin states this in the first post of this thread -

Also please keep discussions in this thread on the topic of this beta release. Off-topic posts will be either ignored or removed, depending on my mood at the time.

Try Diversion thread
https://www.snbforums.com/threads/diversion-the-router-ad-blocker.48538/
Or pixelserv-tls thread
https://www.snbforums.com/threads/pixelserv-a-better-one-pixel-webserver-for-adblock.26114/

 

[skeal]

Before someone tells me to reset, I have and many times. I'm seeing with this client list bug that my dhcp ip address pool keeps resetting to defaults every time I change a setting on that page and apply. Strange behaviour.

 

[Rik Stigter]

Since RMerlin states this in the first post of this thread -

Try Diversion thread
https://www.snbforums.com/threads/diversion-the-router-ad-blocker.48538/
Or pixelserv-tls thread
https://www.snbforums.com/threads/pixelserv-a-better-one-pixel-webserver-for-adblock.26114/

My apologies, I was considering that myself, but sometimes it's difficult to determine which of the threads is best. In this case the problem started with beta1 and the version of amtm that is included in that release.
But I'll ask elsewhere.

 

[dave14305]

Before someone tells me to reset, I have and many times. I'm seeing with this client list bug that my dhcp ip address pool keeps resetting to defaults every time I change a setting on that page and apply. Strange behaviour.

So what happened?

 

[skeal]

So what happened?

I couldn't reproduce the issue so I crossed it out. I called Asus support and went through my discoveries about the client list bug and they are RMAing my router. They said specifically to leave the settings as is when sending mine in. I told them I was using merlin firmware, they said to continue support I would have to flash back to stock. I advised reflashing back to stock was tried several times with the firmware recovery tool and no luck in fixing the client list bug. He then said ok leave it as is and send it in, when you receive the new one.

 

[skeal]

This no client list causes riff raff through the whole system. QOS doesn't work right, you have to manually add IPs to lists that don't use the offline client list feature, etc it really sucks. If I had to give advice, I would say when resetting an AX88U to defaults DO NOT check the box to clear AI-Protection stuff and traffic analyzer stuff. That is where I think this bug was possibly created as it didn't exist before I did that.