Release Asuswrt-Merlin 386.7 is now available for all models

[SomeWhereOverTheRainBow]

Did I mention... I really "Hate" IPv6... ;-)

Nope, not even close.

 

[shabbs]

I'm curious as to what "zero need for it" means in the context of IPv4 v. IPv6 means. Assuming one can get IPv4 addresses from an available IPv4 pool, why would anyone need IPv6? In other words, if someone has no need for it, why would anyone have need for it unless they're in some extremely unique environment?

When I was setting up my Rogers Ignite TV on my network with the Rogers modem in bridged mode, the documents indicated IPv6 was needed. So, I used to have that turned on and enabled. Turns out, it's not really needed and Rogers Ignite TV works just fine with just old school IPv4 alone.

If there's a service or feature I do not need, I'm turning it off. Why enable/run stuff I don't need?

 

[SomeWhereOverTheRainBow]

When I was setting up my Rogers Ignite TV on my network with the Rogers modem in bridged mode, the documents indicated IPv6 was needed. So, I used to have that turned on and enabled. Turns out, it's not really needed and Rogers Ignite TV works just fine with just old school IPv4 alone.

If there's a service or feature I do not need, I'm turning it off. Why enable/run stuff I don't need?

Now you I get. If you don't need it, and your the type to set it and forget it, why bother. Just extra stuff's to complain when crap hits the fan.

 

[AntonK]

When I was setting up my Rogers Ignite TV on my network with the Rogers modem in bridged mode, the documents indicated IPv6 was needed. So, I used to have that turned on and enabled. Turns out, it's not really needed and Rogers Ignite TV works just fine with just old school IPv4 alone.

If there's a service or feature I do not need, I'm turning it off. Why enable/run stuff I don't need?

Hear, hear!

 

[SomeWhereOverTheRainBow]

Did I mention... I really "Hate" IPv6... ;-)

When I was setting up my Rogers Ignite TV on my network with the Rogers modem in bridged mode, the documents indicated IPv6 was needed. So, I used to have that turned on and enabled. Turns out, it's not really needed and Rogers Ignite TV works just fine with just old school IPv4 alone.

If there's a service or feature I do not need, I'm turning it off. Why enable/run stuff I don't need?

Hear, hear!

No offense to the nay say 'ers and non-users of ipv6 , but I believe a good portion of user base enjoy things when they work properly, then are ready to kick it to the curb once it does not serve a purpose. At least that is what a generalized perception is. To be honest, if some people need it, then it should be looked into. However, I do not see it being a nice journey unless someone with the know how can apply appropriate patches or somehow Asus over night decides they want to beta test the 5.+ SDK. Otherwise, the best route is @dave14305 approach. Turn off the dnsfilter option for ipv6.

 

[L&LD]

I don't need it, but I've never had an issue with it either (at least 3 years or more, today).

As soon as the ISP offered it, I had it enabled. Stable, rock-solid. Set-and-forget reliability.

Why?

Because the web is faster with it enabled.

From my perspective, most users that have issues are the ones that don't want to play with the new/required rules of engagement for IPv6 and/or their ISP's weak implementation of the same.

 

[SomeWhereOverTheRainBow]

I don't need it, but I've never had an issue with it either (at least 3 years or more, today).

As soon as the ISP offered it, I had it enabled. Stable, rock-solid. Set-and-forget reliability.

Why?

Because the web is faster with it enabled.

From my perspective, most users that have issues are the ones that don't want to play with the new/required rules of engagement for IPv6 and/or their ISP's weak implementation of the same.

Sorry let me clarify, disabling the new dnsfilter ipv6 options that are not properly working.

 

[shabbs]

No offense to the nay say 'ers ,

Wouldn't say I'm a "nay say 'er" - just someone where it does not fit my use case or my needs.

I may dabble in it later and figure out how to enable it on my Pi-hole's so I can block those requests too. With IPv6 enabled on my network, devices that are able to request it can bypass my Pi-hole at the moment. A project for later.

 

[SomeWhereOverTheRainBow]

Wouldn't say I'm a "nay say 'er" - just someone where it does not fit my use case or my needs.

I may dabble in it later and figure out how to enable it on my Pi-hole's so I can block those requests too. With IPv6 enabled on my network, devices that are able to request it can bypass my Pi-hole at the moment. A project for later.

I was like that once. I figured it out for my setup. I imagine when you get bored enough of the same setup you might decide to tackle it yourself.

 

[Tech9]

I figured it out for my setup.

Share your experience. Perhaps in another thread. It will be interesting discussion.

 

[SomeWhereOverTheRainBow]

Share your experience. Perhaps in another thread. It will be interesting discussion.

Perhaps, or perhaps others are better to fumble through the school of hard knocks. After all, we don't need everyone over the rainbow.

 

[Tech9]

We need to see what do you have and find holes in it. I have some ideas already.

 

[SomeWhereOverTheRainBow]

We need to see what do you have and find holes in it. I have some ideas already.

What took you so long?

 

[dave14305]

First I’ve seen a tcp query also cause it.

Jun 29 22:47:59 kernel: <unknown>: hw csum failure
Jun 29 22:47:59 kernel: CPU: 1 PID: 13525 Comm: dnsmasq Tainted: P           O    4.1.27 #2
Jun 29 22:47:59 kernel: Hardware name: Broadcom-v8A (DT)
Jun 29 22:47:59 kernel: Call trace:
Jun 29 22:47:59 kernel: [<ffffffc0000876d8>] dump_backtrace+0x0/0x150
Jun 29 22:47:59 kernel: [<ffffffc00008783c>] show_stack+0x14/0x20
Jun 29 22:47:59 kernel: [<ffffffc000508b00>] dump_stack+0x90/0xb0
Jun 29 22:47:59 kernel: [<ffffffc0003be90c>] netdev_rx_csum_fault+0x3c/0x50
Jun 29 22:47:59 kernel: [<ffffffc0003b4078>] skb_copy_and_csum_datagram_msg+0xe8/0xf8
Jun 29 22:47:59 kernel: [<ffffffc000446bbc>] tcp_rcv_established+0x4ec/0x6e8
Jun 29 22:47:59 kernel: [<ffffffc0004c5cd0>] tcp_v6_do_rcv+0x218/0x5f0
Jun 29 22:47:59 kernel: [<ffffffc00043b260>] tcp_prequeue_process+0xc8/0x130
Jun 29 22:47:59 kernel: [<ffffffc00043bf88>] tcp_recvmsg+0x520/0xaa8
Jun 29 22:47:59 kernel: [<ffffffc000465464>] inet_recvmsg+0xa4/0xc8
Jun 29 22:47:59 kernel: [<ffffffc0003a1418>] sock_read_iter+0x90/0xc8
Jun 29 22:47:59 kernel: [<ffffffc00013d848>] __vfs_read+0xa8/0x100
Jun 29 22:47:59 kernel: [<ffffffc00013e0d0>] vfs_read+0x78/0x160
Jun 29 22:47:59 kernel: [<ffffffc00013eaf4>] SyS_read+0x44/0xa0

 

[SomeWhereOverTheRainBow]

First I’ve seen a tcp query also cause it.

Jun 29 22:47:59 kernel: <unknown>: hw csum failure
Jun 29 22:47:59 kernel: CPU: 1 PID: 13525 Comm: dnsmasq Tainted: P           O    4.1.27 #2
Jun 29 22:47:59 kernel: Hardware name: Broadcom-v8A (DT)
Jun 29 22:47:59 kernel: Call trace:
Jun 29 22:47:59 kernel: [<ffffffc0000876d8>] dump_backtrace+0x0/0x150
Jun 29 22:47:59 kernel: [<ffffffc00008783c>] show_stack+0x14/0x20
Jun 29 22:47:59 kernel: [<ffffffc000508b00>] dump_stack+0x90/0xb0
Jun 29 22:47:59 kernel: [<ffffffc0003be90c>] netdev_rx_csum_fault+0x3c/0x50
Jun 29 22:47:59 kernel: [<ffffffc0003b4078>] skb_copy_and_csum_datagram_msg+0xe8/0xf8
Jun 29 22:47:59 kernel: [<ffffffc000446bbc>] tcp_rcv_established+0x4ec/0x6e8
Jun 29 22:47:59 kernel: [<ffffffc0004c5cd0>] tcp_v6_do_rcv+0x218/0x5f0
Jun 29 22:47:59 kernel: [<ffffffc00043b260>] tcp_prequeue_process+0xc8/0x130
Jun 29 22:47:59 kernel: [<ffffffc00043bf88>] tcp_recvmsg+0x520/0xaa8
Jun 29 22:47:59 kernel: [<ffffffc000465464>] inet_recvmsg+0xa4/0xc8
Jun 29 22:47:59 kernel: [<ffffffc0003a1418>] sock_read_iter+0x90/0xc8
Jun 29 22:47:59 kernel: [<ffffffc00013d848>] __vfs_read+0xa8/0x100
Jun 29 22:47:59 kernel: [<ffffffc00013e0d0>] vfs_read+0x78/0x160
Jun 29 22:47:59 kernel: [<ffffffc00013eaf4>] SyS_read+0x44/0xa0

Kinda leaning towards packet padding, now. But definitely not fixable since it is on the end of Broadcom.

 

[jata]

Just wanted to chip in on this thread with a non IPv6 comment.

Firstly massive thanks for the work on this project to @RMerlin and the community too.

I have merlin 386.7 clean installed on my AX86S for last 5 days with 3x AiMesh nodes (mixed AC/AX running latest stock) running perfectly. My log is so clean i'm a bit bored! It did take a bit of work to get everything working this smoothly but worth the effort.

I had one node that was playing up (5ghz network dropped after restart) even after a clean install/setup. So I used the firmware restore tool to flash the firmware and it is now working great.

 

[SomeWhereOverTheRainBow]

Just wanted to chip in on this thread with a non IPv6 comment.

Firstly massive thanks for the work on this project to @RMerlin and the community too.

I have merlin 386.7 clean installed on my AX86S for last 5 days with 3x AiMesh nodes (mixed AC/AX running latest stock) running perfectly. My log is so clean i'm a bit bored! It did take a bit of work to get everything working this smoothly but worth the effort.

I had one node that was playing up (5ghz network dropped after restart) even after a clean install/setup. So I used the firmware restore tool to flash the firmware and it is now working great.

Yea, other than the mix of IPV6 talk earlier. @RMerlin firmware flashed rock solid. Network Performance is spot on. Haven't noticed any changes in wireless performance. Overall a great firmware build.

 

[Treadler]

Yea, other than the mix of IPV6 talk earlier. @RMerlin firmware flashed rock solid. Network Performance is spot on. Haven't noticed any changes in wireless performance. Overall a great firmware build.

Yup, no complaints here!

 

[Paliv]

So for those of us with less of a grasp on what has been discussed in terms of IPV6 and DNSFilter, is the issue just for devices that are using only a manually entered IPV6 DNS address? And is it actually breaking resolution/or causing a router issue? I could recreate the log message with my iPhone using a manually entered IPV6 DNS, but resolution was still working. Does that mean DNSFilter is just being bypassed in this case? Sorry for the more basic questions, I was just curious what the issue is to figure out if I should keep DNSFilter and/or IPV6 on for now. So far I've had no issues and even when receiving the log message DNSFilter seems to be working.

 

[rlfromm]

I upgraded 2 RT-AC68U access point units to asuswrt-merlin 386.7. One unit upgraded OK and is working OK. The other unit requested a manual reboot after the upgrade and after the reboot it appears to operate. However, when I try to access its webpage I got the error "192.168.1.3 refused to connect." and therefore I cannot access the administration screens. I can login successfully using SSH and it displays the correct asuswrt-merlin release level. But no access is possible using the webpages.

Does anyone else have this issue? How can I resolve the issue?

I did some investigation connecting via ssh using putty. It turns out httpd was not running and starting httpd manually still did not resolve the issue. However, the refuse to connect error went away and was replaced by a file not found error. So, I decided to factory reset via the reset button. Since it's an access point it only took a minute to reconfigure and it is working normally again at the latest code level.