[Fork] Asuswrt-Merlin 374.43 LTS releases (Archive)

[RogerSC]

Sorry to disappoint, but the ipv6 stack is one of the items that I'm not going to touch on the fork. My feeling from reading the various threads is that whether one implementation works better that another (or conflicts with NAT acceleration) is largely tied to the ISP and their ipv6 implementation. Changing it may make some users happy, but could just as easily create issues for others. Plus, it's such a large change, there's just too much opportunity for error.

I appreciate your response on this one. IPv6 is working really well for me on Comcast...just keeps working. Hate to risk that success, given how sensitive IPv6 support seems to be to change.

Thanks!

 

[LiFePO4]

I have Comcast Business Internet and IPV6 is working here as well. It is not 100% perfect but I know how to work around the ~5% of issues. I am on the latest fork release (not beta) and it is the most stable IPV6 FW yet so I appreciate the stance of not messing with it (thank you!!)...

 

[john9527]

Today, I again tried with RT-AC68U_378.53_0 and I am still seeing this problem

Sorry, wrong thread....this is for a 374 firmware based fork.

 

[RogerSC]

There is one change that would be nice for IPv6, changing from using radvd to dnsmasq for the IPv6 advertising function. RMerlin's version did that some time ago, based on changes by Asus (I understand). However, I don't know how widespread the impact of that would be, or what's required to be in place to port that change. If it's localized, one might look at it, but if not, IPv6 is working well as it is, so it most likely isn't worth the time or risk to stability to do it.

Just a thought, I have a feeling that this may not be an easy change in the framework that you now have, but I thought that I'd mention it.

Thanks.

 

[Deleted member 22229]

I agree with Roger the newer platform for IPv6 is very stable and works better then radvd that said it still works good enough if that change is problematic.

 

[john9527]

There is one change that would be nice for IPv6, changing from using radvd to dnsmasq for the IPv6 advertising function. RMerlin's version did that some time ago, based on changes by Asus (I understand). However, I don't know how widespread the impact of that would be, or what's required to be in place to port that change. If it's localized, one might look at it, but if not, IPv6 is working well as it is, so it most likely isn't worth the time or risk to stability to do it.

Just a thought, I have a feeling that this may not be an easy change in the framework that you now have, but I thought that I'd mention it.

Thanks.

I'm just really nervous about messing around with the IPv6 stack. I did pick up an updated radvd for version 11...let's see how that goes first.

 

[RMerlin]

Since the IPv6 code spreads all over the place, cherry-picking those specific updates might be tricky.

 

[john9527]

Since the IPv6 code spreads all over the place, cherry-picking those specific updates might be tricky.

I picked it up to try and help out a user with a finicky ISP that resulted in radvd hanging. At least with a HE 6in4 tunnel it looks like it's running fine.

Anybody with a native IPv6 connect want to try it out before I make the public posting? Drop me a PM.

 

[rumblpak]

I picked it up to try and help out a user with a finicky ISP that resulted in radvd hanging. At least with a HE 6in4 tunnel it looks like it's running fine.

Anybody with a native IPv6 connect want to try it out before I make the public posting? Drop me a PM.

Actually, I was having a ton of issues with 6to4 due to dnscrypt where it prevented a lot of local traffic. In a nutshell, disabling ipv6 solved a slew of problems including (but not limited to) windows file sharing.

[edit] scratch that. modified the scripts to run opendns properly with a ipv6 local dns and now it works fine.

 

[degen]

With this firmware range is almost back to how it was when I bought the RT-N66U, as others have reported. In fact, 2.4 ghz range is why I bought this router in the first place and not one of the newer wireless AC routers.. I left the transmit power setting at 80 and I'm still seeing range improvements.

Many thanks.

 

[Gunner84]

With this firmware range is almost back to how it was when I bought the RT-N66U, as others have reported. In fact, 2.4 ghz range is why I bought this router in the first place and not one of the newer wireless AC routers.. I left the transmit power setting at 80 and I'm still seeing range improvements.

Many thanks.

Yeah I agree this fork firmware is really good, I've tried a few of the later versions and range does not seem as good, so I always went back to 374.43 and with John's work on security fixes and such is great!

 

[Blinkyz]

Hi John.

I just noticed your Signature is 11J now , not 11B. Any ETA for this release?
After trying 378.53 for a few days, like always im back with the FORK!!! (10j)
RT-AC68u(V10j)

 

[john9527]

I just noticed your Signature is 11J now , not 11B. Any ETA for this release?

Soon

Merlin's comment made me a bit nervous about the RADVD update, so a few kind folks are giving it a smoke test before I post it up (so far, so good).

 

[john9527]

Time for a new release....close to 6 weeks since the last release (how time flies!)

LATEST RELEASE: Update-11
27-April-2015
Merlin fork 374.43_2-11j9527
Download http://1drv.ms/1uChm3J
============================
Update-11 of the 374.43 update fork is now available. This release is primarily a function/fix update, with minimal security updates (only some updated cipher support). Many of Merlin's new functions from 378.52 and 378.53 have been backported, including Dual DNS server support and Policy based routing for OpenVPN connections.

From a fork unique perspective, several new options have been added or externalized in the gui and many gui formatting fixes have been included. The ASUS save configuration now includes the code level in the save filename and the Wireless Log has been 'color coded' for better readability.

Please take a minute to review the release Highlights in the first post, and the Changelog in the download directory for some additional detail.

Thanks again to everyone for their support, and special thanks to Kel-L and GHammer for helping to 'smoke test' the final code before release.

 

[Trebuin]

Everything is stable so far on 11...I'll probably know if something is wrong in about 3 days or earlier. Thanks for the hard work on this! On another note, would it be possible to bring up the ability to create a guest network while operating in repeater mode? I'd like this for security, though it probably won't help being that it will be on the same subnet as the primary network and everything is out the window then.

 

[Chrysalis]

I agree with Roger the newer platform for IPv6 is very stable and works better then radvd that said it still works good enough if that change is problematic.

if you mean dnsmasq then it has a very big flaw it cannot pass on smaller mtu to ipv6 lan devices.

This was an oversight on shibby also in which I posted a bug report which is still left unanswered.

radvd allows mtu manipulation which john kindly implemented.

dnsmasq version of ipv6 is effectively broken on pppoe.

I apologise for never getting round to testing the new radvd, but looks like someone else did, so is good news

 

[Deleted member 22229]

if you mean dnsmasq then it has a very big flaw it cannot pass on smaller mtu to ipv6 lan devices.

This was an oversight on shibby also in which I posted a bug report which is still left unanswered.

radvd allows mtu manipulation which john kindly implemented.

dnsmasq version of ipv6 is effectively broken on pppoe.

I apologise for never getting round to testing the new radvd, but looks like someone else did, so is good news

Thanks for the info. I have only tested DNSMASQ with Comcast native v6 and only general function witch was good,and more importantly the ability for the router to obtain and retain a v6 address consistently. Things may be different with pppoe connections. Probably why getting v6 to work across the board has been proven difficult with home routers.

 

[Chrysalis]

check here for more info.

http://tomato.groov.pl/?page_id=334&bugerator_nav=display&bug_project=1&issue=1706

I have posted in one of merlin's threads to see if he knows if asus has the same issue (my guess is yes as dnsmasq has no mtu overide function).

 

[Deleted member 22229]

check here for more info.

http://tomato.groov.pl/?page_id=334&bugerator_nav=display&bug_project=1&issue=1706

I have posted in one of merlin's threads to see if he knows if asus has the same issue (my guess is yes as dnsmasq has no mtu overide function).

Interesting thanks for posting that.

 

[Chrysalis]

although I guess if someone uses 1500 mtu on pppoe utilising jumbo frames (is added in this firmware) then that should also get pppoe ipv6 via dnsmasq working as well. But most are unaware of this as its a new thing.