[BETA] Asuswrt-Merlin 380.59 Beta 1 is now available

[RMerlin]

I used to use HTB a long time ago in the Linux box I had as a router. The reason I switched to HFSC was that it performs better for VoIP applications, HTB reorders packets very frequently while HFSC doesn't! It's a difference you'll never see while just downloading data but makes a hell of a difference in VoIP apps that are very sensitive in the order that the packets arrive, out of order packets create jitter and that causes delays since the application has to buffer more.

The problem with HFSC though is that it is much more complicated to configure so you won't find it often in generic QoS solutions.

Hm, never had any issue with VoIP here, and I even spent about 2 hours on the phone two weeks ago doing a server migration
.

 

[RMerlin]

If you use the HTTPS interface, you get the following error:


Do you think there is an easy way to fix this? Sadly, the content is not available through https, but it's probably breaking the updateManufacturer part.

Edit: Yeah, it breaks the manufacturer recognition (it gets stuck at "loading manufaturer...").

Edit2: Maybe you could ship a local copy of the latest version with the firmware, and if it fails, it could use that.

It was discussed earlier in the thread. Unfortunately, Asus doesn't support https on their website, and a local copy would take a few megs. Best I could do would be to ditch that and switch it to the IEEE database (which I'm already using everywhere else). I haven't decided what to do about it yet.

The paranoia that is being implemented in browsers lately is starting to annoy me...

 

[charlie2alpha]

Hm, never had any issue with VoIP here, and I even spent about 2 hours on the phone two weeks ago doing a server migration
.

Notice that I said a long time ago, my upload link back then was 384 kbps ... Much more sensitive to the quirks of the packet schedulers.

 

[skeal]

Asuswrt-Merlin 380.59 Beta 1 is now available for most models, except the RT-N66U and RT-AC56U (due to missing GPL components).

The highlights of this release:

• Merged with Asus GPL 380_2697. The notable changes introduced by this are also listed below. Note that this is a BETA GPL, so base features might introduce some new issues.
• (GPL) MU-MIMO beta support for the RT-AC87, AC88, AC3100 and AC5300. This can be enabled under Wireless -> Professional. MU-MIMO-capable clients will be flagged as such on the Wireless Log page.
• (GPL) Fixed IPTV support.
• (GPL) Various security enhancements. For instance, the httpd server is now bound to specific network interfaces, and AiCloud's lighttpd was updated to a newer version.
• In addition, multiple security patches were backported from OpenWRT for Samba.
• A new button on the OpenVPN pages lets you revert the current server/client instance to its factory default settings, also erasing its stored keys/certificates.
• NAT Passthrough settings are now more flexible. You can optionally determine if the NAT helper should also be enabled or not. Previous Asus defaulted to always having the NAT helper enabled whenever passthrough was enabled.
• The official website (https://asuswrt.lostrealm.ca/) is now available through SSL, thanks to Let's Encrypt and my hosting provider officially supporting it. Note that Mediafire still only push downloads over HTTP - there's nothing I can do about this.
• Various changes to Samba. Ina ddition to the security patches mentionned above, SPNEGO support was also enabled, and some performance tweaks were made to its performance on the RT-AC56/AC68/AC87/AC3200.
• Webui enhancements to the DNSFilter and DHCP pages, integrating with Asus's new networkmap functionality.
• Fixed traffic monitoring on ARM models using a BCM5301x switch, through a workaround (actual issue will need to be fixed upstream by Asus)
• Tor updated to 0.2.7.6

Things that need to be tested:

• Test Samba sharing to ensure the security fixes didn't break anything. Test performance if you have one of the router models mentionned above, and report any change in performance.
• People with IPTV or VoIP issues, please check if Asus's IPTV fixes and my NAT helper changes help if you previously had any issue. Test the NAT passthrough in either Enabled or Enabled+Helper modes.
• If you previously used Tor, please confirm that the update didn't break anything
• There were numerous tweaks and enhancements throughout the webui, please report any unusual issue. When you do so, make sure you open your web browser console (usually through F12), reload the page, and report any Javascript error as well as the content of that line.
• Test the DHCP static lease page, as quite a few changes were made to how this page works, to integrate it into Networkmap
• Please provide your feedback regarding specifically the RT-AC68U (new SDK), RT-AC66U (first official merge with GPL 380_xxxx support) and RT-AC3200 (currently using a mix of old and new components due to missing up-to-date GPL components for that model)
• I've made a change to OpenVPN client routing (reverse path filtering was disabled). Please report if it makes any difference if you previously had issues with your OpenVPN client routing when in policy-based mode.

Please keep this thread on the specific topic of these beta builds. Off-topic posts will be ignored, moved or deleted, depending on my mood at the time.

Downloads are here.
Changelog is here.[

Asuswrt-Merlin 380.59 Beta 1 is now available for most models, except the RT-N66U and RT-AC56U (due to missing GPL components).

The highlights of this release:

• Merged with Asus GPL 380_2697. The notable changes introduced by this are also listed below. Note that this is a BETA GPL, so base features might introduce some new issues.
• (GPL) MU-MIMO beta support for the RT-AC87, AC88, AC3100 and AC5300. This can be enabled under Wireless -> Professional. MU-MIMO-capable clients will be flagged as such on the Wireless Log page.
• (GPL) Fixed IPTV support.
• (GPL) Various security enhancements. For instance, the httpd server is now bound to specific network interfaces, and AiCloud's lighttpd was updated to a newer version.
• In addition, multiple security patches were backported from OpenWRT for Samba.
• A new button on the OpenVPN pages lets you revert the current server/client instance to its factory default settings, also erasing its stored keys/certificates.
• NAT Passthrough settings are now more flexible. You can optionally determine if the NAT helper should also be enabled or not. Previous Asus defaulted to always having the NAT helper enabled whenever passthrough was enabled.
• The official website (https://asuswrt.lostrealm.ca/) is now available through SSL, thanks to Let's Encrypt and my hosting provider officially supporting it. Note that Mediafire still only push downloads over HTTP - there's nothing I can do about this.
• Various changes to Samba. Ina ddition to the security patches mentionned above, SPNEGO support was also enabled, and some performance tweaks were made to its performance on the RT-AC56/AC68/AC87/AC3200.
• Webui enhancements to the DNSFilter and DHCP pages, integrating with Asus's new networkmap functionality.
• Fixed traffic monitoring on ARM models using a BCM5301x switch, through a workaround (actual issue will need to be fixed upstream by Asus)
• Tor updated to 0.2.7.6

Things that need to be tested:

• Test Samba sharing to ensure the security fixes didn't break anything. Test performance if you have one of the router models mentionned above, and report any change in performance.
• People with IPTV or VoIP issues, please check if Asus's IPTV fixes and my NAT helper changes help if you previously had any issue. Test the NAT passthrough in either Enabled or Enabled+Helper modes.
• If you previously used Tor, please confirm that the update didn't break anything
• There were numerous tweaks and enhancements throughout the webui, please report any unusual issue. When you do so, make sure you open your web browser console (usually through F12), reload the page, and report any Javascript error as well as the content of that line.
• Test the DHCP static lease page, as quite a few changes were made to how this page works, to integrate it into Networkmap
• Please provide your feedback regarding specifically the RT-AC68U (new SDK), RT-AC66U (first official merge with GPL 380_xxxx support) and RT-AC3200 (currently using a mix of old and new components due to missing up-to-date GPL components for that model)
• I've made a change to OpenVPN client routing (reverse path filtering was disabled). Please report if it makes any difference if you previously had issues with your OpenVPN client routing when in policy-based mode.

Please keep this thread on the specific topic of these beta builds. Off-topic posts will be ignored, moved or deleted, depending on my mood at the time.

Downloads are here.
Changelog is here.

I updated my ac68u with 380.59 and lost the ability to map drives on my computer. As a matter of fact no network components are listed except the computer I use. My media boxes don't show up and the drives connected to my router don't show up either. I noticed this because I use my home network quite a bit. Do you know anything I'm missing? I have a windows 10 computer. It's like I lost my network.

 

[charlie2alpha]

One clarification about the DSL Reports speed/bufferbloat test: I incorrectly stated that with IPv4 I only get 3 streams vs 32 with IPv6. In reality, with IPv4 I get 18 streams but from 3 servers only while with IPv6 I get 32 streams from multiple servers. Noticed it today as the IPv6 test fails currently due to some of these servers not working properly at the moment so I had to fallback to the IPv4 test mode.

In other news, I'm glad to see @RMerlin fine tuning fq_codel support. ECN does seem supported in our kernel though, unsure why you disabled it in fq_codel. A "cat /proc/sys/net/ipv4/tcp_ecn" shows the value if 2 which means "Enable ECN when requested by incoming connections but do not request ECN on outgoing connections" which is the default in the Linux kernel.

Edit: I did some digging, it seems most people recommend ECN off on the egress, I guess you're just following these recommendations. So that answers my question.

 

[bradbort]

I updated my ac68u with 380.59 and lost the ability to map drives on my computer. As a matter of fact no network components are listed except the computer I use. My media boxes don't show up and the drives connected to my router don't show up either. I noticed this because I use my home network quite a bit. Do you know anything I'm missing? I have a windows 10 computer. It's like I lost my network.

On an a88u, I'm not seeing this. I have a windows 10 computer, and all network drives are available after the update. This was not true on the original asus firmware, which would lose various drives randomly.

 

[Veldkornet]

On my 87u, I put my 2.4Ghz wifi on "N-Only" instead of "Auto", but if I use a WiFi scanner (WiFi Explorer on Mac) it shows that the 2.4Ghz is b/g/n still.

So who's not playing nice here, the scanner or the router

Can anyone else test?

I tried a different Netgear AP, set it to only b/g and disabled N; and the scanner picks this change up. So I assume that the scanner is working.


Sent from my  iPhone using Tapatalk

 

[psyc0ld]

On my 87u, I put my 2.4Ghz wifi on "N-Only" instead of "Auto", but if I use a WiFi scanner (WiFi Explorer on Mac) it shows that the 2.4Ghz is b/g/n still.

So who's not playing nice here, the scanner or the router

Can anyone else test?

I tried a different Netgear AP, set it to only b/g and disabled N; and the scanner picks this change up. So I assume that the scanner is working.


Sent from my  iPhone using Tapatalk

Yes I see the same thing. I have N only and seeing b/g/n on the Mac explorer.

 

[RMerlin]

Edit: I did some digging, it seems most people recommend ECN off on the egress, I guess you're just following these recommendations. So that answers my question.

That's correct. ECN would require your upstream to also support it, which is dubious. ECN works best when you control both endpoints.

 

[___Lele___]

Hi sorry for the simple question but is there any thread specifically for building the latest source?

I have already cloned the repo (3 times!) and followed the instructions here (https://github.com/RMerl/asuswrt-merlin/wiki/Compile-Firmware-from-source-using-Ubuntu) but they seem to stop at RT-68U.

I need to build for RT-AC3200 and I understand I need the ARM architecture. I d'd to auswrt-merlin/release/src-rt-7.14.../src and did a

make clean
make -j8 rt-ac3200

Works fine up to a point but then it complains it doesn't find some files (I don't have the log in front of me right now)

Before I spend days debugging this can anybody tell me if there is an updated procedure to follow for AC3200? Any docs at all?

thanks a lot!

 

[defcon]

@RMerlin I'm noticing codel/fq_codel substantially lowers buffer bloat in my tests. I'd like to request if at all possible that we could specify services/ports to get higher QOS priority. Is there a command line way to get this done?
Also can you include busybox or a small nano binary so we can make edits other than using VI? Most of us do not use optware/entware.

Good work so far, thanks for your awesome work.

 

[RMerlin]

I need to build for RT-AC3200 and I understand I need the ARM architecture. I d'd to auswrt-merlin/release/src-rt-7.14.../src and did a

The RT-AC3200 must be built from the src-rt-7.x/ folder. Slightly older SDK than the 7.14, which is for the newer AC88/AC3100/AC5300.

 

[RMerlin]

@RMerlin I'm noticing codel/fq_codel substantially lowers buffer bloat in my tests. I'd like to request if at all possible that we could specify services/ports to get higher QOS priority. Is there a command line way to get this done?

Drop down menu upper right, from there you can access the user rule configuration.

Also can you include busybox or a small nano binary so we can make edits other than using VI? Most of us do not use optware/entware.

Busybox is already there, that's what is providing vi.

I looked at adding nano a few years ago, and it would have added 1-2 MB to the firmware due to its reliance on additional stuff for its term support. Space is starting to be tight on the RT-N66U and RT-AC66U.

 

[RMerlin]

nstalled 380.59 git with QoS patches last night. Everything works fine on my RT-AC56U except one annoying thing, all the LEDs turned off except the power LED. After some tinkering I found that a simple "service restart_wireless" brought all the LEDs back to life so I put that command to the services_start user script to be executed in every boot.

I couldn't reproduce the issue here, so it could be a timing issue.

Try "service restart_leds" instead of restarting the whole wireless stack.

 

[RMerlin]

Regarding MU-MIMO, i can confirm no M flag in 87U wireless log for Galaxy S7 which is supposed to support MU-MIMO.
Not sure which one at fault.

Looks like Quantenna doesn't expose that information, as Asus doesn't report it in the stock firmware either.

 

[___Lele___]

The RT-AC3200 must be built from the src-rt-7.x/ folder. Slightly older SDK than the 7.14, which is for the newer AC88/AC3100/AC5300.

Thanks RMerlin, that helped. I'm running

root@debian:~/asuswrt-merlin/release/src-rt-7.x.main/src# make rt-ac3200

but unfortunately I still can't compile. It keeps crashing at netatalk, see below. I'm running from a fresh install of Debian following the instruction on Git. The DB headers are installed correctly and if I run ./configure from the netatalk folder it runs fine.

Any ideas? Is there a specific config known to work for AC3200?

Thanks again for the help btw, much appreciated.


Error:

...
 CCLD   fce

../../libatalk/.libs/libatalk.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'

  CCLD   afpd

arm-brcm-linux-uclibcgnueabi-gcc: @LIBADD_DL@: No such file or directory

make[8]: *** [afpd] Error 1

make[8]: Leaving directory `/root/asuswrt-merlin/release/src/router/netatalk-3.0.5/etc/afpd'

make[7]: *** [all] Error 2

make[7]: Leaving directory `/root/asuswrt-merlin/release/src/router/netatalk-3.0.5/etc/afpd'

make[6]: *** [all-recursive] Error 1

make[6]: Leaving directory `/root/asuswrt-merlin/release/src/router/netatalk-3.0.5/etc'

make[5]: *** [all-recursive] Error 1

make[5]: Leaving directory `/root/asuswrt-merlin/release/src/router/netatalk-3.0.5'

make[4]: *** [all] Error 2

make[4]: Leaving directory `/root/asuswrt-merlin/release/src/router/netatalk-3.0.5'

make[3]: *** [netatalk-3.0.5] Error 2

make[3]: Leaving directory `/root/asuswrt-merlin/release/src/router'

make[2]: *** [all] Error 2

make[2]: Leaving directory `/root/asuswrt-merlin/release/src-rt-7.x.main/src'

make[1]: *** [bin] Error 2

make[1]: Leaving directory `/root/asuswrt-merlin/release/src-rt-7.x.main/src'

make: *** [rt-ac3200] Error 2

 

[john9527]

root@debian:~/asuswrt-merlin/release/src-rt-7.x.main/src#

Don't switch to src....just /asuswrt-merlin/release/src-rt-7.x.main

 

[___Lele___]

Don't switch to src....just /asuswrt-merlin/release/src-rt-7.x.main

Hi, If I don't switch to source I can't "make". It seems it only contains "src" while the older folders contain the makefiles.

root@ubuntu:~/asuswrt-merlin/release/src-rt-7.x.main# ls
src
root@ubuntu:~/asuswrt-merlin/release/src-rt-7.x.main# make rt-ac3200
make: *** No rule to make target `rt-ac3200'.  Stop.

This is straight from git. Is it corrupt? I figured something changed from 7x onwards

 

[john9527]

Hi, If I don't switch to source I can't "make". It seems it only contains "src" while the older folders contain the makefiles.

Well, I'll be damned.....you are right. It looks like they changed the layout. Sorry for the bad info on my part.

Try running these two commands first before starting the final 'make rt-ac3200'

make cleankernel
make clean

 

[___Lele___]

Well, I'll be damned.....you are right. It looks like they changed the layout. Sorry for the bad info on my part.

Try running these two commands first before starting the final 'make rt-ac3200'

make cleankernel
make clean

no change unfortunately. Looks like there is something wrong with netatalk. Can I disable the module so it is not compiled? If yes, what's the cleanest way of doing it? I don't need compatibility with Macs

What system are you using for compiling the code?

EDIT: Also testing with Ubuntu 14.04. It seems it has passed the netatalk compile. fingers crossed it gets to the end!