[Beta] Asuswrt-Merlin 384.7 Beta is now available

[Grisu]

think only factory reset will solve your issues, maybe second upload of firmware needed too, and doing all settings again without reloading safed config.

 

[The Chief]

I flash 384.7 over 384.6 on my ac86u, all seen good. but after one night, I cannot one the webui, I need ssh into the router to reboot. then webui works again.

thanks a lot!

Got same problem. WEBUI was gone, reboot via ssh helped.

 

[The Chief]

AFAIK, Let's Encrypt requires the use of Asus' own DDNS, because it relies on a special hook to complete the validation process.

No, it's not. I use LE with privatedns.org DDNS service (custom script) with no problem at all. It however requires «white ip» .

 

[PDinDetroit]

You have clients that are trying to use SMBv1. The problem is with these clients, not with the router firmware.

Thank you. I checked and all my Windows Clients had SMB v1 uninstalled, so it must be some other device. I don't have the extra cycles to run it down with Wireshark (started a new project at work). Since I don't use Samba, I simply disabled it.

Problem Solved!

Thank you again for all your hard work, 384.7_beta1 running on RT-AC86U very stable.

 

[Sinner]

Has anyone else had this error since updating to 384.7 Beta1?
I know this error has appeared in the past on 382.2 Beta 1 but it doesn't look like answer for it was found for it and I am wondering if it has something to do with Trend Micros signature update 2.088?

Sep 16 16:32:29 rc_service: httpd 3245:notify_rc restart_qos;restart_firewall
Sep 16 16:32:50 kernel: ERR[parse_qos_conf:932] Can't set new QoS conf while QoS is started!
Sep 16 16:32:50 kernel: ERR[ioctl_iqos_op_config:3592] parse qos_conf error!!
Sep 16 16:32:50 kernel: ioctl_iqos_op_config() fail!
Sep 16 16:32:50 kernel: ERR[qos_start:3344] QoS is already started!
Sep 16 16:32:50 kernel: ioctl_iqos_op_switch(1) fail!

That error has been discussed extensively. Its related to changes in the firmware and trying to load a module before its ready. If it is only present a couple times after a reboot your fine and can ignore it. If repeating constantly then theres possible issue

 

[D_Day]

That error has been discussed extensively. Its related to changes in the firmware and trying to load a module before its ready. If it is only present a couple times after a reboot your fine and can ignore it. If repeating constantly then theres possible issue

I found the answer on Fresh JR’s post on adaptive qos custom script.
Thanks for replying though

 

[RMerlin]

No, it's not. I use LE with privatedns.org DDNS service (custom script) with no problem at all. It however requires «white ip» .

Then I have no idea how the hostname validation is done for these. AsusDDNS relies on a DNS-based authentication, your router sends a special auth token to Asus's server when it updates your IP.

 

[Bob.Dig]

In that case unless Asus decides to keep issuing firmware updates (unlikely based on what's written at the top of that page) then it also means 384.6 will probably be the end of the road for my firmware on that model as well. Too much closed source components, they break backward compatibility every few firmware releases.

To bad, then it is time to switch to dd-wrt.
I had a nice time here, thanks to RMerlin for all the hard work. Bye bye everybody.

 

[Grisu]

To bad, then it is time to switch to dd-wrt.
I had a nice time here, thanks to RMerlin for all the hard work. Bye bye everybody.

Why not useing John's Merlin-fork, he is still updating RT-AC56U.

 

[agilani]

To bad, then it is time to switch to dd-wrt.
I had a nice time here, thanks to RMerlin for all the hard work. Bye bye everybody.

Good luck, I finally got rid of my last dd-wrt router. It was just too much of hassle and kluge of parts and pieces. You'd be better off getting a new router. A real ac68 router sells for all of $40 on ebay right now.

 

[The Chief]

Then I have no idea how the hostname validation is done for these. AsusDDNS relies on a DNS-based authentication, your router sends a special auth token to Asus's server when it updates your IP.

LetsEncrypt has A LOT of validation schemes. One of them is: client requests cert from LE server, LE server connects to host (using FQDN name submitted by client) , sends special https request and waits correct answer from client (this process is so called «ACME challenge»). I have not inspected LE client source code from you code repository, can't say about exactly what types of LE verification was implemented there.

 

[David Tanaka]

I did an SSH restart and that worked! Thanks!

 

[RMerlin]

LetsEncrypt has A LOT of validation schemes. One of them is: client requests cert from LE server, LE server connects to host (using FQDN name submitted by client) , sends special https request and waits correct answer from client (this process is so called «ACME challenge»). I have not inspected LE client source code from you code repository, can't say about exactly what types of LE verification was implemented there.

It's closed source, hence the lack of info.

The lack of a "real" web server seriously limits the possibilities there.

Sent from my P027 using Tapatalk

 

[The Chief]

It's closed source, hence the lack of info.
The lack of a "real" web server seriously limits the possibilities there.
Sent from my P027 using Tapatalk

On Exchange Windows Server, for example, LE ACME client integrates itself into IIS and becomes able to auto-get and auto-renew LE certs. ASUSWRT details of LE client work & possible httpd/firewall integration are to be yet revealed.

What I can definitely claim— it works in my case with custom ddns service/script.

 

[The Chief]

It's closed source, hence the lack of info.

What about this:
https://github.com/RMerl/asuswrt-merlin.ng/tree/master/release/src/router/acme-client-portable

Seems to me it's an acme-portable client using classic linux-way 'webroot' challenge method.

 

[Xentrk]

Hello,
Anyone else seeing the dnsmasq SIGTERM errors in system log?

Sep 21 03:52:05 RT-AC88U-8248 rc_service: service 9996:notify_rc updateresolv
Sep 21 03:52:05 RT-AC88U-8248 dnsmasq[21552]: exiting on receipt of SIGTERM
Sep 21 03:52:05 RT-AC88U-8248 custom_config: Appending content of /jffs/configs/dnsmasq.conf.add.
Sep 21 03:52:05 RT-AC88U-8248 custom_script: Running /jffs/scripts/dnsmasq.postconf (args: /etc/dnsmasq.conf ) - max timeout = 120s
Sep 21 03:52:05 RT-AC88U-8248 ovpn-client4[21126]: SIGTERM[soft,auth-failure] received, process exiting
Sep 21 03:52:05 RT-AC88U-8248 dnsmasq[10003]: started, version 2.80test4 cachesize 1500
Sep 21 03:52:05 RT-AC88U-8248 dnsmasq[10003]: compile time options: IPv6 GNU-getopt no-RTC no-DBus no-i18n no-IDN DHCP DHCPv6 no-Lua TFTP no-conntrack ipset no-auth DNSSEC no-ID loop-detect no-inotify no-dumpfile
Sep 21 03:52:05 RT-AC88U-8248 dnsmasq[10003]: warning: interface tun22 does not currently exist
Sep 21 03:52:05 RT-AC88U-8248 dnsmasq[10003]: warning: interface pptp* does not currently exist
Sep 21 03:52:05 RT-AC88U-8248 dnsmasq[10003]: asynchronous logging enabled, queue limit is 5 messages

 

[M@rco]

Anyone else seeing the dnsmasq SIGTERM errors in system log?

No dnsmasq SIGTERM errors found in my syslog.

 

[Xentrk]

Small typo in the log file.. waiting spelled as waitting

rc_service: waitting "restart_vpnclient4" via httpds ...

 

[Xentrk]

No dnsmasq SIGTERM errors found in my syslog.

Thanks @M@rco, When you have time, logon to SSH session and type service restart_dnsmasq. Then, see if there is a SIGTERM error in the log. Could have something to do with changes I've made testing stubby DNS over TLS.

 

[M@rco]

When you have time, logon to SSH session and type service restart_dnsmasq. Then, see if there is a SIGTERM error in the log.

No errors on service restart_dnsmasq:

Sep 21 14:16:20 rc_service: service 17303:notify_rc restart_dnsmasq
Sep 21 14:16:20 custom_script: Running /jffs/scripts/dnsmasq.postconf (args: /etc/dnsmasq.conf ) - max timeout = 120s
Sep 21 14:16:23 Diversion: restarted Dnsmasq to apply settings, from /jffs/scripts/dnsmasq.postconf
Sep 21 14:16:33 rc_service: watchdog 292:notify_rc start_dnsmasq
Sep 21 14:16:33 custom_script: Running /jffs/scripts/dnsmasq.postconf (args: /etc/dnsmasq.conf ) - max timeout = 120s
Sep 21 14:16:35 Diversion: restarted Dnsmasq to apply settings, from /jffs/scripts/dnsmasq.postconf