Release Asuswrt-Merlin 386.7 is now available for all models

[Wishmaster1965]

Updated my mesh node anac68u wilth no issues, My main router an AX88U again no issues.

 

[sdmf74]

Just flashed Merlin 386.7 and need to know if I should use ATM or PTM w/ Cake qos? (cable service is 260 down / 16 up).

Google did not help other than one person saying that ATM is Only for DSL connections. Not sure if that is accurate

Any other QOS suggestions welcome.

 

[amigohd]

I have lines with: "
kernel: ERR: rdpa_cpu_tx_port_enet_lan#223: rdpa_cpu_tx_port_enet_lan failed. rdd_rc=120 tx_rdd_error_count=1" in the log with 387.7 from time to time. Did not see that message on any beta or older versions.

 

[ColDen]

Performed a dirty upgrade from 386.7_beta2 yesterday on my RT-AC86U:

Still no observable issues to report at this time...

Thanks Éric!

 

[Howard_inGA]

RT-AC5300

2.4 and 5G#2 works fine
5G#1 connects but these no upload data when doing a speed test. Internet doesn't work very well if at all.

.

So, I've the same router and did a dirty upgrade from 386.7_beta2. The stable release went without a single hitch, and my internet speeds both with and without VPN active using the router are the best EVER!
Thanks for a slick upgrade, Eric!!!

 

[Protik]

Dirty upgrade from 386.5_2 to 386.7. The upgrade went well.

Only one cosmetic issue with AiMesh. I have a AiMesh system where AC68U is the router and AX86U is the node. On the AiMesh tab, if I select the 'Network' tab of the AiMesh node, everything is zero. As far as I can tell, the AiMesh is working fine, it just isn't showing anything on that tab.

I know everything about AiMesh is closed source, just wanted to update the folks.

 

[Mutzli]

Just flashed Merlin 386.7 and need to know if I should use ATM or PTM w/ Cake qos? (cable service is 260 down / 16 up).

Google did not help other than one person saying that ATM is Only for DSL connections. Not sure if that is accurate

Any other QOS suggestions welcome.

noatm, I have mine set to normal.

 

[Akore]

Updated from Beta 2 to 386.7 on RT-AC68U dirty with two attached USB drives. Everything seems to be working fine.

Getting these from time to time but had them also on the betas:

Jun 23 12:15:21 DualWAN: skip single wan wan_led_control - WANRED off
Jun 23 12:15:21 WAN_Connection: WAN was restored.
Jun 23 12:15:21 rc_service: wanduck 216:notify_rc stop_ntpd
Jun 23 12:15:21 rc_service: wanduck 216:notify_rc start_ntpd
Jun 23 12:15:21 rc_service: waitting "stop_ntpd" via wanduck ...
Jun 23 12:15:22 ntpd: Started ntpd
Jun 23 12:16:50 DualWAN: skip single wan wan_led_control - WANRED off
Jun 23 12:16:50 WAN_Connection: WAN was restored.
Jun 23 12:16:50 rc_service: wanduck 216:notify_rc stop_ntpd
Jun 23 12:16:50 rc_service: wanduck 216:notify_rc start_ntpd
Jun 23 12:16:50 rc_service: waitting "stop_ntpd" via wanduck ...
Jun 23 12:16:50 ntpd: Stopped ntpd
Jun 23 12:16:51 ntpd: Started ntpd

 

[underdose]

Dirty & seamless upgrade from beta 2. No issues to report.

 

[ebr4him]

I don't intend to use their client design since it's too integrated into OpenVPN's UI. I will have to develop my own webui once their backend code becomes final.

Just give us what Asus has for now or already a setting to enable it while you work on that, please

 

[XIII]

Shouldn't amtm report this new/final version?

awm Asuswrt-Merlin no release yet  386.7.beta2

 

[Smokey613]

Just give us what Asus has for now or already a setting to enable it while you work on that, please

The factory Asus firmware does not have VPN Director in it and it is just one of the reasons I use Merlin firmware. Without the VPN Director functionality, the VPN client is useless to me as I do not want everything going through the VPN like it does on the factory firmware.

 

[pat2]

Hi RMerlin,

dirty upgrade from 386.6. No issues to report, but I found a different behaviour for RAM usage.

On my router RT-AC66U, with the previous fw version 386.6 (differently from the other fw versions I used in the past) I had a lower RAM usage (around 55%) and stable around this value during usage.

I found that with the new version 386.7 RAM usage started from around 75% and after 1 day usage rised to 98% with a behaviour similar to all the previous fw versions (except 386.6).
Please could you explain the reason if you know? And, is it the expected behaviour?

I preferred the 386.6 "not saturated RAM usage" behaviour, even if from a performance standpoint I don't see differences. Shall I have to come back to 386.6 version?

thanks

Note: the router's services I'm using are: media server, 2 external disks with SAMBA, DDNS, port forwarding, DHCP, VPN server, IPSEC server, (Wireless is off).

 

[capncybo]

Hi RMerlin,

dirty upgrade from 386.6. No issues to report, but I found a different behaviour for RAM usage.

On my router RT-AC66U, with the previous fw version 386.6 (differently from the other fw versions I used in the past) I had a lower RAM usage (around 55%) and stable around this value during usage.

I found that with the new version 386.7 RAM usage started from around 75% and after 1 day usage rised to 98% with a behaviour similar to all the previous fw versions (except 386.6).
Please could you explain the reason if you know? And, is it the expected behaviour?

I preferred the 386.6 "not saturated RAM usage" behaviour, even if from a performance standpoint I don't see differences. Shall I have to come back to 386.6 version?

thanks

Note: the router's services I'm using are: media server, 2 external disks with SAMBA, DDNS, port forwarding, DHCP, VPN server, IPSEC server, (Wireless is off).

I think... this behavior has to do with NOT flushing caches & the change (is/was) deliberate. Have you heard the saying Not used RAM is wasted RAM?
But I'll go make some-popcorn & sit waiting for someone such as @RMerlin to explain it LOL

 

[RMerlin]

Just flashed Merlin 386.7 and need to know if I should use ATM or PTM w/ Cake qos? (cable service is 260 down / 16 up).

Neither, just apply the DOCSIS preset from the dropdown.

Just give us what Asus has for now or already a setting to enable it while you work on that, please

I can't, I don't have the up-to-date backend code from RC3, only the initial out of date code from 386_4xxxx.

I found that with the new version 386.7 RAM usage started from around 75% and after 1 day usage rised to 98% with a behaviour similar to all the previous fw versions (except 386.6).
Please could you explain the reason if you know? And, is it the expected behaviour?

It's just disk buffering done for your USB disk, it's perfectly normal.

 

[AntonK]

All good on GT-AX6000

Those are some beautiful speeds!

 

[Wanabo]

Upgraded my RT-AC88U master and imesh slave RT-AC86U_RT-AC2900 from 386.5_2 to 386.7.
All is functioning well. The usb 3.0 bug is still not resolved. All usb drives have their own powersupply. Only with setting usb 2.0 all usb drives are functioning properly. It's a pity I can not make use of the usb 3.0 speed.

 

[isr25]

dnsmasq hasn't changed in many months, so something else on your router is causing it to crash. If you use an addon that modifies its config (like Diversion), try disabling it.
EDIT: There was one change that fixed an issue in DHCPv6's functions. Are you using DHCPv6 on your network?

I uninstalled Diversion a few days prior to installing 386.7 - since I learned that it was causing the dcd / trendmicro crash.

I do have ipv6 enabled (native, ppp, dhcp-pd enabled). ipv6 works just fine though - devices get ipv6 and are connecting to ipv6 sites with no problems.

Dnsmasq is crashing. Do you run any dnsmasq.postconf files? Check those for errors and check your manual ip assignments. Any dnsmasq entries in the log just prior to crash?

I did not manually add/edit the dnsmasq.postconf file, but I do have manual ip assignments via the dhcp server web ui - didn’t change anything. Devices receive their manual ip assignments normally.

Prior to the crash, it was giving an IP address assignment to my work laptop which has a domain - dnsmasq was ignoring said domain. 7 seconds later, it crashed.

15 seconds before that, it was normal dnsmasq startup chatter

 

[Dezzo]

For those of you who were having issues with 5ghz disconnecting with the previous few releases, how is it working with 386.7 build?

I never could solve the disconnects so I've been using 386.3_2 in the meantime.

Didn't someone say the 5ghz issue was recognized by ASUS and that a fix was coming? Weird how not all were experiencing it though. RT-AX68U here.

 

[RMerlin]

I do have ipv6 enabled (native, ppp, dhcp-pd enabled). ipv6 works just fine though - devices get ipv6 and are connecting to ipv6 sites with no problems.

Odd. There could be a bug in the dnsmasq fix then, tho I haven`t seen them make any further change to that patch since it was published weeks ago.

I'll consider reverting it for the next release unless I don`t get any other similar report of dnsmasq crashes.