Asuswrt-Merlin 378.54_2 is now available

[RMerlin]

Asuswrt-Merlin 378.54_1 is now available, for all supported models.

UPDATE 2 (10-June-15): 378.54_2 has been uploaded, fixing OpenVPN config exporting, and igmpproxy not starting when using a PPPoE connection.

UPDATE: 378.54 would crash on some models (notably the RT-N66U) when enabling a feature forcing NAT acceleration to be disabled (like IPTraffic or QOS). All builds were recompiled as 378.54_1, there was no actual code change.

The focus of this release was enhancing and refining the newly added Policy-based VPN routing, as well as merging with Asus GPL 378_5134. This release also marks the move to the latest OpenSSL codebranch (1.0.2).

The highlights:

• Merged with Asus GPL 378.5134. The Quantenna driver from 378_6065 was merged in for the RT-AC87U.
• Numerous enhancements to OpenVPN's policy-based routing: you can now define exception rules (that will always be routed through your ISP), and numerous bugs/quirks resolved. Policy-routing also gets applied at boot time, so if you have it set to block Internet access while the tunnel is down, this will be applied at boot time.
• OpenSSL upgraded from 1.0.0 to 1.0.2. I had already backported numerous performance optimizations from 1.0.2, but since this version is binary compatible with 1.0.0, and 1.0.0 is approaching its end of life, the switch was made to 1.0.2. This adds access to newer ciphers, as well as TLS 1.2. Note that to enable TLS 1.2 support in OpenVPN, you must insert the following in the custom config section:
  

  tls-version-min 1.0

  This will enable negotiation of TLS 1.0 through 1.2 (based on the highest supported by the remote end)
• Miniupnpd was also updated to 1.9.20150430.
• The experimental Trend Micro Ad blocker was removed. It was too buggy, making it impossible to use various mobile applications (as it kept switching you to your web browser to report that a page had been blocked). If you were previously using it, you must disable it over SSH, with the following commands:
  

  nvram set wrs_adblock_popup=0
  nvram set wrs_adblock_stream=0
  nvram commit

  If for some reason you still want to use it, you can enable it manually by setting the above values to "1". There will be *NO* support for it.
• A few tweaks and fixes, such as EMF (multicast routing) which was broken on ARM devices.

See the changelog for more details on the other fixes and tweaks.

Downloads are here.
Changelog is here.

 

[elzounet]

Thx RMerlin for all your hard work !

Do we need to do a reset factory for the RT-AC87U due to the Quantenna driver update ?

 

[RMerlin]

Thx RMerlin for all your hard work !

Do we need to do a reset factory for the RT-AC87U due to the Quantenna driver update ?

Don't think so - at least I didn't do one myself. You might want to disable Beamforming support if you still experience stability issues with some clients.

 

[hungarianhc]

I really appreciate all your hard work, Merlin! I just donated a bit your way! I know it's small, but I'll keep them coming .

 

[dlandon]

When I upload an ovpn file to the OpenVPN Client, the server port resets to 1 and not 1194. If I re-enter 1194 and apply it is all back to normal.

EDIT: Happens on AC68U and AC3200.

 

[RMerlin]

When I upload an ovpn file to the OpenVPN Client, the server port resets to 1 and not 1194. If I re-enter 1194 and apply it is all back to normal.

EDIT: Happens on AC68U and AC3200.

Works fine for me on an RT-AC3200, I just re-tested it with a vpnbook ovpn config. Check your ovpn file, it might have an extra space in there.

 

[dlandon]

Works fine for me on an RT-AC3200, I just re-tested it with a vpnbook ovpn config. Check your ovpn file, it might have an extra space in there.

I am tunneling an AC68U and an AC3200. I exported the ovpn from the server on the server side and then uploaded it on the client side. I don't understand where the extra space could come from.

 

[Haggster]

just remote updated the firmware on my RT-AC87 via OpenVPN and adjusted the settings for TLS as decribed in the inital post. TLS 1.2 working perfectly via OpenVPN, I verified this via checking the logs.

 

[IAAI]

Jun 7 18:00:12 disk_monitor: Got SIGALRM...

no usb plugged in

 

[ObliteRon]

Updated from 378.53 to 378.54 on my RT-AC66 and none of my 5 GHz devices are connecting. 2.4 GHz is connecting fine. I'm not seeing anything of interest in the System log.

 

[mromero]

Jun 7 18:00:12 disk_monitor: Got SIGALRM...

no usb plugged in

Which router?

 

[mromero]

If going from the latest Asus BETA is hard reset required?

Does your release have the latest wireless drivers, Trend Micro version etc?

Asuswrt-Merlin 378.54 is now available, for all supported models.

 

[RMerlin]

If going from the latest Asus BETA is hard reset required?

No.

Does your release have the latest wireless drivers, Trend Micro version etc?

"latest" is a very subjective thing. No, it does not have the 378_61xx versions. It has what's described in the changelog.

 

[RMerlin]

Updated from 378.53 to 378.54 on my RT-AC66 and none of my 5 GHz devices are connecting. 2.4 GHz is connecting fine. I'm not seeing anything of interest in the System log.

Works fine with my RT-AC66U and a Nexus 9. Check your config, and if it still fails do a factory default reset.

The RT-AC66U wireless driver hasn't changed since October 2014.

 

[IAAI]

Which router?

RT-AC68P

 

[RMerlin]

Jun 7 18:00:12 disk_monitor: Got SIGALRM...

no usb plugged in

This is normal.

 

[IAAI]

Jun 7 18:00:09 disk_monitor: Got SIGALRM...

I win, by 3 secs

looooooool , one more time reboot and wait

 

[IAAI]

This is normal.

Thank you , could you please explain what is it ?

 

[hifiwifi]

Attempted upgrade from 53_0 on my RT-N66U. Router went into reboot loop. Was able to get to .54 default settings with WPS+reboot, but I prefer to wait to see if this can be resolved with a patch so I don't have to go through the song & dance of manually reconfiguring. Was able to duplicate this twice (downloaded 54 again to try ruling out bad download). I am NOT using any of the aforementioned adblock stuff.

 

[RMerlin]

Thank you , could you please explain what is it ?

disk_monitor is the service that handles scheduled filesystem scans. SIGALRM is the signal the process receives when it has a scheduled event occurring, which tells it to wake up, and process any scheduled event if there's a disk present before going back into sleep.