Asuswrt-Merlin 380.59 is now available

[stambeccuccio]

Great work mythical Merlin! I really like so much this new firmware.

I would like to report a minor inconvenience / bug, which has always been there so it's not a news.

QoS - Web History:
The Access Time of Web Surfing history it's not always chronologically sorted.

 

[Andrii]

Andrii: Try another web browser... ;-)

Tried. Result the same: "Firmware upgrade unsuccessful. This may result from incorrect image or error transmission. Please check the version of firmware and try again."

 

[punkinduster]

Tried. Result the same: "Firmware upgrade unsuccessful. This may result from incorrect image or error transmission. Please check the version of firmware and try again."

Make sure you removed any USB drives and reboot to free up memory.

 

[rlcronin]

RT-N66U in access point mode running 380.58. Applied 380.59. Got login page and entered credentials. Web UI would not come up. Did factory reset. Initial post-factory-restore Web UI came up fine. Reconfigured and used nvram-restore to set thing back the way they were prior to the upgrade to 380.59. Rebooted. Entered credentials. No Web UI again. Now doing recovery back to 380.58. Something's not right.
--
bc

 

[RMerlin]

1) Sir i would like to know if you can add a feature of Boosting the wireless signals for about 200 MW or more just how it was on hggomes firmware ??

That's not legally allowed.

Installed over the top of beta 1 on AC66U.
Sadly openvpn client would not provide internet.
Client 2, syslog showed initialization complete and all the ip's re-routed, but no external internet.
Internally there was access pc to pc and pc to other devices, only nothing to the outside world.

Check your configuration. There was no recent changes to OpenVPN support, and it works fine here.

Thanks for your great work as always! fq_codel definitely helps me fighting bufferbloat on my AC68U.

Do you have any measurement to confirm it? So far, nobody could provide any numbers showing any improvement. There was next to zero change in ping latency in my own tests.

 

[RMerlin]

in this fw can i set descriptopn for offline devices (when i define dhcp static list ) ?

Functionality of the networkmap is identical to Asus's own.

 

[RMerlin]

Entered credentials. No Web UI again.

Most likely it's either an invalid setting was reapplied, or you have a client on your LAN with an invalid name, breaking the webui.

 

[Weaselboy]

Updated from 380.58 to 380.59 on my AC68P and it works well with one exception. The GUI is painfully slow on Safari (OS X 10.11.4), but is fast and snappy in Chrome. I never had this issue with 380.58 or previous releases.

When I say slow, it takes about 30 seconds to login once I enter the password (just sits at a white screen). Then once logged in if I click from one management page to another, each page takes around ten seconds to load. The same admin pages are very fast in Chrome. I have disabled adblock etc and still the same.

No big deal, I can just admin from Chrome when needed, but it seems odd Safari was fine all along with previous versions.

Edit: Well... ignore my post. In the last few minutes it seems to have fixed itself and is now fast as ever on Safari. Is there some sort of post firmware upgrade cleanup routine that occurs that may account for this? I changed nothing and it just went back to the usual snappy GUI in Safari all on its own. Odd.

 

[rlcronin]

Most likely it's either an invalid setting was reapplied, or you have a client on your LAN with an invalid name, breaking the webui.

So, factory reset and no nvram-restore, then? It was fine after the rescue restore to 380.58, and also after the nvram-restore.
--
bc

 

[bmn1]

Do you have any measurement to confirm it? So far, nobody could provide any numbers showing any improvement. There was next to zero change in ping latency in my own tests.

How would I go about benchmarking this? I have been using Tomato because fq_codel allows me to play games while watching netflix on my terrible 16/1 connection. So my only measurement is that BW limiter at 90% for upload and download plus fq_codel just works perfectly for me without configuring anything (unlike traditional/adaptive QOS).

 

[jb68]

The Safari Bug is still present on this firmware on Mac OS.X 10.11.4, you are unable to change the QOS Settings Like Priority mode for example. Game to media Streaming, when you press apply, it just says applying for 5 secs and refreshes with nothing changed.

Edit: Doesn't work on Safari on iOS 9 either on the iPhone 6

 

[rlcronin]

Updated from 380.58 to 380.59 on my AC68P and it works well with one exception. The GUI is painfully slow on Safari (OS X 10.11.4), but is fast and snappy in Chrome. I never had this issue with 380.58 or previous releases.

When I say slow, it takes about 30 seconds to login once I enter the password (just sits at a white screen). Then once logged in if I click from one management page to another, each page takes around ten seconds to load. The same admin pages are very fast in Chrome. I have disabled adblock etc and still the same.

No big deal, I can just admin from Chrome when needed, but it seems odd Safari was fine all along with previous versions.

Edit: Well... ignore my post. In the last few minutes it seems to have fixed itself and is now fast as ever on Safari. Is there some sort of post firmware upgrade cleanup routine that occurs that may account for this? I changed nothing and it just went back to the usual snappy GUI in Safari all on its own. Odd.

Hm, this sounds a lot like what I was seeing, though I didn't wait as long as you before I panicked and did the factory reset.
--
bc

 

[deathnite]

I disabled the sw_devled watchdog on the RT-AC56U, as I have a feeling this model isn't supposed to use it. This resolves the LEDs turning off at boot time.

Although you fixed the LEDs on AC56U in this release, the entire 380 series still has an issue. The 5GHz wireless LED doesn't blink on activity, it's just solid. The last firmware when it worked is 378.56, I just tested. Something changed starting with 380.
The LEDs on the back are not affected, they continue to blink on activity in 380.59.

The only asus 380 firmware for this model is 380.2695, which has the front LEDs completely off, except for power. And even restarting the wireless service doesn't help to bring them back.

 

[charlie2alpha]

Although you fixed the LEDs on AC56U in this release, the entire 380 series still has an issue. The wireless LED doesn't blink on activity, it's just solid. The last firmware when it worked is 378.56, I just tested. Something changed starting with 380.
The LEDs on the back are not affected, they continue to blink on activity in 380.59.

The only asus 380 firmware for this model is 380.2695, which has the front LEDs completely off, except for power. And even restarting the wireless service doesn't help to bring them back.

Hmm, I do not think that is the case, at least the 2.4GHz LED blinks here. The 5GHz one never blinked much in the first place. I'm satisfied they're at least lit, indicating that the device is up and running.

 

[RMerlin]

The Safari Bug is still present on this firmware on Mac OS.X 10.11.4, you are unable to change the QOS Settings Like Priority mode for example. Game to media Streaming, when you press apply, it just says applying for 5 secs and refreshes with nothing changed.

Edit: Doesn't work on Safari on iOS 9 either on the iPhone 6

Let's just say that my opinion of Safari is fairly low. Apple managed to get a perfectly well supported renderer (Webkit), and somehow managed to break things with sites that work just fine with anything else that also uses Webkit.

 

[RMerlin]

How would I go about benchmarking this? I have been using Tomato because fq_codel allows me to play games while watching netflix on my terrible 16/1 connection. So my only measurement is that BW limiter at 90% for upload and download plus fq_codel just works perfectly for me without configuring anything (unlike traditional/adaptive QOS).

One way is to run multiple streams in parallel, then monitor if your average ping time varies a lot. And also measure the latency of each streams.

It's not something easy to measure, short of going with real-life scenarios, and comparing results. And results will vary greatly, as they are related to the whole environment, including the ISP and the modem.

 

[RMerlin]

So, factory reset and no nvram-restore, then?

Restoring anything after a factory default rest mostly reverts the benefit of the factory default reset.

 

[RMerlin]

The only asus 380 firmware for this model is 380.2695, which has the front LEDs completely off, except for power. And even restarting the wireless service doesn't help to bring them back.

It's a bug in their code. They enable a feature that isn't supported by the RT-AC56U.

 

[Tom Ames]

Hm, this sounds a lot like what I was seeing, though I didn't wait as long as you before I panicked and did the factory reset.
--
bc

And both sound like what I experienced (on both Chrome and Edge), but re-starting my pc seemed to resolve it completely for me. YMMV.

 

[deathnite]

Hmm, I do not think that is the case, at least the 2.4GHz LED blinks here. The 5GHz one never blinked much in the first place. I'm satisfied they're at least lit, indicating that the device is up and running.

I don't know about 2.4GHz as I keep this radio off, but 5GHz does blink on 378.56 and doesn't blink at all on 380 builds. Tested this today.