[Official] RT-AC87U Firmware version 3.0.0.4.378.9177

[ankhazam]

Hello there @ASUS_ASUSWRT !
Thanks for the latest releases we've all been part of the creation and debug process.
Though could you please add to the issues tracker the /60 prefix DHCP-PD complaince? The router gets and ipv6 prefix from my ISP but as it is a /60 prefix. It fails to create a valid someIPV6::1/64 network from that prefix with someIPV6::2/64 gateway and I end up with no IPV6 connectivity.

 

[icemankent]

I ended up having to disable all beamforming options for both 2.4 and 5 Ghz as with them all enabled, both bands would drop connections for no reason.
Also, I still think the 5Ghz performance is not any better than the previous firmwares.
Now that beamforming is disabled I will monitor performance and if it continues to act up I will roll back to my previous firmware.

Sent from my LG G4 via Tapatalk

 

[My_dogs_White]

I am currently using Merlin's 378.55 and am also considering trying the traffic monitor (as well as beamforming).

What are the exact steps that you took to change from 378.55 to 9177? I am a newbie and I would be grateful if you can spell out the details.

Thank you.

- downloaded the file from Asus
- Used winzip to unzip the file
- Logged into router (64b windows 10 computer) and
- Went Admin, Firmware Upgrade, Browse, found the file and double clicked it
- Uploaded the file (no eithernet connection)
- Once upload was complete clicked Restore/Save/Upload setting
- Clicked Restore beside factory default
- Did the quick internet setup and manually inputted all settings

All is working well.

The only question I have has to do with the Traffic Analyzer, what is reception and transmission? It seems very obvious but... I'm showing 26.70GB reception and 21.40GB transmission? 48.10 GB total. That seems very high for one day? What could I possibly be transmitting? I'm using an Apple TV. Weird but.. all in all it's working fine.

 

[RMerlin]

RMerlin: is httpd in this firmware the latest version there is or we could upgrade it somehow?

The httpd server is a custom software, it's not a generic open source project like Apache. There's nothing to upgrade, Asus has to debug it.

 

[Andyf66]

That's what I thought it was at first but it isn't. Even a stock config it still does it (as in factory reset then making zero changes).

Probably stating the obvious, but have people tried deleting cookies/cache or connecting with a 'private' session?

I had a Draytek where the UI went absolutely crazy after every firmware update - problem was it was pulling bits of the UI from the cache and bits from the router.

Admittedly, it shouldn't cause a crash, but it might be a step to deducing what causes the crash.

 

[nagle3092]

Probably stating the obvious, but have people tried deleting cookies/cache or connecting with a 'private' session?

I had a Draytek where the UI went absolutely crazy after every firmware update - problem was it was pulling bits of the UI from the cache and bits from the router.

Admittedly, it shouldn't cause a crash, but it might be a step to deducing what causes the crash.

Tried private sessions, clearing cookies and fresh installs of different browsers.

 

[RMerlin]

I have a feeling the crash comes from a network client (part of networkmap's code lies in httpd, and that code gets run when you access the router's first index page). That would explain why not everyone is affected - a very specific network client case would trigger it. Someone would have to test with just one PC connected, and start connecting other clients until one causes the crash.

Do any of you have Apple devices on your LAN?

 

[RMerlin]

Looks like other people had the same conclusion a year ago and there is still no solution to this problem:

Look at #6 post in this forum:
http://vip.asus.com/forum/view.aspx...929050917433&board_id=11&model=RT-N56U&page=1

That post is from 2014... Totally different issue. The issue being discussed here was introduced with the recent 9000 builds.

 

[RMerlin]

Looking at the code added between 8596 and 9177, I see one case for a potential buffer overrun (as it relies on networkmap doing its own checks first).

Can someone try the httpd build from this zipfile?

http://www.lostrealm.ca/temp/httpd-ac87-a.zip

Unzip it, and copy the httpd file somewhere on your router or a USB disk. Then, run the following command (adjust /tmp/httpd to the location where you copied the new file):

service stop_httpd && mount -o bind /tmp/httpd /usr/sbin/httpd && service start_httpd

Try again logging in. If the router no longer crashes, then please check your syslog for a bunch of lines looking like this:

Oct 11 02:42:34 httpd: device len: 15 - 8/8

Post those lines. If any of them reports a len higher than 15, then that's the cause of the crash.

 

[Jong]

I have a feeling the crash comes from a network client (part of networkmap's code lies in httpd, and that code gets run when you access the router's first index page). That would explain why not everyone is affected - a very specific network client case would trigger it. Someone would have to test with just one PC connected, and start connecting other clients until one causes the crash.

Do any of you have Apple devices on your LAN?

I tried to probe on this earlier:

It seems just too much of a coincidence that 2 different devices owned by one person have the same problem if it were a fault. Surely it must be either a device on the network, something people are configuring during setup, or the way they are connecting to the router.

It seems people have tried making minimal changes after reset, so what about disconnecting all other devices during setup? Try a single wired device to a router with no other devices and using the default Asus SSID. Given this is straight after reset, this has got to work, hasn't it?!

And I got this:

FYI I did try and connect the router directly to my laptop with no other device and Wifi not configured it did not help. Part of troubleshooting any problem is isolating your network to the device giving you problems. If it was a configuration issue why does it only apply to the latest firmware for the RT87U and RT68U? I also tried default ID of ADMIN and just letters in logon. I literally worked on this issue from 3:30 to 10:30 and just gave up. I'm a Physician not a network specialist

Few other things may help. All my computers were orignally Win 7 Ultimate upgraded to Win 10. I tried both Chrome and IE/Edge and firefox. I tried reset serveal times and I also restored one computer to Win 7 using a Acronis Image I had. All other firmwares work fine on both the RT87 and RT68.

 

[mattiL]

Strange thing happened yesterday, all my clients on the network lost connection to internet, everything looked ok in the router GUI, the clients, wired and wireless, were still connected to the router.

I could connect to the router by web GUI and ssh via wireless, and the router even indicated having a WAN address and connection to Asus ddns service.

Nothing in the router log relating to this, only the hourly ntp update.

Ping inside LAN worked but not to WAN.

A reboot of the router solved the issue.

Peculiar thing though, the uptime was almost exactly 6 days, to the minute, when this happened!

 

[krychek]

That post is from 2014... Totally different issue. The issue being discussed here was introduced with the recent 9000 builds.

They're talking about their routers' web service crashing and they being end up in the static "settings have been updated" page. Which is exactly what is happening to me.
Since we don't know what's causing this issue and the symptoms are totally the same then who is to say that it's a totally different issue?

 

[ramp]

It's just information from part of the Bonjour service. Asus have said they will hide in a future version.

Hi Jong

Thanks for that. But if this error is related to Apple what is causing the lag with RDP via Citrix Receiver? This is definitely happened with 9177 and not with 5134

 

[RMerlin]

They're talking about their routers' web service crashing and they being end up in the static "settings have been updated" page. Which is exactly what is happening to me.
Since we don't know what's causing this issue and the symptoms are totally the same then who is to say that it's a totally different issue?

Because the problem appeared only following a code change in the 8000/9000 series of firmware. The firmware releases from 2 months ago do not have the issue.


Sent from my Nexus 9 using Tapatalk

 

[nagle3092]

Looking at the code added between 8596 and 9177, I see one case for a potential buffer overrun (as it relies on networkmap doing its own checks first).

Can someone try the httpd build from this zipfile?

http://www.lostrealm.ca/temp/httpd-ac87-a.zip

Unzip it, and copy the httpd file somewhere on your router or a USB disk. Then, run the following command (adjust /tmp/httpd to the location where you copied the new file):

service stop_httpd && mount -o bind /tmp/httpd /usr/sbin/httpd && service start_httpd

Try again logging in. If the router no longer crashes, then please check your syslog for a bunch of lines looking like this:

Oct 11 02:42:34 httpd: device len: 15 - 8/8

Post those lines. If any of them reports a len higher than 15, then that's the cause of the crash.

I got the output of done done when running the command, though after I cannot access the web ui at all.

 

[VANT]

You have usb device connected to te router ? Try without this device.

 

[nagle3092]

g

You have usb device connected to te router ? Try without this device.

Yes I had the device connected to the router. I've rolled back again so maybe I will try that another day.

 

[RMerlin]

I got the output of done done when running the command, though after I cannot access the web ui at all.

What do you have in the System log after that?

tail /tmp/syslog.log -n 200

 

[nagle3092]

What do you have in the System log after that?

tail /tmp/syslog.log -n 200

Pm'ed the output.

You have usb device connected to te router ? Try without this device.

No change from device connected or not.

 

[VANT]

do You reflash router after disconnect usb device ? Asus write about this in some readme files