Asus RT-AC1200G+ Random Reboots Automatically


LATEST UPDATE:

I am still stay with 9.0.0.4_382_10680 though there is an official 3.0.0.4.382_15525.

New Section:

On 2017/7/5, Unstable 5GHz Issue is Solved by Narrowing  Channel Bandwidth

On 2017/7/3, Unstable 5GHz Issue is Solved by Switching Control Channel

On 2017/7/2, Unstable 5GHz Issue

Official 3.0.0.4.380.7743 Testing by Net Friends

On 2017/6/16, No Reboot for 15 Days But Still Minor Issues

RT-AC1200G+ with RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx Stability Tracking

Follow this post to track!

I bought a new Asus RT-AC1200G+ to replace my RT-AC51U for better signal and Gigabit Ethernet. It’s much faster to copy files to/from QNAP TS-119PII now. But I notice it drops internet connection several times a day, I am sure there is nothing wrong with my tethering device as described in “Android USB Tethering Drops and Data Usage Crashes, try Reset“. All the following time is GMT + 8.

On 2017/3/21, Begin to Track Auto-Restart Issue

RT-AC1200G+ restarts several times. I cannot stand for this anymore. First, I found RT-AC1200G+不定期重啟 in local community group has report the same issue and many others could be googled by “rt-ac1200g+ restarting”. Then I decide to upgrade firmware to  and see if it fix the bug.

All available firmware are on RT-AC1200G+ Driver & Tools. and. I download the latest 3.0.0.4.380.4089 and upload via [Administration] → [Firmware Upgrade] tab. Flash and reboot automatically.

On 2017/4/2, Restarts at 20:54

No, the latest firmware doesn’t solve the auto-reboot issue.

On 2017/4/4, Restarts at 20:54

Because both restarts at the same time, I search for any scheduler-like configurations and disable Enable wireless scheduler on 2.4GHz and 5GHz in [Advanced Settings] → [Wireless] → [Professional] tab.

On 2017/4/10, Reboot at 23:17

[Problem]Router keeps rebooting randomly on Asus forum suggests disabling GRO (Generic receive offload) due to its stability issue which is not available in current firmware. According to Router Guide: NAT Acceleration On or Off, I decide to Disable NAT Acceleration (default Auto) in [Advanced Settings] → [LAN] → [Switch Control] tab.

On 2017/4/11,Fine-tune Mobile Connection

I notice USB computer connection is in Media device (MTP) mode which could identified as a storage device. I decide to manually select Charging mode in the future. The default is Media device (MTP) which I cannot change on this device.

Maybe that’s why I see disk_monitor: Got SIGALRM…. This log is explained in Asuswrt-Merlin 378.54_2 is now available on SNBForums by RMerlin as:

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.

More about Android file transfer is explained in Transfer files between your computer & Android device.

On 2017/4/13, Reboot at 14:41

Another suspect is Enable UPnP Media Server has been turned on in [General] → [USB Application] → [Media Server] tab. It seems disabling NAT acceleration is not enough, I decide to turn it off.

DD-WRT for RT-AC1200G+ is still WIP (working in progress) according to Supported Devices 3.13 Asus on DD-WRT.com. I am not sure what I should turn off next time.

On 2017/4/17, Wireless becomes Unstable and Need a Restart at 23:57

My network becomes unstable and need a restart at 23:57 (GMT+8). I am not sure if it is due to RT-AC1200G+. I use D-Link DWA-171 rev A1 on my HP ProBook 4520s. It reconnect several times today. And Mac mini (Late 2012) also the same around 23:00 at night. Both use 5Ghz. This never happen when download Visual Studio 2017 Community during 2017/4/13 to 4/20. RT-AC1200G+ works okay without any restart.

I notice a new warning “Connection reset by peer” related to Universal Plug and Play begins on 2017/4/17. You may learn more about this in Introduction to UPnP.

According to What does “connection reset by peer” mean? on StackOverflow, it is related to an immediate dropping of the connection. I think it is related to my new download. I use the installation program on SQL Server downloads to download for offline installation. The port 2869 is used by Simple Service Discovery Protocol (SSDP) which is a Windows service for event notification explained on Network Ports Used by Key Microsoft Server Products.

Apr 17 15:50:32 miniupnpd[12235]: upnp_event_send: send(): Connection reset by peer
Apr 17 15:50:32 miniupnpd[12235]: upnp_event_process_notify: connect(192.168.1.122:2869): Connection reset by peer
Apr 17 15:50:32 miniupnpd[12235]: upnp_event_process_notify: connect(192.168.1.122:2869): Connection reset by peer

On 2017/4/18, Fine-tune UPnP, Wireless Configurations, Reset Router

My SQL Server installation program got aborted and need manual restarts several times. Because RT-AC1200G+ generates many “Connection reset by peer” in system log, I decide to re-enable Enable UPnP in [Advanced Settings] → [WAN] → [Internet Connection] tab. This messages are gone after applying but the wireless connection is still unstable.

After more testing, I realize it’s related to 5GHz only. Download speed also gets huge improve when switching to 2.4GHz.

Asus suggest to use channel 149 and 161 on 5GHz in RT-AC58U,RT-AC1300+ 這路由器Wifi 5G. For users in Taiwan, you may use following channels legally according to 電腦王:Wi-Fi 5GHz 你支援嗎?產品支援度混亂,小心頻率不相容:

  1. 5250~5350MHz(CH52 5260MHz、CH56 5280MHz、CH60 5300MHz、CH64 5320MHz)
  2. 5470~5725MHz(CH100 5500MHz、CH104 5520MHz、CH108 5540MHz、CH112 5560MHz、CH116 5580MHz、CH120 5600MHz、CH124 5620MHz、CH128 5640MHz、CH132 5660MHz、CH136 5680MHz、CH140 5700MHz)
  3. 5725~5825MHz(CH149 5745MHz、CH153 5765MHz、CH157 5785MHz、CH161 5805MHz、CH165 5825MHz)

On an informal test, download a LAMP Stack image from TurnKey Linux is around 100 to 200 kbps on 5GHz (802.11ac 1T1R @ 433Mbps with all available 5GHz channels (149, 153, 157, 161, 165)) while 50Mbps on 2.4GHz (802.11n 1T1R 150Mbps). Both with WPA2-Personal enabled and test about the same time.

Conclusion: my RT-AC1200G+ is not only has auto-restart issue but also slow and unstable 5GHz.

I decide to reset everything. Anyway, they already suggest a reset after firmware upgrade which I have never done before on RT-AC1200G常常斷線。WAN Connection: DNS probe failed. It’s said to push the reset button and hold for 10 to 15 seconds while the device is power on. Don’t release until all lights blinking. Then all settings will be reset to factory default. Backup configurations before this reset. Detail with illustration can be found in How to do a firmware update on ASUS wireless router? (ASUSWRT).

Sounds easy but this is what I get after first reset: a firmware upload page! If you see this page, just restart again. It should go back to factory default.

I get a firmware upload page after resetting my device.

Here is a list of my configurations:

  1. In [USB Application] → [Media Services and Servers], disable [Media Server] and [network Place (Samba) Share] which is default on.
  2. In [USB Application] → [3G/4G], Android tethering is identified as USB Modem. Leave it un-touched.
  3. In [Wireless], turn off WPS and set 5GHz with 80MHz and 2.4GHz with 40MHz channel bandwidth. I also assign 161 as the control channel in 5GHz because 149 has conflicts in my house.
  4. In [WAN], UPnP is turned on by default.
  5. In [Firewall], Network Service Filter is turned off by default. But I uncheck all data and apply again.
  6. In [Administration] → [System], Reboot Scheduler is turned off by default.

On 2017/4/19, Fine-tune 5GHz Wireless Configurations

I try to fix the unstable and slow issue on 5GHz today.

First, I check more detail about RT-AC1200G+ and found more information in ASUS RT-AC1200GP on WikiDevi.

  1. Broadcom BCM47189 supports 802.11ac802.11n, and 802.11a @ 5GHz.
  2. Broadcom BCM43217 supports 802.11n802.11g, and 802.11b @ 2.4GHz.

Because my wireless devices are either 802.11ac or 802.11n, I decide to change the 2.4GHz Wireless Mode to [N Only] and 5GHz to [N/AC Mixed] in [Advanced Settings] → [Wireless] → [General] tab. It provides best performance by disabling support for legacy standard.

I also change the 2.4GHz Channel bandwidth to [40MHz] and 5GHz to [80MHz] in the same tab but encounter some stability issue.

On 2017/4/20, Fine-tune 5GHz Wireless Configurations

to continue my experiment. When I set Channel bandwidth to [80MHz] with a fix Control Channel, network is slow and unstable. Even worse, sometimes I cannot see my SSID at 5GHz. I conclude there are conflicts in this combination. Maybe there are too many devices ask for 80MHz. The best solution is to lower bandwidth for a more stable wireless network.

Then I lower the Channel bandwidth to [40MHz] and Control Channel to [Auto] to . Now it works. Fast and stable! One tablet failed to see the SSID at 5GHz is solved by re-enable its WiFi.

I also learn wireless network from following pages to understand Wireless Log in [Advanced Settings] → [System Log] → [Wireless Log] tab.

  1. Best Wireless Routers Now: 20MHz vs 40MHz vs 80MHz vs 160MHz: What Channel Bandwidth Should I Use?
  2. metageek: Understanding WiFi Signal Strength
  3. Oreilly: Chapter 2. The PHY
  4. Wiki: dBm
  5. ZDNet: Gigabit Wi-Fi: 802.11ac is here: Five things you need to know

Here is what I have learned:

  1. When dBm is positive, it’s gain. Negative means loss signal strength.
  2. -80 dBm or less means unreliable.
  3. 80MHz channel bandwidth is an available option to 802.11ac but needs a clear channel range.

When I have to lower the Channel bandwidth to [40MHz] to get a clear channel range for 802.11ac, my connection speed drops from 433Mbps to 200MBps with -55 to -72 dBm on 5GHz. But it downloads a LAMP Stack image from TurnKey Linux at about 4Mbps on 5GHz and 2.4GHz, much improve than 2017/4/18.

BTW, I also add 8.8.8.8 as DNS Server in [Advanced Settings] → [LAN] → [DHCP Server] tab or it will fail to query sometimes.

Therefore, I think this case is closed. Now I have to wait and see if the restart happen again in the next few days.

On 2017/4/24, Reboot at 21:20 and Again at 21:28

As you can see below, time has been first reset after 23:00 and 08:08 again. The later also reset Uptime in [General Log]:

Apr 23 17:03:34 kernel: plcp 000b 0180 || 500f 8f1f 0000 || 7d4f 7fc1 || rxestats2 0000
Apr 23 17:03:34 kernel: txerr_tm 0021 txerr_txbytes 0000 txfunfl 0 8 0 0 0 0 txtplunfl 0
Apr 23 17:03:34 kernel: debug1 7fc1 debug2 0000 debug3 0000 debug4 0000
Apr 23 23:00:28 disk_monitor: Got SIGALRM...
Aug 1 08:00:09 syslogd started: BusyBox v1.17.4
Aug 1 08:00:09 kernel: klogd started: BusyBox v1.17.4 (2016-11-03 15:46:45 CST)
Aug 1 08:00:09 kernel: Linux version 2.6.36.4brcmarm (root@asus) (gcc version 4.5.3 (Buildroot 2012.02) ) #1 PREEMPT Thu Nov 3 15:55:43 CST 2016
...skip...
Aug 1 08:01:17 miniupnpd[426]: add_filter_rule() : chain FUPNP not found
Aug 1 08:07:55 rc_service: httpd 389:notify_rc reboot
Aug 1 08:07:56 kernel: Attempt to kill tasklet from interrupt
Aug 1 08:07:57 iTunes: daemon is stoped
Aug 1 08:07:57 FTP Server: daemon is stoped
Aug 1 08:07:57 Samba Server: smb daemon is stoped
Aug 1 08:07:57 kernel: gro disabled
Aug 1 08:00:09 syslogd started: BusyBox v1.17.4
Aug 1 08:00:09 kernel: klogd started: BusyBox v1.17.4 (2016-11-03 15:46:45 CST)

I notice the difference because Uptime in [General] → [System Log] → [General Log] tab has been reset. Usually it should keep counting from the time RT-AC1200G+ being turned on. And there is a lot of warning related to MiniUPnP, too.

Is there any relation because I switch to another Android device for tethering? I don’t know. But I decide to change Select USB device in [USB Application] → [3G/4G] from USB Modem back to Android phone to mach current scenario, it was identified as a USB Modem with another Android device on 2017/4/18.

On 2017/4/26, New Leads

I notice some suspicious message in system log.

Apr 26 20:14:58 rc_service: httpd 389:notify_rc restart_wireless
Apr 26 20:15:03 kernel: wl_module_init: passivemode set to 0x0
Apr 26 20:15:03 kernel: wl_module_init: txworkq set to 0x0
Apr 26 20:15:04 kernel: eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
Apr 26 20:15:04 kernel: External imprecise Data abort at addr=0x2ac8a3b0, fsr=0x1c06, pc=0x2ac6de20 lr=0x9eb0 ignored.
Apr 26 20:15:04 kernel: eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
Apr 26 20:21:22 kernel: txerr valid (1) reason 0100 tst 2222 pctls 1cc2 1c0a 0022
Apr 26 20:21:22 kernel: plcp 000b 03ff || dc07 8f2a 0000 || 74c1 77ad || rxestats2 0000
Apr 26 20:21:22 kernel: txerr_tm 0033 txerr_txbytes 0000 txfunfl 0 1 0 0 0 0 txtplunfl 0
Apr 26 20:21:22 kernel: debug1 77ae debug2 0000 debug3 0000 debug4 0000

notify_rc restart_wireless means wireless devices are restart. I wonder if this is the reason for my restart issue. Both of them have been restart without interrupt my internet connection but no reboot.

But on QNAP TS-119PII with Different Wireless Device on Mac OSX File Transfer Performance Report, I have check the internal of RT-AC1200G+ as below:

According to ASUS RT-AC1200GP on WikiDevi, some important features include:
1. Broadcom BCM47189 supports 802.11ac802.11n, and 802.11a@ 5GHz.
2. Broadcom BCM43217 supports 802.11n802.11g, and 802.11b @ 2.4GHz.
3. Max speed for 802.11n is 300Mbps with 2T2R.
4. Max speed for 802.11ac is 867Mbps with 2T2R.
5. Gigabit Ethernet.

There is no BCM43227 which supports 802.11n802.11g, and 802.11b @ 2.4GHz and BCM43c8 which I cannot find any related information.

The last four repeat from 20:21:22 to 21:56:40. I have no ideas at all.

It’s a strange log and I will keep my eyes on them.

Around 2017/5/5, Unstable 5GHz without Reboot

I notice my 802.11n @ 5GHz becomes unstable because connections break and reconnect within 2 minutes.

I cannot remember the exact date and time. Just around 2017/5/5.

I also notice there is a lot of similar kernel message from 10:46:35 to 12:05:16:

May  5 10:46:35 kernel: txerr valid (1) reason 0100 tst 2222 pctls 3cc2 5c1a 0022
May  5 10:46:35 kernel: plcp 000b 0180 || 5e0f 8f1f 0000 || 92b4 9472 || rxestats2 0000
May  5 10:46:35 kernel: txerr_tm 0030 txerr_txbytes 0000 txfunfl 0 1 0 0 0 0 txtplunfl 0
May  5 10:46:35 kernel: debug1 9472 debug2 0000 debug3 0000 debug4 0000

On 2017/5/7, Unstable 5GHz without Reboot

It’s the same to 2017/5/5. My 5GHz wireless connections break and reconnect within 2 minutes at about 23:35.

It’s the same kernel message again:

May  7 23:37:11 kernel: txerr valid (1) reason 0100 tst 2222 pctls 3cc2 5c1b 0022
May  7 23:37:11 kernel: plcp 000b 0063 || 430f 8f07 0000 || a457 a623 || rxestats2 0000
May  7 23:37:11 kernel: txerr_tm 0020 txerr_txbytes 0000 txfunfl 0 1 0 0 0 0 txtplunfl 0

On 2017/5/8, Reboot at 10:16

Finally, it reboots after 14 days. The Uptime in [General Log] has been reset.

May  8 09:20:07 kernel: txerr valid (1) reason 0100 tst 2222 pctls 1cc2 1c13 0022
May  8 09:20:07 kernel: plcp 000b 00e1 || 1b07 9f09 0000 || 3c70 3e22 || rxestats2 0000
May  8 09:20:07 kernel: txerr_tm 003a txerr_txbytes 0000 txfunfl 0 1 0 0 0 0 txtplunfl 0
May  8 09:20:07 kernel: debug1 3e22 debug2 0000 debug3 0000 debug4 0000
...skip...
May  8 09:24:15 kernel: txerr valid (1) reason 0100 tst 2222 pctls 1cc2 1c13 0022
May  8 09:24:15 kernel: plcp 000b 039f || cd07 9f26 0000 || bd6e be48 || rxestats2 0000
May  8 09:24:15 kernel: txerr_tm 003b txerr_txbytes 0000 txfunfl 0 14 0 0 0 0 txtplunfl 0
May  8 09:24:15 kernel: debug1 be48 debug2 0000 debug3 0000 debug4 0000
May  8 10:00:20 disk_monitor: Got SIGALRM...
Aug  1 08:00:09 syslogd started: BusyBox v1.17.4

PawelW from Poland share a new beta firmware RT-AC1200G+_9.0.0.4_380_7415-gc949b277fca.trx which works great on his machine in the comments of this post.

I share this good news to Asus Forum with subject “哪裡可以下載 RT-AC1200G+ 台灣新 Beta 韌體 380_7415?” and ask for local beta firmware because every country has its public channel sets. Asus reply there is no official link for new beta firmware.

小昱 who read my post and decide to give it a try. This firmware works great for him. Everything seems okay now.

Okay, now we have a new beta firmware. There are two people go for it. I decide to go another way. See if I may find tune my configuration. I found many interested information today.

On Asus RT-AC1200G, BCM47189+ BCM43217 chipset on OpenWrt, I learn new to commends to explore RT-AC1200G+ through ssh.

  1. Enable SSH Daemon in [Advanced Settings] → [Administration] → [System] tab for [LAN only]. If you want to access from WAN, read Enable SSH on Asus Routers with(out) SSH Keys on HTPC Guides
  2. Use dmesg to display kernel messages. I notice same eth1 and eth2 messages on 2017/4/26. And also many error.
  3. Use nvram show to display configuration in flash.
  4. Use uname -r to get Linux version 2.6.36.4brcmarm on RT-AC1200G+.

I also learn following facts from the thread:

  1. Broadcom BCM47189 is similar to BCM53573 with 100MHz higher frequency and both are ARM Cortex A architecture. They are both identified as arch/arm/mach-bcm53xx as a Broadcom SoC with ARM Cortex A9 CPU in OpenWRT kernel.
  2. changeway believe A7 is similar to A9 arch, but need to patch some driver to support BCM47189 and ARM Cortex A7.
  3. changeway believe the wireless may support by b43.

Then I read Broadcom BCM47xx on OpenWrt to learn more about the Linux wireless driver, it seems b43 is used to support BCM47189 @ 5GHz according to this. When I follow the lead to read b43 on Linux Wireless, I found BCM43217 @ 2.4GHz is on the supported list but no BCM47189 @ 5GHz. And there is a know problem looks similar to my wireless issue on 2017/4/17 :

Scanning/connectivity problems: caused by b43 bug exposed with kernel 3.8 (by 0172bb7). Fixed in kernel 3.14 (by 64e5acb) and backported to 3.13.2, 3.12.10, 3.10.29.

I think maybe the kernel message and unstable 5GHz is related to support for BCM47189 @ 5GHz in wireless driver.

I decide to drop the support for 802.11ac by setting Wireless Mode to [N only] in [Advanced Settings] → [Wireless] → [General] tab.

On 2017/5/10, Reboot at 13:15

The Uptime in [General Log] has been reset.

I drop 5GHz by setting Enable Radio on  5GHz to [No] in [Advanced Settings] → [Wireless] → [Professional] tab. Then I keep on receiving similar messages going on all day:

May 10 13:24:40 kernel: txerr valid (1) reason 0100 tst 2222 pctls 3cc2 5c1b 0025
May 10 13:24:40 kernel: plcp 000b 006f || 3f0f 8f08 0000 || 722b 7288 || rxestats2 0000
May 10 13:24:40 kernel: txerr_tm 001c txerr_txbytes 0000 txfunfl 0 1 0 0 0 0 txtplunfl 0
May 10 13:24:40 kernel: debug1 7288 debug2 0000 debug3 0000 debug4 0000

txerr should be Transmit Error but I have no ideas why.

Once I drop 2.4GHz by setting Enable Radio on  2.4GHz to [No] in [Advanced Settings] → [Wireless] → [Professional] tab, all above messages stops.

Anyway, I will start by disabling 5GHz. If it doesn’t work, I will stop 2.4GHz and re-enable 5GHz next time.

I also get a respond from Hong Kong local community. fatdog suggest to use firmware 3.0.0.4.380_3971-g516a6d6 and schedule reboot every day in Asus RT-AC1200G+不定期重啟 #11. He has no problems at all. I will give it a try after above option. (Update on 2017/5/13: due to 40 Asus RT Router Models Are Vulnerable to Simple Hacks on BleepingComputer, I will give up this suggestion).

Here I got two bad messages. PawelW reports that the beta firmware RT-AC1200G + _9.0.0.4_380_7415-gc949b277fca.trx eventually crash after 19 days. nikisi on Asus suppport forum also reports in [Sharing]A diary to solve Asus Rt-AC1200G+ reboot issue that the link to beta firmware is no longer available.  (Update on 2017/5/12: PawelW provides a new link to download in his new reply)

On 2017/5/12, Freeze at 13:13

I see warning message continuously generated after disabling 5GHz from 2017/5/10. Finally it freeze today. For those already connected, they won’t be able to access internet. Others won’t see the SSID. I have to restart RT-AC1200G+ with its power button.

May 12 11:22:19 kernel: txerr valid (1) reason 0100 tst 2222 pctls 3cc2 5c1b 0022
May 12 11:22:19 kernel: plcp 000b 0474 || 6b0f 8f5f 0000 || 0581 5dc2 || rxestats2 0000
May 12 11:22:19 kernel: txerr_tm 001c txerr_txbytes 0000 txfunfl 0 248 1 0 0 0 txtplunfl 0
May 12 11:22:19 kernel: debug1 5dc2 debug2 0000 debug3 0000 debug4 0000

Now, I have no disable 2.4GHz and re-enable 5GHz in [Advanced Settings] → [Wireless] → [Professional] tab.

On 2017/5/13, Change Network Topology

Some of my device doesn’t support 5GHz. Therefore I disable 5GHz and DHCP of RT-AC51U. Use LAN to LAN to connect RT-AC51U to RT-AC1200G+. The former will be responsible for 2.4GHz while the later is 5GHz. How to Connect Two Routers on wikiHow has great illustration and explain in plain English.

Both of my NAS are connected to the Gigabit Ethernet of RT-AC1200G+ while Mac mini is 802.11n @ 5GHz with 2T2R at 300Mbps. Copying a single 3.87Gb file is about 25 MB/s. Only those connected to RT-AC51U will be limited by the slow Fast Ethernet.

I may print to my wireless printer which only has 2.4GHz capability through this LAN to LAN topology.

On 2017/5/15, Reboot at 10:12 But Reconnect WAN Automatically

It’s a different experience. I notice my internet connection has been dropped. Mac mini detects no 5GHz from RT-AC1200G+ and switch to 2.4GHz on RT-AC51U but fails. After a while, I may connect to 2.4GHz and login to RT-AC1200G+. I notice the Uptime in [General Log] has been reset. Strangely, WAN was restored automatically. No need to reset tethering on Android mobile.

May 15 10:00:34 disk_monitor: Got SIGALRM...
Aug 1 08:00:09 syslogd started: BusyBox v1.17.4
...skip...
Aug 1 08:01:07 WAN Connection: WAN was restored.
Aug 1 08:01:09 ntp: start NTP update
May 15 10:13:26 rc_service: ntp 783:notify_rc restart_upnp

Conclusion

Even turning off wireless features on RT-AC1200G+, the best is to restore WAN automatically. There is no way to stop the reboot.

1st RMA: From 2017/5/15 till Now

On 2017/5/15, I contact Asus and get a RMA (return merchandise authorization) code. Print this post together with my RT-AC1200G+ to send back to Asus at 15:30.

On 2017/5/16, a tech support form Kaohsiung call me at 17:37. He ask me several questions which I have already provide the detail in my post. Maybe it’s the routine conversation or he doesn’t have time to read this long report. At the end, he decide to send me a new replacement and upgrade to the latest firmware.

On 2017/5/17, the same tech support contact me that a research and development engineer will investigate my case. They need to keep my RT-AC1200G+ and there will be no replacement during that period.

It’s a good news to me because a replacement won’t solve the reboot issue which I have learned from many others before.

On 2017/5/19, the same tech support contact me that he cannot reproduce reboot with GbE. I instruct him to reproduce reboot in 3 days by:

  1. Enable 2.4GHz and WPA2. Errors may be found in system log immediately.
  2. Enable 5GHz with 80MHz bandwidth and WPA2. Wireless network becomes unstable within 3 days.
  3. Use Google Drive to monitor if it’s still connected to Internet.

You have to be a good tech support to instruct them step-by-step with patient and love when a tech support calls. Don’t forget to summarize at the end or they will forget everything. Most people won’t read the notes even it is in printed material.

On 2017/5/22, the same tech support contact me that he finally reproduce a reboot on 5GHz in the weekend. This is a great news and I can see how hard he work on this issue. Because the reboot has been confirmed by tech support, it’s officially confirmed my RT-AC1200G+ has random reboot. Now research and development engineer finally may investigate my issue.

On 2017/5/23, the same tech support contact me that THE RESEARCH AND DEVELOPMENT ENGINEER REFUSED TO CHECK UNLESS I PROVED VIDEO OF RT-AC1200G+ AND SCREEN RECORDING ON ALL MY DEVICE AND NAS AT THE CRASH POINT TO PROOF IT’S NOT WORKING.

I told the tech support it’s impossible to provide screen recording because it against security and governance. I may setup a video camera on the router and clip only the crash point and upload to YouTube.

The only way to stop is to return you device for refund or repair within warranty.

I will continue to fight for a stable machine! Put your RMA image link here and join!

On 2017/5/24, I setup a Surveillance System to Proof Crash of Asus RT-AC1200G+. You  may follow my steps to create your own video. Upload to YouTube and share with communities. I will share more tips to pick up the crash point when I get back my RT-AC1200G+.

BTW, local community replied they will ask customers to record videos but not the case which Asus has reproduce in office.

On 2017/5/25, I got a call from a senior tech support engineer at 11:17. He offer an Asus RT-AC66U as a replacement. He also explains they only reproduce 5GHz restart successfully but not other reboot. I ask if I may accept the offer but keep my RT-AC1200G+ and continue working with them to find a SOP to reboot.

He calls again at 13:35PM and say if I accept the offer, I have to return my RT-AC1200G+. Therefore, I refuse the offer because I am more interested in HELPING PEOPLE IN THE WORLD THAN A FREE UPGRADE. We both agree to close the case. He will pass my contact information to the research and development engineer. If he needs my help to collect more information at the crash point, he will contact me.

If you just want to get a stable machine, show my post and highlight what happen today to Asus local support team and see if you could get another model. Leave me a message with a photo to proof if you get one to help others asking for another model.

I will continue working on this issue. I am both a CRM independent consultant and software developer. I understand how hard it could be to debug, especially random errors. Therefore, I would like to establish a form/guideline/framework to capture. Customers could provide better feedback and engineers may capture what happens more preciously. We may work together to solve problems.

Eventually, I get a different RT-AC1200G+ in the 1st RMA.

On 2017/5/25, Prepare to Capture Next Reboot

Upgrade to Latest Firmware

On 2017/5/25, I got my replaced RT-AC1200G+ and decide to upgrade to the latest firmware. It comes with 3.0.0.4.380_3865 but only find 3.0.0.4.380_3971 the latest. I upgrade to what it found the latest and manually upload 3.0.0.4_380_4089.

Then push the reset button for 5 seconds to restore to factory default. How to do a firmware update on ASUS wireless router? (ASUSWRT) has detail step-by-step instructions.

Minima Configuration

Login and setup. To keep most configuration as default. I made following changes:

  1. Change login password
  2. Change SSID and password for 2.4GHz and 5GHz.
  3. Connect the LAN on RT-AC51U to WAN on RT-AC1200G+.
  4. Enable MAC filter for 2.4GHz and 5GHz.

Now, I connect my Mac mini (Late 2012) via 5GHz, pull the cable of Asustor AS-5102T and QNAP TS-251A on RT-AC51U and push to RT-AC1200G+.

Reboot

Everything is okay until I turn on Asustor AS-5102T and QNAP TS-251A, hola! Reboot, again! I notice it at 17:46 which is about 4 minutes after unplug from RT-AC51U. Yes, RT-AC1200G+ is fine but RT-AC51U reboot and Uptime has been reset.

May 25 15:29:10 kernel: MT7610ap:ASSOC - Assign AID=2 to STA a8:8e:24:a4:57:dd
May 25 16:46:55 kernel: MT7610ap:ASSOC - Assign AID=1 to STA ec:22:80:7e:03:19
May 25 17:19:22 kernel: MT7610ap:ASSOC - Assign AID=2 to STA a8:8e:24:a4:57:dd
May 25 17:26:00 kernel: MT7610ap:ASSOC - Assign AID=2 to STA a8:8e:24:a4:57:dd
May 25 17:42:28 kernel: ESW: Link Status Changed - LAN4 (P4) Link DOWN
May 25 17:42:29 kernel: ESW: Link Status Changed - LAN3 (P3) Link DOWN
Aug 1 08:00:14 syslogd started: BusyBox v1.17.4
Aug 1 08:00:14 kernel: klogd started: BusyBox v1.17.4 (2017-03-16 04:33:04 CST)
Aug 1 08:00:14 kernel: Linux version 2.6.36 (root@asus) (gcc version 4.2.4) #1 Thu Mar 16 04:36:47 C
...skip...
Aug 1 08:00:35 kernel: MT7610ap:ASSOC - Assign AID=1 to STA ec:22:80:7e:03:19
Aug 1 08:02:04 kernel: MT7610ap:ASSOC - Assign AID=2 to STA a8:8e:24:a4:57:dd
...skip...
Aug  1 08:03:42 WAN Connection: WAN was restored.
Aug  1 08:03:45 ntp: start NTP update
May 25 17:46:28 rc_service: ntp 559:notify_rc restart_upnp
May 25 17:46:28 miniupnpd[558]: shutting down MiniUPnPd
May 25 17:46:28 miniupnpd[642]: version 1.9 started
May 25 17:46:28 miniupnpd[642]: HTTP listening on port 42515
May 25 17:46:28 miniupnpd[642]: Listening for NAT-PMP/PCP traffic on port 5351
May 25 17:46:43 crond[194]: time disparity of 955303 minutes detected

Now, I shutdown AS-5102T and TS-251A and turn on both again when they are connected to AS-5102T.

Okay, it’s not because of power-on. I restore the condition by pull the cable of Asustor AS-5102T and QNAP TS-251A on RT-AC1200G+ and push to RT-AC51U. Then pull the cable of Asustor AS-5102T and QNAP TS-251A on RT-AC51U and push to RT-AC1200G+. Everything is fine.

May 25 18:14:51 kernel: ESW: Link Status Changed - LAN4 (P4) Link UP
May 25 18:14:54 kernel: ESW: Link Status Changed - LAN3 (P3) Link UP
May 25 18:15:39 kernel: ESW: Link Status Changed - LAN4 (P4) Link DOWN
May 25 18:15:39 kernel: ESW: Link Status Changed - LAN3 (P3) Link DOWN

Now, turn on Asustor AS-5102T and QNAP TS-251A, safe. I repeat the same procedures for 5 times but there is no reboot. Why RT-AC51U reboots? It never happens. Why I cannot reproduce reboot? I have no ideas and cannot reproduce.

Minima Configuration, Continue

Because I cannot reproduce the reboot, I continue finish configuration:

  1. Set LAN IP as 192.168.120.1
  2. Enable MAC and IP address Binding on AS-5102T and TS-251A

What’s the Difference

Compare to previous tests, I give up USB features. RT-AC1200G+ will get internet access from RT-AC51U via WAN.

I will begin to record when it reboot next time.

On 2017/5/26, Research on syslog.log, Change Network Map, Testing Cooling System, and an Unstable 5GHz Issue

Research on syslog.log

On 2017/5/26, I enable ssh service on RT-AC1200G+ and found two syslog.log:

admin@(none):/tmp# find / -iname syslog.log
/jffs/syslog.log
/tmp/syslog.log
admin@(none):/tmp# ls -l /jffs/syslog.log 
-rw-rw-rw- 1 admin root 24675 May 26 13:21 /jffs/syslog.log
admin@(none):/tmp# ls -l /tmp/syslog.log 
-rw-rw-rw- 1 admin root 24675 May 26 13:21 /tmp/syslog.log

RMerlin explain the different of them in “Where is the ASUS RT-AC68U storing all it’s logs?” on SNBForum:

/tmp/syslog.log is the main, “live” log.

/jffs/syslog.log is a backup made so that when your router reboots, it won’t lose all past log entries (/tmp is a volatile ramdisk).

If it generate any log before it crash, it’s gone with /tmp/syslog.log once reboot. josh.zyman notice /jffs/syslog.log is update every minute.

FATMIN’s Configure Syslog Logging Levels on the Asus RT-AC66U Router is very useful if you want to get more logs. I follow his instruction to get current log_level and set new parameters. Unlike the comments to the post, it didn’t reset to default when I reboot my RT-AC1200G+:

admin@(none):/tmp# nvram show | grep log_level
size: 40172 bytes (25364 left)
log_level=6
admin@(none):/tmp# nvram set log_level=2
admin@(none):/tmp# nvram commit

RMerlin replied in “Default message log level” & “Log only messages more urgent than” in Asuswrt-Merlin 378.56 on SNBForums, there is no difference of this parameter because Asuswrt just ignore it:

Default message log level is the level used by the router’s own logging. Unfortunately, Asuswrt uses the same level for all of the log entries it generates itself, rather than distinguishing between notices, and critical errors when logging. This can be problematic for people using a centralized syslog daemon, as by default everything carried a priority of emergency, making it impossible to filter things. This setting allows you to assign a more sane default level so not everything from the firmware comes in as being an emergency.

The various daemons and services (dnsmasq, miniupnpd, etc…) will typically use an appropriate level when generating log entries.

Log only messages more urgent than is the default filter level for the router’s own syslog (what is shown on the System Log page). If set to “All”, then everything sent to the router’s syslog gets shown. If set to a different level, then you can ensure that, for example, only errors get shown (once again, see the limitation in the previous paragraph regarding the firmware’s own log entries).

Besides, RMerl replied in “spontaneous reboot on RT-AC87U v.380.59 #941” on GitHub:

The router will reboot whenever the kernel or the init process crashes. The only way to debug these is through a serial cable – a crashed kernel is unable to write to syslog.

Conclusion: syslog.log won’t help to collect any useful information on crash point because nothing will be written to it. You need a serial hook to the router.

Change Network Map

In Setup a Surveillance System to Proof Crash of Asus RT-AC1200G+, I use D-Link DCS-930L Wireless N Network Camera to capture videos and stored to AS-5102T. Therefore, I cannot connect AS-5102T to RT-AC1200G+. One it crash, DCS-930L won’t be able to save video stream to AS-5102T.

Here is my modified network map:

20170526 Network Map to Capture Crash Moment of Asus RT-AC1200G+
20170526 Network Map to Capture Crash Moment of Asus RT-AC1200G+

As you may see, DCS-930L will save video stream to AS-5102T through RT-AC51U even RT-AC1200G+ is crash. If RT-AC1200G+ has reboot, I should get a log in RT-AC51U because the Fast Ethernet has been reconnect or just go offline. Here is a sample log in RT-AC51U when RT-AC1200G+ is connected to RT-AC51U:

May 27 10:42:38 kernel: ESW: Link Status Changed - LAN1 (P1) Link UP
May 27 10:43:10 kernel: ESW: Link Status Changed - LAN1 (P1) Link DOWN
May 27 10:43:12 kernel: ESW: Link Status Changed - LAN1 (P1) Link UP

With the help of RT-AC51U, we may double check when RT-AC1200G+ has reboot.

A Cooling System Won’t Make It Stable

I got so many suggestion that RT-AC1200G+ is so hot that you MUST need a cooling system to help it stable. Okay, let’s try. I put RT-AC1200G+ on top of 友情牌 10 吋壁掛循環扇 KG-1064 which is a Vornado-like fan. yes, it’s c0ol. No, RT-AC1200G+‘s 5GHz wireless becomes unstable between 23:03 to 23:21 on 2017/5/26.

Fact: a cooling system won’t help your RT-AC1200G+ becomes more stable.

Asus RT-AC1200G+ Cooled by a Vornado-like 10" Fan at the bottom.jpg
Asus RT-AC1200G+ Cooled by a Vornado-like 10″ Fan at the bottom.jpg

Unstable 5GHz Issue

At 23:03, I switch connection to RT-AC51U from RT-AC1200G+. There is a log in RT-AC51U to proof:

May 26 23:03:29 kernel: MT7610ap:ASSOC - Assign AID=2 to STA a8:8e:24:a4:57:dd

At 23:19, I switch connection to RT-AC1200G+ from RT-AC51U but fail. I cannot connect – something wrong. I reconnect to RT-AC51U and make sure video has been kept. There is a log in RT-AC51U to proof:

May 26 23:19:41 kernel: MT7610ap:ASSOC - Assign AID=2 to STA a8:8e:24:a4:57:dd

Video captures of lights on RT-AC1200G+ at 23:19 on 2017/5/26:

At 23:20, I switch connect to RT-AC1200G+ from RT-AC51U again. I cannot connect. Here is a screen capture to proof:

Asus RT-AC1200G+'s 5GHz is gone
Asus RT-AC1200G+’s 5GHz is gone

Video captures of lights on RT-AC1200G+ at 23:20 on 2017/5/26:

Video captures of lights on RT-AC1200G+ at 23:21 on 2017/5/26:

In the next few minutes, I may connect to RT-AC1200G+ but it’s not stable. Sometimes I cannot browse any web site.

At 23:31, I decide to ssh login to RT-AC1200G+ and execute reboot command. There is log in RT-AC1200G+ to proof but I didn’t see any log in RT-AC51U:

May 26 23:00:59 disk_monitor: Got SIGALRM...
May 26 23:31:14 dropbear[9529]: Password auth succeeded for 'admin' from 192.168.120.91:61508
Aug  1 08:00:09 syslogd started: BusyBox v1.17.4

At 23:37, I decide to power reset RT-AC1200G+. There is log in RT-AC1200G+ and RT-AC51U to proof. Here is the log in RT-AC1200G+ :

May 26 23:31:14 dropbear[9529]: Password auth succeeded for 'admin' from 192.168.120.91:61508
Aug 1 08:00:09 syslogd started: BusyBox v1.17.4
...skip...
Aug 1 08:00:26 dhcp client: bound 192.168.1.120 via 192.168.1.51 during 86400 seconds.
Aug 1 08:00:27 ntp: start NTP update
May 26 23:38:19 rc_service: ntp 594:notify_rc restart_upnp

Here is the log in RT-AC51U:

May 26 23:37:21 kernel: ESW: Link Status Changed - LAN1 (P1) Link DOWN
May 26 23:37:24 kernel: ESW: Link Status Changed - LAN1 (P1) Link UP
May 26 23:37:57 kernel: ESW: Link Status Changed - LAN1 (P1) Link DOWN
May 26 23:37:59 kernel: ESW: Link Status Changed - LAN1 (P1) Link UP

On 2017/5/28, Upgrade to Un-official RT-AC1200G+_9.0.0.4_380_7415-gc949b277fca.trx

From RMerl’s reply, I realize it doesn’t make sense to focus on syslog.log anymore because it won’t help to what cause kernel crash. I need to try something else.

How to Solve Reboot

  1. Call customer service and show them my post. Ask to change another model. Please share your success story in comments to help others.
  2. Get a serial and hack to RT-AC1200G+. If Asus blocks the communication port on shipping version, we cannot do anything.
  3. Try new firmware and configuration.

I decide to continue with 3 and also study 2.

Learn from Known

I read reboot history and notice BCM47189 has more problems with bandwidth 80MHz. Lower to 40MHz makes RT-AC1200G+ more stable but still reboot.

For kernel crash, get a new Linux (or new driver compile to kernel) or disable problem functions and drivers should help. I decide to give RT-AC1200G+_9.0.0.4_380_7415-gc949b277fca.trx a shot.

Before upgrading, run dmesg and nvram show in ssh to keep important information in firmware RT-AC51U_3.0.0.4_380_7378-g7a25649.trx. It helps to compare the difference.

Upgrade Firmware

Upgrade is easy. Follow comments in this post to download the mystery RT-AC1200G+_9.0.0.4_380_7415-gc949b277fca.trx which is not available on official web site. You also need to clear NVRAM (Non-volatile random-access memory) by push-and-hold WPS button then turn on power until power light is blinking. Or you may run mtd-erase2 after configuration:

admin@(none):/tmp/home/root# mtd-erase2
usage: mtd-erase2 [device]
admin@(none):/tmp/home/root# mtd-erase2 nvram
admin@(none):/tmp/home/root# 

Read “RT-AC66U Reset/Clear NVRAM” on SNBForum for more instructions.

Here are my modified configurations to test for stability. Others are kept default.

  1. In [Wireless], set Channel bandwidth on 5GHz to [40MHz], Control Channel to [Auto], and Wireless Mode to [N only].
  2. In [Wireless] → [Professional] tab, disable 2.4GHz radio signal.
  3. Set Enable UPnP Media Server on [Media Server] tab and Enable Share on [Network Place (Samba) Share] tab both to Off in [USB Application]
  4. Keep Enable SSH to [No] in  [Administration] → [System] tab. Turn on only when need it.

For my personal requirement:

  1. In Quick Internet Setup, change login password.
  2. In Quick Internet Setup, change SSID and password for 2.4GHz and 5GHz.
  3. In [Wireless] → [Wireless MAC Filter] tab, enable Enable MAC Filter for 5GHz.
  4. In [LAN] → [LAN IP] tab, Set LAN IP as 192.168.120.1.
  5. In [Network Map], Enable MAC and IP address Binding on TS-251A.

Difference

Run dmesg on the new firmware. I found following differences (total 7):

380_7378 380_7415
Linux version 2.6.36.4brcmarm (root@asus) (gcc version 4.5.3 (Buildroot 2012.02) ) #1 PREEMPT Thu Nov 3 15:55:43 CST 2016 Linux version 2.6.36.4brcmarm (root@asus) (gcc version 4.5.3 (Buildroot 2012.02) ) #1 PREEMPT Mon Mar 27 10:53:10 CST 2017
.data : 0x803fc000 – 0x8041ef40   ( 140 kB) .data : 0x803fc000 – 0x8041ef80   ( 140 kB)
MPCORE Global Timer Clock 35987Hz MPCORE Global Timer Clock 35995Hz
0x0000001ef240-0x000000d30000 : “rootfs” 0x0000001ef354-0x000000d30000 : “rootfs”
External imprecise Data abort at addr=0x2acc7000, fsr=0x1c06, pc=0x2aacae10 lr=0x2aacae5c ignored. External imprecise Data abort at addr=0x2ae18000, fsr=0x1c06, pc=0x800ca8e0 lr=0x800ca998 ignored.
N/A br0: port 2(eth1) entering forwarding state
device eth1 left promiscuous mode
br0: port 2(eth1) entering disabled state
br0: port 3(eth2) entering forwarding state
device eth2 left promiscuous mode
br0: port 3(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2ae49004, fsr=0x1c06, pc=0x2aec2014 lr=0xecec ignored.
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x803bf308 lr=0x80212cac ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device eth1 entered promiscuous mode
br0: topology change detected, propagating
br0: port 2(eth1) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 3(eth2) entering forwarding state
br0: port 3(eth2) entering forwarding state
N/A autodet uses obsolete (PF_INET,SOCK_PACKET)

gro disabled
br0: port 2(eth1) entering forwarding state
device eth1 left promiscuous mode
br0: port 2(eth1) entering disabled state
br0: port 3(eth2) entering forwarding state
device eth2 left promiscuous mode
br0: port 3(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x2abc2e20 lr=0x9eb0 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device eth1 entered promiscuous mode
br0: topology change detected, propagating
br0: port 2(eth1) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 3(eth2) entering forwarding state
br0: port 3(eth2) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth1 left promiscuous mode
br0: port 2(eth1) entering disabled state
br0: port 3(eth2) entering forwarding state
device eth2 left promiscuous mode
br0: port 3(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x91ba0 lr=0x2ae68254 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device eth1 entered promiscuous mode
br0: topology change detected, propagating
br0: port 2(eth1) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 3(eth2) entering forwarding state
br0: port 3(eth2) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth1 left promiscuous mode
br0: port 2(eth1) entering disabled state
br0: port 3(eth2) entering forwarding state
device eth2 left promiscuous mode
br0: port 3(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x2abc2e20 lr=0x9eb0 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 2(eth2) entering forwarding state
br0: port 2(eth2) entering forwarding state
br0: port 2(eth2) entering forwarding state
device eth2 left promiscuous mode
br0: port 2(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x2abc2e20 lr=0x9eb0 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 2(eth2) entering forwarding state
br0: port 2(eth2) entering forwarding state

gro disabled
Attempt to kill tasklet from interrupt
device eth0 left promiscuous mode
br0: port 1(vlan1) entering forwarding state
device eth0 entered promiscuous mode
br0: topology change detected, propagating
br0: port 1(vlan1) entering forwarding state
br0: port 1(vlan1) entering forwarding state
br0: port 2(eth2) entering forwarding state
br0: port 1(vlan1) entering forwarding state
device eth0 left promiscuous mode
device vlan1 left promiscuous mode
br0: port 1(vlan1) entering disabled state
device eth2 left promiscuous mode
br0: port 2(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x7168c lr=0x2ae63254 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device vlan1 entered promiscuous mode
device eth0 entered promiscuous mode
device eth1 entered promiscuous mode
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 3(eth2) entering forwarding state
br0: port 3(eth2) entering forwarding state
br0: topology change detected, propagating
br0: port 2(eth1) entering forwarding state
br0: port 2(eth1) entering forwarding state
br0: topology change detected, propagating
br0: port 1(vlan1) entering forwarding state
br0: port 1(vlan1) entering forwarding state

Disable SSH Daemon

caliban suggest to set Enable SSH to [No] in  [Administration] → [System] tab and my auto-restart history also support this suggestion: no-reboot between 2017/4/24 to 5/8 when SSH Daemon is off. Then it reboots every 2 to 3 days after 2017/5/8.

I will enable it only when needed.

On 2017/5/29, Reboot, Identify Crash Time with Another Router

How I catch Reboot time

With the new network map on 2017/5/26, RT-AC51U will notice LAN has down once RT-AC1200G+ reboot. That’s why I got the exact time 15:27:19. Unfortunately, I didn’t turn on my surveillance system and miss the video.

Here is the log in RT-AC51U:

May 29 15:21:08 kernel: MT7610ap:LOG#6 b4:18:d1:c3:a1:b4 has disassociated
May 29 15:27:19 kernel: ESW: Link Status Changed - LAN1 (P1) Link DOWN
May 29 15:27:21 kernel: ESW: Link Status Changed - LAN1 (P1) Link UP
May 29 15:27:27 kernel: MT7610ap:ASSOC - Assign AID=1 to STA a8:8e:24:a4:57:dd
May 29 15:27:54 kernel: ESW: Link Status Changed - LAN1 (P1) Link DOWN
May 29 15:27:57 kernel: ESW: Link Status Changed - LAN1 (P1) Link UP
May 29 15:34:37 kernel: MT7610ap:LOG#7 a8:8e:24:a4:57:dd has disassociated

On the other hand, you only get useless log in RT-AC1200G+ because nothing will write to syslog.log when kernel panic happens.

May 29 14:22:28 rc_service: waitting "start_lltdc" via httpd ...
Aug 1 08:00:09 syslogd started: BusyBox v1.17.4
Aug 1 08:00:09 kernel: klogd started: BusyBox v1.17.4 (2017-03-27 10:42:39 CST)
...skip...
Aug 1 08:00:25 miniupnpd[701]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 1 08:00:26 ntp: start NTP update
May 29 15:28:16 rc_service: ntp 696:notify_rc restart_upnp
May 29 15:28:16 miniupnpd[701]: shutting down MiniUPnPd

On 2017/5/30, Pressure Test, 5GHz Radio Jamming, Wireless Technologies

Pressure Test

I design a pressure test to make BCM47189 and GbE busy. Therefore, I prepare 57 files for total 154.73GB. Copying them between QNAP TS-251A and Mac mini (Late 2012) at the same time. QNAP TS-251A uses a wired GbE connection to RT-AC1200G+ while Mac mini is 802.11n 2T2R @ 5GHz.

Here are some screen captures to understand how busy they are:

CPU and Memory Usage in Asus RT-AC1200G+
CPU and Memory Usage in Asus RT-AC1200G+
5GHz Wireless Traffic in Asus RT-AC1200G+
5GHz Wireless Traffic in Asus RT-AC1200G+
Network Traffic Monitor in QNAP T251A
Network Traffic Monitor in QNAP T251A

I run twice and it takes for more than 8 hours to finish. RT-AC1200G+ pass the test without any reboot or unstable wireless network. I was very impressed.

5GHz Radio Jamming

During pressure test, my tablet which connects to RT-AC51U on 5GHz is so slow that won’t be able to load any web page. I suspect it is a radio jamming issue. Here are the current configurations:

  1. RT-AC1200G+ is using channel 153 with bandwidth 40MHz.
  2. RT-AC51U is using channel 157 with bandwidth 20/40/80MHz.

When I change RT-AC51U to channel 149, everything back to normal. My tablet works as normal and the test is still going on.

Study Kernel Panic and Wireless Technologies

I spend my whole day studying kernel panic, interference , and many related terminologies. Here are some useful posts I found:

  1. Linux Shtuff: Troubleshoot Linux Kernel Panics – understand what are kernel panics and how to trace them.
  2. StackOverflow: Generate kernel oops or crash in the code – generate kernel panic in code.
  3. Unix & Linux: What’s the difference between a kernel oops and a kernel panic? – they are more precise than reboot.
  4. Embedded System Blog: Debugging Analysis of Kernel panics and Kernel oopses using System Map: – debug kernel panic and soft panic step-by-step.
  5. StackOverflow: Kernel panic error in ARM board – makers meet kernel panic more often than desktop.
  6. StackOverflow: Arguments to ARM Linux’s arm_pm_restart() – from the functions here I begin to read Linux kernel source code.
  7. free electrons: /arch/arm/kernel/reboot.c – source code for static void __soft_restart(void *addr)
  8. free electrons: /kernel/reboot.c – source code for void do_kernel_restart(char *cmd) and void migrate_to_reboot_cpu(void)
  9. O’Reilly: Learning Debian GNU/Linux: 7.4 Viewing System Messages and Logs – learn dmesg and /var/log/message.
  10. My Cyber Universe: Full controlling the ASUS router via command line – learn commands used in Asuswrt.
  11. Cisco: Meraki: Channel Planning Best Practices – learn basic wireless terminologies and concepts.
  12. Extreme Networks: How to create a 5GHz WiFi Channel Plan – learn things needs to take into consideration when design 5GHz channel.
  13. Networking Computing: Dynamic Frequency Selection Part 3: The Channel Dilemma – I love the 802.11ac channel allocation illustration.
  14. 電腦王:無線路由器怎麼設定?拆機器、教術語,基本功一次搞懂 – basic wireless terminologies and concepts in Traditional Chinese.
  15. 聊Wi-Fi – advanced topics on wireless concept and implement design in Traditional Chinese.
  16. 無線網路新趨勢–漫談 802.11ac 新協定 by 臺灣大學計算機及資訊網路中心游子興 – learn bandwidth, spatial streams, MU-MIMO, beam forming, 256-QAM modulation in Traditional Chinese.

Configuration Update

To make it unstable and generate a reboot more quickly, I made following changes::

  1. In [Wireless], set Channel bandwidth on 5GHz to [80MHz], Control Channel to [149], and Wireless Mode to [N/AC mixed].
  2. Set Enable SSH to [Yes] in  [Administration] → [System] tab.

On 2017/5/31, 5GHz Unstable, Reboot

Finally, I caught a reboot and more facts about the unstable situation.

5GHz Unstable

I see many unstable wireless in different time period today. Here is my story during 13:26 ~ 13:34.

First is the video capture. I didn’t see anything special nor know how to decode these lights.

Here is the log in [System log] → [General Log] tab during 13:26 to 13:58. I reset power once but doesn’t work. Therefore, I pull the power completely. That’s why you see syslogd started: BusyBox v1.17.4 twice.

May 31 09:19:58 miniupnpd[705]: Listening for NAT-PMP/PCP traffic on port 5351
May 31 09:20:00 disk monitor: be idle
Aug 1 08:00:09 syslogd started: BusyBox v1.17.4
Aug 1 08:00:09 kernel: klogd started: BusyBox v1.17.4 (2017-03-27 10:42:39 CST)
... skip ...
Aug 1 08:00:25 miniupnpd[689]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 1 08:00:27 ntp: start NTP update
May 31 13:52:15 rc_service: ntp 687:notify_rc restart_upnp
May 31 13:52:15 miniupnpd[689]: shutting down MiniUPnPd
May 31 13:52:15 miniupnpd[704]: version 1.9 started
May 31 13:52:15 miniupnpd[704]: HTTP listening on port 41021
May 31 13:52:15 miniupnpd[704]: Listening for NAT-PMP/PCP traffic on port 5351
May 31 13:52:16 rc_service: ntp 687:notify_rc restart_diskmon
May 31 13:52:16 disk_monitor: Finish
May 31 13:52:19 disk monitor: be idle
Aug 1 08:00:09 syslogd started: BusyBox v1.17.4
Aug 1 08:00:09 kernel: klogd started: BusyBox v1.17.4 (2017-03-27 10:42:39 CST)
... skip ...
Aug 1 08:00:24 miniupnpd[689]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 1 08:00:27 ntp: start NTP update
May 31 13:58:25 rc_service: ntp 687:notify_rc restart_upnp
May 31 13:58:25 miniupnpd[689]: shutting down MiniUPnPd

As you may see in screen captures on WiFi Scanner / Analyzer, Asus_5G from RT-AC1200G+ is available and disappear quickly. There is no other visible SSID. It’s the same with Wifi Analyzer:

Screenshot_2017-05-31-13-28-10: Asus_5G is available for a while
Screenshot_2017-05-31-13-28-10: Asus_5G is available for a while
Screenshot_2017-05-31-13-28-33: Asus_5G is disappear now
Screenshot_2017-05-31-13-28-33: Asus_5G is disappear now

Using Network Preference in macOS Sierra, I found more interesting stuff. As mentioned on 2017/5/28, I Set LAN IP as 192.168.120.1 and the range should be 192.168.120.2 to 192.168.120.254. What is 169.254.26.113? It’s not even within the range.

Lenaius Chang explain in his comment:

The 169.254.26.113 you mentioned is likely the pc’s local auto-assinged IP before the router’s DHCP Server finish initializing. When physical(either wired or wireless) connection is established, the OS needs an IP address to communicate, but if DHCP Requests don’t return with a valid ip address, the PC can only create it’s own for temporary use. When the router’s DHCP server finishes it’s preparing sequence, it will return(assign) an IP address in range to your PC on receiving a DHCP request.

Screen Shot 2017-05-31 at 1.26.52 PM: Lucky me, I connect to As_5G but who is 169.254.26.113?
Screen Shot 2017-05-31 at 1.26.52 PM: Lucky me, I connect to As_5G but who is 169.254.26.113?
Screen Shot 2017-05-31: normal IP from Asus RT-AC1200G+ at 1.29.23 PM
Screen Shot 2017-05-31: normal IP from Asus RT-AC1200G+ at 1.29.23 PM
Screen Shot 2017-05-31 at 1.30.11 PM: okay, no IP. Seems DHCP server is not functional
Screen Shot 2017-05-31 at 1.30.11 PM: okay, no IP. Seems DHCP server is not functional

At 23:24, 5GHz becomes unstable again and I use a wired GbE cable to connect RT-AC1200G+ and Mac mini. RT-AC1200G+ is still functional but fail to provide wireless connection.

Here is what I collect in [System Log] → [Wireless Log] tab about wireless:

2.4 GHz radio is disabled

SSID: "ASUS_5G"
RSSI: 0 dBm SNR: 0 dB noise: -87 dBm Channel: 149/80
BSSID: 60:45:CB:8E:3C:8C Capability: ESS
Supported Rates: [ 6(b) 9 12(b) 18 24(b) 36 48 54 ]
VHT Capable:
Chanspec: 5GHz channel 155 80MHz (0xe09b)
Primary channel: 149
HT Capabilities:
Supported MCS : [ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 ]
VHT Capabilities:
Supported VHT (tx) Rates:
NSS: 1 MCS: 0-9
NSS: 2 MCS: 0-9
Supported VHT (rx) Rates:
NSS: 1 MCS: 0-9
NSS: 2 MCS: 0-9

Mode : AP Only

Stations List
----------------------------------------
idx MAC Associated Authorized RSSI PSM SGI STBC Tx rate Rx rate Connect Time

For comparison, here is what I collect in [System Log] → [Wireless Log] tab when Mac mini connects to RT-AC51U with 1T1R @ 5GHz:

MAC address : F0:79:59:77:F9:78
OP Mode : AP
Phy Mode : 11n
Channel : 9

Stations List
----------------------------------------
MAC PSM PhyMode BW MCS SGI STBC Rate Connect Time
0C:84:DC:C2:5A:C6 NO HTMIX 20M 7 NO NO 65M 00:01:12

MAC address : F0:79:59:77:F9:7C
OP Mode : AP
Phy Mode : 11n/ac
Channel : 149

Stations List
----------------------------------------
MAC PSM PhyMode BW MCS SGI STBC Rate Connect Time
A8:8E:24:A4:57:DD NO HTMIX 40M 7 NO NO 135M 00:01:07

And there is nothing special in [System log] → [General Log] tab till 2017/6/1 00:34:

ay 31 18:43:23 rc_service: ntp 688:notify_rc restart_diskmon
May 31 18:43:23 disk_monitor: Finish
May 31 18:43:25 disk monitor: be idle
May 31 19:00:25 disk_monitor: Got SIGALRM...
May 31 23:00:25 disk_monitor: Got SIGALRM...

Reboot

During 18:42 and 18:43, RT-AC1200G+ reboots but the Uptime keeps going on. Begin from 00:28 in the video:

Here is the log in RT-AC51U. It helps to catch the reboot time. a8:8e:24:a4:57:dd is my Mac mini. It seems he connected to RT-AC51U due to lost connection with Asus_5G from RT-AC1200G+ automatically. And the later eventually crash and reboot after about 33 minutes.

May 31 18:09:47 kernel: MT7610ap:ASSOC - Assign AID=1 to STA a8:8e:24:a4:57:dd
May 31 18:42:26 kernel: ESW: Link Status Changed - LAN1 (P1) Link DOWN
May 31 18:42:28 kernel: ESW: Link Status Changed - LAN1 (P1) Link UP
May 31 18:43:01 kernel: ESW: Link Status Changed - LAN1 (P1) Link DOWN
May 31 18:43:03 kernel: ESW: Link Status Changed - LAN1 (P1) Link UP
May 31 19:09:25 kernel: MT7610ap:ASSOC - Assign AID=1 to STA a8:8e:24:a4:57:dd

Here is the log in RT-AC1200G+. Same, nothing useful.

May 31 14:05:09 rc_service: httpd 409:notify_rc start_miniupnpc
May 31 14:05:09 rc_service: waitting "start_lltdc" via httpd ...
Aug 1 08:00:09 syslogd started: BusyBox v1.17.4
Aug 1 08:00:09 kernel: klogd started: BusyBox v1.17.4 (2017-03-27 10:42:39 CST)
... skip ...
Aug 1 08:00:24 miniupnpd[690]: Listening for NAT-PMP/PCP traffic on port 5351
Aug 1 08:00:25 ntp: start NTP update
May 31 18:43:22 rc_service: ntp 688:notify_rc restart_upnp
May 31 18:43:22 miniupnpd[690]: shutting down MiniUPnPd

On 2017/6/1, Upgrade to RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx

I got a call from Dolly Chu who work in Asus Customer Service, she provides me a link to a new beta firmware RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx. I decide to give it another try.

Difference

Run dmesg on the new firmware and found following differences (total 14):

380_7415 380_10680
Memory: 125488k/125488k available, 5584k reserved, 0K highmem Memory: 125464k/125464k available, 5608k reserved, 0K highmem
Virtual kernel memory layout:
vector : 0xffff0000 – 0xffff1000 ( 4 kB)
fixmap : 0xfff00000 – 0xfffe0000 ( 896 kB)
DMA : 0xf7e00000 – 0xffe00000 ( 128 MB)
vmalloc : 0x88800000 – 0xf0000000 (1656 MB)
lowmem : 0x80000000 – 0x88000000 ( 128 MB)
modules : 0x7f000000 – 0x80000000 ( 16 MB)
.init : 0x80008000 – 0x80038000 ( 192 kB)
.text : 0x80038000 – 0x803e3000 (3756 kB)
.data : 0x803fc000 – 0x8041ef80 ( 140 kB)
Virtual kernel memory layout:
vector : 0xffff0000 – 0xffff1000 ( 4 kB)
fixmap : 0xfff00000 – 0xfffe0000 ( 896 kB)
DMA : 0xf7e00000 – 0xffe00000 ( 128 MB)
vmalloc : 0x88800000 – 0xf0000000 (1656 MB)
lowmem : 0x80000000 – 0x88000000 ( 128 MB)
modules : 0x7f000000 – 0x80000000 ( 16 MB)
.init : 0x80008000 – 0x80039000 ( 196 kB)
.text : 0x80039000 – 0x803e5000 (3760 kB)
.data : 0x803fe000 – 0x80420f40 ( 140 kB)
N/A * Invalid signature of oopsbuf: 00-10-00-87-77-FE-7F-4F (len 2396913050)
MPCORE Global Timer Clock 35995Hz MPCORE Global Timer Clock 35987Hz
N/A PCIE PLL is locked
Creating 5 MTD partitions on “bcmsflash”:
0x000000000000-0x000000040000 : “boot”
0x000000040000-0x000000ff0000 : “linux”
0x0000001ef354-0x000000d30000 : “rootfs”
0x000000d30000-0x000000ff0000 : “jffs2”
0x000000ff0000-0x000001000000 : “nvram”
Creating 5 MTD partitions on “bcmsflash”:
0x000000000000-0x000000040000 : “boot”
0x000000040000-0x000000ff0000 : “linux”
0x0000001ef910-0x000000da0000 : “rootfs”
0x000000da0000-0x000000ff0000 : “jffs2”
0x000000ff0000-0x000001000000 : “nvram”
Freeing init memory: 192K Freeing init memory: 196K
eth0: Broadcom BCM47XX 10/100/1000 Mbps Ethernet Controller 9.10.178.27 (r584393)
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
PCI: Enabling device 0001:01:00.0 (0140 -> 0142)
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2ae18000, fsr=0x1c06, pc=0x800ca8e0 lr=0x800ca998 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
eth0: Broadcom BCM47XX 10/100/1000 Mbps Ethernet Controller 9.10.178.61 (r667680)
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x0
PCI: Enabling device 0001:01:00.0 (0140 -> 0142)
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.61 (r667680)
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.61 (r667680)
JFFS2 version 2.2. (NAND) © 2001-2006 Red Hat, Inc.
device vlan1 entered promiscuous mode
device eth0 entered promiscuous mode
device eth1 entered promiscuous mode
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 3(eth2) entering forwarding state
br0: port 3(eth2) entering forwarding state
br0: topology change detected, propagating
br0: port 2(eth1) entering forwarding state
br0: port 2(eth1) entering forwarding state
br0: topology change detected, propagating
br0: port 1(vlan1) entering forwarding state
br0: port 1(vlan1) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth1 left promiscuous mode
br0: port 2(eth1) entering disabled state
br0: port 3(eth2) entering forwarding state
device eth2 left promiscuous mode
br0: port 3(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2ae49004, fsr=0x1c06, pc=0x2aec2014 lr=0xecec ignored.
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x803bf308 lr=0x80212cac ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device eth1 entered promiscuous mode
br0: topology change detected, propagating
br0: port 2(eth1) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 3(eth2) entering forwarding state
br0: port 3(eth2) entering forwarding state
N/A
SCSI subsystem initialized
Initializing USB Mass Storage driver…
usbcore: registered new interface driver usb-storage
USB Mass Storage support registered.
N/A
usbcore: registered new interface driver usblp usbcore: registered new interface driver cdc_acm
cdc_acm: v0.26:USB Abstract Control Model driver for USB modems and ISDN adapters
nf_conntrack_rtsp v0.6.21 loading
nf_nat_rtsp v0.6.21 loading
autodet uses obsolete (PF_INET,SOCK_PACKET)gro disabled
br0: port 2(eth1) entering forwarding state
device eth1 left promiscuous mode
br0: port 2(eth1) entering disabled state
br0: port 3(eth2) entering forwarding state
device eth2 left promiscuous mode
br0: port 3(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x2abc2e20 lr=0x9eb0 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device eth1 entered promiscuous mode
br0: topology change detected, propagating
br0: port 2(eth1) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 3(eth2) entering forwarding state
br0: port 3(eth2) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth1 left promiscuous mode
br0: port 2(eth1) entering disabled state
br0: port 3(eth2) entering forwarding state
device eth2 left promiscuous mode
br0: port 3(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x91ba0 lr=0x2ae68254 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device eth1 entered promiscuous mode
br0: topology change detected, propagating
br0: port 2(eth1) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 3(eth2) entering forwarding state
br0: port 3(eth2) entering forwarding state
br0: port 2(eth1) entering forwarding state
device eth1 left promiscuous mode
br0: port 2(eth1) entering disabled state
br0: port 3(eth2) entering forwarding state
device eth2 left promiscuous mode
br0: port 3(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x2abc2e20 lr=0x9eb0 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 2(eth2) entering forwarding state
br0: port 2(eth2) entering forwarding state
br0: port 2(eth2) entering forwarding state
device eth2 left promiscuous mode
br0: port 2(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x2abc2e20 lr=0x9eb0 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
device eth2 entered promiscuous mode
br0: topology change detected, propagating
br0: port 2(eth2) entering forwarding state
br0: port 2(eth2) entering forwarding stategro disabled
Attempt to kill tasklet from interrupt
device eth0 left promiscuous mode
br0: port 1(vlan1) entering forwarding state
device eth0 entered promiscuous mode
br0: topology change detected, propagating
br0: port 1(vlan1) entering forwarding state
br0: port 1(vlan1) entering forwarding state
br0: port 2(eth2) entering forwarding state
br0: port 1(vlan1) entering forwarding state
device eth0 left promiscuous mode
device vlan1 left promiscuous mode
br0: port 1(vlan1) entering disabled state
device eth2 left promiscuous mode
br0: port 2(eth2) entering disabled state
wl_module_init: passivemode set to 0x0
wl_module_init: txworkq set to 0x1
eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.27 (r584393)
External imprecise Data abort at addr=0x2abdf3b0, fsr=0x1c06, pc=0x7168c lr=0x2ae63254 ignored.
eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.27 (r584393)
JFFS2 version 2.2. (NAND) © 2001-2006 Red Hat, Inc.
N/A SCSI subsystem initialized
Initializing USB Mass Storage driver…
usbcore: registered new interface driver usb-storage
USB Mass Storage support registered.
usbcore: registered new interface driver usblp
nf_conntrack_rtsp v0.6.21 loading
nf_nat_rtsp v0.6.21 loading

I think the most important differences are:

  1. Linux version 2.6.36.4brcmarm is still the same
  2. Broadcom BCM43227 802.11 Wireless Controller 9.10.178.61 (r667680) has been upgrade from 9.10.178.27 (r584393) in 3.0.0.4_380_4089 and 9.0.0.4_380_7415.
  3. Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.61 (r667680) has been upgraded from 9.10.178.27 (r584393) in 3.0.0.4_380_4089 and 9.0.0.4_380_7415.

As we suspect drivers are the source to reboot. Now we finally see new some changes here.

I am also interested in * Invalid signature of oopsbuf: 00-10-00-87-77-FE-7F-4F (len 2396913050) because it looks like for kernel soft panics.

2.4GHz SSID Disappear Issue

caliban reported fail to find 2.4GHz SSID after applying. I think it might be related to 2.4GHz channel issue. In PCMagazine’s “Totolink A2004NS AC1200 Wireless Dual Band Gigabit Router With USB Port”:

To finish installation, you set the SSID name for each band, the encryption level, and a passphrase. As I discovered with the N100RE, the A2004NS’s default region is Europe. I had to switch it to United States/Canada. Why? Because different countries use different wireless channels. If you have the wrong region set, you may be using the wrong wireless channel and as a result get poor wireless connectivity—or none at all.

caliban shared his solution: just reset your power and everything works normal.

Mike shared his solution: run mtd-erase2 as below and use power button to reset. Then you 2.4GHz SSID won’t disappear after a soft reboot.

admin@(none):/tmp/home/root# mtd-erase2
usage: mtd-erase2 [device]
admin@(none):/tmp/home/root# mtd-erase2 nvram
admin@(none):/tmp/home/root# 

On 2017/6/7, Update Network Map, MAC filter list Issue

Because there is no reboot for 6 days after upgrading to RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx and nobody feedback any issue, I restore original network map again on 2017/6/7. Make RT-AC1200G+ the only one wireless router.

During setup, I notice some issue with the icon and name in MAC filter list in [Wireless MAC Filter] tab. It would be displayed as New device with default icon after modified and apply until connected. In some case, your edit won’t be apply for no reason.

For example, I add a Windows device, choose a Windows icon, and change it to SurfaceRT. Then hit Apply button twice. It is displayed as New device with a default laptop icon. When this Windows device connects to RT-AC1200G+, the name is changed to WinRT with a Windows Desktop icon. I try to change it again but fails. It won’t save or apply.

I will continuous monitoring if there is any issue in the next 3 weeks.

On 2017/6/16, No Reboot for 15 Days But Still Minor Issues

It’s 15 days after upgrading to RT-AC1200G+_9.0.0.4_382_10680- gc72d1c8.trx without reboot issue. There are three minor issue as below:

  1. caliban reports a 20 seconds wireless drop and reconnect once.
  2. Available channels doesn’t fit EU regulation.
  3. Need NVRAM reset after upgrade or 2.4GHz SSID might be unavailable.

I will continuous monitoring if there is any issue in the next 2 weeks.

On 2017/7/2, Unstable 5GHz Issue

I have experience slow 5GHz connection around 2017/7/2 23:30 to 2017/7/3 00:30. Here is what happen to me:

  1. 5GHz SSID is missing sometimes.
  2. Even connect successfully, transfer rate is extremely slow.
  3. 2.4GHz SSID has no connection or slow issue at the same time.
  4. Very few visible 5GHz SSID around.
  5. Reset USB Tethering doesn’t improve.
  6. Reboot via web interface and power button don’t improve.
  7. Change 5GHz from channel 149 to auto will be assigned to 36 and disappear from Surface RT.
  8. Switch back from auto to 149. After 10 minutes, it seems become stable again.

It might be 5GHz radio jamminginterference or firmware issue. Therefore, I decide to extent stability test and will change channel to 153 or 157 next time. For radio jamming and interference, should be resolved by assigning a different channel.

Here is a partial wireless log during 5GHz unstable period:

SSID: "MyHome_5G"
RSSI: 0 dBm	SNR: 0 dB	noise: -92 dBm	Channel: 149/80
BSSID: 60:45:CB:8E:3C:8C	Capability: ESS 
Supported Rates: [ 6(b) 9 12(b) 18 24(b) 36 48 54 ]
VHT Capable:
	Chanspec: 5GHz channel 155 80MHz (0xe09b)
	Primary channel: 149
	HT Capabilities: 
	Supported MCS : [ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 ]
	VHT Capabilities: 
	Supported VHT (tx) Rates:
		NSS: 1 MCS: 0-9
		NSS: 2 MCS: 0-9
	Supported VHT (rx) Rates:
		NSS: 1 MCS: 0-9
		NSS: 2 MCS: 0-9

Mode	: AP Only

On 2017/7/3, Unstable 5GHz Issue is Solved by Switching Control Channel

I experience the same slow and missing 5GHz connection around 2017/7/3 23:30 again. WiFi Scanner / Analyzer found no other visible 5GHz SSID in my house. Because 2.4GHz is still working properly, I use my mobile connect to RT-AC1200G+ and change the Control Channel at 5GHz to 153. After applying change and restarting Broadcom BCM47189Mac mini connects to RT-AC1200G+ again and everything goes back to normal.

I am not sure it is a radio jamming or interference but sure it can be solve by changing Control Channel.

Here is a partial wireless log after changing Control Channel at 5GHz to 153:

SSID: "MyHome_24G"
SSID: "MyHome_5G"
RSSI: 0 dBm	SNR: 0 dB	noise: -92 dBm	Channel: 153/80
BSSID: 60:45:CB:8E:3C:8C	Capability: ESS 
Supported Rates: [ 6(b) 9 12(b) 18 24(b) 36 48 54 ]
VHT Capable:
	Chanspec: 5GHz channel 155 80MHz (0xe19b)
	Primary channel: 153
	HT Capabilities: 
	Supported MCS : [ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 ]
	VHT Capabilities: 
	Supported VHT (tx) Rates:
		NSS: 1 MCS: 0-9
		NSS: 2 MCS: 0-9
	Supported VHT (rx) Rates:
		NSS: 1 MCS: 0-9
		NSS: 2 MCS: 0-9

Mode	: AP Only

On 2017/7/5, Unstable 5GHz Issue is Solved by Narrowing  Channel Bandwidth

Again, I experience the same slow and missing 5GHz connection around 2017/7/5 17:58 again. Change Control Channel at 5GHz to others doesn’t work. Therefore, I lower the Control Channel Bandwidth at 5GHz from 80MHz (auto select) to 40MHz. Although the 802.11ac 2T2R top connection speed drops to 400Mbps, it’s stable and good enough for my 802.11n devices.

As you may see below, EC:22:80:7E:03:19 is D-Link Wireless AC Dual Band USB Adapter DWA-171 which is a 1T1R 802.11ac device has been shrink to 200Mbps. A8:8E:24:A4:57:DD is Mac mini which is the same as usual.

 

If you experience radio jamming or interference, change Control Channel or Control Channel Bandwidth will help. RT-AC1200G+ won’t do it for you automatically.

SSID: "MyHome_5G"
RSSI: 0 dBm	SNR: 0 dB	noise: -91 dBm	Channel: 149l
BSSID: 60:45:CB:8E:3C:8C	Capability: ESS 
Supported Rates: [ 6(b) 9 12(b) 18 24(b) 36 48 54 ]
VHT Capable:
	Chanspec: 5GHz channel 151 40MHz (0xd897)
	Primary channel: 149
	HT Capabilities: 
	Supported MCS : [ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 32 ]
	VHT Capabilities: 
	Supported VHT (tx) Rates:
		NSS: 1 MCS: 0-9
		NSS: 2 MCS: 0-9
	Supported VHT (rx) Rates:
		NSS: 1 MCS: 0-9
		NSS: 2 MCS: 0-9

Mode	: AP Only

Stations List                           
----------------------------------------
idx MAC               Associated Authorized    RSSI PSM SGI STBC Tx rate Rx rate Connect Time
    EC:22:80:7E:03:19 Yes        Yes         -68dBm No  Yes Yes     200M    200M 00:00:53
    A8:8E:24:A4:57:DD Yes        Yes         -61dBm No  Yes Yes     300M    300M 00:19:53

Summarize Facts

  1. Reboot is not due to pressure/high traffic.
  2. Reboot is not due to 40/80 channel bandwidth.
  3. 5GHz unstable is not related to heat.
  4. 5GHz unstable is not related to 40/80 channel bandwidth.
  5. Wired GbE is okay while wireless in unstable.
  6. Didn’t see any meaningful light signal when wireless becomes unstable.
  7. Blinks as power on when reboot.

Let’s wait and see more reboot and unstable stories!

RT-AC1200G+ with RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx Stability Tracking

Name Stable Days Issue and Notes
Alew Waiting for reply
Amigo 52 (Update 2017/7/17)
  • MAC filter list display issue
  • Experience 5GHz slow and missing issue on 2017/7/2 to 3.
  • Experience 5GHz slow and missing issue on 2017/7/3. Solved by changing control channel.
AsusRTUsers 3 (Update 2017/6/9)
AsusRT1200g+User Waiting for reply
caliban 14 (Update 2017/7/6)
  • 2.4GHz SSID disappear issue needs a power off and on via button on the back to solve.
  • Use in AP-mode. Experience line drops for 20 seconds but reconnects automatically and cannot login to router within 20 seconds. Take a note and see if it happens again.
  • 120, 124, 128 which are included in EU 5GHz regulation are not available in AsusWRT
  • As we always need four channels for 80MHz bandwith, you can see why manual selection stops at 112: after that there never are four next to each other. For 40MHz we need 2 channels – there it stops at 136. Only when setting it to 20 MHz we can use 140. And if everything is on auto most of the time it uses 20 or 40 MHz, never 80 -> auto-channel goes up to 136/140.
  • 5GHz – 3 Missing Channels in Europe on WifiNigel explains why the 3 channels are left out. Important is the part about DFS wait time of 10 minutes.
Charles Waiting for reply
  •  Running 3.0.0.4.380.7743 since 2017/6/23. 7743 reboots after 4 to 5 days.
Fernando Puech Waiting for reply
hitchhiker 5 (Update 2017/6/27)
Kobe Waiting for reply
Lenaius Chang Waiting for reply
massimo Waiting for reply
Mike 32 (Update 2017/7/11)
  •  Using wifi scheduler doesn’t trigger random reboots to the router.
nikisi 33 (Update 2017/7/5)
  •  Strage log: Jun 3 13:33:57 rc_service: httpd 213:notify_rc start_lltdc
  • A slightly higher RAM usage after the update (before: ~38% – now: ~42% ).
  • After 10 days uptime, RAM usage is 46% ( 56mb ). Ram usage is raising with 1-2% every few days.
Nandolv Waiting for reply
Norbert Waiting for reply
PawelW Waiting for reply
Rokapet Waiting for reply
RT-AC1200G+ 35 (Update 2017/7/14)
  • RAM usage 37-39 MB without firewall、TX rate、USB and others.
tycjan 50 (Update 2017/7/23)
賴雨賢 Waiting for reply

Official 3.0.0.4.380.7743 Testing by Net Friends

On 2017/6/23, mouse remind me there is a new official  3.0.0.4.380.7743 release to public on 2017/6/22. Because I am testing 9.0.0.4_382_10680, Charles kindly agree to join the test and will feedback here. caliban quickly compare the difference and share with us. Therefore, I crease this section to keep track of the new official firmware.

Name Stable Days Issue and Notes
caliban 0 (Update 2017/6/23)
  •  Both BCM43227 and BCM43c8 802.11 Wireless Controller are 9.10.178.27 while 10680 is 9.10.178.61.
  • 2.4GHz missing issue cannot be solved completely by resetting power and clearing NVRAM.
  • Client list is broken and shows no devices.
  • Reboot after activating QoS and 2.4GHz SSID is missing.
  • Stop testing and switch to 10680 on 2017/6/23.
Charles Waiting for reply
  • Running 3.0.0.4.380.7743 since 2017/6/23.
ducklingasa 4 (Update 2017/7/3)
  • Running 3.0.0.4.380.7743 since 2017/6/28.
  • Reboot on 2017/7/2.
James 4 to 5 days
  • Reboot after 4 to 5 days.
JG 16 (Update 2017/7/23)
  • Reboot after 9 days (reported on 2017/7/7)
mouse Waiting for reply
  • First to share the latest firmware is available officially.

RT-AC1200G+ Auto-restart History

Version On Issue and Notes
3.0.0.4.380.3310 2017/3/10 New installation.
3.0.0.4.380.4089 2017/3/21 Reboot several times.

Install new firmware.

x 2017/4/2 Reboot at 20:54.
x 2017/4/4 Reboot at 20:54.

Disable Enable wireless scheduler on 2.4GHz and 5GHz

x 2017/4/10 Reboot at 23:17 on 2017/4/10.

OSX Finder warns Server connections interrupted which seems due to restart.

Disable NAT acceleration and need to restart.

x 2017/4/11 Manually select USB computer connection in Charging mode.
x 2017/4/13 Reboot at 14:41.

Manually select USB computer connection in Charging mode.

x 2017/4/17 Unstable network and need a restart at 23:57.
x 2017/4/18 Re-enable Enable UPnP in WAN.
x 2017/4/19 Reset and reconfiguration:

  1. In [USB Application] → [Media Services and Servers], disable [Media Server] and [network Place (Samba) Share] which is default on.
  2. In [USB Application] → [3G/4G], leave Select USB Device to [USB Modem].
  3. In [Wireless], turn off WPS and set Channel bandwidth on 5GHz for [80MHz] and on 2.4GHz for [40MHz]; set Wireless Mode on 2.4GHz to [N Only] and on 5GHz to [N/AC Mixed].
  4. In [WAN], UPnP is turned on by default.
  5. In [Firewall], Network Service Filter is turned off by default. Uncheck all selection and apply again.
  6. In [Administration] → [System], Reboot Scheduler is turned off by default.
x 2017/4/20 In [Wireless], set Channel bandwidth on 5GHz to [40MHz] and Control Channel on 5GHz to [Auto].
x 2017/4/24 Reboot at 21:20 and again at 21:28.

Change Select USB device in [USB Application] → [3G/4G] from USB Modem back to Android phone.

x 2017/5/8 Reboot at 10:16.

Set Wireless Mode on 5GHz to [N only] in [Advanced Settings] → [Wireless] → [General] tab.

x 2017/5/10 Reboot at 13:15.

Set Enable Radio on  5GHz to [No] in [Advanced Settings] → [Wireless] → [Professional] tab.

x 2017/5/12 Freeze at 13:13.

Set Enable Radio on  5GHz to [Yes] but 2.4GHz to [No] in [Advanced Settings] → [Wireless] → [Professional] tab.

x 2017/5/15 Reboot at 10:12 but WAN was restored automatically.
x 2017/5/15 ~ 25 1st RMA
x 2017/5/26 Unstable network and need a restart at 23:37.
9.0.0.4.380.7415 2017/5/28 Upgrade to new firmware and reset NVRAM.

Apply following configurations:

  1. In [Wireless], set Channel bandwidth on 5GHz to [40MHz], Control Channel to [Auto], and 5GHz to [N only].
  2. In [Wireless] → [Professional] tab, disable 2.4GHz radio signal.
  3. Set Enable UPnP Media Server on [Media Server] tab and Enable Share on [Network Place (Samba) Share] tab to off in [USB Application]
  4. Keep Enable SSH to [No] in  [Administration] → [System] tab. Turn on only when need it.
  5. In Quick Internet Setup, change login password.
  6. In Quick Internet Setup, Change SSID and password for 2.4GHz and 5GHz.
  7. In [Wireless] → [Wireless MAC Filter] tab, enable Enable MAC Filter for 5GHz.
  8. In [LAN] → [LAN IP] tab, Set LAN IP as 192.168.120.1.
  9. In [Network Map], Enable MAC and IP address Binding on TS-251A.
x 2017/5/29 Reboot at 15:27:19.
x 2017/5/30 Radio jamming @ 5GHz
x 2017/5/31 First unstable network and need restart twice between 13:26 to 13:58.

Reboot at 18:42.

More unstable wireless connection at night.

9.0.0.4.380.10680 2017/6/1 Upgrade to new firmware without reset NVRAM.

Current Configuration

After install to official latest stable version RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx, reset and configure as below:

  1. In [Wireless], keep Channel bandwidth on 5GHz to [20/40/80MHz]. Change Control Channel to [149], and Wireless Mode to [N/AC mixed]
  2. In [Wireless], keep Channel bandwidth on 2.4GHz to [20/40/80MHz] and Control Channel to [Auto]. Change Wireless Mode to [N Only]
  3. In [Wireless], enable Enable MAC Filter on 5GHz and 2.4GHz and set MAC Filter Mode to [Accept]. Add my devices in the list.
  4. In [Wireless], keep Enable wireless scheduler in [Professional] tab as [No] on both bands.
  5. In [USB Application] → Media Services and Servers, disable Enable UPnP Media Server on [Media Server] tab, Enable Share on [Network Place (Samba) Share] tab, and Enable FTP on [FTP Share] tab.
  6. In [Administration] → [System] tab, disable Enable Reboot Scheduler.

Note:

Here are some other personal configurations. Not necessary if you don’t need them.

  1. During Quick Internet Setup, I have change login password, change SSID names, setup [WPA2-Personal] key.
  2. In [LAN] → [LAN IP] tab, set LAN IP as 192.168.120.1.
  3. In [Network Map], Enable MAC and IP address Binding on my wired devices.
  4. I use 4G USB Tethering to access Internet.

Test Report in Comments

  1. nikisi: set Enable TX Bursting to [Disable] and Preamble Type to [Short] in [Wireless] → [Professional] tab last for 2 days.
  2. tycjan: set Preamble Type to [Short] in [Wireless] → [Professional] tab doesn’t work.
  3. hitchhiker and PamelW: the new firmware doesn’t stop reboot.
  4. tycjan and caliban: set Enable SSH to [No] in  [Administration] → [System] tab improves stability for more days.
  5. caliban: if you encounter 2.4GHz SSID disappear issue after firmware upgrade, just reset your power.

Reference

  1. Android USB Tethering Drops and Data Usage Crashes, try Reset
  2. Apple: Mac mini (Late 2012) – Technical Specifications
  3. Apple: macOS Sierra
  4. Asus Forum: [Problem]Router keeps rebooting randomly
  5. Asus Forum: RT-AC58U,RT-AC1300+ 這路由器Wifi 5G
  6. Asus Forum: RT-AC1200G常常斷線。WAN Connection: DNS probe failed
  7. Asus Forum: 哪裡可以下載 RT-AC1200G+ 台灣新 Beta 韌體 380_7415?
  8. Asus Support: How to do a firmware update on ASUS wireless router? (ASUSWRT)
  9. Asus Support: Introduction to UPnP
  10. Asus Support Forum: nikisi
  11. Asus Support Forum: [Sharing]A diary to solve Asus Rt-AC1200G+ reboot issue
  12. Asus RT-AC1200G+
  13. Asus RT-AC1200G+ Driver & Tools
  14. Asus RT-AC51U
  15. Asus RT-AC66U
  16. Asustor AS-5102T
  17. Best Wireless Routers Now: 20MHz vs 40MHz vs 80MHz vs 160MHz: What Channel Bandwidth Should I Use?
  18. BleepingComputer: 40 Asus RT Router Models Are Vulnerable to Simple Hacks
  19. Cisco: Meraki: Channel Planning Best Practices
  20. DD-WRT
  21. DD-WRT: Supported Devices 3.13 Asus
  22. D-Link DCS-930L Wireless N Network Camera
  23. D-Link Wireless AC Dual Band USB Adapter DWA-171
  24. Embedded System Blog: Debugging Analysis of Kernel panics and Kernel oopses using System Map:
  25. Extreme Networks: How to create a 5GHz WiFi Channel Plan
  26. Facebook: iTUP 資訊專業人員交流協會(挨踢趴)
  27. FATMIN: Configure Syslog Logging Levels on the Asus RT-AC66U Router
  28. free electrons: /arch/arm/kernel/reboot.c
  29. free electrons: /kernel/reboot.c
  30. GitHub: spontaneous reboot on RT-AC87U v.380.59 #941
  31. Google Play: Wifi Analyzer
  32. Google Play: WiFi Scanner / Analyzer
  33. Google: Nexus Help: Transfer files between your computer & Android device
  34. HKEPC: Asus RT-AC1200G+不定期重啟 #11
  35. HKEPC: fatdog
  36. HTPC Guides: Enable SSH on Asus Routers with(out) SSH Keys
  37. Huawei MediaPad M1
  38. Linux kernel source tree: cfg80211: use DS or HT operation IEs to determine BSS channel
  39. Linux Shtuff: Troubleshoot Linux Kernel Panics
  40. Linux Wireless: b43
  41. Linux Wireless: System on a chip
  42. LWN.net: JLS2009: Generic receive offload
  43. metageek: Understanding WiFi Signal Strength
  44. Microsoft: Network Ports Used by Key Microsoft Server Products
  45. Microsoft: SQL Server downloads
  46. MiniUPnP Project HomePage
  47. Mobile01: RT-AC1200G+不定期重啟
  48. My Cyber Universe: Full controlling the ASUS router via command line
  49. Networking Computing: Dynamic Frequency Selection Part 3: The Channel Dilemma
  50. OpenWrt: Asus RT-AC1200G, BCM47189+ BCM43217 chipset
  51. OpenWrt: Broadcom BCM47xx
  52. OpenWrt: changeway
  53. OpenWrt: NVRAM usage on OpenWrt ‘WhiteRussian’
  54. O’Reilly: Chapter 2. The PHY
  55. O’Reilly: Learning Debian GNU/Linux: 7.4 Viewing System Messages and Logs
  56. PCMagazine: Totolink A2004NS AC1200 Wireless Dual Band Gigabit Router With USB Port
  57. QNAP TS-119PII with Mac OSX File Transfer Performance Report
  58. QNAP TS-119PII with Different Wireless Device on Mac OSX File Transfer Performance Report
  59. QNAP TS-251A
  60. Router Guide: NAT Acceleration On or Off
  61. Setup a Surveillance System to Proof Crash of Asus RT-AC1200G+
  62. SNBForums: Asuswrt-Merlin 378.54_2 is now available
  63. SNBForum: Clear NVRAM
  64. SNBForum: RT-AC66U Reset/Clear NVRAM
  65. SNBForum: Where is the ASUS RT-AC68U storing all it’s logs?
  66. SNBForums: “Default message log level” & “Log only messages more urgent than” in Asuswrt-Merlin 378.56
  67. StackOverflow: Arguments to ARM Linux’s arm_pm_restart()
  68. StackOverflow: Generate kernel oops or crash in the code
  69. StackOverflow: Kernel panic error in ARM board
  70. StackOverflow: What does “connection reset by peer” mean?
  71. Trusted Review: Computex 2017: AMD, Nvidia, Asus and Intel set to impress
  72. Vornado: Circulators & Fans
  73. TurnKey Linux: LAMP Stack – Web Stack (MySQL)
  74. Unix & Linux: What’s the difference between a kernel oops and a kernel panic?
  75. WifiNigel: 5GHz – 3 Missing Channels in Europe
  76. Wiki: Customer relationship management
  77. Wiki: dBm
  78. Wiki: dmesg
  79. Wiki: Dynamic Host Configuration Protocol
  80. Wiki: Fast Ethernet
  81. Wiki: IEEE 802.11a-1999
  82. Wiki: IEEE 802.11ac
  83. Wiki: IEEE 802.11b-1999
  84. Wiki: IEEE 802.11g-2003
  85. Wiki: IEEE 802.11n-2009
  86. Wiki: Interference (communication)
  87. Wiki: Gigabit Ethernet
  88. Wiki: Kernel panic
  89. Wiki: Local area network
  90. Wiki: MAC address
  91. Wiki: Non-volatile random-access memory
  92. Wiki: Radio jamming
  93. Wiki: Reboot (computing)
  94. Wiki: Return merchandise authorization
  95. Wiki: Secure Shell
  96. Wiki: Service set (802.11 network)
  97. Wiki: Standard operating procedure
  98. Wiki: Universal Plug and Play
  99. Wiki: Wide area network
  100. Wiki: Wi-Fi Protected Access
  101. WikiDevi: ASUS RT-AC1200GP
  102. WikiDevi: Broadcom
  103. wikiHow: How to Connect Two Routers
  104. WikiDevi: D-Link DWA-171 rev A1
  105. YouTube
  106. ZDNet: Gigabit Wi-Fi: 802.11ac is here: Five things you need to know
  107. 電腦王:Wi-Fi 5GHz 你支援嗎?產品支援度混亂,小心頻率不相容
  108. 電腦王:無線路由器怎麼設定?拆機器、教術語,基本功一次搞懂
  109. 友情牌 10 吋壁掛循環扇 KG-1064
  110. 閒聊Wi-Fi
  111. 無線網路新趨勢–漫談 802.11ac 新協定 by 臺灣大學計算機及資訊網路中心游子興
  112. Feature Image: Wikiimages on pixabay
Advertisements

517 thoughts on “Asus RT-AC1200G+ Random Reboots Automatically

  1. RT-AC1200G+ here, same behavior.
    Not sure if it adds something, but I`m using my RT-AC1200G+ in AP mode (connected to a Mikrotik router) and it also is rebooting (once or twice a day)… so I guess the problem is not related to any router feature but the wifi itself.

    I just changed the 5G band to 40Mhz too, let`s see what happens.

    Like

    • Hi,

      Thanks for sharing! I do see less warning with my latest configuration but it reboots last night again. I will update my post later.

      Keep in touch! 🙂

      Like

      • sorry, I dont now why the replly button does not appear the last post you talked to me, but here it is:
        firmware 10680 running
        I lost the countermark because of an power outage, but since then it is running stable in the last 5 days 4 hours 1 minute(s) 34 seconds. It sounds very promising.

        Liked by 1 person

      • Hi hitchhiker,

        About reply, that’s a bad design of comments in WordPress.com. You have to use the reply button on the top (not the first) comments and it will append after.

        About stable report, thanks. I have update your record. It’s okay without continuous run time. I turn mine off every night. 🙂

        Have a nice day!

        Like

      • Hey Amigo,
        A new firmware version has been released.

        Version 3.0.0.4.382.15525
        2017/08/1513.63 MBytes
        ASUS RT-AC1200G+ Firmware version 3.0.0.4.382.15525
        Release note:
        – Improved system stability
        – Please note, for security concern, login from WAN with http was removed. Please using https.

        Has anyone tried it?

        Regards,
        Jefferson

        Liked by 1 person

      • Hi Jefferson,

        As you may see, there are many people share their experience with this new official firmware. It seems stable but high RAM usage. I haven’t see any reboot issue yet.

        Like

      • Hi Amigo,

        Just to report:
        I upgraded my two routers to the new 3.0.0.4.382.15525 firmware version.

        1 day and 19 hours up and running perfectly…

        It sounds promising…

        Liked by 1 person

      • Hi hitchhiker,

        Sounds good. But let’s test it for a few more weeks, some has report the new driver is same to unpublic firmware which should resolved the stability issue in previous firmware.

        Like

    • Hi, just FYI. It’s almost 8 days since last configuration. No restart. Is there any good news from you?

      Like

      • Yes!!!
        It is weird because just like in your case, in the following 4 days after I modified the 5GHz Channel bandwidth to 40Mhz it keep rebooting some times… and then it some day suddenly stopped…
        The current Uptime is 4 days 1 hours 55 minute(s)… let`s hope it stays like that… I will keep monitoring it in the next weeks.

        Liked by 1 person

    • Hi hitchhiker,

      PawelW offers a link to a new beta firmware. One reports it works great in Taiwan. I check with Asus Taiwan there is no official link to this firmware to public.

      After 13 days, it finally crash. I check many threads and think maybe it’s related to the wireless driver in Linux kernel. Therefore, I set 5GHz to 802.11n only. See if the compatibility gets any improvement.

      You may check my latest update in this post.

      Cheers!

      Like

    • Hi hitchhiker,

      How is your router? I have update to 2017/5/13 and disable 2.4GHz on RT-AC1200G+ for the last test. I use LAN to LAN to use my old RT-AC51u to provide 2.4GHz capability.

      Wish to hear form you soon!

      Like

      • Hi Amigo,

        Unfortunatelly after some days it restarted again. I`m following your thread including the beta firmware PamelW mentioned, but I didn`t apply it to my config. I`m setting up a new network (cabled one, with the mikrotik as main router and two RT-AC1200G+ working as Acces Points). After it is ready I`ll let you know if it is working fine. I hope Asus is paying attention to this problem too… It looks like they don`t give the same priority to this entry-level router that they do to the high-end ones. Well… I hope I`m wrong and they shut my mouth up…

        Cheers

        Liked by 1 person

      • Hi hitchhiker,

        Thanks for your comments. I have update Test Report in Comments to include your contribution.

        mikrotik is famous for its quality. I also change my network as mentioned in Change Network Map. Unfortunately, 5GHz becomes unstable within 36 hours as recorded in On 2017/5/26, Research on syslog.log, Change Network Map, Testing Cooling System, and an Unstable 5GHz Issue.

        I remember you said 40MHz is more stable. Are you going to use this option?

        Have a nice day!

        Like

      • Hi hitchhiker,

        May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

        If you haven’t test the new version, tycjan has shared the new firmware in his comment, here, and there.

        Have a nice day!

        Like

    • Hi hitchhiker,

      It has been 8 days seems our last talk. The new firmware 380_10680 seems fine to me. Do you want to give it a try? It works great on my RT-AC1200G+ for 4 days. No reboots yet.

      tycjan has shared a link in his comments. You may try if it still works.

      Wish it helps!

      Like

      • Hi Amigo,

        Just a question before I try the last firmware:
        Does the “RT-AC1200G+ with RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx” firmware could be installed in any RT-AC1200G+? Mine was bought in Europe, not sure if it is compatible. Is that firmware in english?

        Best Regards!

        Liked by 1 person

      • Hi hitchhiker,

        We have some feedback from Europe. I think it should work. Please check the RT-AC1200G+ with RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx Stability Tracking section. caliban has detail explain about EU regulation and channels.

        I am currently using English interface. Therefore, It does provides English menu.

        Wish it helps!

        Like

      • Forget my previous message… I just did it.
        “RT-AC1200G+ with RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx” is up and running, I will give you a feedback in the next weeks.

        Thanks!

        Liked by 1 person

      • Hi hitchhiker,

        I think it’s a good question. Maybe someone read the post have doubt on it.

        Have a nice day!

        Like

  2. Hello gentlemen.
    I have the same problem as you. Basically I had it. I think the problem has been solved. No restart for two weeks. I turned to Asus for help. They sent me a link to two beta firmware.
    https://www.asuswebstorage.com/navigate/s/09BC5426FD364BB486805459E402EDAFY
    First I uploaded newer firmware (RT-AC1200G + _9.0.0.4_382_9959-g648a694.trx), but I quickly resigned from this version because no device saw the ASUS network at 2.4GHz. The ASUS5 band on the 5GHz band was clearly visible.
    I have downloaded the second, older firmware (RT-AC1200G + _9.0.0.4_380_7415-gc949b277fca.trx) and there are no problems with this release.
    Both networks (2.4 and 5) are available. Restart for two weeks lack. The router is working stably.

    Sorry for my English. (Google translator … :))

    Like

    • Hi PawelW,

      Thank you for sharing the good news of beta firmware. I expect to see a stable version for local user soon!

      For those who want to use the link, please check with your local wireless regulations. Selectable channels are probably different.

      With my own configuration set, it doesn’t restart for 13 days with firmware 3.0.0.4.380.4089 yet.

      I will wait for two more weeks. If there is no restart, I will write a new post and share my configurations in a better way. This post looks more like a diary and difficult to know what I have set. 😦

      Please keep in touch!

      Have a nice day!

      Best regards,

      Amigo

      Like

  3. Unfortunately after 19 days of work the router has restarted … 😦
    This beta firmware does not solve the problem.

    Like

    • Hi,

      Which link? If you mean firmware, it seems to be removed by someone. Nothing I can do. 😦

      Do you have the same reboot problem?

      Have a nice day!

      Like

    • Yes, it has been removed. Please use the “Reply” under your first comment for future questions. It would be easier to read.

      Like

  4. i have Yellow led right (ì) blinking in router menu’ is a problem???i have called italian asus support in no….

    all you have same??? my router is attacched on modem router via wan port.. thanks

    Like

    • Hi massimo,

      Try to reset firmware to factor default which I have explain in “On 2017/4/18, Fine-tune UPnP, Wireless Configurations, Reset Router”.

      Use reply in your thread instead of creating a new thread or nobody knows what’s going on.

      Like

    • Hi massimo,

      I didn’t hear any thing from you for a long time. Did you solve your reboot? If not, come back and see our latest test configuration and share your experience, please.

      Have a nice day!

      Like

    • Hi massimo,

      May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

      If you haven’t test the new version, tycjan has shared the new firmware in his comment, here, and there.

      Have a nice day!

      Like

    • Hi nikisi,

      Because PawelW has report it reboots after 19 days and both you and massimo are testing the new firmware, I decide to print this post (14 pages, OMG) together with the machine for RMA. I will keep track of my RMA results and see if we may solve this issue together.

      Keep walking! 🙂

      Like

  5. Hello All,

    I have got the same issue I manage to replace my device in the shop however new device got the same issues… I was trying to play with several settings. I will try to replace firmware form above links.

    Like

    • Yes, I believe a hardware replacement won’t work. I have even disable 2.4GHz and 5GHz and still reboot.

      Like

      • Can you backup the complete programmer firmware from SSH to me?
        FULLFlash BACKUP
        你能不能備份一下你路由器的CFE,或者備份你路由器的完整編程器固件.不是升級固件謝謝。

        Like

      • No, I cannot. It’s already in RMA now. I didn’t have my router and don’t know when it will return. They want to keep it for study. You may try HKEPC. Maybe local community may help.

        Like

    • Hi Alew,

      May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

      If you haven’t test the new version, tycjan has shared the new firmware in his comment, here, and there.

      Have a nice day!

      Like

  6. Hello Amigo !

    Did you inform this Asus research and development engineer that this issue is not only for your unit, but the problem is global for all RT-AC1200G+ units all over the world?

    Liked by 1 person

    • Hi nikisi,

      No, I didn’t. But I did inform support engineer that this is a worldwide issue and a more stable firmware are available in my post. I don’t have a chance to talk to any research and development engineer yet. Do you have any RMA receipt that I may proof you have the same trouble? They need reproducible issues or official tickets as leads.

      Anyone follow this post, please paste a link to your case ticket image. It helps to collect facts to proof.

      Are you going to join?

      Like

    • Hi nikisi,

      chinwah at HKEPC think it’s just to me, a single issue, instead of a global issue because his is fine.

      Wish it helps!

      Like

      • Hi Amigo!

        I am following your blog post about this router for a while now, and I can confirm you that the reboots you had are NOT a single issue! I live in Hungary, and have this very router model in the family, and we experience the very same reboots regularly, using the latest stable firmware!

        Moreover, the forum thread of this router on a local, popular hungarian IT site is full of people who also suffer from this issue, starting from the first firmware version, until the very last betas posted by Pawel above. It is DEFINITELY a global issue, that ASUS should be aware of!

        A temporary solution (or more like a workaround) for many people here was to configure a scheduled daily reboot of the router, to a time when the router is not used by anyone (like 2AM). But that is hardly a fix!

        Liked by 1 person

      • Hi Rokapet,

        Thank you for your kindly support! Could you please help me to invite to leave a message about their country and crash model on my post? I want to make this post as a hub to collect facts to proof this is not a single issue.

        About your temporary solution, I power off before sleep and turn it on every morning. Still doesn’t work in my case. 😦

        Fernando from Spain suggest me to set “short” in Preamble Type in [Wireless] → [Professional] tab in a reply to my thread. Maybe it helps!

        Anyway, I am so happy to read your comment. Please follow my post, the war is not finish yet. 🙂

        Have a nice day!

        Like

      • Hi Rokapet,

        May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

        If you haven’t test the new version, tycjan has shared the new firmware in his comment, here, and there.

        Have a nice day!

        Like

      • Hi Amigo!

        Unfortunately I can not test beta firmwares. This router is owned by someone else in the family (with no IT knowledge on networks, she can only do power down – power up if needed, nothing more), and it is in active daily use,with a scheduled restart every night, to avoid issues as much as possible. So I will only be able to upgrade to an issue-free stable FW, if there will ever be one! Else, there will be a family war if something is not working. 😉

        Sorry!

        Liked by 1 person

      • Hi Rokapet,

        It’s okay. I understand. 🙂

        Maybe you may follow up this post to track the numbers of day-without-reboot. I didn’t know when Asus will release this firmware on its web site.

        Keep in touch!

        Like

      • Hi Rokapet,

        mouse has report there is a new firmware 3.0.0.4.380.7743 is available on Asus’s web site at 2017/6/22. Do you want to try the new firmware?

        Have a nice day!

        Like

  7. Hello Amigo,

    As I stated in ASUS support forum you’re not the only one to have the random reboots for no apparent reason. In my case happens every one or two days so its definitely a bug in the firmware of the router (I haven’t a clue whether it might be a wireless driver issue or something else) The point is that the router reboots when it shouldn’t. I haven’t sent the router back to ASUS because I read your post about the poor customer support you received. My router has 3 wired devices connected: a NAS, my home alarm system and an IPTV device provided by movistar (my ISP). 2.4GHz Wifi is used by my laptop and smartphone, while the 5GHz wifi is used by my wife’s laptop and smartphone. Every now and then family and friends come by connecting to the router via 2.4GHz or 5GHz wireless so at peak usage times there are 10~15 devices connected which I find that shouldn’t be too demanding.
    The reboots happen even without being connected to the wifi network.
    I found in a smallnetbuilder.com forum someone stating that stability came back changing wifi preamble from long to short, which I’m currently testing (so far 1 day and 2 hours without reboot, which is not yet conclusive)
    ASUS, please, at least have the guts to debug your own firmware code instead of telling your customers to have the router video recorded to check if it is really rebooting…

    Liked by 1 person

    • I have setup preamble to short on the first device and it did not helped. Like Amigo I was fighting with Asus support… Now I am wondering if I should keep this device or simply return it pay additional price and get brand new router from another company. The shop told me that there is such option. I really like this device however such issue is not acceptable. I cannot imagine that non technical user is playing with professional settings spent time with Asus support… This kind of device should be up all the time and you should not remember where you have it. I had an old Linksys wrt54g and it was up for ages..

      Liked by 1 person

      • Hi tycjan,

        Thanks for your report. Linksys wrt54g is a marvelous device.

        About upgrade, yes, it’s absolutely a good idea. I have encourage people to change a different model in 1st RMA: From 2017/5/15 till Now section:

        If you just want to get a stable machine, show my post and highlight what happen today to Asus local support team and see if you could get another model. Leave me a message with a photo to proof if you get one to help others asking for another model.

        I refuse this offer because I want to fix the problem and learn how to reproduce random reboots. I don’t have enough knowledge but google and wiki are my friends. I learn a lot from researching this issue.

        Although some others think we just want free upgrade, no, it’s not. We just need a workable and stable machine. Besides, we already pay and suffer from random reboots for a long time. It’s time to get what you pay for. Go and get another model if you can!

        Please leave me a message with photo about your new model. It helps others to apply a free upgrade by your case.

        If you could help me to answer these questions, it helps me to summarize.

        1. Do you use the USB port? For what?

        2. Where do you get internet signal? cable modem from WAN? cable modem from LAN? USB Tethering?

        Thanks in advance!

        Like

      • Hello,
        I am not using USB port i disabled everything related to it. My ISP provides cable modem via DSL link.

        Liked by 1 person

      • Hi tycjan,

        Woo! Now we have the same network architecture. Check my latest update!

        At least we know USB Tethering is not the key to reboot. 🙂

        Thanks for your quick reply!

        Like

    • Hi Fernando,

      In my reply to tycjan, I have wrote a post Android USB Tethering Drops and Data Usage Crashes, try Reset, Android reboots when it cannot save any more data usage log. Maybe it helps if we remove all these logs? It explains why we have a stable machine for more days after a factory reset.

      I have many device but at most 2 NAS, 2 computers, and less than 3 tablets connecting to RT-AC1200G+ at the same time. If I can perform pressure tests, I will know its limit.

      If he want video, we give it via YouTube and share. I understand random reboot is difficult to capture but it’s not customer’s responsibility. Especially you already confirm it happens by your tech support engineer. If you want us to help, at least you should have a Dr. Watson build-in. The system log doesn’t provide enough information.

      Anyway, my machine is back and I may do more different testing. Stay and follow this post!

      Together we stand!

      Have a nice day!

      Like

    • Hi Fernando Puech,

      May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

      If you haven’t test the new version, tycjan has shared the new firmware in his comment, here, and there.

      Have a nice day!

      Like

  8. Hello,

    I have got the same issue. I was playing with a lot of different settings. After upgrade to 9.0.0.4.380_7415 it was up for four days… Additionally I can say it is my second device, replaced by the shop where i bought it. By the way! I have had open ticket to Asus support and I already informed them that according to the Internet posts i am not the only one person with such issue.

    Seems like they simply don’t care!

    Liked by 1 person

    • Amigo,
      For your information I have reopened my ticket and provided a link to this thread. I admire your involvement. I will connect my RPI directly and I will set up remote syslog on it. I did that when I had issues with my first device.

      Liked by 1 person

      • Hi tycjan,

        This is a great idea! I know how to setup a syslog server on my Asustor AS-5102T. To configure syslog in RT-AC1200G+, check [Administration] → [System] tab. There is a Remote Log Server.

        BTW, as I have mentioned in Android USB Tethering Drops and Data Usage Crashes, try Reset, Android reboots when it cannot save any more data usage log. Maybe it helps if we remove all these logs? It explains why we have a stable machine for more days after a factory reset.

        Don’t forget, RMerlin refuse to support RT-AC1200G+ because its limited storage. Maybe we should try a factory reset every few days to clean the mess. Or maybe we may use ssh to delete these logs?

        Thanks for your good idea!

        Like

      • I will post a full log via pastebin after few days or next reboot. Polish Asus support is going to send me the same beta firmware you altredy posted… ;/

        Liked by 1 person

    • Hi tycjano,

      Because it’s here and there, they may ignore single case. That’s why I encourage people with the same problem to leave a message here. Together we stand!

      Have a nice day!

      Like

    • Hi nikisi,

      Thanks for your report. If you could help me to answer these questions, it helps me to summarize.

      1. Do you use the USB port? For what?

      2. Where do you get internet signal? cable modem from WAN? cable modem from LAN? USB Tethering?

      Thanks in advance!

      Like

      • Hello !

        1. Im not using USB port i disabled everything related to it.
        2. My internet is from optical fiber net.
        Optical cable —> media converter —> LAN cable —> LAN 100Mbit Switch —> LAN cable —> RT-AC1200G+ WAN port ( WAN Connection Type: Automatic IP )

        Usualy i have connectet to router:
        1x Desktop PC ( LAN port )
        1x RaspberryPI as home medica player ( LAN port )
        1x Ipad Mini ( Draft N 5Ghz )
        1x Smartphone ( AC 5Ghz )
        2x Smartphone ( Draft N 2.4Ghz )

        Liked by 1 person

      • Hi nikisi,

        Thanks for your sharing!

        I think we may confirm that USB tethering is not the key factor.

        BTW, do you know how to access RT-AC1200G+ via serial? My wireless connection lost when reboot. I want to capture more information before crash. syslog.log doesn’t help. I notice you have Raspberry Pi. Do you have any experience debugging embedded Linux?

        You are so cool to run Raspberry Pi at home!

        Like

  9. In my opinion its obvious that FOR SURE THERE IS a problem with hardware and firmware in this model.
    On daily basics i am dealing as IT Administrator hence I am a bit more involved in the router configuration. I suppose that more people simply don’t see that their router just went down for few seconds even so its up and they are still connected… They are browsing websites so they don’t feel that something went wrong. Even we manage to work out a working solution after changing each value/parameter it would not be a professional approach from the Asus side. They ignore all information like”i am not the only one person witch such issue” Yesterday the guy from the support side told me to do not count on the Internet posts. LOL!
    Anyway I will continue to keep an eye on my device. Rsyslog is collecting messages.

    Liked by 1 person

    • 我的1200+剛剛真的自己重新開機,手機玩到一半wifi消失,重連登進去看…設備開機時間不到一分鐘,時間歸回2015

      這邊沒提我還真的沒去注意過@@

      Liked by 1 person

      • Hi Kobe,

        May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

        If you haven’t test the new version, tycjan has shared the new firmware in his comment, here, and there.

        Have a nice day!

        Like

    • Hi tycjan,

      Thanks for your investigation!

      I think you are right. If he is not always online, he probably won’t notice the difference if WAN is reconnected or wireless becomes stable again when he need them. I do see freeze once but not as often as reboot.

      Let me quote from MIB – Man In Black:

      The only way these people can get on with their happy lives is that they “do not know about it!”.

      That’s why I post my receipt to proof my RMA. Together we stand!

      Have a nice day!

      Like

    • Hi nikisi,

      PawelW has test this. He said 2.4GHz SSID is not visible. Usually it’s related to channel. Every country has different regulation, I mean open channel. Therefore, if you cannot see SSID, try different channels.

      Wish it helps!

      Like

    • Hi tycjan,

      I update your suggestion to disable SSH. It match my experience. My longest no-reboot period is between 2017/4/24 to 5/8 for 14 days. Then I enable SSH daemon and it reboots every 2 to 3 days after 2017/5/8.

      Thanks for sharing!

      Like

  10. Amigo,

    love your dedication on trying to solve this. 🙂

    However, I haven’t read anything about setting “Authentification Method” to http.

    I’m not talking about Wifi-Settings, but the setting in Administration -> System. (secure login to router or not). SSH-daemon is disabled as well.

    This tip was mentioned in Asus-forums a few months ago and with it, I got up-times as high as 10 days with latest official firmware. Sometimes it will reboot after a day or two and then it stays on for another week. Tried both beta-firmwares, but they didn’t change anything.

    Not a solution (I don’t think there ever will be – Asus clearly has no interest, which everybody should remember when buying new stuff), but it made a difference.

    Liked by 1 person

    • Hi caliban,

      Thanks for your suggestion. My auto-restart history also support your idea. I didn’t enable SSH Daemon until 2017/5/8. My longest no-reboot period is between 2017/4/24 to 5/8 for 14 days. Then it reboots every 2 to 3 days after 2017/5/8.

      Together we stand!

      Have a nice day!

      Like

      • For me setting auth-method to “http” clearly made the difference, SSH was disabled anyway as I don’t need it.

        Regarding number of running processes:

        I’m sure it helps keeping them as low as possible. But I’m using this thing in AP-mode from the beginning, yet I still got the same frequency of reboots as everybody else (from twice a day to every second day). I really dumbed it down as much as possible, all USB-Apps disabled, no NAS just some Wifi-Clients on both bands.

        So there’s a lot less happening compared to a system in full Router-mode, but still reboots galore. To me that says the main bug sits elsewhere, but more processes don’t add to stability, of course.

        Liked by 1 person

      • Thanks for your sharing. It match my test configurations from 2017/3/21 till now. There must be something wrong with kernel, maybe drivers compiled into kernel. Because of the protection ring, syslog has nothing to do with errors in kernel (ring 0) or driver (ring 1).

        As I have explained in 2017/5/8, b43 supports BCM43217 @ 2.4GHz but no BCM47189 @ 5GHz. That’s why I think it’s related to driver issue.

        For drivers in DOS era, we may force it to comparability mode to skip modern features. If we are lucky, bugs in modern functions won’t be execute.

        Unfortunately, reboot happens even I disable one of the radio (2.4GHz or 5GHz). I am running out of hope.

        Therefore, I decide to test these two un-official firmware to see if the kernel has been upgrade. Detail in my update on 2017/5/28.

        I believe we have try our best. Asus request to provide video captures of lights and I did (on 2017/5/26). Let’s wait and see.

        Keep in touch!

        Like

      • BTW, if you don’t like it, do you know you may ask Asus to exchange for another model? they want to offer me another model but I want to test more and refused. You may show them my post and ask for another model if you like.

        Like

      • Hi tycjan,

        Do you mean to keep as less running process/daemon as possible?

        Have a nice day!

        Like

    • Woo, sounds like my issue on 2017/5/12. It requires a power reset to make it work again. Thanks for sharing!

      Like

      • Yes, less processes is equla to “more up time”. Maybe memory issue? What about alternative firmware? I haven’t found any other software…

        Liked by 1 person

    • Hi 賴雨賢,

      May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

      If you haven’t test the new version, tycjan has shared the new firmware in his comment, here, and there.

      Have a nice day!

      Like

  11. Hey Amigo,

    It’s Norbert here from Hungary. A guy linked your site on a popular local IT/HW site here at Hungary, that’s how I ended up reading your posts. I can confirm there are plenty of guys here facing stability issues with this router here as well – many of them got rid of it already due to the issues.
    I can join the community both by owning this device, and by facing the random restart issues as well.
    It started with the stock fw, I kept upgrading whenever new official fws were released, currently running with 3.0.0.4.380_4089 but of course still having the random restarts – however, I admit it is less frequent with this fw compared to the stock one. (with stock fw it was very frequent, sometimes multiple times within a day, and largest uptime was around 5-7 days ever. With current fw my personal record was around 15 days as I recall.)
    I applied a weekly automatic restart – can’t really tell if it makes it any more stable or not, probably this is just a bitter action as I’m running out of ideas. Sometimes it still restarts in between the weekly restarts, but I’m not monitoring the uptime anymore, I only notice it when it happens while I’m actually online.
    Mine also running with 5 GHz currently off, as I was not using it nowdays, but it’s not making any difference as I see.
    Earlier a few times I had the impression that the restart happened when I was causing peaking traffic for a longer period, but it might be coincidental. In some cases when having high traffic it did not restart but simply froze and needed a hard reset. (could not load the management page while the device was answering to ping, and net traffic was dead)
    It is really a shame, because other than this it’s a lovely device for the price. It’s my first ASUS router, but I’m kind of sure that the last one as well. 🙂
    Looks like it’s a general issue, ASUS should have recalled the whole product line if the issue is hardware related and not being able to fix it with fw updates – which is is my sad assumption now.

    Like

    • Hi Norbert,

      May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

      If you haven’t test the new version, tycjan has shared the new firmware in his comment, here, and there.

      Have a nice day!

      Like

      • Hi Amigo,

        sorry, haven’t checked back here in a while.
        Can’t tell yet I’ve just applied this fw a few mins ago, but now I did and turned off the weekly scheduled restart I had previously. Will get back to you later with my findings.
        Thanks for your endless efforts on this!

        Liked by 1 person

      • Hi Norbert,

        It’s nice to hear from you again! Remember to reset NVRAM and use physical button to power-off and on your device.

        Have a nice day!

        Like

  12. Just a small quote…..
    The 169.254.26.113 you mentioned is likely the pc’s local auto-assinged IP before the router’s DHCP Server finish initializing.
    When physical(either wired or wireless) connection is established, the OS needs an IP address to communicate, but if DHCP Requests don’t return with a valid ip address, the PC can only create it’s own for temporary use.
    When the router’s DHCP server finishes it’s preparing sequence, it will return(assign) an IP address in range to your PC on receiving a DHCP request.

    Like

    • Hi Lenaius,

      Thanks you! I add your answer to my post. Please help me to correct more if you find any thing misleading or needs more explain. I appreciate your help!

      Have a nice day!

      Like

    • Hi tycjan,

      You are awesome!

      I have compare the difference with 380_7415:

      1. kernel: Linux version 2.6.36.4brcmarm (root@asus) (gcc version 4.5.3 (Buildroot 2012.02) is still the same
      2. kernel: eth1: Broadcom BCM43227 802.11 Wireless Controller 9.10.178.61 (r667680) has been upgrade from 9.10.178.27 (r584393)
      3. kernel: eth2: Broadcom BCM43c8 802.11 Wireless Controller 9.10.178.61 (r667680) has been upgraded from 9.10.178.27 (r584393)

      Seems we may have hope on this new firmware.

      I will update my post tonight. Too many issue to write that I have no time to feedback to community. Could you please help me to spread the word in Europe? I don’t know local language and communities. I will focus on the post today!

      Anyway, thank you for your great news and support! Together we stand!

      Like

      • Above Changelog plus they added Repeater-Mode.

        But what I really don’t understand: If this is a beta-version, intended for testing and bugfixing – why is the syslog still pretty empty?!

        I finally contacted Asus-Techsupport and got the standard reply about flashing latest firmware, doing a hardreset blabla – well, I played along, they have their playbook.

        Took it about 24 hours to reboot, so they got a very informative syslog from me:

        May 31 23:02:34 disk_monitor: Got SIGALRM…
        Aug 1 02:00:09 syslogd started: BusyBox v1.17.4
        Aug 1 02:00:09 kernel: klogd started: BusyBox v1.17.4 (2016-11-03 15:46:45 CST)
        Aug 1 02:00:09 kernel: Linux version 2.6.36.4brcmarm (root@asus) (gcc version 4.5.3 (Buildroot 2012.02) ) #1 PREEMPT Thu Nov 3 15:55:43 CST 2016
        Aug 1 02:00:09 kernel: CPU: ARMv7 Processor [410fc075] revision 5 (ARMv7), cr=10c53c7f
        Aug 1 02:00:09 kernel: CPU: VIPT nonalia
        |…|
        Aug 1 02:00:10 ntp: start NTP update
        Jun 1 09:34:09 rc_service: ntp 362:notify_rc restart_diskmon

        Which means the reboot happened at around 9:30, but there’s no indication what happened before. Just flashed the new beta and still the same? What developer doesn’t want any feedback apart from “nope, still rebooting”?!

        Liked by 1 person

      • Hi caliban,

        If you need a more precise time, check the “On 2017/5/29, Reboot, Identify Crash Time with Another Router” section of this post. With two router connected with each other, you may identify the exact time by logs about Link Status Changed.

        For syslog.log, check the section “Research on syslog.log”. It is useless to catch kernel panic because kernel panic happens in protection ring 0 and 1, while syslogd is on 3. There won’t be any message sent to syslog once kernel panic happens.

        But I did use a surveillance system to capture the crash moment which the R&D engineers ask for, it’s in the “Reboot” section of 2017/5/31. It doesn’t provides any meaningful light patterns to me.

        For kernel panic, you need a serial cable direct connect to router. This is how embedded system developers communicate with the circuit board, like Raspberry Pi.

        Do they offer you any option to other models?

        Keep in touch!

        Like

      • Nope!

        You are awesome 😉 I just upgraded firmware without investigation whats new on there 😉
        I have got this firmware from the Asus support (PL) I have re-opened my incident.

        Liked by 1 person

    • Good news !

      Who provide this new firmware to you ?
      Do you know something about changelog and whats new in this f/w ?
      Do you notice some changes on first sight ?
      We waiting for first impressions from you.

      Liked by 1 person

      • Hi nikisi,

        I got call from Dolly Chu who works for Asus this afternoon. She provides me the same file. Therefore, I may confirm RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx does exist but not public yet.

        I didn’t see any change-log or release notes. But after following my steps 0n 2017/5/28, I found some importance change as mentioned above reply to tycjan.

        I notice at 17:06:42, my mac mini has been disconnected by RT-AC1200G+ and reconnect to RT-AC51u without any notice. Therefore, I don’t think it’s ready as a beta. Maybe just a night build.

        I will keep my eyes on it and update my post ASAP. But I don’t have time to sync with communities, if you could help me to post any update to local community, it helps to catch everybody’s attention.

        I do worry that Asus is playing hunger marketing. Make us feel we are the one who have special access to early beta and get free marketing by us.

        Hunger marketing is a technique very popular in Asia especially China, Taiwan, and Japan. They sell prototypes with limited quantity. Young people are crazy for them because it makes them looks different and special in community or among friends. They don’t care the quality because they pay for rare rather than useful.

        Therefore, until I see a stable, official, public firmware, I encourage anyone with the same issue to send back for RMA immediately. Show them my post to save your time.

        Just my two cents.

        Like

  13. @Amigo:

    Okay, useless syslog is a design failure of Asuswrt then. ^^ Even if you can’t catch a kernel panic, one would think they’d like to get as much info as possible about what happens right before.

    Regarding your question if they offered me another model:
    Nope, not yet. But I didn’t get an answer to my report incl. syslog neither so far. Also, the Tech who answered me didn’t write anything that would imply it’s a known problem, they’re working on it or that he’s had a look at your diary or even the threads in the Asus community. Classic “let’s first rule out it’s anything you did” standard text. 😉 Maybe that will change now.

    Liked by 1 person

    • Hi caliban,

      I think syslog is for application errors while debugging a kernel panic needs a cable to direct communicate with router. It’s what I learned from another wireless engineer in local group and RMerlin. When we are talking about reboot, syslog is the first thing customer service needs to check but R&D engineers know it’s worthless.

      I am very familiar with service workflow in CRM. I think the first tier support needs follow company SOP. If you refuse to close the case, it will get promoted after days. Then a senior support will check. Be sure to leave your case in open status. When they see many open case with the same product model, a higher level manager will get warning from system automatically. If he didn’t solve it, his KPI will become very ugly.

      Therefore, I encourage people to return or replace a new one ASAP. Once they notice the increasing tickets are all related to RT-AC1200G+, he will begin to investigate. Remember to show them my post to help the supervisor get a lead to follow.

      Once it becomes a burden cost to his department, the Customer Service VP will bring these reports to talk to CEO or R&D VP. Then the product manager will be assigned to handle.

      The key is to show them my post and video to save your time. There is no way to invest a surveillance system to capture and your precious time to create another diary like this.

      Just my two cents.

      Like

  14. Okay, just noticed that no device can see 2.4 GHz Wifi with 9.0.0.4_382_10680.
    AFAIK someone had the same problem with 9959, did depend on region maybe (Europe here).

    Liked by 1 person

    • Can you try to manually assign channel in 2.4GHz? I also use Android WiFi scanner to scan visible SSID to help as mentioned in 2017/5/31 section in this post.

      You may find available 2.4GHz channels in your local wireless regulation government department web site or try “Channel spacing within the 2.4 GHz band” in Wiki in your local language.

      Wish it helps!

      Like

    • Can you call local tech support and ask for this firmware for your country? Don’t give up! We are here to help!

      Like

      • Thanks, but it’s definitely a firmware thing.
        After finally figuring out that only flashing any older firmware (incl betas) plus resetting NVRAM via WPS-button (important!) brings back 2.4GHz for me, I tried again to make sure:
        During initial setup with v10680 (used reset-button after flash) the 2.4GHz-Network is there, but as soon as it reboots it’s gone (and stays gone) until you flash older FW and clear NVRAM. Reset-button is useless here.

        Also, I’ve never seen it use channel 132 on 5GHz before. That’s perfectly legal here in the EU, but auto-setting never went higher than something in the 120s. Manual tops at channel 112. So something is different with this beta. 🙂

        Liked by 1 person

      • I just tried again – it works for me too if I just flash the FW and let it be.

        If I use reset-button or clear NVRAM after flash, 2.4-band is gone (Router says it’s broadcasting, but nothing can see it).

        Liked by 1 person

      • Hi caliban,

        That’s great! I have add your find out in related sections. Thank you!

        Let’s try how long it won’t reboot!

        Like

      • Hi AsusRT1200g+User,

        Thanks for sharing! Did you reset NVRAM? caliban reported that if just flash without resetting NVRAM, it would be fine. Could you please confirm this?

        Have a nice day!

        Like

      • This is kinda emberassing, but the missing 2.4-SSID after resetting 9.0.0.4_382_10680 (or clearing NVRAM) is solved very easily:

        – power button off
        – power button on

        Who thinks of that after just resetting the damn thing to what’s basically factory defaults?! 🙂

        Well, at least I have a clean v10680 running, let’s see how it goes. Guess nobody had a reboot since it popped up ?

        Liked by 1 person

      • Hi caliban,

        Do you mean even flash the new firmware and reset NVRAM, you just need a power reset to see 2.4GHz SSID? No need to keep NVRAM after upgrade firmware? Please confirm!

        I haven’t see any reboot yet and is monitoring if there is any drop line.

        Have a nice weekend!

        Like

      • Yes, what I did was:

        – flash new firmware
        – clear NVRAM
        – go through inital setup (router-login, set internet connection, set SSIDs)
        – router reboots
        – 2.4-SSID is not visible
        – simple power reset
        – 2.4-SSID is visible to all devices

        Currently I’m playing around with channel-settings for 5GHz. Noticed that connection is really slow on channels 100-112. It starts at 866 mbit with my laptop after handshake, but quickly goes down to ~200, then even lower to 54 and 19,5. Never gets higher than 200 during use and than back down.

        All is fine on channels 36-64 (between 650 and 866 mbit, depending on location). Also, on the 100-channels the router-UI is laggy. But hey, it’s a beta. 🙂

        Good to hear it’s stable for you so far. Should be running around 36 hours now – not unheard of, but still promising.

        Liked by 1 person

      • Hi caliban,

        Thanks for your feedback. I have update your solution in post.

        Your speed test is excellent! I didn’t notice the difference switching channels. It’s very useful information!

        Hace a nice day!

        Like

      • And another detail about missing 2.4-SSID:

        If you at some point let it reboot via UI, it will be gone again. Now power the router off and on via HW-button on the back, it’s visible again.

        Like

      • Hi AsusRT1200g+User,

        May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

        Have a nice day!

        Like

    • Hi Lenaius Chang,

      Nice sharing!

      In “Research on syslog.log” section, I found log_level won’t be restored after reboot. Therefore, I think it might depends on models.

      Have a nice weekend!

      Like

    • Hi Lenaius Chang,

      May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

      If you haven’t test the new version, tycjan has shared the new firmware in his comment, here, and there.

      Have a nice day!

      Like

  15. Hello !

    Just upgrade to 9.0.0.4.382_10680 without nvram clear and reset. All SSIDs work fine for me. Im from Europe. Im testing and will provide feedback here.

    Liked by 1 person

    • Hi nikisi,

      Thanks for your feedback! Let’s see if the new driver make it more stable!

      Have a nice weekend!

      Like

    • Today I recieved this message in general log:

      Jun 3 13:33:57 rc_service: httpd 213:notify_rc start_lltdc

      I have no idea what it means. I have this massage for first time.

      Liked by 1 person

      • Hi nikisi,

        It’s a Link Layer Topology Discovery Daemon. The source code is here.

        It’s a reference implementation of the Ortelius topology discovery protocol.

        Just my two cents.

        Like

    • Hi nikisi,

      May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

      Have a nice day!

      Like

    • Hi Charles,

      It works great for me! No reboots after 2017/6/1. I do have suspicious for wireless drop but couldn’t confirm.

      Could you please leave me a comments if you have any reboot?

      Have a nice day!

      Like

    • Hi Charles,

      May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

      If you haven’t test the new version, tycjan has shared the new firmware in his comment, here, and there.

      Have a nice day!

      Like

  16. ,,Thanks for sharing! Did you reset NVRAM? caliban reported that if just flash without resetting NVRAM, it would be fine. Could you please confirm this?”

    I didn’t reset NVRAM. I just only flash new firmware and my router working great.

    Liked by 1 person

    • Hi AsusRT1200g+User,

      Thanks for your confirm! It’s good to know everything works fine.

      Have a nice day!

      Like

  17. Hi Amigo,

    I’m using AC1200g+ router as well. Where can I download “RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx” ? Can you share me the link?

    Thanks in advanced

    Liked by 1 person

  18. Hi AsusRTUsers,

    Taiwan. My RTM has been sent to Kaohsiung. I got call from Taipei on 2017/6/1. Seems they have two different cities support my case.

    Have a nice day!

    Like

    • I contacted Singapore ASUS and they are not aware of this current issue.

      Advice was given is to send into their service centre to get a 1 to 1 replacement.

      Anyway, I’ve upgraded to the beta firmware and will monitor if there is any reboot

      Thanks Amigo

      Liked by 1 person

      • Hi AsusRTUsers,

        You may show them my post and encourage to confirm with Dolly Chu in Taipei.

        Yes, 1 to 1 replacement is the SOP for tech support but it won’t work.

        Could you please feedback if you have any issue or leave a comments if everything is okay after days? it would help others who read my post. Thanks in advance!

        Have a nice day!

        Like

      • Hi AsusRTUsers,

        May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

        Have a nice day!

        Like

    • Hi tycjan,

      It’s good to know everything is fine from you! Congratulation!

      May I know why you want to disable 5GHz? Is there anything wrong? I, on the other side, disable 2.4GHz but keep 5GHz.

      Thanks for sharing!

      Like

    • Hi tycjan,

      What’s your firmware xx82? It seems you have a different firmware. I am using 10680. Previous 7415 and 4089 are unstable to me. Do you mean they give you another 10682?

      Could you please confirm your firmware?

      Like

      • Sure! I am using; 9.0.0.4.382_10680 link should be posted here. This fw has been provided by polish asus supporter.

        Liked by 1 person

      • So far so good.
        5 dni 0 godziny 47 minuty 28 sekundy
        Please note I disabled 5GHz after current firmware installation due to fact I have got less devices that support this range.

        Liked by 1 person

      • Hi tycjan,

        Thanks for your feedback.

        I just create a tool to help heat dissipation, reset and reconfigure RT-AC1200G+ for next test.

        I notice the displayed icon and device name in MAC filter won’t be update util a device connected to RT-AC1200G+ even you have already change. I will check it again when more device connected to RT-AC1200G+ tomorrow.

        I will update my status every week if no reboot nor any other issues.

        Keep in touch!

        Liked by 1 person

      • Hi tycjan,

        May I know how many days without reboot till today after applying the new firmware 9.0.0.4_382_10680? I want to make a list to share.

        Have a nice day!

        Like

  19. ASUS support isn’t the same everywhere.
    Finally got an answer from so-called “Tech-Support” in Germany and I’m slightly pissed. They told me I have to RMA it to the dealer, meaning: “We’re not interested, good day to you.”

    Since I bought it more than 6 months ago, Amazon could very well say “prove to us that the problem existed from the beginning”, and even if they play nice and send me a new one, that wouldn’t solve the reboot-bug until a new FW is officially released.

    What they’re actually trying to do is close my ticket and mark it “solved”. The latest beta may be looking good, but ASUS won’t be on the list for future tech-purchases. At least I got this Router pretty cheap, so the damage is limited.

    Liked by 1 person

    • Hi caliban,

      True! I know it’s the SOP. That’s why I write such a long post detail describing every test and publish videos on YouTube. I have a good story. Then use my CRM professional to accelerate. Finally, it catch Asus’s eye.

      If the latest firmware is stable, we are lucky.

      BTW, I have google before buying RT-AC1200G+ but didn’t see any media review talk about reboot. Maybe I should check communities instead.

      Anyway, it’s a good lesson to me. Although I have learned a lot of wireless technologies and kernel panic during writing, it’s better to avoid.

      Keep in touch!

      Like

      • Got another reply, after basically asking them “what the f…?!”
        They say they hadn’t read my other reply, where I talk about latest beta, when they told to me RMA it.

        But, and here’s the kicker, they go on telling me that I can continue using the beta, if I want (never told them I am). And if I have any other question, don’t hesitate to contact us.

        That’s it. Nothing more. Just “use the beta, if you want”.
        They didn’t even send me the link for the beta or told me that they’re working on it. Guess from their point of view it’s better to keep me in a state where I load firmware from some “dubious” web source… (should there ever be a hw-defect: “sorry, 3-year-warranty is void, unofficial firmware found”)

        Ah, the beta: Still working fine, at least that’s a plus. 🙂

        Liked by 2 people

      • Hi caliban,

        It’s nice to hear that the new firmware is stable till now. May I know how many days without reboot till 2017/6/7 after applying the new firmware? I want to make a list to share.

        Have a nice day!

        Like

  20. Feedback from me for the new uptime table:

    -4 days no restart from update to 9.0.0.4_382_10680 (since morning on 03.06.17)

    -I noticed a slightly higher RAM usage after the update ( before was ~38% – now its ~42% ).

    Liked by 1 person

      • Seven days! I am not sure if its because of disabling 5GHz od new firmware but after few more days i am going to enable SSH and enable disbaled range. We will say!

        Liked by 1 person

      • Hi RT-AC1200G+,

        Thanks for your feedback. I have update your data. May I know how many days without reboot?

        Have a nice day!

        Like

      • Hi RT-AC1200G+,

        It’s okay. I don’t need continuous up-time because it reboots even I turn it off every day. But you might need to check syslog to see if any un-normal reboots, or Internet drop.

        I have update your report in the table.

        Thanks for your feedback!

        Like

  21. Hi Amigo,

    I’ve also experienced auto-reboot issues with my router(same model). My timezone is GMT+2, DST start(3m, 4thSun, 2h), DST end(10m, 5thSun,2h).

    In “Current Configuration” section you mention that now you have “3.0.0.4.380.7415” firmware applied on the router but in latest posts you said you have 10680. So what firmware do you run now?

    I”m also running now 10680 but, in compare with all of you, I encounter reboots each night:
    “Jun 8 06:00:31 disk_monitor: Got SIGALRM…”
    What I have different than you are the following (comparing to your “current configuration” section):
    1. 5Ghz was disabled (but I’ve enabled now, however I can’t select control channel to 149, maximum is 112, I”ve left it to auto)
    2. 2.4Ghz was not disabled (I still have wifi devices not having the 5Ghz support), but I”ve disabled now
    Most imporant here: I had a wireless schedule set: wifi turning off at 23:00 and on at 7:00 AM. I’m setting the same schedule now for 5ghz wifi.

    3. Enable UPnP Media Server on [Media Server] tab and Enable Share on [Network Place (Samba) Share] tab were ON in [USB Application] – I”ve set them now to OFF.
    4. ssh was already off
    7. wifi mac filter was off, I’ve set it to on and added some wifi macs in order to be able to connect
    8. in LAN/LAN-IP tab I have 192.168.1.1, why do you need 192.168.120.1? I”ve left it as before
    9. I have 2 mac to ip address bindings in LAN/DHCP Server (according to you, there should be a switch in “Network Map” to enable the binding – “9. In [Network Map], Enable MAC and IP address Binding on TS-251A”)
    10. on point 4 you say to have ssh disabled but here you say to have it enabled… I kept it disabled.

    I’ve noticed something strange: in “Network Map”, System Status, the lan and wifi2.4ghz mac address are the same. Do you have the same on your router?

    Also,when I have 2.4ghz enabled, I notice the same missing SSID after a soft reboot (same case as @caliban), which comes back after a hard reset (manual power off/on the router).

    Just to be clear: after enabling the 5ghz and soft reboot, the SSID is visible (no need to hard reboot as for 2.4ghz). However, it connects at 78Mpbs(with “control channel” on auto and connected on 52), while, with older firmwares, it connected at more than 500Mbps. After setting the “control channel” manually to 112, I got a connection of 234Mbps. So it seems this setting matters a lot, but how should I find the optimal value other than checking manually for each value?

    THanks,
    Mike

    Liked by 1 person

    • Hi Mike,

      Here is a quick reply to you. I will write a more complete reply later.

      1. I am using 10680 without any reboot for 7 days now.
      2. Did you power off and up with physical button after upgrade? One of us has SSID issue solved by reset physical power.
      3. May I know which country of your RT-AC1200G+? Wireless channels are different by countries.
      4. Could you please help me to check if you have schedule reboot? It’s in [Administration] -> [System] tab -> Enable Reboot Scheduler.
      5. Do you have an Android Tablet? Could you please follow On 2017/5/31, 5GHz Unstable, Reboot section and show me your WiFi Analyzer/Scanner screen capture? It helps me to know your 2.4GHz and 5GHz channels and if any interference.

      We’re here to help. Don’t worry!

      Have a nice day!

      Like

      • Hi Amigo,

        My replies:
        2. Yes, I did. SSID came up after I’ve powered via physical button but disappears each time I do a soft reboot (available only for 2.4ghz, not also for 5Ghz).
        3. I’m from Romania.
        4. Oh, it seems I have the reboot scheduler enabled, date to reboot: all days, time of day to reboot: 05:30 – I’ve disabled now. It seems I’ve only synchronised with your configuration mentioned in “Current Configuration” section.
        5. I don’t have a tablet, but I suppose a smartphone will also work. I’ll post in several hours the screen capture with wifi signals using wifi analyzer for 2.4 and 5Ghz (currently I’m at work).

        So, if auto-reboot will be solved by point 4, the only problem remains that 2.4ghz gets not available after a soft reboot of the router and the low value of Mbps the 5Ghz gets connected to.

        Thanks and have a nice day!
        Mike

        Liked by 1 person

      • Hi Mike,

        Reply to your answer:

        2. Okay, got it. Seems soft reboot will result in 2.4GHz disappear. (P.S. you may use reboot command for a soft reboot in ssh or just click in the upper middle Reboot button.).

        3. Thanks. Seems Europe have different available wireless channels.

        4. I would like to hear from you after verify your reboot scheduler configuration.

        5. I use WiFi Scanner/Analyzer to monitor visible SSID and its channel. This helps me to identified a Radio Jamming.

        I made a list tracking how many days without reboot in RT-AC1200G+ with RT-AC1200G+_9.0.0.4_382_10680-gc72d1c8.trx Stability Tracking section. May I invite you to join?

        Have a nice day!

        Like

      • Hi Amigo,

        4. Sure, I’ll keep you posted with the status of auto-reboot
        5. I’ve installed the app on my smartphone. Here are the screenshots and a wifi.txt file exported via the Wifi ANalyzer application:
        https://drive.google.com/open?id=0B-POc-shPSSzSEFpNWd0WXFXalU
        https://drive.google.com/open?id=0B-POc-shPSSzRzRMaTNYdlBZTzQ
        https://drive.google.com/open?id=0B-POc-shPSSzOS1hd0pkNnVKZVk
        https://drive.google.com/open?id=0B-POc-shPSSzbENkX3NublZ5Rjg
        https://drive.google.com/open?id=0B-POc-shPSSzOC1QUU9Ua0pKTTQ

        Sure, you can invite me to join the stability tracking section, but first I’ll need to play a little more with the router (clearing NVRAM and setting wifi scheduler – it is easier for me to set the scheduler rather than to power on/off the router each time when I go to bed).

        Liked by 1 person

      • Hi Mike,

        Because I didn’t go to bed at the same time every day, I choose to use the power button. 🙂

        From your screenshot, it seems 2.4GHz is very crowded and all weak signals. mike_mma_BUC5 is 5.7GHz is clear and has strong signal.

        FYI, my 2.4GHz signal is -40dBm while 5.74GHz is close to -50 dBm and the router is about 3 meters to my MediaPad M1.

        I also add you to Stability Tracking list. When you are comfort, leave me a comments and I will update your record. No need for uninterrupted uptime because it reboots even I turn it off every day.

        Keep in touch!

        Like

      • Hi Amigo,

        Uptime: 2 days 13 hours (since I”ve cleared NVRAM and reconfigured my router). Also the router has wifi scheduler active (auto power off at 23:00 and auto power on at 07:00).

        It seems that the error:
        “Jun 9 23:02:09 disk_monitor: Got SIGALRM…”
        was not related to auto-reboot (as I initially thought) but could be related to the wifi scheduler (taking into consideration it appeared around 23:00) (although I should have found this error each time at 07:00 and 23:00) BUT not only related to this case, example:

        Jun 9 09:59:45 rc_service: ntp 399:notify_rc restart_upnp
        Jun 9 09:59:45 miniupnpd[401]: shutting down MiniUPnPd
        Jun 9 09:59:45 miniupnpd[424]: version 1.9 started
        Jun 9 09:59:45 miniupnpd[424]: HTTP listening on port 47756
        Jun 9 09:59:45 miniupnpd[424]: Listening for NAT-PMP/PCP traffic on port 5351
        Jun 9 09:59:46 rc_service: ntp 399:notify_rc restart_diskmon
        Jun 9 09:59:46 disk_monitor: Finish
        Jun 9 09:59:47 disk monitor: be idle
        Jun 9 09:59:47 rc_service: zcip 426:notify_rc start_firewall
        Jun 9 09:59:47 zcip client: configured
        Jun 9 09:59:48 miniupnpd[424]: shutting down MiniUPnPd
        Jun 9 09:59:48 start_nat_rules: apply the nat_rules(/tmp/nat_rules_ppp0_eth0)!
        Jun 9 09:59:48 miniupnpd[446]: version 1.9 started
        Jun 9 09:59:48 miniupnpd[446]: HTTP listening on port 46662
        Jun 9 09:59:48 miniupnpd[446]: Listening for NAT-PMP/PCP traffic on port 5351

        Jun 9 10:00:47 disk_monitor: Got SIGALRM…

        Jun 9 10:21:43 WAN Connection: Ethernet link down.
        Jun 9 10:21:43 stop_nat_rules: apply the redirect_rules!
        Jun 9 10:22:48 pppd[253]: Serial link appears to be disconnected.
        Jun 9 10:22:51 miniupnpd[446]: upnp_event_process_notify: connect(:2869): No route to host
        Jun 9 10:22:54 pppd[253]: Connection terminated.
        Jun 9 10:22:54 pppd[253]: Modem hangup
        Jun 9 10:23:03 WAN Connection: Fail to connect with some issues.
        Jun 9 10:23:05 pppd[253]: Connected to via interface eth0
        Jun 9 10:23:05 pppd[253]: Connect: ppp0 eth0
        Jun 9 10:23:08 pppd[253]: PAP authentication succeeded

        By the way, the log above is my last log from the router – nothing was reported in the log today and yesterday.

        Regards,
        Mike

        Liked by 1 person

      • Regarding channels on 5GHz:
        It’s always been like that, even on official FWs from ASUS-website: on auto-setting it will go up to 149, but manually you can’t go higher than 112. (I’m in Europe too, so same regulations)

        Keep in mind that with auto-channel you may still experience connection losses. Not because the router restarts, but if it changes channel on 5 GHz (because there’s another network using the same channel and the router decides to switch), it takes up to a minute before it’s visible again. That’s normal, because on channels higher than 48 it has to check for radar first (TPC and DFS) but it may be long enough that a download fails.

        Slow connection on channels higher than 64 seems to be an issue with this beta only, experienced the same.

        You should probably think about starting from scratch by clearing NVRAM (power off – keep wps-button pressed – power on – wait until power led blinks rapidly – release wps-button – follow setup). That way you can be sure that absolutely no remnants of an older FW are left. While figuring out the missing 2.4-SSID I noticed that even reset-button doesn’t clear everything, only clearing NVRAM does.

        Sidenote:
        Since you asked amigo about his IP-config (which you wouldn’t if you knew what you’re doing ^^), may I ask if you’re running it as router behind a pure modem (did you enter your provider-credentials in RT-AC1200G+) or are you running it in router-mode behind another router? In that case I’d recommend turning it to Access Point-Mode. You’ll lose some of the functions (QoS and such), but avoid the potential headache of double NAT.

        Liked by 1 person

      • Hi caliban,

        I use Networking Computing: Dynamic Frequency Selection Part 3: The Channel Dilemma to learn available channels. To push to 802.11ac 867MBps, you will need 80MHz bandwidth. My Mac Mini 2012 Late has 802.11n 2T2R @ 5GHz which only need 40MHz.

        I don’t have enough wireless knowledge but I do learn a lot in this event. And also from professional like you, tycjan, and Lenaius Chang to leave comments to help.

        Sometimes auto-channel are not smart enough like my radio jamming in this post. Then a WiFi Scanner/Analyzer would help. But I don’t know what would be displayed with hidden SSID?

        I have to use high channels or Huwawei MediaPad M1 won’t see SSID.

        About cleaning NVRAM, yes, that’s what I think the best to restart. You may also use command: mtd-erase2 nvram in ssh as mentioned in Upgrade Firmware section. I don’t know the difference of reset button, WPS button, and mtd-erase2 with removing configurations/NVRAM.

        Thanks for sharing. I do learn a lot!

        Like

      • About difference between Reset-button and clearing NVRAM with WPS:

        I’d expect a Reset-Button to do just that, a reset to factory standards of the current firmware. But it clearly doesn’t, although it looks like it does (initial setup on boot).
        When I experienced missing 2.4-SSID the first time, I flashed older betas and even latest official FW, always doing a “normal” reset after flash. It didn’t help.

        Only when I read up and cleared NVRAM via WPS-button it showed up again. A bit later I found out that a simple Power Reset would’ve helped too. But most of the time before I was thinking “this is fucking great, did I break 2.4 now or what?!” ^^

        So, if 10680 left something behind in NVRAM when doing just a “normal” reset, it’s also possible older firmwares do the same when upgrading to 10680.

        And I really, really hope that clearing via command line does not do even more, like:
        Reset button -> hyper reset via WPS-button -> ludicrous reset via command line??? Kinda dumb… ^^

        Like

    • Hi Mike,

      Here is a detail reply to your comment.

      1. I have update Current Configurations section to match 2017/6/7. Sorry for misleading you.

      2. About Control Channel in 5GHz, it’s different from countries. 149 is to match local wireless regulation and prevent from radio gaming as I mentioned in 5GHz Radio Jamming section. It’s okay to leave it auto.

      3. About wireless scheduler, I never use it because I power off RT-AC1200G+ while sleeping. Also, check your [Administration] → [System] tab, there is a Reboot Scheduler. Make sure it’s off if you don’t need auto-reboot.

      4. You may leave your favorite application/daemon on or off. If they are the reason to reboot, I should be able to see their logs in syslog. But for driver (ring protection 1) and kernel (ring protection 0), it’s impossible to keep the log. Kernel panics before sending any message to syslog. If you are interested, Research on syslog.log and Study Kernel Panic and Wireless Technologies sessions have many useful information.

      5. You may setup your own MAC filter, LAN IP, and etc. I have rewrite that section because it’s misleading.

      6. About ssh, it’s my wrong. Thanks for correction!

      7. About MAC of LAN and 2.4GHz, mine is the same. Then it should use the same chip. It’s strange because according to WikiDevi, they use different chips.

      8. About 2.4GHz SSID disappear issue, Utes, it’s reported by caliban on 2017/6/5.

      9. About 5GHz connection speed, you need to study 11 to 13 in Study Kernel Panic and Wireless Technologies session. But I suggest you to check it again after power off/on, it seems a power off/on is necessary after firmware upgrade.

      Wish it helps!

      Like

      • HI Amigo,

        After disabling “reboot scheduler” (on point 3) I have now an uptime of 1 day and 3 hours. Of course, I’ll try now the nvram reset first with your solution via ssh and then with caliban’s instructions – so that I have everything cleared – this will reset my uptime to 0.
        It seems that the only thing that bothers me now is that I need to manually reboot router in order to have 2.4ghz up and running (my intention is to seal the router in a plastic box inside the wall – which will make difficult to access it manually).

        Liked by 1 person

      • Hi Mike,

        I have update your record in the table.

        If you leave RT-AC1200G+ inside the wall, you might have heat issue and weaker signal. In other words, life shorten and slower speed.

        Are you sure the temperature is comfortable inside the wall? I even want to invest a fan to help dispatch dissipation.

        P.S. It’s said the power amplifier in home wireless router is about 2 to 3 years in normal use.

        Have a nice day!

        Like

      • HI Amigo,

        It will not be sealed inside the wall, just in a plastic box inline with the wall having a plastic cover which, at the moment, is not mounted over the box (at the moment, the router stays inside this plastic box). The idea is to have as many items around the room.

        Liked by 1 person

  22. Hi Caliban,

    Thanks for your info.
    What would be the advantages of clearing the NVRAM? It may solve the slow connection on 5ghz or fix the missing 2.4-SSID after soft reboot? Anyway, if you haven’t already tried it, then I can try to see if it fixes these 2 issues.
    Related on my router, I simply use it on PPOE (WAN/Basic Config/WanConnection Type: PPOE), not even behind a pure modem: I receive an ethernet cable in my apartment which I connect on the blue separated ethernet port and provide the credentials from my internet provider) to the router.

    Liked by 1 person

    • Okay, then forget about AP-mode.

      @NVRAM:
      Nope, clearing NVRAM won’t fix slow connection on specific 5GHz-Channels (read below) or missing 2.4-SSID, as they’ve been introduced by latest beta.
      But since you seem to experience other “funny stuff”, what it will do is make sure that nothing from an older firmware interferes with it. It’s a beta, it has its own quirks and possibly nothing will be solved by it, but at least you then know for sure that it’s all caused by the current firmware alone.

      Regarding that slow connection on certain 5GHz-channels:
      – While typing this on my laptop I’ve set it from auto to manual channel 64 / 80 MHz -> normal* link speed between 520 and 650 mbit
      – changed it to channel 112 / 80 MHz -> the same, between 520 and 650 mbit

      Don’t ask me, I have absolutely no idea why it’s working now. ^^

      And remember about auto-channel: Depending on your surroundings you may lose connection now and then. But not because of the router rebooting, but because it’s checking if the new channel (above 48) is clear of weather and airport radar.

      * I’m not in the same room as the router and wireless energy savings are set to max on this laptop while on battery. Has some impact too.

      Liked by 1 person

      • Hi caliban,

        Yes, 5GHz signal strength drop very quickly if there is any obstacles in between.

        Have a nice day!

        Like

      • Hi caliban,

        Thanks for the info. I’ll try to play with some manual values for “control channel” and track the results.

        Liked by 1 person

      • Hi caliban,

        playing with 5Ghz settings, having: wireless mode n/ac mixed, channel bandwidth 80 and control channel 112 allowed me to connect with my Note4 on 780Mbps to the router (staying 1 meter near the router).

        However, if I connect from another room to the router, I get connected one conencted with 585Mbps but immediately dropped to around 300Mbps but then trying reconnecting several times I got a stable connectino of 702Mbps.

        Liked by 1 person

      • Hi Mike,

        Channel 112 is not available in Taiwan. It’s better to use 149 and above to prevent conflict or interference with radar and DFS.

        Wish it helps!

        Like

      • Hi Amigo,

        According to the region I”ve set, on wifi “control channel” I have only these options available: 36, 40, 44, 48, 52, 56, 60, 64, 100, 104, 108, 112 (for 80Mhz), 36..136 (for 40Mhz) and 36..140(for 20Mhz).
        So I can’t choose in any way 149 unless I switch maybe to Taiwan region.

        Liked by 1 person

      • Hi Mike,

        I do hear about region issue with limited channels. Here is a link about changing region code in Asuswrt. Do you want to try?

        I suggest to test stability and don’t do anything else until no reboots after 4 weeks. It helps to focus on reboot issue. My longest non-reboot record is 14 days with previous firmware and that’s why I think 4 weeks is good enough.

        Have a nice day!

        Like

      • Hi,

        At the moment I’ll try to focus on stability, as you said, so won’t try anything exotic. I’ll keep you posted with the uptime.

        Liked by 1 person

      • Yeah, well, changing region has two downsides:

        – It may operate outside regulations, which isn’t funny. If authorities identify you as a jammer, it won’t be cheap. Frequency-bands are watched closely everywhere. May take them a while, but once you’re found out – hammer falls!
        – You lose more channels than you gain. Europe has the broadest spectrum for 5GHz-Wifi worldwide. (enough space for 4 SSID on 80 MHz next to each other and then there’s still some leftover, I think 40 MHz) With this router you can use three of these 80MHz-slots.

        Good info! I never noticed that you get more channels on 20 and 40 MHz for manual setting than on 80 MHz. I wonder what’s the reason for that.

        Liked by 1 person

      • Hi caliban,

        I think if he may find Romania, maybe he may change the country code to match. I don’t mean to change to US. And I agree with your point – follow the local regulation.

        About channels, if you choose different bandwidth, available channel numbers will be different. It’s how these channel numbers are organized. I learn this from 5GHz implement guide.

        Your comments are highly appreciated!

        Like

      • Romania is part of the EU -> has implemented EU-regulations -> makes no difference if he sets it to Romania, Germany, Italy, whatever. Always the same channels. 🙂

        Liked by 1 person

    • Hi Mike,

      Because the data format might be different after upgrade, some old configurations might mislead.

      As I mentioned in On 2017/5/8, Reboot at 10:16 section, you may use command:nvram show to display configuration in flash.

      I am not sure if clearing NVRAM would help on slow connection at 5GHz because there are many potential problems. For example, wrong channel, interference, radio jamming, and etc. But we may try one by one, takes some time and we all learn from experiments. I suggest to read Channel Planning Best Practices and google more related posts would help. Wireless is more complex than wired.

      For your reference, I have wrote QNAP TS-119PII with Different Wireless Device on Mac OSX File Transfer Performance Report to share my wireless transfer rate.

      PPPoE needs a dial-up and login process. I didn’t use it for a long time. Your RT-AC1200G+ will get a physical IP after being authenticated successfully on WAN. RT-Ac1200G+ will use its DHCP server to assign an IP to device behind. I used to write Connect NAS to your Computer Directly about this.

      Have a nice day!

      Like

  23. I have same 5ghz issue.
    When 5ghz ssid on auto chanal(auto select chanal above 100) i have low link speed with my ac devices ( uder 100Mbit).
    When 5ghz ssid on maual chanal ot lower than 100 there is no problems with link speed.

    Liked by 1 person

    • Hi nikisi,

      I think it’s safe to use 5GHz channels between 35 to 48 and 149 to 165 with less interference with DFS and Radar.

      I recommends to lean 5GHz implement from 11 to 13 in Study Kernel Panic and Wireless Technologies session. They are very helpful.

      Wish it helps!

      Like

      • Depends on where he’s located. In Europe it stops with channel 140 (US has even less available). But on the other hand up to that we have far more channels available here than you in Taiwan (?). 🙂

        Anyhoo, trying to go around that slow-connection-bug on certain channels is a mute point, I think. It only happens with latest beta – let’s assume they will fix it with next beta or final (if there ever will be one). Also, I was the first to report it – yet when I played around with it again yesterday it suddenly worked fine on manual channel 112. It’s not really consistent.

        Liked by 1 person

      • Hi Mike,

        Yes, it’s not easy for wireless implement. Sometimes experiment helps to find the best parameters.

        You are good!

        Like

      • Hi caliban,

        Thanks for your great explain! I learned a lot.

        As you said, let’s focus on stability now.

        Have a nice weekend!

        Like

  24. Hi all,

    Good news, at least for me: it seems 2.4GHz is visible now after soft reboot of the router.
    Steps to reproduce:
    1. mtd-erase2 nvram in ssh (from Amigo)
    2. power off – keep wps-button pressed – power on – wait until power led blinks rapidly – release wps-button – follow setup – (from caliban)
    3. normal setup: setting up router admin pwd, wifi pwd, connection credentials, etc. similar to “Current configuration” section, with differences to:
    – 5ghz is disabled (until I won’t upgrade my laptop wifi Lenovo E520 to a dual-band wifi I will stick to 2.4ghz only) – I”ve performed this setting afte step 5 below
    – wifi wireless scheduler – enabled (turning wifi automatically off at 23:00 is a sign for me to go to bed 😉 )
    4. connected via ssh and run a “reboot” command – surprise, after reboot, 2.4ghz available
    5. from router web interface performed reboot: same thing: 2.4 ssid available after reboot.

    I’ll keep you posted with news.

    Liked by 1 person

    • Hi Mike,

      I have add your solution in “2.4GHz SSID Disappear Issue” section to share.

      Thanks and good job!

      Like

      • you can try both steps on your router and confirm or not whether the 2.4ghz is visible after a soft reset of the router either via reboot on ssh or via gui.
        I’ve tried both steps because you said you’ve tried the second one and still 2.4ghz was not visible.

        Liked by 1 person

      • Hi Mike,

        I agree. First, it’s better to learn from experiment. And you may use “nvram show” command to compare yourself. Second, one of a feedback from local communities told me they are not always the same to clean all NVRAM which someone work in Asus share with him.

        Therefore, I agree Mike’s opinion n this case.

        We may confirm this by reading source code but it (this new 10680 version) is not released yet.

        Have a nice day!

        Like

      • Hi caliban,

        Theoretical, yes. But I got feedback from local communities that he suggest me to do both because they are not exactly the same, it’s from his friend who works in Asus.

        We may read the source to proof assumption.

        Thanks for reading comments throughly and provides your suggestion!

        Like

      • @Amigo:
        Wait, so it’s really like I mocked about earlier?!

        || – Reset-button
        || – Clear NVRAM with WPS-Button
        V – Clear even more NVRAM via SSH-command

        And each “resets” a little more? Is there anything else, like a “super-duper-now-for-sure-cross-my-heart-reset”??? ^^

        Liked by 1 person

      • Hi caliban,

        I don’t know the difference but we may read the source code to verify. It’s available on Asus’s Official Web Site. 🙂

        Have a nice day!

        Like

      • Hi tycjan,

        I don’t know. But I do confirm an assumption: it’s easier to catch eyes to blog in English than local languages.

        I blog in English and share my post in English and local communities with partial translation. Most of the feedback from local communities suggest me to walk away, give up, leave it alone while you guys from Europe discuss and share your experience with me. I choose to work on this issue with you rather than just complain.

        Dolly told me it’s my post in local communities to catch their eyes. They want to know what I am going to do. They care about the feedback and discussion in local communities. I told her if the latest firmware is good, I will share my experience in communities. Case close. If not, you will see me again and again. I just refused to give up.

        This is just a lucky case. I learn from all of you a lot. Not just tech but also philosophy of life.

        Thank you my good friend!

        Liked by 1 person

      • Hi AsusRTUsers,

        How many days without reboot till today? 5 days?

        Thanks for your feedback!

        Like

  25. Something weird just happened.

    While downloading an update on my smartphone, suddenly Wifi connection broke off for some seconds (Phone showed LTE-connection). My laptop was running as well, but there Wifi was shown still connected (Win10 is a bit slow detecting Wifi-changes). But when I tried to open Router settings in browser, it didn’t load. (stuck on “connecting to…”)

    20 seconds later my Smartphone re-connected (download failed by then) and I was finally able to open router settings on laptop. But there is absolutely nothing to see… No restart, no wireless-restart in syslog, no channel-switch – nothing.

    Strange.
    Can’t be just the Phone acting up, because then I should’ve been able to connect to it from my laptop.

    Liked by 1 person

    • Hi caliban,

      I have experience similar case in “5GHz Radio Jamming” section.

      My situation: iPad mini connects to RT-AC51u @ 5GHz. It’s okay until I begin my pressure test between Mac mini and RT-AC1200G+ @ 5GHz. iPad’s internet becomes extreme slow. I cannot launch the admin page like you.

      Things go back to normal when I manually configure channels on both wireless routers to avoid radio jamming.

      I am not sure if you experience the same radio jamming. You may use WiFi Scanner/Analyzer to get a picture of who is using 5GHz near you. Maybe someone consumes all the bandwidth.

      Just my two cents.

      Like

      • Nah, it’s set to channel 48 manually, and there are only 2 other 5GHz-networks in my neighbourhood on 64 and 112. They’re not even strong in my flat, it’s almost a cosmetic thing to avoid their channels.

        Liked by 1 person

      • Hi caliban,

        Then it’s not a radio jamming. Do you have any ideas?

        Have a nice day!

        Like

      • Nope, not really.
        Maybe it was just a “hiccup”? ^^

        It was pure coincidence that I spotted it, as there are no log-entries or a runtime reset as marker.
        – How often has it happened before without me noticing? (without a failing download as indicator)
        – Will it happen again?
        – Does it happen to all of you guys running latest beta and if yes – how would you know?
        – It’s running in AP-mode, not Router – does that make a difference? (less sys-load, so should be even more stable if there’s any connection)

        If it happens more often (can’t find out unless I set up some logger – but that’s what people at ASUS get paid for, not me), only thing I could think of is playing around with WMM-settings (APSD – power save, maybe?).

        Liked by 1 person

      • Hi caliban,

        To monitor connection drops, we may setup a continuous download or upload at low speed from your desktop to a storage device connected to router. Thai is what I have done during pressure test section – Mac mini 2012 Late and a NAS. One the copy stops, you will get notified.

        I can only proof it happens but probably cannot get enough data. It’s too fast to reconnect (20 seconds). Maybe WiFi Scanner/Analyzer? And there is nothing to see in syslog.

        AP-mode is different to router. At least the DHCP server will stop working in AP-mode. But I have never use AP-mode, maybe there are more difference.

        I haven’t experience connection drops with new firmware yet with router mode.

        Have a nice day!

        Like

      • Hi caliban,

        Probably related to APSD or some kind of regular sync like the encryption password? I am not sure.

        Like

      • Thing is, I’m still not sure if there really is some sort of bug going on or if it’s been a one-time thing. Never noticed it before and didn’t ever since.
        Encryption interval as a cause sounds possible at first, but it’s set to 3600 seconds – every hour. Never heard of that causing connection-resets before, and I think there would be much more reports if it could cause that.

        Should I ever spot it again, I’ll try to think fast enough to check with Wifi Analyzer if the SSIDs are still up.
        Yes, I’m connecting to it via Wifi, but those 20-30 seconds don’t give me much time to grab my laptop, run to the ASUS connect it to LAN and try if the router is accessable via cable. ^^

        It happened once on my setup now, obviously to nobody else so far. It’s been reported, let’s wait if there ever will be another report. It’s intriguing that you experienced the same in the past, but the conditions seem to be completely different (jamming).

        AP-mode is different, yes, but only because router-stuff gets deactivated. There’s something missing in USB-apps, QoS-stuff isn’t there, in LAN you can only set router-IP – subnet – gateway – DNS, WAN is completely missing -> about 75% of the options are not there/not needed. Basically you’re dumbing it down, but that also means there’s much less stuff going on that could cause errors. (Wireless-settings are untouched)

        I’m doing this because the modem/router I got from my provider can’t be set to single user/modem-only. I could connect the ASUS as exposed host in full router-mode, it works fine. But I’d still have a router-cascade, double-NAT etc, there’s nothing to gain.

        Liked by 1 person

      • Hi caliban,

        No, it’s not radio jamming. I call it unstable. sometimes I can see my connection is okay but cannot access anything. And next few seconds, my connection drops. It’s recorded in “On 2017/5/31, 5GHz Unstable, Reboot” section. But it’s with the old firmware and I have to reset power, I am not sure if it’s the same to your case.

        If it happens again, you will need a plan and tools to capture the moment. If the WiFi Scanner/Analyzer shows your SSID appear and disappear at the moment, it’s same to my experience with old firmware.

        Have a nice day!

        Like

    • Hi caliban,

      One question, when the connection drops between your RT-AC1200G+ and laptop, how do you open router settings with a browser? Do you change from wireless to a wired connection before opening?

      I have the same experience and found when the wireless refuse to connect or drop, I may still access the router with a wired connection. But I have no idea how it works when it’s in AP-mode, my experience is in router mode.

      Have a nice day!

      Like

  26. 7 day’s working like beast on steroids. Great firmware(4 phone+2pc never lost connection), i believe this firmware= Fix all problems and make this router great again 🙂

    Liked by 1 person

  27. Hello !

    Today is 10-th day with no restart since update to latest beta.
    Amigo, do you give feedback to asus support for results with this firmware ?

    Liked by 1 person

    • Hi nikisi,

      Good news!

      I write an email to Debby every week. She also follow my post. And I share these numbers (days-without-reboot) to communities, too.

      Have a nice day!

      Like

  28. Few days ago I changed my internet plan and restart router by my own(to apply new internet settings).
    But stil 13 days no spontaneously selfreboots. 🙂

    Liked by 1 person

  29. Hi @Amigo,

    I decided to give full router mode instead of AP mode a try, which included a reboot. But still, up to that 13 days it was stable. Except that strange 20-sec-hiccup, which I didn’t notice again ever since.

    But may I ask how you’ve come to the conclusion that channels don’t fit EU-regulations?
    Because they perfectly do, they just completely excluded possible channels 120-124-128, which are completely in the range of Terminal Doppler Weather Radar (TDWR) and have some extra rules.

    Understanding that I finally know why manual channel-list stops at channel 112 when using AC @ 80 Mhz, but goes up to channel 140 when using 20/40 Mhz.

    These are the available 5GHz-channels in Europe:

    36
    40
    44
    48

    52
    56
    60
    64

    100
    104
    108
    112

    116
    120 – not available in AsusWRT
    124 – not available in AsusWRT
    128 – not available in AsusWRT
    132
    136
    140

    As we always need four channels for 80MHz bandwith, you can see why manual selection stops at 112: after that there never are four next to each other. For 40MHz we need 2 channels – there it stops at 136. Only when setting it to 20 MHz we can use 140.

    And if everything is on auto most of the time it uses 20 or 40 MHz, never 80 -> auto-channel goes up to 136/140.

    Liked by 1 person

  30. Hi Amigo,

    Problem detected in my case: it took 11 days for my router in order to reboot. However I need to investigate whether it was a power cut in my flat (I’ll get back to this comment when I’ll have more info). Until then, see below the link to the log of my router (the log was too big to pollute the forum):
    https://drive.google.com/open?id=0B-POc-shPSSzTjI2WTlZLTVWNVE

    Analyzing the log, it appears that the problem starts here:
    Jun 20 10:41:16 miniupnpd[1817]: Listening for NAT-PMP/PCP traffic on port 5351
    Feb 14 02:00:10 syslogd started: BusyBox v1.17.4
    (a power cut is plausible according to the logs)

    but also here:
    Feb 14 02:02:08 ntp: start NTP update
    Jun 20 20:34:03 rc_service: ntp 424:notify_rc restart_upnp
    Jun 20 20:34:03 miniupnpd[426]: shutting down MiniUPnPd
    Jun 20 20:34:03 miniupnpd[457]: version 1.9 started
    Jun 20 20:34:03 miniupnpd[457]: HTTP listening on port 57761
    Jun 20 20:34:03 miniupnpd[457]: Listening for NAT-PMP/PCP traffic on port 5351
    Jun 20 20:34:04 rc_service: ntp 424:notify_rc restart_diskmon
    Jun 20 20:34:04 disk_monitor: Finish
    Jun 20 20:34:05 disk monitor: be idle
    (at the moment the router reports uptime 0 days, 13hours, 15 min according to the restart at 20 Jun 20:34)

    Liked by 1 person

    • Hi Mike,

      From your information:

      Feb 14 02:02:08 ntp: start NTP update
      Jun 20 20:34:03 rc_service: ntp 424:notify_rc restart_upnp

      Yes, there is a reboot or reconnect to internet before Jun 20 20:34:03.

      From your log:

      Jun 20 10:41:16 miniupnpd[1817]: Listening for NAT-PMP/PCP traffic on port 5351
      Feb 14 02:00:10 syslogd started: BusyBox v1.17.4

      It should happen after Jun 20 10:41:16.

      But with your uptime information “at the moment the router reports uptime 0 days, 13hours, 15 min according to the restart at 20 Jun 20:34”, it should happen at 20 Jun 07:19.

      They are ambiguous.

      Please confirm if this is due to power cut.

      Waiting for your reply!

      Like

      • Hi Amigo,
        I confirm it was a power cut. Related to your comments, be aware of the timezone I have (gmt+2 – my router reports now: 1day,1 hour, 56 min uptime and my hour is 22:26 –> power cut yesterday at around 20:30 my hour).
        So, you may update the stability tracking table with my status: 12 days.

        Thanks!
        Kind regards,
        Mike

        Liked by 1 person

      • Hi Mike,

        Thanks for your quick feedback. Now I know it’s due to power failure and update your latest non-reboot record.

        Have a nice day!

        Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s