Release New AX88U Firmware v3.0.0.4.386.41700

[Gravityz]

installed 41700 over the previous version

al seems well but i see these messages in the log
for the record, i do not have a mesh system nor have i nodes which are mesh compatible.
device with ip address 92 is an iphone xs

i get the not a mesh client a lot for all different devices.
is there a setting which should be off or is the router checking every device to check if it can be used as a mesh client


Jan 23 16:26:01 kernel: 192.168.1.92 is used by someone else, can't use it
Jan 23 16:26:14 kernel: XX:XX:XX:XX:XX:XX not mesh client, can't update it's ip
Jan 23 16:26:14 kernel: 192.168.1.92 is used by someone else, can't use it

 

[Natey2]

Just curious, but why is v3.0.0.4.386.41700 29MB larger for the AX88U than it is for the AC88U on the Asus download site?

Thinking of updating my AC88U from the recent 41535 to 41700...

 

[juched]

@ASUSWRT_2020 I am seeing a kernel system crash even when using 41700 (saw it on 41535 as well). Seems to be in mcpd crashing when enabling some hardware acceleration for mcast (ie. Mdns).

It seems to be fairly reproducible when my Google Home and Chromecast products connect to the network. They do make extensive use of mdns, and in Wireshark I can see mdns requests occurring quite regularly, but nothing over the top.

What is the best way to get you the information you need?

Spoiler: Crash Dump

May 5 01:05:02 crashlog: <1>Unable to handle kernel NULL pointer dereference at virtual address 0000005c
May 5 01:05:02 crashlog: <1>pgd = ffffffc02d537000
May 5 01:05:02 crashlog: <1>[0000005c] *pgd=000000002bddb003, *pud=000000002bddb003, *pmd=000000002bd0f003, *pte=0000000000000000
May 5 01:05:02 crashlog: <0>Internal error: Oops: 96000007 [#1] PREEMPT SMP
May 5 01:05:02 crashlog: <4>CPU: 0 PID: 1884 Comm: mcpd Tainted: P O 4.1.51 #2
May 5 01:05:02 crashlog: <4>Hardware name: Broadcom-v8A (DT)
May 5 01:05:02 crashlog: <4>task: ffffffc02febb580 ti: ffffffc02e9b0000 task.ti: ffffffc02e9b0000
May 5 01:05:02 crashlog: <4>PC is at fc_mcast_activate_hw.constprop.20+0x60/0x118 [pktflow]
May 5 01:05:02 crashlog: <4>LR is at fc_mcast_activate_hw.constprop.20+0x84/0x118 [pktflow]
May 5 01:05:02 crashlog: <4>pc : [<ffffffbffc3d16a0>] lr : [<ffffffbffc3d16c4>] pstate: a0000145
May 5 01:05:02 crashlog: <4>sp : ffffffc02e9b3850
May 5 01:05:02 crashlog: <4>x29: ffffffc02e9b3850 x28: ffffffc02ff4a800
May 5 01:05:02 crashlog: <4>x27: 00000000fffffffc x26: 0000000000000002
May 5 01:05:02 crashlog: <4>x25: 00000000fbffffef x24: ffffffc03023d300
May 5 01:05:02 crashlog: <4>x23: 0000000000000000 x22: 00000000ffffffff
May 5 01:05:02 crashlog: <4>x21: ffffffc03023d300 x20: ffffffc031c4ba00
May 5 01:05:02 crashlog: <4>x19: 0000000000000000 x18: 0000000000000000
May 5 01:05:02 crashlog: <4>x17: 0000000000000000 x16: ffffffc000400a28
May 5 01:05:02 crashlog: <4>x15: 0000000000000000 x14: 0000000000000001
May 5 01:05:02 crashlog: <4>x13: 0000000000000001 x12: 0000000000000001
May 5 01:05:02 crashlog: <4>x11: 0000000000000001 x10: ffffffbffc2c8568
May 5 01:05:02 crashlog: <4>x9 : ffffffc00bd67ec0 x8 : 0000000000000001
May 5 01:05:02 crashlog: <4>x7 : 0000000000067ec0 x6 : ffffffbffc13fd84
May 5 01:05:02 crashlog: <4>x5 : 0000000000000000 x4 : ffffffc030284ba0
May 5 01:05:02 crashlog: <4>x3 : ffffffc030280000 x2 : 0000000000000000
May 5 01:05:02 crashlog: <4>x1 : ffffffbffc3f3040 x0 : 00000000ffffffff
...
May 5 01:05:02 crashlog: <0>Call trace:
May 5 01:05:02 crashlog: <4>[<ffffffbffc3d16a0>] fc_mcast_activate_hw.constprop.20+0x60/0x118 [pktflow]
May 5 01:05:02 crashlog: <4>[<ffffffbffc3d3114>] fc_config+0x57c/0x988 [pktflow]
May 5 01:05:02 crashlog: <4>[<ffffffc0003fab58>] blog_activate+0x60/0xa8
May 5 01:05:02 crashlog: <4>[<ffffffbffc52d0f4>] bcm_mcast_blog_process+0xa3c/0x1010 [bcmmcast]
May 5 01:05:02 crashlog: <4>[<ffffffbffc529884>] bcm_mcast_igmp_add+0x484/0x6d8 [bcmmcast]
May 5 01:05:02 crashlog: <4>[<ffffffbffc526734>] bcm_mcast_netlink_process_igmp_snoop_entry+0x1b4/0x4a8 [bcmmcast]
May 5 01:05:02 crashlog: <4>[<ffffffbffc5252c8>] bcm_mcast_netlink_rcv_msg+0x30/0x50 [bcmmcast]
May 5 01:05:02 crashlog: <4>[<ffffffc000420e74>] netlink_rcv_skb+0xc4/0xf8
May 5 01:05:02 crashlog: <4>[<ffffffbffc5252f8>] bcm_mcast_netlink_rcv_skb+0x10/0x20 [bcmmcast]
May 5 01:05:02 crashlog: <4>[<ffffffc000420758>] netlink_unicast+0x150/0x228
May 5 01:05:02 crashlog: <4>[<ffffffc000420c20>] netlink_sendmsg+0x2f0/0x350
May 5 01:05:02 crashlog: <4>[<ffffffc0003bf658>] sock_sendmsg+0x18/0x58
May 5 01:05:02 crashlog: <4>[<ffffffc0003c01cc>] ___sys_sendmsg+0x26c/0x280
May 5 01:05:02 crashlog: <4>[<ffffffc0003c2a88>] __sys_sendmsg+0x40/0x80
May 5 01:05:02 crashlog: <4>[<ffffffc000400a38>] compat_SyS_sendmsg+0x10/0x18
May 5 01:05:02 crashlog: <0>Code: b4000301 f9400e60 b40002c0 12800016 (39417262)
May 5 01:05:02 crashlog: <4>---[ end trace 3a78fa886dc6ec86 ]---
May 5 01:05:02 crashlog: <0>Kernel panic - not syncing: Fatal exception in interrupt

 

[juched]

Also still seeing regular packet loss with devices connected to AiMesh nodes (which are AC68U nodes). Packet loss can sometimes be very low or sometimes as high as 30-40%, measured by a regular ping occurring.

Seems to happen if a GuestNetwork is setup which is both SYNC to all nodes and ENABLE intranet. If one of those is off the issue of packet loss goes to 0.

I use to think that it was related to Ethernet backhaul being enabled, but I can reproduce it without it on as well.

@ASUSWRT_2020 Any insight?

 

[neil0311]

not mesh client, can't update it's ip

I updated my router this morning and did a full reset but see that in the log as well. I do not have any AIMesh nodes or clients enabled. Just the router in AP mode.

Otherwise everything seems ok. Power on 2.4GHz band is better.

 

[Gravityz]

I updated my router this morning and did a full reset but see that in the log as well. I do not have any AIMesh nodes or clients enabled. Just the router in AP mode.

Otherwise everything seems ok. Power on 2.4GHz band is better.

ok sounds like a normal message

 

[RMerlin]

Just curious, but why is v3.0.0.4.386.41700 29MB larger for the AX88U than it is for the AC88U on the Asus download site?

Completely different SDK and kernel, 64-bit kernel, radio firmware that has to support AX in addition to AC, and so on.

 

[brummygit]

Are you using trend micro stuff or anything related to it? I asked because I compile my own beta 5 based on GPL41700, if you like I can share.

Thanks, at the moment I think I'm OK to wait as I usually have Trend Micro AiProtection activated. However if I get desperate to get Cake back up and running I will reach out to you. This GPL is still running very well for me.

 

[Treadler]

Thanks, at the moment I think I'm OK to wait as I usually have Trend Micro AiProtection activated. However if I get desperate to get Cake back up and running I will reach out to you. This GPL is still running very well for me.

Did you try the factory adaptive qos? If so, how was/is it?

 

[Gravityz]

still seems to run fine.
i am getting this message on 1 device

Jan 24 09:18:43 kernel: 192.168.1.92 is used by someone else, can't use it

i assign most of my devices with a DHCP assigned list.
somehow this device (iphone XS) is using ip address 92 like it should but the router is complaining that he can not use it.

i have a RT-AC66U connected as a repeater but at the moment of this message it is completely turned off.
maybe there is some old cache in the router which is causing this.

i did not do a full factory reset since i was on the previous 386 build and i hate to configure it again.

 

[LimJK]

I would like to report:

I have been running smoothly with 3.0.0.4_386_41700-gb567bc9 for about 2 hours. I did Factory reset and use Stock default NVRAM settings except for items I required for my environment.

Hoping that it can be a useful comparison for others with similar configuration.

My Configuration:

• AiMesh Router: AX88U 3.0.0.4_386_41700-gb567bc9 +
• 2 x AiMesh Nodes: AC86U 3.0.0.4_386_41634-g08f88ae
• Backhaul (Priority:"Auto", Type:"Wired"), Asus GX-U1051 5-Port Gigabit Switch from Router to Nodes, ISP: 1 Gbps Fibre

My steps to Upgrade from Merlin 386.1_beta4b -> 3.0.0.4_386_41634-g08f88ae:

(1) Factory Reset AiMesh Nodes with FW 3.0.0.4_386_41634-g08f88ae:

• AiMesh -> Topology -> (Select a Node) -> Management -> Remove Node
• Repeat for all Nodes

(2) Upgrade & Factory Reset AiMesh Router:

• Administration -> Upgrade Firmware -> Select and Upgrade 3.0.0.4_386_41700-gb567bc9
• Administration -> Restore/Save/Upload Setting -> Restore with (X) Initialize all the settings

(3) I try to leave NVRAM with Stock FW default values, except with the following changes:

• Enabled AiProtection: ON
  • No issue with build-in Speedtest or Speedtest on my MacBookPro
• USB Application
  • UPnP Media Server: OFF
  • Samba Share: OFF
• Wireless:
  • 2.4 GHz: Fixed Channel Bandwidth 20MHz, Control Channel: Auto, Explicit & Universal Beamforming: Disable
  • 5.0 GHz: Fixed Channel Bandwidth 80MHz, Control Channel: Auto
• Guest Network (1) on 5GHz Band
  • No Spewing of error messages in System Log
• WAN DDNS:
  • Using Asus DDNS Service with Let’s Encrypt (Let’s Encrypt working now)
• IPv6: Enable
  • Native
• VPN Servers (Connecting one at a time from the following clients):
  • OpenVPN works with MacBookPro 3.8.5beta02 (build 5620), iPhoneXsMax iOS OpenVPN Connect 3.2.2
  • IPSec works with MacBookPro & iPhoneXsMax Native Configurations
  • Instant Guard works with iPhoneXsMax iOS Instant Guard 1.0.9

(6) Reboot AiMesh Network

• AiMesh -> System Settings -> Reboot (AiMesh Router & Nodes)

Update: My AiMesh Network has been running smoothly for about 5.5 days. My Configuration in my Signature.

 

[Gravityz]

i decided to do a factory reset after all.
i redid the configuration manually.

i noticed the 2.4Ghz range seems a litttle bit better than previous version 41249(this version should also have had better range but did not notice this)

one important thing i did different than last time

on the first upgrade to 41249 i did a factory reset from within the gui.
This time i did a factory reset by turning the router off,
holding the WPS button
turn the power on
wait until the power led goes off
release the WPS button
Turn the router off
turn the router on

so it seems that the WPS factory reset works better.

 

[OzarkEdge]

so it seems that the WPS factory reset works better.

better than no reset?

OE

 

[neil0311]

One thing I'm seeing with this firmware that I have not seen before with the Asuswrt-Merlin builds....MAC addresses in the log that I don't recognize, that look like they're connecting or trying to connect.

Anyone know exactly what this means? A concern or just noise in the log? They happen over and over and over, every few seconds.

Jan 23 12:09:09 wlceventd: wlceventd_proc_event(526): eth6: Auth XX:XX:XX:XX:XX, status: Successful (0), rssi:0
Jan 23 12:09:09 wlceventd: wlceventd_proc_event(555): eth6: Assoc XX:XX:XX:XX:XX, status: Successful (0), rssi:0
Jan 23 12:09:13 wlceventd: wlceventd_proc_event(490): eth6: Deauth_ind XX:XX:XX:XX:XX, status: 0, reason: Deauthenticated because sending station is leaving (or has left) IBSS or ESS (3), rssi:0

 

[jass]

still seems to run fine.
i am getting this message on 1 device

Jan 24 09:18:43 kernel: 192.168.1.92 is used by someone else, can't use it

i assign most of my devices with a DHCP assigned list.
somehow this device (iphone XS) is using ip address 92 like it should but the router is complaining that he can not use it.

i have a RT-AC66U connected as a repeater but at the moment of this message it is completely turned off.
maybe there is some old cache in the router which is causing this.

i did not do a full factory reset since i was on the previous 386 build and i hate to configure it again.

My guess,
your iphone is set to shift mac adresses randomly, and for some reason not ip... then the router gets contacted by a new mac adress to use the already assigned ip/mac combo for a new mac adress.

 

[Gravityz]

better than no reset?

OE

Better than gui reset and Yes also better than no reset.

it seems the 2.4Ghz range increased a bit.
now it could be that build 386_41249 which should have the better range as well still got a bug regarding this.

after a while i still got this message again
Jan 24 09:18:43 kernel: 192.168.1.92 is used by someone else, can't use it

i think i know what is causing this(but this should not create this message)

the device with IP 92 first get's it's ip assigned when it is making contact with my RT-AC66U repeater.
Then when this device moves downstairs and makes contact with the RT-AX88U wifi it tries to get it's mac address assigned ip address but somehow thinks it is in use.
yes it is in use but by the same device which got the ip address assigned when it conmnected to the RT-AC66U

looks like a bug because the repeater is using the dhcp server from the RT-AX88U so it apparently lost track of it.

anyway no harm since the device still keeps it's ip address (92)

 

[Gravityz]

My guess,
your iphone is set to shift mac adresses randomly, and for some reason not ip... then the router gets contacted by a new mac adress to use the already assigned ip/mac combo for a new mac adress.

no that's not it. that feature(assign a random mac address) is turned off) even if it was turned on then it should get an ip address assigned from the free dhcp pool because it does not match any record in the manual assignment list. i posted a possible explanation in another answer

 

[Gravityz]

running for a couple of days now and the router seems very stable
no strange messages enymore and also the wifi seems more stable.
The connectivity time for devices like wifi camera's is very high, this means they do not disconnect/reconnect from time to time

 

[neil0311]

My AX88U has been on the latest st firmware for 6 days, and for 4 days has been on 160MHz channel bandwidth on the 5 GHz network.

Been rock solid. No issues, except for the 1st day on 160MHz I used “auto” channel selection. After setting a fixed control channel it’s been rock solid.

I was using Asuswrt-Merlin but didn’t use any of the added features. I plan to stay on this stock firmware. Less bloat and very stable.

 

[Gregory Phillips]

I just took my new AX88U out of the box a short while ago and immediately updated to the current firmware so I don't have any time running it but I did have to manually tell the router to look for new firmware. Normally the yellow exclamation mark is blinking when my routers connect to the net to let me know something is available.