AX88U strange log entries - help needed

[Sabbath81]

Hello guys,

im running latest stable Merlin WRT on my AX88U and i need some advice here.

i would like to ask for some assistance on strange log entries, some of which cause the device to stuck, and needs rebooting:

1. Which entries are problems from the ones below?
2. These started to occur recently , how can i fix them? Most likely they are the reason for the device to hung.
May 17 17:10:06 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x8
May 17 17:10:06 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0
May 17 17:21:57 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x1
May 17 17:21:57 kernel: bcm63xx_nand ff801800.nand: intfc status f00000e0
May 17 17:57:42 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x1
May 17 17:57:42 kernel: bcm63xx_nand ff801800.nand: intfc status f00000e0
May 17 17:59:56 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x4
May 17 17:59:56 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0
May 17 18:15:03 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x1
May 17 18:15:03 kernel: bcm63xx_nand ff801800.nand: intfc status f00000e0
May 17 19:33:36 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x1
May 17 19:33:36 kernel: bcm63xx_nand ff801800.nand: intfc status f00000e0
3. What are these:
May 5 08:05:08 wsdd2[1252]: error: wsdd-mcast-v4: wsd_send_soap_msg: send
May 5 08:05:08 wsdd2[1252]: error: wsdd-mcast-v4: wsd_send_soap_msg: send
May 5 08:05:12 wsdd2[1252]: Terminated received.
4. After boot i notice transmission reports errors, although there are such dirs and there is some mount:
May 17 20:23:53 transmission-daemon[3307]: UDP Failed to set receive buffer: requested 4194304, got 1048576 (tr-udp.c:97)
May 17 20:23:53 transmission-daemon[3307]: watchdir:inotify Failed to setup watchdir "/mnt/WDHDD/Torrents/Watch": No such file or directory (2) (watchdir-inotify.c:171)
May 17 20:23:53 transmission-daemon[3307]: watchdir Failed to open directory "/mnt/WDHDD/Torrents/Watch" (2): No such file or directory (watchdir.c:358)

Thank you!

 

[JGrana]

I would hazard a guess that your routers NAND flash chip is borderline defective. This memory contains the /jffs directory and other critical areas used by the firmware.

Another user of a similar class router had similar timeouts/failures and pinged the Linux kernel developers for some help:

On Sun, Feb 06, 2022 at 09:57:15AM -0500, MP wrote:
>
> Hello all,
>
> Just looking for a bit of guidance here and a suggestion on proper next
> steps. I think there may be an issue within the below broadcom nand driver
> implementation.
>
> I have an Asus GTE-AX11000 and it reboots\crashes multiple times a week,
> upon inspection of snmp logging you can see the below entries, when the
> kernel panic occurs the entire router locks up, and sometimes reboots.
> Sometimes it doesn't reboot and the router hangs completely and drops all
> network devices, and removes wireless SSIDs and all of your devices lose
> their connection.
>
> Not sure who to turn to but Asus support has not grasped the gravity of this
> problem and I am thinking this can only really be resolved with a
> firmware\driver\linux update, so just hoping for some additional help or
> guidance from this group.

This is a very old, obsolete, and known buggy kernel version running any
number of unknown closed source kernel modules. Only the manufacturer
can help you here, there is nothing the community can do at all, sorry.

Best of luck!

greg k-h

For starters, you might want to try and re-format the NAND flash. Go to the Administration->System tab in the GUI and select “Format JFFS partition at next boot” Yes.
Reboot and hope it’s maybe a bad block that needs to be marked during format.
Note - you will lose anything you have put in /jffs - Entware, Addons etc.
So, be prepared to setup the router again from scratch.
If this fixes things - great. If you continue to get bcm63xx_nand errors, time to RMA the router…

 

[Sabbath81]

I would hazard a guess that your routers NAND flash chip is borderline defective. This memory contains the /jffs directory and other critical areas used by the firmware.

Another user of a similar class router had similar timeouts/failures and pinged the Linux kernel developers for some help:



For starters, you might want to try and re-format the NAND flash. Go to the Administration->System tab in the GUI and select “Format JFFS partition at next boot” Yes.
Reboot and hope it’s maybe a bad block that needs to be marked during format.
Note - you will lose anything you have put in /jffs - Entware, Addons etc.
So, be prepared to setup the router again from scratch.
If this fixes things - great. If you continue to get bcm63xx_nand errors, time to RMA the router…

Hello,

thank you. The device is 12months old, it used to work flawlessly until last 2 weeks, when this happened.

Is it possible the problem to be caused by the following: my SSD flash drive which was used to run entware and add-ons failed, so i replaced it with a new one, installed everything again but without formatting JFFS?

As for the transmission - problem solved.

 

[L&LD]

Anything is possible. With such an easy way to check, just do it.

Safely remove the new drive from the router and format it on a PC with NTFS file format. Save any files you want to save, but do not restore them for testing purposes below until after you are satisfied the hardware is working properly first.

Format the JFFS partition (click the button to format on next boot, be sure you hit 'Apply' at the bottom of that GUI page, then reboot the router 3 times in the next 15 minutes, making sure to allow the router to idle 5 to 10 minutes between reboots.

Now, insert the drive you formatted above, ssh into the router, and via the amtm menu, format it to Ext4 with journaling, add a 5 or 10GB swap file, and then re-install the scripts you want, from scratch.

Monitor the router/network to see if the above log entries return.

 

[Sabbath81]

I will do the suggested actions if this happens again. After the last reboot, 3 days ago, log is empty.

 

[Sabbath81]

Hello Guys,
im running 386.8 Merlin on my AX88U, with 2 USBs connected - one for the Merlin, and one for Data and i notice some log entries upon reboots:

1. Some lines with this output - is it a problem, those are at the beginning of the log?

May 1 05:37:10 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x4
May 1 05:37:11 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0
May 1 05:37:12 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x4
May 1 05:37:12 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0
May 1 05:37:13 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x4
May 1 05:37:13 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0
May 1 05:37:17 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x4
May 1 05:37:17 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0
May 1 05:37:17 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x8
May 1 05:37:17 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0
May 1 05:42:14 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x1
May 1 05:42:14 kernel: bcm63xx_nand ff801800.nand: intfc status f80000e0
May 1 05:44:01 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x4
May 1 05:44:01 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0
May 1 05:49:17 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x4
May 1 05:49:18 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0
May 1 05:49:20 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x4
May 1 05:49:21 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0
May 1 05:55:57 kernel: bcm63xx_nand ff801800.nand: timeout waiting for command 0x4
May 1 05:55:57 kernel: bcm63xx_nand ff801800.nand: intfc status c80000e0

2. Transmission deamon says it cannot see the Watchdir, however, there is definitelty such under WDHDD (external USB drive). How can i fix this ?
Aug 15 11:57:25 transmission-daemon[3542]: watchdir:inotify Failed to setup watchdir "/mnt/WDHDD/Torrents/Watch/": No such file or directory (2) (watchdir-inotify.c:171)
Aug 15 11:57:25 transmission-daemon[3542]: watchdir Failed to open directory "/mnt/WDHDD/Torrents/Watch/" (2): No such file or directory (watchdir.c:358)

3. wsdd2 Terminated received? what is it ?

4. Avahi issue?
Aug 15 11:57:30 avahi-daemon[1145]: Alias name RT-AX88U: avahi_server_add_cname failure: The requested operation is invalid because redundant


Thank you guys,

 

[RMerlin]

1. Some lines with this output - is it a problem, those are at the beginning of the log?

Problems accessing the router's flash. No idea if this is the sign of a hardware issue, or a software issue that would need to be addressed by Broadcom.

3. wsdd2 Terminated received? what is it ?

That's the WS-Discovery daemon, which basically replaces the deprecated network discovery mechanism of SMB1.

 

[Volt]

You can be sure that your router is OK, because I also see these kernel: bcm63xx_nand ff801800.nand messages on my RT-AX56U on stock firmware when copying large files to the USB drive over Samba. I googled them and one of the replies was that these messages look like informational messages from an "ancient" Linux kernel, nothing to worry about. I think that the Linux version used on the router does not like when one CPU core has long-lasting 100% load.

 

[Sabbath81]

Problems accessing the router's flash. No idea if this is the sign of a hardware issue, or a software issue that would need to be addressed by Broadcom.


That's the WS-Discovery daemon, which basically replaces the deprecated network discovery mechanism of SMB1.

Thank you for your comments.

Can i do something for 1. ? Or hope the device will not die. Its working with no issues.

how about 2.? How to fix this transmission error?
And 4.?

 

[ColinTaylor]

3. and 4. are normal messages as part of the boot up process.

 

[Sabbath81]

3. and 4. are normal messages as part of the boot up process.

Ant idea how to resolve 2?

 

[RMerlin]

Can i do something for 1.

Probably not.

how about 2.? How to fix this transmission error?

No idea, I don't use Transmission. Running a torrent client on a router is a bad idea, both from a security point of view as well as a performance point of view. If you absolutely must, look for possibly a Pi-powered solution instead of putting your primary network security device into that kind of duty.

Based on the error message, my guess is the path is incorrect.

 

[Sabbath81]

Probably not.


No idea, I don't use Transmission. Running a torrent client on a router is a bad idea, both from a security point of view as well as a performance point of view. If you absolutely must, look for possibly a Pi-powered solution instead of putting your primary network security device into that kind of duty.

Based on the error message, my guess is the path is incorrect.

I understand your point, thanks for the advice.

The strange thing here is that the path is correct, may i presume that transmission tries to locate the external HDD before its fully mounted? In this scenario, can i add a script or something , so to delay transmission start until the attached USB HDD is fully mounted? If yes, where? I already tried under JFFS -> pre-mount to add sleep 10 , seems to have no effect?

@RT-AX88U:/tmp/home/root# cd //mnt/WDHDD/Torrents/Watch/

@RT-AX88U:/tmp/mnt/WDHDD# ls
Asus WS PC Torrents
@RT-AX88U:/tmp/mnt/WDHDD#
@RT-AX88U:/tmp/mnt/WDHDD/Torrents# ls
Completed Incomplete Squid.Game.S01.KOREAN.WEBRip.x264-ION10 Watch Wrath Of Man 2021 720p WEB-DL DD+5.1 H.264-TEPES watchdir

"watch-dir": "/mnt/WDHDD/Torrents/Watch/",
"watch-dir-enabled": true

 

[Sabbath81]

Any ideas guys for 4.?

 

[JGrana]

Maybe a problem with symbolic links?

/mnt is a symbolic link to /tmp/mnt

Try to change “watchdir” to “/tmp/mnt/WDHDD/Torrents/Watch”

 

[BreakingDad]

Eric @RMerlin Do you know what this is on the logs? only seen it since latest f/w and can't find any info on google. or forum. All I've worked out is FCache is file cache in python.

kernel: ^[[0;33;41mFCACHEfc_vblog_list_add ERROR: Duplicate blog: list blog<0xffffffc015a61300> JOIN blog<0xffffffc015a8ee40>^[[0m

 

[RMerlin]

Eric @RMerlin Do you know what this is on the logs? only seen it since latest f/w and can't find any info on google. or forum. All I've worked out is FCache is file cache in python.

kernel: ^[[0;33;41mFCACHEfc_vblog_list_add ERROR: Duplicate blog: list blog<0xffffffc015a61300> JOIN blog<0xffffffc015a8ee40>^[[0m

Related to a proprietary Broadcom component, no idea what it means.

 

[BreakingDad]

Related to a proprietary Broadcom component, no idea what it means.

Thank you, I can stop worrying now.

 

[Sabbath81]

Maybe a problem with symbolic links?

/mnt is a symbolic link to /tmp/mnt

Try to change “watchdir” to “/tmp/mnt/WDHDD/Torrents/Watch”

Tried this, didn'e help, just the message is that i cannot find /tmp/mnt/WDHDD/Torrents/Watch