Linksys WRT32X on davidc502 builds keeps restarting automatically on its own after crashing

Installing and Using OpenWrt Network and Wireless Configuration
#1

Hi guys, I am new to this forum, don't mind. I have been using LEDE davidc502 (2 June 2018; Lede SNAPSHOT r7093-4fdc6ca31b) builds for the Linksys WRT32X and I have found out that the wireless router keeps restarting on its own for no apparent reason as well crashing on its own. Is there any solution to this problem?

And how do I post my System and Kernel log on this forum? I seem to see a few warnings that might catch your eye.

#1:
From Kernel Log

[    0.010368] cpuidle: using governor ladder
[    0.010500] mvebu-pmsu: CPU hotplug support is currently broken on Armada 38x: disabling
[    0.010506] mvebu-pmsu: CPU idle is currently broken on Armada 38x: disabling

From System Log

Fri Jun  1 19:19:40 2018 kern.info kernel: [    0.010368] cpuidle: using governor ladder
Fri Jun  1 19:19:40 2018 kern.warn kernel: [    0.010500] mvebu-pmsu: CPU hotplug support is currently broken on Armada 38x: disabling
Fri Jun  1 19:19:40 2018 kern.warn kernel: [    0.010506] mvebu-pmsu: CPU idle is currently broken on Armada 38x: disabling

Additional Information: WRT32X v1 with Spansion NAND and am using a WRT32X-AH SKU.

Is there anyway this issue can be fixed? Thank you!

#2

cat /sys/kernel/debug/crashlog

#3

Apparently, this command led the router saying
cat: can't open '/sys/kernel/debug/crashlog': No such file or directory

Despite it rebooting by itself randomly.

Is there any other way to show error codes?

#4

No, because all those would get stored on the RAM disk and get lost in reboot.

Those CPU error lines have likely no relevance. Typical lines.

Most likely something consumes RAM and the router runs out of memory.
You should monitor that.

And routers usually do reboot automatically after crash. That is also typical.

#5

I only enabled wireless for this router and disable DHCP and WAN interface, OpenVPN, Dynamic DNS, Adblock is not enabled, so what could be the cause of so much RAM being eaten up?

I have been monitoring when the WRT32X crash and the LuCI RAM monitor states that there is still 50% RAM available?

#6

The crashlog will be preserved after the crash-autoreboot, if the kernel is compiled with crashlog support

#7

In the dmesg log, it states that there is a "Crashlog allocated RAM at 0x3f00000"

So how do I access that?

#8

Reproduce the crash, you can access /sys/kernel/debug/crashlog after its autoreboot.
If the crashlog is still not there, the autoreboot might not be caused by kernel crash, but some userspace program.

APPEND: Is it watchcat that causes autoreboots? watchcat is enabled in davidc502's builds.
And according to its description -

Allows to configure a periodically reboot, or after losing internet connectivity. Configured through UCI /etc/config/system.

#9

Oh thanks! Ok I will watch to see if my router reboots by itself again.

#10

This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.