Kernel Panic "echo 0 > /proc/sys/kernel"
I've been getting an unusual panic during the restart cycle, also during normal operation, it says the following :
Code:
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. Code:
vi /etc/sysctl.conf Code:
vm.dirty_background_ratio = 5 Code:
May 7 17:21:59 odroid kernel: [54375.164266] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. I also wanted to add on the way this kernel panic happens, it just doesn't suddenly lock-up and only remedy is cold reboot via power button, first the network goes down, I notice this because my ssh session times out, then I am unable to open/write/read to any hard drive / ssd. Then everything locks up except the mouse and I am able to hover over the power icon and restart / shutdown, the restart doesn't work as mentioned, the shutdown sometimes works, sometimes doesn't. And I am unable to dmesg or syslog the event as those two seem to fail as well. |
The log entries report on a problem with a Realtek r8152 network card.
|
Quote:
Quote:
|
I just looked at Realtek's site. The only 8152 they mention is an Ethernet (=wired) card. You have a separate card for wireless if you have wireless. You can turn off or not-load the 8152 if you don't need it. If it isn't working you should disable or not-load it in any case.
Code:
lsusb Code:
lspci |
I think the the realtek 8152 is a USB-ethernet adapter chip.
|
Quote:
|
An update, I got my wifi going, and I left it on wifi when I went to sleep, I woke up to find my system non-responsive again, again with the Workqueue: events rtl_work_func_t [r8152] kernel panic,
Code:
May 8 08:33:37 odroid kernel: [22354.813321] kworker/2:3 D 0 11174 2 0x80000000 Also I forgot to add one important detail, I upgraded my connection to 250/100 it used to be 60/10 , could this be the cause for my ethernet kernel panic? This is exactly when this problem started. |
Quote:
Code:
man modprobe Quote:
Quote:
Quote:
|
As it turns out the iwlwifi was listed in the /etc/modprobe.d/blacklist.conf. When I was getting rid of wifi capabilities I must of put it in there and forgot about it, I now have a .txt file with any modifications / updates and so on I do in case I have to go back on something I did a few months ago and revert my action. So I # the iwlwifi mod in the blacklist.conf and I rebooted and wifi was available, I guess I can use the same process to blacklist the r8152 driver, but something caught my eye in my current lsmod
Code:
lsmod By the way /etc/rc.d/rc.modules doesn't exist on this installation Code:
cd /etc/rc.d Code:
nano /etc/rc.d/rc.modules |
Quote:
Quote:
Quote:
Quote:
There may be a way to throttle the r8152's bandwidth. |
I am sort of stuck with the adapter, it not only has 1 RJ45 ethernet port but it also contains 3 USB 3.0 ports which I use, this model http://2click.fen.pl/files/image/Newsletter_Y-30951.jpg I don't think if I blacklist the r8152 module the USB part of the hub will stop to function. That's a good idea that you posted *throttling the r8152* what if I throttled the program that uses that much bandwidth, could that be a solution ? Or do we need to throttle the r8152 at the source ?
Quote:
|
Quote:
Quote:
Quote:
in /usr/src/linux-4.16.8/drivers/net/usb/r8152.c I see Code:
#define RTL8152_MAX_TX 4 |
Quote:
Code:
ls -a Code:
/usr/src/linux-headers-4.13.0-39-generic/drivers/net/usb$ ls Code:
config USB_RTL8152 Code:
#define RTL8152_MAX_TX 4 And update, with wifi enabled and ethernet enabled I still got a kernel panic, with both r8152 and all the wifi modules listed in the call trace, it always happens every 18 hours it seems. I am going to blacklist the r8152 and see if that helps. |
Quote:
Quote:
Quote:
I used Skype when it first came out. It runs at 48K, not alterable. My hardware wouldn't support audio at that speed, garbled every conversation. I re-wrote the driver to make the maximum sampling rate a parameter. I could unload it then reload it with, say, 5K as the maximum sampling rate (as written it returned 0 - no maximum), which worked. POTS is 2.3K, so this handled phone conversations, all I wanted. |
I may have finally found a solution to my problem. I am going to post it here in case anyone else comes across this issue. I noticed that the panics always happened when I left my laptop on for extended periods of time without any user interaction so I started came to the conclusion that perhaps the usb was getting deadlocked in runtime suspend and causing a panic. So I did the following: First I prepared some info that I will need later on,
Code:
cd /sys/bus/pci/devices/ and then you ls (list) Code:
lspci Code:
# echo on | tee /sys/bus/usb/devices/*/power/control Code:
# echo on > /sys/bus/pci/devices/<bus_name>/power/control |
All times are GMT -5. The time now is 05:24 AM. |