1

I think I have a issue related to random networking issues, that affect and make my php processes to get stuck in D state.

This prints a kernel error:

[Tue May 17 11:24:55 2022] INFO: task php-fpm:3017789 blocked for more than 120 seconds.
[Tue May 17 11:24:55 2022]       Not tainted 4.15.0-176-generic #185-Ubuntu
[Tue May 17 11:24:55 2022] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.

Is this considered a kernel panic that work with automatic reboot after kernel panic

If this doesn't work I was thinking in just creating a while loop checking for that error message in dmesg log, then reboot if found, but I think there's probably a safer and more elegant way to achieve this

1 Answer 1

1

I found a answer on very good blog by Jethro Carr

You just need to add to /etc/sysctl.conf:

# Panic if a hung task was found
kernel.hung_task_panic=1

# Reboot 5 seconds after panic
kernel.panic=5

Just sharing because who knows how long that blog will be around, but if this is not allowed in this community I am sorry.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .