Public bug reported:
The following very simple bash script results in an unkillable process topping
the CPU.
```
#!/bin/bash
while true; do
echo -n "."
if echo "hw-tc-offload: off [fixed]" 2>&1 | grep 'x-checksumming' ; then
echo "should never get here...."
fi
done
```
I can reproduce the issue by starting a few (8) concurrent executions and wait
a few minutes. The system used is:
```
> cat /etc/os-release
PRETTY_NAME="Ubuntu 21.10"
NAME="Ubuntu"
VERSION_ID="21.10"
VERSION="21.10 (Impish Indri)"
VERSION_CODENAME=impish
ID=ubuntu
ID_LIKE=debian
HOME_URL="https://www.ubuntu.com/";
SUPPORT_URL="https://help.ubuntu.com/";
BUG_REPORT_URL="https://bugs.launchpad.net/ubuntu/";
PRIVACY_POLICY_URL="https://www.ubuntu.com/legal/terms-and-policies/privacy-policy";
UBUNTU_CODENAME=impish
```
With kernel:
```
> uname -a
Linux ubuntu 5.13.0-1011-raspi #13-Ubuntu SMP PREEMPT Fri Nov 19 18:40:23 UTC
2021 aarch64 aarch64 aarch64 GNU/Linux
```
The issue reported over at [1] and it seems to not be present on debian
[2].
At the time the issue is triggered top shows the script using one of the CPUs:
```
top - 16:39:24 up 21 min, 10 users, load average: 1.86, 3.00, 3.52
Tasks: 194 total, 3 running, 191 sleeping, 0 stopped, 0 zombie
%Cpu(s): 11.1 us, 32.1 sy, 0.0 ni, 56.8 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
MiB Mem : 7808.5 total, 5194.8 free, 873.9 used, 1739.7 buff/cache
MiB Swap: 0.0 total, 0.0 free, 0.0 used. 6682.4 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
78617 ubuntu 20 0 8836 408 0 R 99.7 0.0 9:09.02 repro.sh
```
The stack of the CPU as reported by sysrq with the "l" option:
```
[ 876.197594] sysrq: Show backtrace of all active CPUs
[ 876.202717] sysrq: CPU1:
[ 876.202724] Call trace:
[ 876.202727] dump_backtrace+0x0/0x1ec
[ 876.202738] show_stack+0x24/0x30
[ 876.202743] showacpu+0x84/0x94
[ 876.202751] flush_smp_call_function_queue+0xd8/0x260
[ 876.202757] generic_smp_call_function_single_interrupt+0x20/0x2c
[ 876.202762] do_handle_IPI+0x1d8/0x310
[ 876.202769] ipi_handler+0x28/0x40
[ 876.202774] handle_percpu_devid_irq+0x94/0x270
[ 876.202780] __handle_domain_irq+0x88/0xec
[ 876.202787] gic_handle_irq+0x5c/0xf0
[ 876.202791] el1_irq+0xc0/0x15c
[ 876.202795] schedule_tail+0x12c/0x1fc
[ 876.202802] ret_from_fork+0x4/0x3c
```
The stack of the stuck:
```
> cat /proc/78617/stack
[<0>] __switch_to+0x138/0x190
[<0>] exit_to_kernel_mode+0x28/0x80
```
This is causing MicroK8s to fail after a few days. Thank you for your
help.
[1] https://github.com/ubuntu/microk8s/issues/2280#issuecomment-866912974
[2] https://github.com/ubuntu/microk8s/issues/2280#issuecomment-931306599
** Affects: linux-raspi (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1955174
Title:
Unkillable process pegging CPU
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-raspi/+bug/1955174/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
