Two boxes (lxd-armhf{1,2}) have been running for three days with cking's
kernel, 3 is running the standard xenial kernel. 1 and 3 don't respond
to ssh connections any more; 2 still does, but several processes are in
'D' state:

root        47  0.0  0.0      0     0 ?        D    May13   0:00 [fsnotify_mark]
root     12892  0.0  0.0      0     0 ?        D    May13   0:00 [kworker/1:0]
root     26785  0.0  0.0      0     0 ?        D    May13   0:00 [kworker/1:2]

 (uninterruptible kernel sleep), such as lxd or systemd (pid 1). Calling
"top" hangs as well (but "ps" works), and there are a ton of zombie
processes (including lxd, check-new-release, sshd, socat).

All three continue to have lots of dmesg like

203722.697873] rcu_sched kthread starved for 15002 jiffies! g50919 c50918 f0x0 
s3 ->state=0x1
[209670.455074] INFO: rcu_sched detected stalls on CPUs/tasks:
[209670.456446]         3-...: (126 GPs behind) idle=a7e/0/0 
softirq=30811/30811 fqs=1 
[209670.457910]         (detected by 2, t=15002 jiffies, g=50962, c=50961, 
q=8377)
[209670.459430] Call trace:

So I'd say that this kernel did not really help (but also did not make
things worse for sure).

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1531768

Title:
  [arm64] locks up some time after booting

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1531768/+subscriptions

-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to