Number of CPUs reported by /proc/stat fluctuates, causing issues

LXCFS
21

I’m getting exactly version 24164 but I should be getting a higher one?

22

24164 is fine

Stéphane

1 Like
23

Yes, that seems to have done the trick, the issue can’t be reproduced anymore :slight_smile:
Thank you for getting in a fix so quickly. Once again I’m very impressed by the LXD team!

2 Likes
24

After running a few days /proc/cpuinfo and the other lxcfs mounts became unreadable:

df -h
df: /proc/cpuinfo: Transport endpoint is not connected
df: /proc/diskstats: Transport endpoint is not connected
df: /proc/loadavg: Transport endpoint is not connected
df: /proc/meminfo: Transport endpoint is not connected
df: /proc/slabinfo: Transport endpoint is not connected
df: /proc/stat: Transport endpoint is not connected
df: /proc/swaps: Transport endpoint is not connected
df: /proc/uptime: Transport endpoint is not connected
df: /sys/devices/system/cpu/online: Transport endpoint is not connected
df: /var/snap/lxd/common/var/lib/lxcfs: Transport endpoint is not connected

LXCFS did crash:

show_signal_msg: 14 callbacks suppressed
lxcfs[3219179]: segfault at 0 ip 00007f8084afdf81 sp 00007f8084a2e780 error 6 in libc-2.31.so[7f8084a94000+178000]
Code: 00 00 4c 89 ef 4c 89 4c 24 08 e8 3a 68 00 00 48 89 e9 4c 89 e2 48 89 ee 48 8d 05 2a d2 15 00 4c 89 ef 48 89 84 24 e8 00 00 00 <c6> 45 00 00 e8 06 7e 00 00 89 d9 4c 89 fa 4c 89 f6 4c 89 ef e8 c6
25

Hi @zrav,

this means that lxcfs fuse daemon crashed for some reason.
I think it’s better to fill an issue on Github Issues · lxc/lxcfs · GitHub

I’ll take a look and try to figure out a reason.

You’ll need to restart all containers to make lxcfs work again.

26

@zrav please, provide the following information:

  1. service apport status
  2. cat /proc/sys/kernel/core_pattern
  3. ls -la /var/crash
  4. ls -la /var/lib/apport/coredump/
  5. cat /var/log/apport.log
  6. journalctl -u snap.lxd.daemon -n 200
27