0

I discovered that on my Ubuntu 16.04 VPS a process is getting killed due to CPU usage.

Running ulimit -t shows 86400.

systemctl show | grep LimitCPU shows

DefaultLimitCPU=18446744073709551615
DefaultLimitCPUSoft=86400

I talked to the provider support and they said they don't set any limits for the OpenVZ container.

I looked in /etc/systemd/system.conf and /etc/systemd/user.conf but I didn't find anything.

Any idea where this limit could be set? It's driving me mad, especially since I didn't add this limit and I have another VPS with same software & settings that doesn't have this limit.

Thomas
  • 4,225
  • 5
  • 23
  • 28
Gaspy
  • 3
  • 1
  • What is the proof, that the process gets killed due to CPU usage? As a workaround you could renice the process or split the work in two if possible. – Tomáš Pospíšek Dec 15 '17 at 16:37
  • I saw in syslog `systemd[1]: my.service: Main process exited, code=killed, status=24/XCPU` The process needs to run uninterrupted and with normal priority. – Gaspy Dec 15 '17 at 16:40
  • ah, I see here: https://serverfault.com/questions/888289/what-is-status-24-xpu-in-syslog – Tomáš Pospíšek Dec 15 '17 at 16:46
  • Your VPS provider probably doesn't know what they're doing. This is just one more reason to avoid OpenVZ. – Michael Hampton Dec 15 '17 at 20:30
  • Possible, but I have another OpenVZ container, with the same provider, same Ubuntu version, pretty much the same software installed, just on a different datacenter, that does not have this limitation. – Gaspy Dec 15 '17 at 20:56

1 Answers1

0

I think what you are looking for is here: https://openvz.org/Resource_shortage

Check if you have failcnt > 0 if you do this:

cat /proc/user_beancounters
Tomáš Pospíšek
  • 233
  • 1
  • 7