0

I have a web server: ubuntu apache+php app+postgresql and a stats server: ubuntu apache+php - piwik and munin2 installed.

The communication for munin2 is made through ssh.

In munin i see a lot of system cpu activity, that I assume it it because of apache (i see 5 or 6 apache instances using ~5% CPU on top)

I was not having this system CPU activity before.

Does anyone knows how can I see where that comes from?

EDIT: some munin graphs munjin graphs

jperelli
  • 223
  • 5
  • 10
  • You were not having this system CPU activity befor[e] *what*? – womble Jul 02 '12 at 14:26
  • 2
    Leap Second Bug http://serverfault.com/questions/403732/anyone-else-experiencing-high-rates-of-linux-server-crashes-during-a-leap-second ? – Yohann Jul 02 '12 at 14:29
  • good question... I don't know really, but I moved piwik to the other server, installed mod_pagespeed, installed curl (php app uses it a little). Maybe I did more thinks, not sure. – jperelli Jul 02 '12 at 14:30
  • It might be that leap sec... not sure. I don't fully understand it. I'll attach munin graphs. – jperelli Jul 02 '12 at 14:40
  • @Yohann Please make that comment an answer and I will accept it. I finally updated the kernel, rebooted, and it works all ok. Because of the date (as seen on munin graphs) I'll supose that the problem was that leap second. – jperelli Jul 02 '12 at 15:05

2 Answers2

1

Check out the TOP command to see system resource utilization.

#man top
Univ426
  • 2,139
  • 14
  • 26
1

This is considered a bug. See Anyone else experiencing high rates of Linux server crashes during a leap second day?

Upgrading the kernel and restarting the system should solve the problem.

Community
  • 1
mailq
  • 16,882
  • 2
  • 36
  • 66