2

I have a Cent OS 6.4 Apache Nginx VPS. I have WHM cpanel. How do I check past 24 hours or 48 hours this server RAM usage?

I use new relic server monitor web application. It reported that, this VPS peak time RAM usage less than 30%. Also WHM service status shows similar figures. But when I'm contacting my VPS company, they mentioned that, it uses more than 95% RAM (peak time). I'm totally new to the server management.

Updated!

When I run free -m command, I receive following output. But at the same time WHM cPanel service status shows, Memory Used 16.88% (442,464 of 2,621,440). Also new relic shows similar statistics (Similar to the WHM).

root@host [~]# free -m
             total       used       free     shared    buffers     cached
Mem:          2560       1117       1442          0          0        748
-/+ buffers/cache:        368       2191
Swap:            0          0          0
root@host [~]#
7
  • What is the reason for down vote?
    – MaxBro57
    Commented Feb 2, 2014 at 8:53
  • 1
    Does this 'relic server monitor' tell you total RAM consumption? Or just active RAM use? It's important to note the difference as Linux tends to cache large volumes of stuff in RAM, but not 'actively use it'. This results in performance metrics than can appear skewed. (E.g. your server might be consuming 95% RAM but only actively using 30%). I know this isn't answering your actual question - but it could help explain the difference you're seeing. Help us out by posting the output of "free -m".
    – blacklight
    Commented Feb 2, 2014 at 9:14
  • Yes Blacklinght, it shows overall statistics about the server. This is a wired issue. When I try free -m it shows total-2560 used-1117 free-1442. But WHM cPanel Service Status shows 338,696 of 2,621,440.
    – MaxBro57
    Commented Feb 2, 2014 at 10:27
  • 338 megs of memory is used by actual programs/processes on your system. 748 megs is used to cache things like recently opened files. The cache can go away without bad things happening. If non-cache memory runs out, bad things happen (namely: processes start to get killed by the kernel until there's enough free memory again)
    – samiam
    Commented Feb 2, 2014 at 10:44
  • You mean cached memory + actual programs/process memory=used memory (in SSH) Therefore, WHM and new relic statistics and SSH statistics are same?
    – MaxBro57
    Commented Feb 2, 2014 at 10:51

1 Answer 1

0

Is there a Solus or equivalent control panel with your VPS account? Most VPS control panels provide 1-day memory usage graphs. For example, with Solus (a popular VPS control panel), click on the "statistics" tab near the bottom of the control panel.

As an aside, if you would like to learn how to manage things without needing to depend on WHM cpanel, and are comfortable with learning how to use a command-line terminal, you are welcome to ask questions over at http://unix.stackexchange.com -- it's a smaller and tighter community than serverfault, with a lot of really experienced old-school UNIX gurus.

3
  • I have WHM. Its service status shows 338,696 of 2,621,440 and new relic also reporting same figures. But Muni service monitor and SSH shows more than 1GB RAM usage (same time). Thank you samiam your suggestion.
    – MaxBro57
    Commented Feb 2, 2014 at 10:31
  • What is the exact output of free (edit and add this to the question)
    – samiam
    Commented Feb 2, 2014 at 10:33
  • just updated the question.
    – MaxBro57
    Commented Feb 2, 2014 at 10:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .