uptime information...

uptime information...

Post by Gary Ta » Wed, 12 Jun 2002 22:47:15



I've been getting the following output from uptime:
9:33pm up 11 days 9:26, 1 user, load average: 194.32, 189.00, 176.46

From everything I've read, these numbers are showing extreme load - or are
they accurate? The server is being used as a web server and nothing else.
What do these numbers mean, and what is considered "normal"?

Thanks,

Gary

 
 
 

uptime information...

Post by Paul Kimot » Wed, 12 Jun 2002 23:55:10



> I've been getting the following output from uptime:
> 9:33pm up 11 days 9:26, 1 user, load average: 194.32, 189.00, 176.46

> From everything I've read, these numbers are showing extreme load - or are
> they accurate? The server is being used as a web server and nothing else.
> What do these numbers mean, and what is considered "normal"?

These numbers are supposed to be taken from /proc/loadavg, and, quoting the
proc(5) manpage:

:        loadavg
:              The load average numbers give the number of jobs in
:              the run queue (state R) or waiting for disk I/O
:              (state D) averaged over 1, 5, and 15 minutes.  They
:              are the same as the load average numbers given by
:              uptime(1) and other programs.

Probably top(1) and ps(1) can help you tell whether there are a lot of
processes trying to run, or trying to do I/O, or something else.  Do you
think that the system is running poorly?

I imagine that "normal" would be somewhere between 0 and a small multiple
of the number of processors installed.

--
Paul Kimoto
This message was originally posted on Usenet in plain text.  Any images,
hyperlinks, or the like shown here have been added without my consent,
and may be a violation of international copyright law.

 
 
 

uptime information...

Post by Yuan Li » Thu, 13 Jun 2002 04:19:21



> I've been getting the following output from uptime:
> 9:33pm up 11 days 9:26, 1 user, load average: 194.32, 189.00, 176.46

> From everything I've read, these numbers are showing extreme load - or are
> they accurate? The server is being used as a web server and nothing else.
> What do these numbers mean, and what is considered "normal"?

Depending on what's happening to the Web server, it's possible.  Yeap,
that's hell.  I ran into this situation before, with extremely low
hardware resource and extremely high hit rate.  But the box survived for
a long time.

Yuan Liu

Quote:> Thanks,

> Gary

 
 
 

uptime information...

Post by Michael Heimin » Thu, 13 Jun 2002 04:43:12



Quote:> I've been getting the following output from uptime:
> 9:33pm up 11 days 9:26, 1 user, load average: 194.32, 189.00,
> 176.46

> From everything I've read, these numbers are showing extreme load
> - or are they accurate? The server is being used as a web server
> and nothing else. What do these numbers mean, and what is
> considered "normal"?

No, it's really high, in general running with a load of around 1
per CPU is considered "healthy". You should check using
top/iostat/vmstat what's going on and resolve the problem, dynamic
pages written like 'shit' and/or the box is constantly swapping, or
there's just a runaway proc, that needs to be killed, who knows,
multiple possibilities. On the other hand it shows how reliable the
kernel runs....;-)

Michael Heiming
--
Remove the +SIGNS case mail bounces.

 
 
 

1. Help! Uptime information

Hello Unix Programmer:
    I am looking for more information about Uptime and Downtime of system.
Does anyone know about software that can keep records of uptime and downtime
system? I am writing routines to keep them and put them into its file. MY
Unix manuals do not tell me more information well. I complaint that lack
of information for accessing to kernel area. Because, service level agreement
require that.

-- Tim Stark
   Computer Programmer/Analyst / System Adminstrator

--

4754 Kenmore Ave, Apt 203   Home: (703) 212-9731 TDD/TTY  BIX:   tstark
Alexandria, Va. 22304-1275  Work: (202) 377-9606 TDD/TTY  GEnie: T.STARK1
Coming to your near e-mail service - Portal and FAX!

2. modules char-major-6?

3. Uptime rollover, How to tell real uptime?

4. Portable Unix shell scripts

5. How does uptime.netcraft.com know my uptime?

6. 19.15 - 10.00 - 9.15 not equal to 0?

7. Custom Logs With Web App Specific Information That Is NON Cookie NON Header Information?

8. On A Resume, A Little Unix Goes A Long Way - Webmonkey

9. looking for information on how to query a pc for system hardware information

10. getting the network interface information and default gateway information ??

11. HELP: /var filesystem fills. df information != du information

12. Segfaults on Server after long uptime using 2.2.x