11-13-2007
First off: if you tar something, the command will read something on your disk (a bunch of files) and then create an archive from it, writing that to the disk (most of times). What else then seeing activity on your disk do you suppose will happen?
Additionally i would like to clarify what the "% tm_act" field in the output of iostat means:
The OS has a sensor, regularily asking the disk if it is busy or not. When the disks aswers half of the times "I'm busy", then the "% tm_act" will be 50%. If the disk answers every time "I'm busy" then tm_act will be 100%, etc.. A disk answers with "busy", when there are requested operations not yet fulfilled, read or write. If many very small requests go to the disk the chance of the sensor asking exactly when one such operation is still open goes up - much more so than the real activity of the disk.
So, "100% busy" does not necessarily mean the disk is at the edge of its trasnfer bandwidth. It could mean either that because the disk is getting relatively few but big requests (example: stream I/O) but it could also mean that the disk is getting a lot of requests which are relatively small so that the disk is occupied most of the time, but not using its complete transfer bandwith.
To find out which is the case analyse the corresponding "Kb_read" and "Kb_wrtn" column from iostat. You know how much a modern disk drive can approximately handle (~17MB/second) physically and bypassing any cache. Compare your data to this (rule-of-thumb-)value and you will get a more detailed picture.
bakunin
Last edited by bakunin; 11-13-2007 at 08:42 AM..
10 More Discussions You Might Find Interesting
1. UNIX for Dummies Questions & Answers
hi,
Can anyone tell me how I make a tree on the console showing the largest folders on my UNIX system?
I want to know where are located the biggest files so I can free some disk space.
thanks a lot. (3 Replies)
Discussion started by: gandoura
3 Replies
2. HP-UX
I want to get hard disk usage
hard disk usage= already allocated/total allocated
first I use ioscan -kfnC disk get disk name c#t#d#
then I use diskinfo /dev/rdsk/c#t#d#
but this command can only get total disk size,not disk usage
has any command can get that
thanks:) (2 Replies)
Discussion started by: alert0919
2 Replies
3. Solaris
I have a SparcStation5 that is making sounds that make me think the disk drive(s) may be on the verge of quiting. What is the command to list the disk types and sizes? I'm thinking I can possibly pick up another drive or two and compy the exisiting drives while they are still working. (1 Reply)
Discussion started by: muletrainman
1 Replies
4. Shell Programming and Scripting
Hi,
I need a command that gives me the user who is using more space in the unix box (1 Reply)
Discussion started by: abinaya
1 Replies
5. UNIX for Dummies Questions & Answers
How to determine the total percentage cpu usage of a server having 4 cpu cores(quad core)? (3 Replies)
Discussion started by: proactiveaditya
3 Replies
6. Linux
Hi all,
I'm kind of new to programming in Linux & c/c++. I'm currently writing a FileManager using Ubuntu Linux(10.10) for Learning Purposes. I've got started on this project by creating a loopback device to be used as my virtual hard disk. After creating the loop back hard disk and mounting it... (23 Replies)
Discussion started by: shen747
23 Replies
7. Linux
Hi,
Am doing an enhancements related to monitoring a Linux disk I/O statistics.
The /proc/diskstats file is used to get the each disk I/O statistics. But, It returns the raw value.
How to calculate the Disk Queue Length and Disk Busy time from the raw values.
Guide me. (1 Reply)
Discussion started by: maruthu
1 Replies
8. AIX
Please help with command to check the busy % for a hard disk device like hdisk5
Best regards,
Vishal (7 Replies)
Discussion started by: Vishal_dba
7 Replies
9. UNIX for Dummies Questions & Answers
Let's say i have 20 users logged on Server. How can I know how much memory percent used each of them is using with system time in each user? (2 Replies)
Discussion started by: roy1912
2 Replies
10. Shell Programming and Scripting
Given this directory /web
I need to get the current usage (in %) on Linux and Unix both using the same command on bash shell ?
The command i tried was working on Unix (solaris) but does not filter the desired same value when run of Linux.
My command df -h /web | awk '{print $5}' | sed -n... (5 Replies)
Discussion started by: mohtashims
5 Replies
LEARN ABOUT DEBIAN
watch-multiple-mimedefangs
WATCH-MULTIPLE-MIMDEFANGS(8) System Manager's Manual WATCH-MULTIPLE-MIMDEFANGS(8)
NAME
watch-multiple-mimdefangs.tcl - Keep an eye on a cluster of MIMEDefang machines
SYNOPSIS
watch-multiple-mimdefangs.tcl [-archive] machine-1 machine-2 ... machine-N
DESCRIPTION
watch-multiple-mimdefangs.tcl is a Tk script that graphically displays the status of mimedefang-multiplexor(8) on a cluster of machines.
Note that Tcl/Tk 8.4 or higher is required to run watch-multiple-mimdefangs.tcl.
If you supply the -archive command-line option, then watch-multiple-mimdefangs.tcl logs the output of md-mx-ctrl rawload for each machine.
The output for machine_name is logged in:
$HOME/.watch-multiple-mimedefangs/machine_name/data.
PREREQUISITES
To use watch-multiple-mimdefangs.tcl to monitor a cluster of machines, you need the following prerequisites:
o A UNIX-like machine with Tcl/Tk 8.4, the X window system and an SSH client. You also need "xterm" to use some of the features.
o The ability to SSH as root to all of the machines you want to monitor. Ideally, you should be using the SSH agent so that you can
SSH to the monitored machines as root without being prompted for a password or pass phrase.
o The program md-mx-ctrl on root's path on all the machines you want to monitor.
DISPLAY
For each machine specified on the command line, watch-multiple-mimdefangs.tcl creates a chart with five columns. The columns are:
o A button with the name of the machine. Clicking the button pops up a menu that lets you take various actions, as described later.
If all slaves on the machine are busy, the button turns yellow.
o A label showing the number of busy slaves in the form "busy/total", where total is the total number of slaves.
o A label showing the average number of messages per second over the last 10 seconds.
o A label showing the average number of milliseconds per scan over the last 10 seconds.
o A chart that graphs the average number of busy slaves, the average number of messages per second and the average scan time in mil-
liseconds, all averaged over the last 10 seconds.
MACHINE MENU
If you click on a machine name, a menu with three options pops up:
SSH Open an xterm session and ssh as root to the machine.
Busy Slaves
Monitor the busy slaves on the machine. If you click on the process-ID of a slave, an xterm will open up and the command "strace -s
100 -t -p pid" will be executed on the remote machine. This is Linux-specific, but you can edit watch-multiple-mimdefangs.tcl to
replace the command with your particular system's command for tracing system calls.
Delete Remove the machine from the list of machines being monitored.
ADDING A MACHINE
If you need to add a machine to the display, simply type the name of the machine in the "Add Machine:" box and press Enter.
AUTHOR
watch-multiple-mimdefangs.tcl was written by David F. Skoll.
SEE ALSO
mimedefang.pl(8), mimedefang-filter(5), mimedefang(8), mimedefang-protocol(7), md-mx-ctrl(8), watch-mimedefang(8)
4th Berkeley Distribution 12 January 2007 WATCH-MULTIPLE-MIMDEFANGS(8)