Interpreting TOP and GLANCE outputs

[Musa I. S.]

Hello,

I need a detailed explanation on the output of TOP and GLANCE.

For example;
>324 processes: 278 sleeping, 41 running, 1 zombie

what does zombie mean?

[john123 · Accepted solution]

Zombies are the processes which actually completed the work and exited but its parent process is still waiting for the zombie process to exit..

HTH

John

[Shrikant Lavhate]

Hi,

====zombie process====
The name given to a process which terminates for any reason, but whose parent process has not yet waited for it to terminate (via wait(2)). The process which terminated continues to occupy a slot in the process table until its parent process waits for it. Because it has terminated, however, there is no other space allocated to it either in user or kernel space. It is therefore a relatively harmless occurrence which will rectify itself the next time its parent process waits. The ps(1) command lists zombie processes as defunct.

From glossery page : http://docs.hp.com/en/B2355-60105/glossary.9.html


About each values displayed in top, go through man page of top

#man top

http://www.docs.hp.com/en/B2355-90690/top.1.html

[Jeeshan]

A zombie process is a process that has finished execution but whose parent has not issued a "wait()" The zombie process will go away as soon as the parent process does the "wait()". If they are staying around for a while, then you have an application that is not behaving properly and you should try and investigate which one it is. It is worth pointing out that unless you get lots and lots of these, they aren't taking up resources.

[Musa I. S.]

>Load averages: 0.41, 0.32, 0.28

how do i interpret the line above. which values should i benchmark that against?

[Venkatesh BL]

They will be listed as '' in the ps output. But, you cannot kill something that is already dead.

[Shrikant Lavhate]

You should go through man page completly. It reflects:

Load averages in the last one, five, and fifteen minutes of all the active processors in the system.

[Kenan Erdey]

load avarage is number of processes in run queue in a period of time.

in top output: numbers are for 1 minute, 5 min and 15 minute avarage. if these numbers are high it may be a performance bottleneck.

[Musa I. S.]

Thank you all. The man pages really helped.