Difference Between Orphans and Zombies

[Karthick S.]

1841HRS 29 Jun 2004

Hi,

Can anyone tell me some good references to materials on Zombies
and Orphans? I have asked this question to many people and have
received different replies from each of them. Some of the answers even
cost me interviews.

So can anyone give me the name of some resource (in the net or
otherwise) about this.

If you are giving some direct replies, it would be better if you
can substantiate your replies with references to books, just so that
we can get more info on the subject that way.

Thanks in advance.
with warm regards,
Karthick S.

[Lew Pitcher]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Karthick S. wrote:

> 1841HRS 29 Jun 2004
>
> Hi,
>
> Can anyone tell me some good references to materials on Zombies
> and Orphans?

Google is your friend. STFW, and you shall find all that you seek.
Or, you could read a good book on Unix, or on OS design, such as
- - "Advanced Programming in the Unix Environment" by W. Richard Stevens
- - "Operating Systems Design and Implementation" by Andrew Tanenbaum and
Albert Woodhull

> I have asked this question to many people and have
> received different replies from each of them. Some of the answers even
> cost me interviews.

The definition of an orphan and a zombie process is well known, and has
been standardized for decades. I'm surprised that you find a variance in
the definitions, wide enough to cause problems at job interviews

> So can anyone give me the name of some resource (in the net or
> otherwise) about this.
>
> If you are giving some direct replies, it would be better if you
> can substantiate your replies with references to books, just so that
> we can get more info on the subject that way.

Briefly,
- - Each process has a parent process, culminating with the init (pid 1)
process

- - The parent process is expected to monitor the health of any children
processes it spawns.

- - If the parent process terminates before the children processes do,
then the children processes become 'orphans', and are 'adopted' by the
init process (pid 1)

- - Orphan processes are 'live'; they are executing, and take up both a
process table slot, and real system resources (CPU, memory, I/O)

- - If the child process terminates before the parent process does, then
the child process becomes a 'zombie' until the parent process
retrieves the child process' status. Normally, this interval is so
small that it is not noticable in human time. However, when a parent
process refuses to reap it's dead children's status, or defers that
activity for an extended period of time, the child process is shown
as a 'zombie' process.

- - Zombie processes are 'dead', but not buried. They are no longer
taking up real system resources (CPU, memory, I/O), but /do/ take up
a process table slot. They will give up even this, once their parent
(or init) reaps their process status.


[url]http://www-cdf.fnal.gov/offline/UNIX_Concepts/concepts.zombies.txt[/url]
[url]http://www.cs.princeton.edu/courses/archive/fall03/cs217/precepts/12/orphanzombie.pdf[/url]

> Thanks in advance.
> with warm regards,
> Karthick S.

- --

Lew Pitcher, IT Consultant, Enterprise Application Architecture
Enterprise Technology Solutions, TD Bank Financial Group

(Opinions expressed here are my own, not my employer's)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (MingW32)

iD8DBQFA4XC+agVFX4UWr64RAsMqAKC58ypw2YPERR0l1CyGe9 EHYYzIEwCeNaej
lgjxQOnID+dA3JjKBnkedyI=
=TCQh
-----END PGP SIGNATURE-----

[Davide Bianchi]

Karthick S. <tosk@rediffmail.com> wrote:

> If you are giving some direct replies, it would be better if you
> can substantiate your replies with references to books, just so that

See [url]http://www.cseg.ca/recorder/199802/3.html[/url]
And a quick search with google would have given you a lot more.

Davide


--
| Command, n.: Statement presented by a human and accepted by a computer
| in such a manner as to make the human feel as if he is in control.
|
|