[ale] Absolutely deplorable Linux quiz

[Jim Kinney]

On the average, the answers are mostly correct but for some very old
setups. Some of the explanations show a fundamental misunderstanding
of a (modern) implementation and/or a lack of technical proficiency in
English. With a name like Rupali Kapil I expect English to be second
language and the wording of things backs this up.

As it's set several people's teeth on edge, how about a similar set of
Q&A's we can post on the ALE server with a permanent link? I have a
"Intro to Linux" training program I wrote for work (I'm cleaning out
the work specifics from first) and will post that into a plain web or
wiki site. It's originally a wiki.

Hmm. Maybe I should get my new web server system actually finished and
online. Then I'll have the horsepower and space to have a wiki site.

On Sat, Feb 26, 2011 at 1:09 AM, Michael B. Trausch <mike at trausch.us> wrote:
> On Fri, 2011-02-25 at 23:57 -0500, Pat Regan wrote:
>> Stumbleupon just gave me this page.  It is horrendous.  After seeing
>> how poorly worded some of the question were I just had to click through
>> for my score.  I can forgive them for a technicality on question number
>> 1...
>>
>> The first oddly phrased question is probably number 4, I had no idea
>> what they were really trying to get out of question 5.  Getting marked
>> wrong for number 9 when I picked the correct answer (none of the above)
>> is just ridiculous...
>>
>> http://www.careerride.com/Linux-Interview-Questions.aspx
>
> Okay, there are so many issues with that page that I don't even know
> where to begin.
>
> So, uh, I won't.  It's awful, and that's it.
>
> I lied.  "Every user has a defined quota in Linux. This is done mainly
> for the security, as some users have only limited access to files."
> What?  What the bloody hell does a quota have to do with filesystem
> permissions?
>
> Then there's the second one.  While dmesg is, in certain system
> configurations, a file stored in /var/log that stores kernel messages,
> dmesg is firstly a command that displays (and optionally clears) the
> kernel's ring buffer.  Wow.
>
> Next! There is nothing unique to Linux about its security model in the
> default configuration, and for that matter PAM is neither kernel-space
> nor is it Linux-only.  And one could make a perfectly valid argument
> that the design of PAM is in fact detrimental to the goal of having a
> secure system---at least in the default configuration where shadow
> password files are used---because applications can only authenticate if
> they have the ability to read the shadow file!  Either that or you have
> to have to fuss around with permissions and the odds of screwing that up
> aren't exactly zero.
>
> Alright, someone REALLY needs to hit this idiot on the head.  THIS KIND
> OF STUPIDITY IS EXACTLY WHY WE HAVE SUCH A HARD TIME EXPLAINING
> NETWORKING TO PEOPLE:
>
>> Yes a Linux machine can be made a router. This is called "IP
>> Masquerade." IP Masquerade is a networking function in Linux similar
>> to the one-to-many (1: Many) NAT (Network Address Translation) servers
>> found in many commercial firewalls and network routers. The IP
>> Masquerade feature allows other "internal" computers connected to this
>> Linux box (via PPP, Ethernet, etc.) to also reach the Internet as
>> well. Linux IP Masquerading allows this functionality even if the
>> internal computers do not have IP addresses.
>
> WHAT?!  The kernel can route, and it can also do NAT, but one doesn't
> have any bloody thing to do with the other!
>
> And NAT isn't a server!  UGH!
>
> Minimum number of partitions?  1.  Period.  You can put swap on a
> filesystem.  It's not as efficient.  It's not as safe.  But it can be
> done.  Hell, for that matter, one can use LVM and boot directly onto a
> logical volume---logical volumes aren't partitions!
>
> Oh, finally they mention the ring buffer.  And then they go and say that
> pipes allow things to be viewed one screen at a time.  Oops.
>
> logrotate, sure.  But it's not the only game in town.  Oy.
>
> "the mail server"?  'scuse me, which one are we talking about?
>
> nice does not make a process use less CPU time.  It makes it get less
> CPU time slices as compared to other processes.  That just means that
> its job is going to take longer than it did before; it won't use any
> more or less CPU time, just wall-clock time.
>
> "How to change window manager by editing your home directory?"  I heard
> that some older UNIX systems let you view and edit directories directly,
> but I'm pretty sure that Linux does not allow that type of behavior...
>
> And is ~/.xinitrc even used anymore, anyway?  I was pretty sure that
> things are all done with .desktop files now, unless you go out of your
> way to do it differently.
>
> And why is this page saying *anything* about whether or not
> documentation is installed at the time of a package's installation?
> Yeah, sure, for some things.  For others, the documentation is freakin'
> huge and is either not packaged at all or is packaged separately.
>
> I can't even make it through the rest.  It hurts too much.  I mean,
> seriously.  What is this supposed to be, as accurate as a CompTIA exam?
>
>        --- Mike
>
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> http://mail.ale.org/mailman/listinfo/ale
> See JOBS, ANNOUNCE and SCHOOLS lists at
> http://mail.ale.org/mailman/listinfo
>



-- 
--
James P. Kinney III
I would rather stumble along in freedom than walk effortlessly in chains.