LVS
lvs-users
Google
 
Web LinuxVirtualServer.org

[lvs-users] RHEL Cluster Suite. Linux J, Nov 2007, p 66

To: lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Subject: [lvs-users] RHEL Cluster Suite. Linux J, Nov 2007, p 66
Cc: rnibali@xxxxxxxxx, Roberto Nibali <ratz@xxxxxxxxxxxx>
From: Joseph Mack NA3T <jmack@xxxxxxxx>
Date: Sun, 7 Oct 2007 09:17:36 -0700 (PDT)
I mention this article because I think RHELCS has something 
to do with LVS, although this being Linux J, you won't find 
out from the article (I'm dropping my subscription, by the 
way, after 3 bad articles on LVS).

The only clue you get about a possible involvement of LVS is 
on p2 of the article where, after waxing lyrical in 
market-speak on the virtues of RHELCS, the author finally 
reveals what it does - it's high availability with load 
balancing - and the load balancing used to be called 
Piranha. Piranha has a special spot in my heart.

http://www.austintek.com/LVS/LVS-HOWTO/HOWTO/LVS-HOWTO.unsupported.html#pbs_nutshell

The references at the bottom of the RHELCS article are all 
to Linux-HA websites and none to LVS.

The article aside, it would seem that RHELCS has all the gui 
bells, whistles and monitoring tools that LVS lacks and that 
are an essential part of a commercial product. But 
underneath RHELCS is likely functionally equivalent to 
ultramonkey. (Why isn't RH sending patches back to us?)

The author, Daniel Bartholomew, lives in NC (as I do). A 
google search finds that he's a member of TriLUG (as I am, 
but I don't go to meetings much and I don't recall having 
met him). His postings to TriLUG show him to be a normal 
helpful nice guy. google reveals no hits for the author with 
RHEL Cluster Suite but he did post to one thread in the 
Linux-HA mailing list.

Other articles in this month's Linux J, written by people 
who aren't involved in the project they're writing about, 
include an article on Beowulfs, by David Morton, who doesn't 
post to the Beowulf mailing list.

So why have I, after reading 3 bad articles on LVS in Linux 
J, not written a good one myself?

The bigger question then is why don't people involved in 
projects write about them for magazines. An even bigger 
question is why do I subscribe to Linux J, when I can't 
think of anything I've learned from it in 10-15yrs of 
subscribing (if I want to understand/install/use a project, 
I usually download the docs). I enjoy reading Maddog's 
articles in Linux J - he has a more global picture of Linux 
and the social effects of computers than any of the coders.

o After the first bad article on LVS, I wrote to the 
editors, listing the missconceptions and errors and offering 
the services of the LVS team to review any further articles 
on LVS. I got an unconvincing "sure, sure, we'll contact you 
next time", and they never did.

o It takes time and after answering posts here and 
maintaining the HOWTO, I want to do other things with my 
time. Recently I posted a presentation I gave on LVS at UNC. 
They've been asking for a presentation for over a year and I 
haven't wanted to do it. Heaven's knows, if anyone should be 
able to give a presentation on LVS, I should and anyone who 
wants a presentation on LVS would reasonably wonder why I 
don't have one ready to go and be ready to give next week. 
The presentation took about a month of spare time to write.

o There seems to be little overlap between people who write 
for magazines and people involved in the projects written 
about. When I post here, it doesn't take real long (since I 
know, before I post, whether I have the answer) and I'm 
attempting to solve someone's problem. Most of the time I 
get a reply telling me whether it was useful or not and I 
learn something. Writing for a magazine would take a long 
time and I won't hear anything back from the readers. Is 
anyone any better off, or did the magazine just fill another 
issue with content that takes up a subscriber's spare time, 
but doesn't help them?

I wonder if all the other (non-technical) magazines I 
subscribe to have as little to do with the topics of the 
articles as does Linux J? Maybe I should drop all my 
subscriptions.

Books on the other hand are usually written by experts in 
the field. At least the books I read are written by 
academics who are expected to write as part of their job and 
the books enhance their reputation. I doubt if anyone 
writing for magazines gets these benefits.

Joe

-- 
Joseph Mack NA3T EME(B,D), FM05lw North Carolina
jmack (at) wm7d (dot) net - azimuthal equidistant map
generator at http://www.wm7d.net/azproj.shtml
Homepage http://www.austintek.com/ It's GNU/Linux!


<Prev in Thread] Current Thread [Next in Thread>
  • [lvs-users] RHEL Cluster Suite. Linux J, Nov 2007, p 66, Joseph Mack NA3T <=