CentOS - May 2011 - So sorry! was: Re: EL 6 rollout strategies? (Scientific Linux)

As the OP for this thread, it saddens me to see that the thread I started has 
now been used as a forum for behavior of the worst kind seen in professional 
circles. 

I'm a longtime user of CentOS and merely wanted to know of users' past 
experiences transitioning between SL and CentOS. My first experience with 
CentOS was as a transition from Whitebox to CentOS, and the process was as 
simple as swapping out perhap 3 RPMs and a "yum update". That's
it, and that
same server is still chugging along today many years later. 

Both CentOS and WBEL had the same goal, near-perfect compatibility with RHEL, 
but SL has slightly different goals. What are users' past experiences doing
a
"hot swap out" of SL for CentOS and/or vice versa? 

I have tremendoud respect for the CentOS project. The developers here cast an 
extremely long shadow that we should all admire and respect. This is a bit of 
a dark time here, prompted by a delayed release of EL6 that was already 
delayed by Red Hat for their own (probably similar) reasons. SysAdmin 
frustration is at a high point as a result of these combined factors, causing 
many involved to question even the very best of decisions. 

I suspect that the many requests for ETAs are exacerbating the issue that the 
CentOS developers simply cannot predict a final release date, which is further 
exacerbated by the otherwise noteworthy, high-quality QA now in place, which 
is causing requests for build changes that delay the release and cause need 
for further QA checks on packages rebuild due to deps changes. 

It's a perfect storm of factors that, together, result in the current 
situation. I'm a developer; this type of situation is unfortunately common
and
can be frustrating for all involved, and just cannot be completely avoided! 

The choices are clear, however: 

1) Stick w/CentOS, get a high quality, highly compatible release at little/no 
cost, with an uncertain release date.
2) Switch to SL, get a high quality, reduced compatibility release (due to 
slightly different goals and qualities) at little/no cost, right now; 
3) Switch to RHEL, get a quality release at significant cost, right now. 
4) Complain loudly about how it's not perfect and/or get angry at people who
question your motives and/or competence after years of otherwise successful 
cooperation. 

I wish option #4 was not so commonly exercised here. It really might be a good 
time to consider moderation. Anybody want to volunteer as a moderator? 

-Benjamin Smith

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.centos.org/pipermail/centos/attachments/20110516/d458b82b/attachment-0005.html>

On Mon, May 16, 2011 at 11:17 PM, Benjamin Smith
<lists at benjamindsmith.com> wrote:
> The choices are clear, however:
>
> 1) Stick w/CentOS, get a high quality, highly compatible release at
> little/no cost, with an uncertain release date.
I would say the "uncertain release date" is pretty much moot now, as
CentOS has reached QA. According to the calendar, the date CentOS
(tentatively) is going to start syncing to the mirrors is May 31st.

http://qaweb.dev.centos.org/qa/

-- 
RonB -- Using CentOS 5.6

On Tuesday, May 17, 2011 12:17 PM, Benjamin Smith wrote:
> I wish option #4 was not so commonly exercised here. It really might be
> a good time to consider moderation. Anybody want to volunteer as a
> moderator?
The Centos ML does quite well without a moderator imho. No need to go 
draconian like other projects/'communities'