[linux-elitists] What to do about cluebatting such companies, that require possibly *YEARS* old Distros

Aaron Sherman ajs@ajs.com
Wed Jan 26 13:32:56 PST 2005


On Wed, 2005-01-26 at 16:14, Rick Bradley wrote:

> In a properly managed production environment software upgrades of any
> size should not be an ordeal in the least.

This is a rose-colored view of the world. "Properly managed" here
assumes that it's possible to apply that term to software which is, at
best, difficult to manage. I have never in my life seen a
systems/operations team that could honestly say that an upgrade of even
a medium sized environment (say 500-1000 machines which is medium in my
experience, but many people have scoffed at my thinking so small in the
past) was straight-forward.

>  One can run their own
> package repositories for distro of choice, and segregate testing from
> production for the various components being run.

You make that sound like a fairly simple proposition, and I guess if
you're running some kind of well-isolated, homogeneous application then
that's workable. In the real world, no test environment is perfectly
representative of production (a bastardized word at best). Load can
never be perfectly simulated, users can never be perfectly simulated,
buying a complete duplicate of all hardware is almost always cost
prohibitive, monitoring differs by necessity, and most important of all:
customers aren't calling you at 3AM to tell you that your test
environment is broken and customers are the most aggressive QA in the
world.

You are also ignoring the fact that we're talking about not upgrading
systems in 3 years.... 3 years... that's a blink of the eye on most
corporate timelines. Why on earth would you expect software to be
upgraded that fast when it comes with nearly no benefit to the company
that does so (and before you say there are tons of benefits, please show
me the examples of the companies that are running Red Hat/SuSE 7 that
are being beaten out of their markets by the guys running the latest and
greatest).

> Then again, I suppose we're talking about improperly administered
> environments almost by definition in this thread, so, um, carry on.

If we're talking about successful companies, then I debate your use of
the word "improperly". The company is not in business to upgrade an OS.
They are in business to make money. Demonstrate the cost/benefit
analysis that shows them making more money using kernel 2.6 everywhere
and I'm sure they'll move faster.

-- 
☎ 781-324-3772
✉ ajs@ajs.comhttp://www.ajs.com/~ajs




More information about the linux-elitists mailing list