Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 5 May 2009 18:55:24 +0200
From:      Paul Schenkeveld <fb-chat@psconsult.nl>
To:        freebsd-chat@freebsd.org
Subject:   Re: End of Life is Meaningless
Message-ID:  <20090505165523.GA90651@psconsult.nl>
In-Reply-To: <49FF8F2E.60800@highperformance.net>
References:  <49FF8F2E.60800@highperformance.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, May 04, 2009 at 05:58:22PM -0700, Jason C. Wells wrote:
> That should be read as "End of Life" is meaningless.  Not end of "Life is 
> Meaningless."  Life is still meaningless, as is this post if you disagree.
> 
> It mystifies me that there is this recent tendency for people to get 
> concerned about EOL. "What do I do?"  My answer, "Do nothing."  Just 
> because a FreeBSD version is EOL doesn't mean you have to stop using it.  
> It doesn't mean that your particular version is suddenly prone to downtime. 
>  It doesn't mean you can't install patches even though the secteam won't be 
> updating CVS. It doesn't mean you can't continue to develop applications 
> for a major version.
> 
> EOL is a tool for FreeBSD to manage its own house.  It is in no way a 
> directive on how you should manage your house. Queue someone still running 
> 2.1.5 with uptime stats. Come on.  You know you want to show off.
> 
> To the people who have to manage limited resources and must therefore 
> implement an EOL policy.  I commend you on the balancing act. Good on ya 
> mates.  Your doing a fine job.

I have several servers and a couple of old notebooks running versions
of FreeBSD that have been EOL'ed long time ago.  They are not accessible
from the Internet and I've got no potentially dangerous local users to
worry about.

On the other hand I administer dozens of FreeBSD systems for my own
company and for customers that are connected to the Internet or other
hostile networks.

I prefer to upgrade those systems every now and then and certainly
before they reach their EOL.  Not upgrading for a long time, past EOL
means that if a major vulnerability gets exposed I cannot simply do a
small upgrade using a published fix but that I have to either fix it
by hand, which can be time consuming if the sources have changed a lot
since, or do a big upgrade probably skipping multiple releases to get to
a supported (fixed) release exposing many challenges that I would have
faced one by one and not under time pressure if I had done incremental
upgrades.

So I think the EOL issue is not a black and white issue, it all depends
on many things.  Whether you run a handful of systems or a whole lot of
all different ones, what those systems are exposed to, the importance
of the systems and data on them, whether you own the systems of you are
responsible of someone elses systems and so on.

My $0.02

Paul Schenkeveld



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20090505165523.GA90651>