[fbsd] HEADS UP: FreeBSD 5.3, 5.4, 6.0 EoLs coming soon

Doug Barton dougb at FreeBSD.org
Wed Oct 11 15:15:30 PDT 2006


Jason Stone wrote:

> I realize that resources to keep chasing this stuff are in limited 
> supply,

You just hit the nail on the head. The vast majority of FreeBSD 
developers (including but not limited to the committer community) have 
moved on. If you (meaning the people that want continued support for 
4.x) want to see this supported, you're going to have to step up and 
support it.

I see at least 2 areas that will require work:
1. Ports -- the portmgr team for sure, and to a large measure the port 
committers, do not have the resources necessary to continue support 
4.x. The cluster (read, package building) resources could almost work 
if 5.x support is totally dropped (which is an intriguing idea), but 
the manpower to keep things working (ports compiling, etc.) isn't 
there right now, and would have to be supplied from this community. I 
like the idea proposed by another poster that requiring perl, openssl, 
and gcc 3.4 from ports makes most things work. Perhaps something can 
be added to bsd.port.mk to make that happen transparently? My point 
here is that the resources have to come from somewhere, because they 
are running very thin now, and will get much thinner soon.

2. Security updates -- some committers are still interested in this, 
but not many. Work would have to come from the community to keep key 
things updated (and of course, the SO team would have to sign off on 
that effort).

In order to facilitate this effort, I'd like to suggest that a new 
mailing list be created, freebsd-releng4. That would allow the 
interested folks to get together, pool resources, and decide what is 
possible.

One last suggestion, for those of you who are still using 4.x for 
performance reasons, perhaps you could dedicate a system or two, and 
some of your resources, to helping determine where and how things need 
to be improved. I could make an argument that doing this would 
actually give you a better ROI than putting work into trying to keep 
4.x (and the old hardware it runs on) alive, but I won't.  :)

Finally, I think it's important to keep in mind that unless the 
personpower comes from somewhere, official support for 4.x WILL go 
away. It's nothing personal, it's simply a question of resources.

Doug

-- 

     This .signature sanitized for your protection


More information about the freebsd-security mailing list