httpd-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Brett @Google" <brett.maxfi...@gmail.com>
Subject Re: [users@httpd] Apache 2.4 - non adoption reasons??
Date Tue, 22 Apr 2014 00:35:55 GMT
Apache 2.4 is technically better and many commonly use components have been
pulled into the Apache core, which had to be manually compiled before, but
pulling in some modules and changing them for Apache 2.4 broke
compatibility with said external modules, ie. importing mod_proxy_html but
not mod_proxy_xml, but having a different API and default charset etc.,
also using an external mod_proxy_html/xml for compatibility is not possible
due to conflicts with the internal one.

But IMHO the problem is mainly that if you have a legacy web site / reverse
proxy, or a site which requires particular versions of a custom Apache
module (mod_perl comes to mind) you need to stick with 2.2.x, until the
third party module SW is upgraded. This not a reflection on the Apache
team, only that software dependencies are sometimes complex, and many
modules are supported by free by third parties in their spare time (or not)
and need to be patched significantly for significant changes in api. There
can be up to 2 layers on to of Apache, all with their own API compatibility
issues at each level, not to mention the dependencies of Apache itself.

The content creation model is often to make a web site as a single pay for
development item, and then host it as an ongoing item, the hosting not
related to the individual developer. So the site needs to be deployed on a
server version which has a compatible ABI as it was developed against, as
the hosting organization likely has no relationship to the original
developer. This prevents the application being upgraded without having a
relationship to the original developer (and related SW stack).

Companies who manage a whole stack of application(s), and also host them
are pretty rare these days. Most apps get hosted on commercial web hosts.
There is no impetus of the original developer to upgrade the site, as they
are working elsewhere.

I don't see how the Apache developers can make such significant
improvements, without changing the API, so it is wise that they keep the
previous stable version, especially when the software interface changes
significantly. This is not an issue of the apache developers, simply that
software changes over time, and keeping changes in sync across non-related
projects (apr/apr-util/apache/mod_perl/perl/perl modules/Embperl as an
example) is difficult. But BTW Embperl just yesterday released a beta for
apache 2.4 - which i look forward to trying :)

On Fri, Apr 11, 2014 at 3:43 AM, Joey J <Joey@buymro.net> wrote:

> Apache 2.4 has had a stable release out for over 2 years but is only used
> by 2.5% of active Apache sites.   Why is the adoption so low?? The Apache
> foundation has been recommending upgrading to 2.4 for some time and looking
> at the improvements I see significant value in several.  I don't see any
> reason why anybody wouldn't want to use it but the community seems to think
> it's bad.
>
> What am I missing??
>
> Current market share:
> http://w3techs.com/technologies/details/ws-apache/2.4/all
>
>  -Joey J
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
> For additional commands, e-mail: users-help@httpd.apache.org
>
>


-- 
Whenever you find yourself on the side of the majority, it is time to pause
and reflect.

- Mark Twain

Mime
View raw message