You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by "William A. Rowe, Jr." <wr...@rowe-clan.net> on 2008/04/11 16:39:35 UTC

Ship 1.3.0 apr in httpd 2.2.9

So here's a thought from Davi's^WColm's APR talk in speaking with
chipig, jorton and others...

We have several inheritance quirks addressed by 1.3.0 apr, and the
new memcache, ssl and other features are present in 1.3.  The features
in APR[-util] 1.3 would be a definite benefit, today.

When we add a few new functions to HTTPD - we bump the MMN and then
guarantee that these functions are binary compatible from that given
MMN bump ... all older builds against that HTTPD are binary forward
compatible until we move along to 2.4.

In APR - we guarantee backwards binary compat... apr 1.{any} will
continue to behave once we we bump the version minor.

Combine these, and it seems that with a Minor MMN bump, we can throw
the switch to require APR 1.3.x (or later) for the remaining life
of httpd 2.2.

The other side effects of requiring apr 1.3 over 1.2.x all seem to be
build related and really have no binary compatibility issues whatsoever.

So if APR ships 1.3.0 - would we be prepared to accept 1.3.0 as the new
minimum version, modulo the minor MMN bump?

Bill

Re: Ship 1.3.0 apr in httpd 2.2.9

Posted by Nick Kew <ni...@webthing.com>.
On 11 Apr 2008, at 15:39, William A. Rowe, Jr. wrote:

> So if APR ships 1.3.0 - would we be prepared to accept 1.3.0 as the  
> new
> minimum version, modulo the minor MMN bump?

-0.9

People who want 1.3 can get it for themselves.

People who want packages want stable.  And there's stuff in 1.3
that falls way short of what wants to go into a packaged products.

Example: apr_dbd_freetds is wide open to SQL injection if used
as you would use AN Other driver, because of the prepared statements
implementation.

-- 
Nick Kew

Re: Ship 1.3.0 apr in httpd 2.2.9

Posted by Jim Jagielski <ji...@jaguNET.com>.
+1... Would make things easier for the backport of
some stuff to 2.2 as well :)

PS: I'd like for us to consider a 2.2.9 "soonish"... I offer
to RM again ;)

On Apr 11, 2008, at 10:39 AM, William A. Rowe, Jr. wrote:
> So here's a thought from Davi's^WColm's APR talk in speaking with
> chipig, jorton and others...
>
> We have several inheritance quirks addressed by 1.3.0 apr, and the
> new memcache, ssl and other features are present in 1.3.  The features
> in APR[-util] 1.3 would be a definite benefit, today.
>
> When we add a few new functions to HTTPD - we bump the MMN and then
> guarantee that these functions are binary compatible from that given
> MMN bump ... all older builds against that HTTPD are binary forward
> compatible until we move along to 2.4.
>
> In APR - we guarantee backwards binary compat... apr 1.{any} will
> continue to behave once we we bump the version minor.
>
> Combine these, and it seems that with a Minor MMN bump, we can throw
> the switch to require APR 1.3.x (or later) for the remaining life
> of httpd 2.2.
>
> The other side effects of requiring apr 1.3 over 1.2.x all seem to be
> build related and really have no binary compatibility issues  
> whatsoever.
>
> So if APR ships 1.3.0 - would we be prepared to accept 1.3.0 as the  
> new
> minimum version, modulo the minor MMN bump?
>
> Bill
>


Re: Ship 1.3.0 apr in httpd 2.2.9

Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
Plüm wrote:
>  
> It has not be the right way for long and the fixes that need to be
> done are small. So I see no issue here having this done in 2.2.9 with
> apr 1.2.x. We can align it with trunk 2.2.10 or later.

fair enough

>>   * users looking at the mmn expect a certain baseline; explaining why
>>     they cannot compile this module which is purported to run with httpd
>>     ver 2.2.9 or later would become too complicated.
> 
> Do we really need a bump anyway when we only ship 2.2.x with 1.3.x but
> do not make it dependent on it?
> We could do the bump later once we depend on 1.3.x.

this makes things more difficult on third party module authors, but not
saying that makes it impossible.

You are right that we can ship apr-1.3 claiming apr-1.2 readiness, and
bump to 1.3 prereq+mmn bump a bit later.

>>> Furthermore I could image adding apr-1.2.x / apr-util-1.2.x 
>> directories
>>> to our tarballs for the next release to make it easy for 
>> users to fall
>>> back to 1.2.x if there is some kind of regression in 1.3.0.
>> Now that the code is branched, we should ensure there is fundamentally
>> no difference in existing 1.2.x functionality; we can't break abi or
>> even significantly change behaviors, only add "new things"
> 
> In general yes. All the measures above should only ensure that the users
> have a quick fallback in case something is broken. I still think that
> this approach gives us the best of both:
> 
> * A broad test audience for 1.3.0 and thus giving us a much better feeling
>   for the quality of 1.3.x.

goodness, yes

> * A very quick possibility to resolve possible regression issues in 1.3.0
>   Of course these possible regression issues should be solved quickly
>   1.3..x

yes

let's see what others have to say over the next day or two, if they can
stand the frustration of waiting 2 more months to take advantage of the
api changes.

Re: Ship 1.3.0 apr in httpd 2.2.9

Posted by Plüm, Rüdiger, VF-Group <ru...@vodafone.com>.
 

> -----Ursprüngliche Nachricht-----
> Von: William A. Rowe, Jr. 
> Gesendet: Montag, 14. April 2008 12:49
> An: dev@httpd.apache.org
> Betreff: Re: Ship 1.3.0 apr in httpd 2.2.9
> 
> Ruediger Pluem wrote:
> > 
> > here is my idea:
> > 
> > Ship 1.3.0 with httpd 2.2.x, but do *not* make httpd 
> dependent on 1.3.x
> > now. Wait for this until 1.3.x has settled a bit more and 
> things like
> > the ones Nick mentioned are fixed.
> 
> I'm not entirely keen on this idea; the reasons being
> 
>   * we want to pick up 1.3 api's to fix log file creation 
> 'the right way'

It has not be the right way for long and the fixes that need to be
done are small. So I see no issue here having this done in 2.2.9 with
apr 1.2.x. We can align it with trunk 2.2.10 or later.

>   * users looking at the mmn expect a certain baseline; explaining why
>     they cannot compile this module which is purported to run 
> with httpd
>     ver 2.2.9 or later would become too complicated.

Do we really need a bump anyway when we only ship 2.2.x with 1.3.x but
do not make it dependent on it?
We could do the bump later once we depend on 1.3.x.


> 
> > Furthermore I could image adding apr-1.2.x / apr-util-1.2.x 
> directories
> > to our tarballs for the next release to make it easy for 
> users to fall
> > back to 1.2.x if there is some kind of regression in 1.3.0.
> 
> Now that the code is branched, we should ensure there is fundamentally
> no difference in existing 1.2.x functionality; we can't break abi or
> even significantly change behaviors, only add "new things"

In general yes. All the measures above should only ensure that the users
have a quick fallback in case something is broken. I still think that
this approach gives us the best of both:

* A broad test audience for 1.3.0 and thus giving us a much better feeling
  for the quality of 1.3.x.
* A very quick possibility to resolve possible regression issues in 1.3.0
  Of course these possible regression issues should be solved quickly
  1.3..x


> 
> > But it might be also ok to require the users in these cases 
> to download
> > 1.2.x separately drop in the sources and call buildconf.
> 
> well if there is no difference between 1.2 and 1.3, and we recognize
> that the first 1.3.0 is entirely not solving the problem for memcached
> for example, we could change the prereq to 1.3.1 for the next release
> to ensure they have the bug-free implementation for the next 
> httpd tag.

Sure. I guess this needs to be done anyway regardless of my proposal.

Regards

Rüdiger


Re: Ship 1.3.0 apr in httpd 2.2.9

Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
Ruediger Pluem wrote:
> 
> here is my idea:
> 
> Ship 1.3.0 with httpd 2.2.x, but do *not* make httpd dependent on 1.3.x
> now. Wait for this until 1.3.x has settled a bit more and things like
> the ones Nick mentioned are fixed.

I'm not entirely keen on this idea; the reasons being

  * we want to pick up 1.3 api's to fix log file creation 'the right way'
  * users looking at the mmn expect a certain baseline; explaining why
    they cannot compile this module which is purported to run with httpd
    ver 2.2.9 or later would become too complicated.

> Furthermore I could image adding apr-1.2.x / apr-util-1.2.x directories
> to our tarballs for the next release to make it easy for users to fall
> back to 1.2.x if there is some kind of regression in 1.3.0.

Now that the code is branched, we should ensure there is fundamentally
no difference in existing 1.2.x functionality; we can't break abi or
even significantly change behaviors, only add "new things"

> But it might be also ok to require the users in these cases to download
> 1.2.x separately drop in the sources and call buildconf.

well if there is no difference between 1.2 and 1.3, and we recognize
that the first 1.3.0 is entirely not solving the problem for memcached
for example, we could change the prereq to 1.3.1 for the next release
to ensure they have the bug-free implementation for the next httpd tag.

Bill

Re: Ship 1.3.0 apr in httpd 2.2.9

Posted by Ruediger Pluem <rp...@apache.org>.

On 04/11/2008 04:39 PM, William A. Rowe, Jr. wrote:
> So here's a thought from Davi's^WColm's APR talk in speaking with
> chipig, jorton and others...
> 
> We have several inheritance quirks addressed by 1.3.0 apr, and the
> new memcache, ssl and other features are present in 1.3.  The features
> in APR[-util] 1.3 would be a definite benefit, today.
> 
> When we add a few new functions to HTTPD - we bump the MMN and then
> guarantee that these functions are binary compatible from that given
> MMN bump ... all older builds against that HTTPD are binary forward
> compatible until we move along to 2.4.
> 
> In APR - we guarantee backwards binary compat... apr 1.{any} will
> continue to behave once we we bump the version minor.
> 
> Combine these, and it seems that with a Minor MMN bump, we can throw
> the switch to require APR 1.3.x (or later) for the remaining life
> of httpd 2.2.
> 
> The other side effects of requiring apr 1.3 over 1.2.x all seem to be
> build related and really have no binary compatibility issues whatsoever.
> 
> So if APR ships 1.3.0 - would we be prepared to accept 1.3.0 as the new
> minimum version, modulo the minor MMN bump?

I thought about this for a while. To be honest I always mistrust software
that was not widespread and thus tested in real life before (that has
nothing to do with APR or httpd and is no insult on either project), but
yes we need to start this at some point of time.
With APR I guess it would not be helpful to just release 1.3.0 or do some
alpha / beta release and wait for the reports coming in. We actually need
to ensure that it is used otherwise we do not get sufficient test feedback.

So here is my idea:

Ship 1.3.0 with httpd 2.2.x, but do *not* make httpd dependent on 1.3.x
now. Wait for this until 1.3.x has settled a bit more and things like
the ones Nick mentioned are fixed.
Furthermore I could image adding apr-1.2.x / apr-util-1.2.x directories
to our tarballs for the next release to make it easy for users to fall
back to 1.2.x if there is some kind of regression in 1.3.0.
But it might be also ok to require the users in these cases to download
1.2.x separately drop in the sources and call buildconf.

Regards

Rüdiger



Re: Ship 1.3.0 apr in httpd 2.2.9

Posted by Guenter Knauf <fu...@apache.org>.
> So if APR ships 1.3.0 - would we be prepared to accept 1.3.0 as the new
> minimum version, modulo the minor MMN bump?

+1

BTW. wouldnt it make sense to introduce a similar versioning as we now use with httpd?
I mean that current trunk 1.3.0-unstable becomes 1.4.0-stable, and we start on 1.5.0-unstable?

Guenter.



Re: Ship 1.3.0 apr in httpd 2.2.9

Posted by Justin Erenkrantz <ju...@erenkrantz.com>.
On Fri, Apr 11, 2008 at 4:39 PM, William A. Rowe, Jr.
<wr...@rowe-clan.net> wrote:
>  So if APR ships 1.3.0 - would we be prepared to accept 1.3.0 as the new
>  minimum version, modulo the minor MMN bump?

+1.  -- justin

Re: Ship 1.3.0 apr in httpd 2.2.9

Posted by "Akins, Brian" <Br...@turner.com>.
On 4/11/08 10:39 AM, "William A. Rowe, Jr." <wr...@rowe-clan.net> wrote:

> So if APR ships 1.3.0 - would we be prepared to accept 1.3.0 as the new
> minimum version, modulo the minor MMN bump?

+1

-- 
Brian Akins
Chief Operations Engineer
Turner Digital Media Technologies