You are viewing a plain text version of this content. The canonical link for it is here.
Posted to cvs@httpd.apache.org by wr...@apache.org on 2001/11/27 06:19:40 UTC

cvs commit: httpd-2.0 ROADMAP STATUS

wrowe       01/11/26 21:19:39

  Modified:    .        STATUS
  Added:       .        ROADMAP
  Log:
    OK... we keep deferring these issues, it's time for a ROADMAP.
  
    Jump in everyone.
  
  Revision  Changes    Path
  1.343     +3 -14     httpd-2.0/STATUS
  
  Index: STATUS
  ===================================================================
  RCS file: /home/cvs/httpd-2.0/STATUS,v
  retrieving revision 1.342
  retrieving revision 1.343
  diff -u -r1.342 -r1.343
  --- STATUS	2001/11/21 18:19:06	1.342
  +++ STATUS	2001/11/27 05:19:39	1.343
  @@ -1,5 +1,5 @@
   APACHE 2.0 STATUS:						-*-text-*-
  -Last modified at [$Date: 2001/11/21 18:19:06 $]
  +Last modified at [$Date: 2001/11/27 05:19:39 $]
   
   Release:
   
  @@ -60,8 +60,8 @@
         administrator to order filters, beyond the order of filename (mime)
         extensions.  It isn't clear if Set...Filter(s) should be inserted 
         before or after the Add...Filter(s) which are ordered by sequence of
  -      filename extensions.  Add...FilterByType will add to this quandry.
  -      Some sort of resolution needs to be proposed, 
  +      filename extensions.  At minimum, some sort of +-[0-10] syntax seems
  +      like the quickest fix for a 2.0 gold release.
   
       * mod_negotiation needs a new option or directive, something like
         ForceLanguagePriority, to fall back to the LanguagePriority
  @@ -69,12 +69,6 @@
           Status: Bill has some code in his tree that accomplishes
                   this, and will commit it Friday after it's tested.
   
  -    * revamp the input filter syntax to provide for ordering of
  -      filters created with the Set{Input|Output}Filter and the
  -      Add{Input|Output}Filter directives.  At minimum, some sort
  -      of insert first v.s. append to end or 'relative to x' is
  -      probably needed.
  -
       * revamp the input filter behavior, per discussions since
         February (and especially at the hackathon last
         April). Specifically, ap_get_brigade will return a brigade with
  @@ -132,11 +126,6 @@
                   with limited OS exposure.
   
   RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
  -
  -    * Source code should follow style guidelines.
  -        This shouldn't wait until we have a 2.0-gold release because 
  -        then style corrections will conflict with bug fixes found after
  -        release which is not nice.
   
       * Allow the DocumentRoot directive within <Location > scopes?  This
         allows the beloved (crusty) Alias /foo/ /somepath/foo/ followed
  
  
  
  1.1                  httpd-2.0/ROADMAP
  
  Index: ROADMAP
  ===================================================================
  APACHE 2.1+ ROADMAP:
  
  Last modified at [$Date: 2001/11/27 05:19:39 $]
  
  DEFERRRED FOR APACHE 2.1
  
      * Source code should follow style guidelines.
        OK, we all agree pretty code is good.  Probably best to clean this
        up by hand immediately upon branching a 2.1 tree.
        Justin's voulenteered to hand-edit the entire source tree ;)
  
      * revamp the input filter syntax to provide for ordering of
        filters created with the Set{Input|Output}Filter and the
        Add{Input|Output}Filter directives.  A 'relative to filterx' 
        syntax is definately preferable, but not realistic for 2.0.
  
      * Platforms that do not support fork (primarily Win32 and AS/400)
        Architect start-up code that avoids initializing all the modules 
        in the parent process on platforms that do not support fork.
        Better yet - not only inform the startup of which phase it's in,
        but allow the parent 'process' to initialize shared memory, etc,
        and create a module-by-module stream to pass to the child, so the
        parent can actually arbitrate the important stuff.
  
      * Replace stat [deferred open] with open/fstat in directory_walk.
        Justin, Ian, OtherBill all interested in this.  Implies setting up
        the apr_file_t member in request_rec, and having all modules use
        that file, and allow the cleanup to close it [if it isn't a shared,
        cached file handle.]
  
      * Refactor auth into auth protocols and auth database stores.
        Many interested hackers, too destabilizing for 2.0 inclusion.
  
  DEFERRRED FOR APACHE 3.0
  
      * The Async Apache Server implemented in terms of APR.
        [Bill Stoddard's pet project.]
  
      
  
  
  

Re: cvs commit: httpd-2.0 ROADMAP STATUS

Posted by Ryan Bloom <rb...@covalent.net>.
On Tuesday 27 November 2001 08:31 pm, Justin Erenkrantz wrote:
> On Tue, Nov 27, 2001 at 07:45:38PM -0800, Ryan Bloom wrote:
> > My guess is that once 2.0 goes GA, we'll all be busy fixing bugs and
> > working on more fun stuff for a while.  Also, take a look at everything
> > in the ROADMAP file.  99% of it can be done as a dot release in the 2.0
> > tree.  Most of it doesn't require large portions of the code to be
> > re-written.
>
> Well, that's an issue of the contents in ROADMAP not its location.  =)
>
> My original concern about ROADMAP being separated from a versioned
> source repository still stands.  -- justin
>
> P.S.  IMHO, the idea is that when we go GA, we're pretty damn
> certain there aren't a lot of bugs in it.  When we say 2.0 is GA,
> we MUST mean that we 100% recommend switching from 1.3 to 2.0.
> I haven't seen any major showstoppers from 2.0.28, so we're on the
> right track.  2.0.29 should have performance improvements, but may
> have some bugs due to that...

I think you are kidding yourself.  Take a look at 1.3.  That was supposed to
be an interim release, in between 1.2 and 2.0.  The real bugs won't really 
start to come out of 2.0 until it hits GA and millions of sites start to migrate to
it.  Also, remember that most sites aren't going to migrate to 2.0 until it has
been out for a very long time.

I'm all for starting to look at 2.1 and 3.0, but let's not jump the gun too much.
I would thoroughly expect at least 22 2.0 GA releases, just like we had with
1.3, and I would expect that a lot of the advancements people want to make
can be done very easily in a 2.0 framework.

Even the async work can be done inside of 2.0.  It won't be fully async, but
we can do a lot of the low hanging fruit.

Ryan
______________________________________________________________
Ryan Bloom				rbb@apache.org
Covalent Technologies			rbb@covalent.net
--------------------------------------------------------------

Re: cvs commit: httpd-2.0 ROADMAP STATUS

Posted by "William A. Rowe, Jr." <wr...@covalent.net>.
From: "Ryan Bloom" <rb...@covalent.net>
Sent: Tuesday, November 27, 2001 9:45 PM


> My guess is that once 2.0 goes GA, we'll all be busy fixing bugs and
> working on more fun stuff for a while.  Also, take a look at everything in
> the ROADMAP file.  99% of it can be done as a dot release in the 2.0
> tree.  Most of it doesn't require large portions of the code to be re-written.

The only bit that causes problems is introducing/rearchitecting modules, such
as the auth stuff.  Yes, incremental changes could go into 2.0.x, but dropping
out a module [such as splitting mod_auth in two] or the apr_file_t member of the
request rec are really significant enough to warrent the 2.1 bump.

The splitting-a-module means their old config files are [probably] broken.  For
the user's sanity, a 2.1 bump makes sense.

The apr_file_t member means modules must be recrafted to pay attention to this
open file handle.  I think the _real_ benefit is to split out the filesystem as
it's own module, much like mod_dav_fs.  

In any case, plenty of folks regularly complain that their suggestions are
ignored, or are discussed and just drop.  That isn't a good way to attract
contributors.  ROADMAP allows us to do some planing/architecting into the
future, beyond the event horizion.

I'm really beginning to believe that nayoga's bugzilla might be [part of] the 
answer to the headaches.

Bill





Re: cvs commit: httpd-2.0 ROADMAP STATUS

Posted by Justin Erenkrantz <je...@ebuilt.com>.
On Tue, Nov 27, 2001 at 07:45:38PM -0800, Ryan Bloom wrote:
> My guess is that once 2.0 goes GA, we'll all be busy fixing bugs and
> working on more fun stuff for a while.  Also, take a look at everything in
> the ROADMAP file.  99% of it can be done as a dot release in the 2.0
> tree.  Most of it doesn't require large portions of the code to be re-written.

Well, that's an issue of the contents in ROADMAP not its location.  =)

My original concern about ROADMAP being separated from a versioned
source repository still stands.  -- justin

P.S.  IMHO, the idea is that when we go GA, we're pretty damn 
certain there aren't a lot of bugs in it.  When we say 2.0 is GA,
we MUST mean that we 100% recommend switching from 1.3 to 2.0.
I haven't seen any major showstoppers from 2.0.28, so we're on the
right track.  2.0.29 should have performance improvements, but may 
have some bugs due to that...


Re: cvs commit: httpd-2.0 ROADMAP STATUS

Posted by Ryan Bloom <rb...@covalent.net>.
On Tuesday 27 November 2001 07:37 pm, Justin Erenkrantz wrote:
> On Tue, Nov 27, 2001 at 05:19:40AM -0000, wrowe@apache.org wrote:
> > wrowe       01/11/26 21:19:39
> >
> >   Modified:    .        STATUS
> >   Added:       .        ROADMAP
> >   Log:
> >     OK... we keep deferring these issues, it's time for a ROADMAP.
> >
> >     Jump in everyone.
>
> I think the ROADMAP file should be on the website.  It shouldn't
> be tied in with any specific httpd version (i.e. living in the
> httpd-2.0 repository).
>
> See httpd-site/xdocs/dev/project-plan.html
>
> (That file could/should be migrated over to the new XML format...)
>
> My guess is once 2.0 goes GA, we'll open up httpd-2.1.  -- justin

My guess is that once 2.0 goes GA, we'll all be busy fixing bugs and
working on more fun stuff for a while.  Also, take a look at everything in
the ROADMAP file.  99% of it can be done as a dot release in the 2.0
tree.  Most of it doesn't require large portions of the code to be re-written.

Ryan

______________________________________________________________
Ryan Bloom				rbb@apache.org
Covalent Technologies			rbb@covalent.net
--------------------------------------------------------------

Re: cvs commit: httpd-2.0 ROADMAP STATUS

Posted by Justin Erenkrantz <je...@ebuilt.com>.
On Tue, Nov 27, 2001 at 05:19:40AM -0000, wrowe@apache.org wrote:
> wrowe       01/11/26 21:19:39
> 
>   Modified:    .        STATUS
>   Added:       .        ROADMAP
>   Log:
>     OK... we keep deferring these issues, it's time for a ROADMAP.
>   
>     Jump in everyone.

I think the ROADMAP file should be on the website.  It shouldn't
be tied in with any specific httpd version (i.e. living in the
httpd-2.0 repository).

See httpd-site/xdocs/dev/project-plan.html

(That file could/should be migrated over to the new XML format...)

My guess is once 2.0 goes GA, we'll open up httpd-2.1.  -- justin