cvs commit: apache-2.0 STATUS
stoddard00/02/24 19:44:45 Modified:.STATUS Log: More stuff to de before release Revision ChangesPath 1.29 +13 -1 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.28 retrieving revision 1.29 diff -u -r1.28 -r1.29 --- STATUS2000/02/22 23:00:53 1.28 +++ STATUS2000/02/25 03:44:44 1.29 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 2000/02/22 23:00:53 $] +Last modified at [$Date: 2000/02/25 03:44:44 $] Release: @@ -8,6 +8,15 @@ 2.0 : later in 2000 RELEASE SHOWSTOPPERS: +* Fix lingering close +Status: + +* MPMs should have responsibility for closing sockets and +registering the socket cleanup. Stop registering multiple +cleanups for each socket. Move socket close code out of +http_connection.c and into the MPMs. + Status: + * Reliable piped logs look broken everywhere. Each MPM includes essentially identical code to ap_register_other_child(), etc. Most of this code can be moved out of the MPMs and into some common file (http_core.c?). @@ -29,6 +38,9 @@ * suEXEC doesn't work Status: Manoj has posted an patch to fix this. [EMAIL PROTECTED] + +* Win32: Enable the Windows MPM to honor max_requests_per_child +Status: Bill will fix this. * Win32: Fix Win9* specific code in the winnt MPM Status:
cvs commit: apache-2.0 STATUS
stoddard00/02/22 15:00:53 Modified:.STATUS Log: Add a couple of todos. Revision ChangesPath 1.28 +6 -2 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.27 retrieving revision 1.28 diff -u -r1.27 -r1.28 --- STATUS2000/02/16 18:29:20 1.27 +++ STATUS2000/02/22 23:00:53 1.28 @@ -1,10 +1,10 @@ Apache 2.0 STATUS: -Last modified at [$Date: 2000/02/16 18:29:20 $] +Last modified at [$Date: 2000/02/22 23:00:53 $] Release: Vague plan, assuming more features aren't demanded for the 2.0 release -2.0a1/b1: January 2000 +2.0a1 : March 2000 2.0 : later in 2000 RELEASE SHOWSTOPPERS: @@ -59,6 +59,10 @@ resolver calls, RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP: + +* Win32: Reuse accept socket after transmitfile/close + This is not a bug, but would be nice to get this feature in + before ship. * Win32: Enable the winnt MPM to use the new scoreboard API
cvs commit: apache-2.0 STATUS
stoddard00/02/16 08:48:34 Modified:.STATUS Log: Some status updates Revision ChangesPath 1.26 +22 -15apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.25 retrieving revision 1.26 diff -u -r1.25 -r1.26 --- STATUS2000/02/08 14:19:59 1.25 +++ STATUS2000/02/16 16:48:32 1.26 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 2000/02/08 14:19:59 $] +Last modified at [$Date: 2000/02/16 16:48:32 $] Release: @@ -8,6 +8,10 @@ 2.0 : later in 2000 RELEASE SHOWSTOPPERS: +* DSO hooks registered during the preflight call to ap_read_config + are not 'unregistered' during the preflight clean-up. Thus, when + Apache comes up, the hook tables contain two entries for each DSO hook, + one of which is bogus, leftover from the preflight. * Pipes to CGI scripts are not being timed out Status: code has been added to APR to support timing out pipes. @@ -22,16 +26,19 @@ Status: Manoj has posted an patch to fix this. [EMAIL PROTECTED] -* Windows NT port isn't done -Status: Bill [EMAIL PROTECTED] is working on MPM and APR. - Remaining work: -- Fix Win9* specific code in the winnt MPM -- Test access logging with multiple threads. Will the native file I/O - calls serialize automagically like the CRT calls or do we need to - add region locking each time we write to the access/error logs? - - Piped logging is broken - - DSO support is broken +* Win32: Fix Win9* specific code in the winnt MPM + Status: +* Win32: Test access logging with multiple threads. Will the + native file I/O calls serialize automagically like the + CRT calls or do we need to add region locking each time + we access the logs? + Status: + +* Win32: Smoke test all ported modules to make sure they work + correctly under Windows. + Status: + * Current 2.0 code is not tested on many Unix platforms. Make 2.0 work on most, if not all the systems 1.3 did Status: Autoconf and APR will get us most of the way there. @@ -49,6 +56,10 @@ RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP: +* Win32: Non-blocking CGI pipes + +* Win32: Graceful restart of a service is broken (code does shutdown then start) + * Dynamically loadable MPMs Enable users of the binary distributions of Apache to be able to dynamically load the MPM. @@ -76,12 +87,8 @@ * There are still a number of places in the code where we are loosing error status (i.e. throwing away the error returned by a system call and replacing it with a generic error code) - -* APRize the stat function. CRT stat() is about 30% slower on Windows than - the equivalent native Windows call. The APR call should return the Unix - style stat structure, just to keep it familier. -* Implement reliable piped logs on Windows +* Win32: Implement reliable piped logs on Windows Status: Bill [EMAIL PROTECTED] has prototype code (not reliable) in 1.3. Should be much cleaner with APR in 2.0.
cvs commit: apache-2.0 STATUS
stoddard00/02/16 10:29:21 Modified:.STATUS Log: More stuff to do. Revision ChangesPath 1.27 +7 -1 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.26 retrieving revision 1.27 diff -u -r1.26 -r1.27 --- STATUS2000/02/16 16:48:32 1.26 +++ STATUS2000/02/16 18:29:20 1.27 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 2000/02/16 16:48:32 $] +Last modified at [$Date: 2000/02/16 18:29:20 $] Release: @@ -8,6 +8,10 @@ 2.0 : later in 2000 RELEASE SHOWSTOPPERS: +* Reliable piped logs look broken everywhere. Each MPM includes essentially + identical code to ap_register_other_child(), etc. Most of this code can + be moved out of the MPMs and into some common file (http_core.c?). + * DSO hooks registered during the preflight call to ap_read_config are not 'unregistered' during the preflight clean-up. Thus, when Apache comes up, the hook tables contain two entries for each DSO hook, @@ -55,6 +59,8 @@ resolver calls, RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP: + +* Win32: Enable the winnt MPM to use the new scoreboard API * Win32: Non-blocking CGI pipes
cvs commit: apache-2.0 STATUS
stoddard00/02/08 06:20:00 Modified:.STATUS Log: Work done, bugs found. Revision ChangesPath 1.25 +16 -12apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.24 retrieving revision 1.25 diff -u -r1.24 -r1.25 --- STATUS1999/12/22 06:20:11 1.24 +++ STATUS2000/02/08 14:19:59 1.25 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/12/22 06:20:11 $] +Last modified at [$Date: 2000/02/08 14:19:59 $] Release: @@ -9,10 +9,6 @@ RELEASE SHOWSTOPPERS: -* Complete HAVE_SENDFILE work. Need to add iovec head and tail pointers to - the iol sendfile API and make it work under Unix. -Status: Bill [EMAIL PROTECTED] is working on this - * Pipes to CGI scripts are not being timed out Status: code has been added to APR to support timing out pipes. This needs to be used in Apache now. @@ -29,13 +25,12 @@ * Windows NT port isn't done Status: Bill [EMAIL PROTECTED] is working on MPM and APR. Remaining work: -1. Add back ability to run Apache as a service -2. Fix Win9* specific code in the winnt MPM -3. Test access logging with multiple threads. Will the native file I/O - calls serialize automagically like the CRT calls or do we need to - add region locking each time we write to the access/error logs? -4. Fix the DOS hole in AcceptEx (need to time out connections over - which data is never sent). +- Fix Win9* specific code in the winnt MPM +- Test access logging with multiple threads. Will the native file I/O + calls serialize automagically like the CRT calls or do we need to + add region locking each time we write to the access/error logs? + - Piped logging is broken + - DSO support is broken * Current 2.0 code is not tested on many Unix platforms. Make 2.0 work on most, if not all the systems 1.3 did @@ -53,6 +48,15 @@ resolver calls, RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP: + +* Dynamically loadable MPMs + Enable users of the binary distributions of Apache to be able to + dynamically load the MPM. + +* Platforms that do not support fork (primarily Win32 and AS/400) + Consider introducing HAVE_FORK feature macro. Architect start-up code + that avoids initializing all the modules in the parent process on + platforms that do not support fork. * Clean the code. There are a lot of places we used APR but didn't remove the hacks that were required for the cross-platform code in
Re: cvs commit: apache-2.0 STATUS
+* Platforms that do not support fork (primarily Win32 and AS/400) + Consider introducing HAVE_FORK feature macro. Architect start-up code + that avoids initializing all the modules in the parent process on + platforms that do not support fork. APR already has an APR_HAS_FORK feature macro. It is just a question now of using that macro in Apache. Ryan Come to the first official Apache Software Foundation Conference!!! http://ApacheCon.Com/ ___ Ryan Bloom [EMAIL PROTECTED] 2121 Stonehenge Dr. Apt #3 Raleigh, NC 27615 Ryan Bloom -- thinker, adventurer, artist, writer, but mostly, friend. ---
cvs commit: apache-2.0 STATUS
manoj 99/12/21 22:20:12 Modified:.STATUS Log: I don't have to get up before 2PM for almost 2 weeks! Revision ChangesPath 1.24 +4 -4 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.23 retrieving revision 1.24 diff -u -d -u -r1.23 -r1.24 --- STATUS1999/12/21 08:26:49 1.23 +++ STATUS1999/12/22 06:20:11 1.24 @@ -1,11 +1,11 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/12/21 08:26:49 $] +Last modified at [$Date: 1999/12/22 06:20:11 $] Release: -Assuming more features aren't demanded for the 2.0 release +Vague plan, assuming more features aren't demanded for the 2.0 release 2.0a1/b1: January 2000 -2.0 : Early 2000 +2.0 : later in 2000 RELEASE SHOWSTOPPERS: @@ -85,7 +85,7 @@ Status: Manoj has placed an initial version into the 2.0 repository. Todos include: - - make install and make depend should work + - make install should work - a tool to simplfy third-party module building should be written. Something like apxs - Add a lot more checks to satisfy the various platforms, e.g.
cvs commit: apache-2.0 STATUS
manoj 99/12/21 00:26:50 Modified:.STATUS Log: [insert weather comment here] Revision ChangesPath 1.23 +16 -24apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.22 retrieving revision 1.23 diff -u -d -u -r1.22 -r1.23 --- STATUS1999/12/13 15:46:45 1.22 +++ STATUS1999/12/21 08:26:49 1.23 @@ -1,14 +1,11 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/12/13 15:46:45 $] +Last modified at [$Date: 1999/12/21 08:26:49 $] Release: - -2.0: In pre-alpha development - -Plan: -* Vague goal of an alpha or beta release in 1999. Commit-then-review - is active. +Assuming more features aren't demanded for the 2.0 release +2.0a1/b1: January 2000 +2.0 : Early 2000 RELEASE SHOWSTOPPERS: @@ -25,9 +22,6 @@ * Port mod_rewrite Status: Paul Reder [EMAIL PROTECTED] is working on this. -* Port mod_mime_magic -Status: Paul Reder [EMAIL PROTECTED] is working on this. - * suEXEC doesn't work Status: Manoj has posted an patch to fix this. [EMAIL PROTECTED] @@ -68,31 +62,29 @@ were told these would be considered for inclusion in Apache 2.0, it would be nice to actually do so. -* Win32: Migrate the MPM over to use APR thread/process calls. This would eliminate - some code in the Win32 branch that essentially duplicates what is in APR. - Status: Bill [EMAIL PROTECTED] is working on this. +* Win32: Migrate the MPM over to use APR thread/process calls. This + would eliminate some code in the Win32 branch that essentially + duplicates what is in APR. +Status: Bill [EMAIL PROTECTED] is working on this. * Move I/O layering into APR. -* There are still a number of places in the code where we are loosing error - status (i.e. throwing away the error returned by a system call and replacing - it with a generic error code) +* There are still a number of places in the code where we are + loosing error status (i.e. throwing away the error returned by a + system call and replacing it with a generic error code) * APRize the stat function. CRT stat() is about 30% slower on Windows than the equivalent native Windows call. The APR call should return the Unix style stat structure, just to keep it familier. * Implement reliable piped logs on Windows -Status: Bill [EMAIL PROTECTED] has prototype code (not reliable) in 1.3. -Should be much cleaner with APR in 2.0. +Status: Bill [EMAIL PROTECTED] has prototype code (not +reliable) in 1.3. Should be much cleaner with APR in 2.0. * Switch to autoconf, et al. for configuration. Status: Manoj has placed an initial version into the 2.0 repository. Todos include: - - Support selection of modules (including MPMs; only Dexter -works right now). Probably involved creating a directory for -each module - make install and make depend should work - a tool to simplfy third-party module building should be written. Something like apxs @@ -117,9 +109,9 @@ behavior. Dean has suggested getting rid of regex containers completely. -* SIGSEGV on Linux seems to only kill a thread, not a whole process; - we need to work around this, probably by bouncing the signal to - the sigwait thread. But this will hurt debugability. +* SIGSEGV on Linux (glibc 2.1.2) isn't caught properly by a + sigwaiting thread. We need to work around this, perhaps unless + there is hope soon for a fixed glibc. Other features that need writing:
Re: cvs commit: apache-2.0 STATUS
Release: - -2.0: In pre-alpha development - When did we get out of pre-alpha development? Last time I checked, we hadn't released an alpha, and that means this is still pre-alpha code. I would like this put back, because there are still some changes that need to be made before we officially release an alpha. It isn't so much functionality, as it is making the code use APR all over the place, but if we make the change after the alpha release, we make it much harder for people to port modules. -Plan: -* Vague goal of an alpha or beta release in 1999. Commit-then-review - is active. +Assuming more features aren't demanded for the 2.0 release +2.0a1/b1: January 2000 +2.0 : Early 2000 I guess I'm wondering when this was decided. I would like to see an alpha in January, a beta in February, and a release when it is ready. We released 7 beta's for 1.3, I expect almost as many if not more for 2.0. I don't want to make grand plans to release 2.0 in early 2000 if we can't meet it. Ryan ___ Ryan Bloom [EMAIL PROTECTED] 6209 H Shanda Dr. Raleigh, NC 27609 Ryan Bloom -- thinker, adventurer, artist, writer, but mostly, friend. ---
cvs commit: apache-2.0 STATUS
stoddard99/12/13 07:46:45 Modified:.STATUS Log: Update STATUS Revision ChangesPath 1.22 +4 -3 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.21 retrieving revision 1.22 diff -u -r1.21 -r1.22 --- STATUS1999/12/02 06:33:05 1.21 +++ STATUS1999/12/13 15:46:45 1.22 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/12/02 06:33:05 $] +Last modified at [$Date: 1999/12/13 15:46:45 $] Release: @@ -37,10 +37,11 @@ Remaining work: 1. Add back ability to run Apache as a service 2. Fix Win9* specific code in the winnt MPM -3. Get the MPM working in multi process mode (one parent one child) -4. Test access logging with multiple threads. Will the native file I/O +3. Test access logging with multiple threads. Will the native file I/O calls serialize automagically like the CRT calls or do we need to add region locking each time we write to the access/error logs? +4. Fix the DOS hole in AcceptEx (need to time out connections over + which data is never sent). * Current 2.0 code is not tested on many Unix platforms. Make 2.0 work on most, if not all the systems 1.3 did
cvs commit: apache-2.0 STATUS
manoj 99/12/01 22:33:06 Modified:.STATUS Log: A thread (of messages) in comp.programming.threads reminded me of this. Blh Revision ChangesPath 1.21 +6 -1 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.20 retrieving revision 1.21 diff -u -d -u -r1.20 -r1.21 --- STATUS1999/11/30 00:52:30 1.20 +++ STATUS1999/12/02 06:33:05 1.21 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/11/30 00:52:30 $] +Last modified at [$Date: 1999/12/02 06:33:05 $] Release: @@ -51,6 +51,11 @@ Status: Ben Laurie is working on this. * Modules. Which modules do not work yet, can we get a list? + +* We need a thread-safe resolver, at least on Unix. +Status: The best known candidate would be something from BIND +(v8 or v9?) The only other option would be to mutex all the +resolver calls, RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
cvs commit: apache-2.0 STATUS
manoj 99/11/29 16:10:24 Modified:.STATUS Log: Autoconf can heal your family strife. And it's also a floor wax! Revision ChangesPath 1.19 +12 -1 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.18 retrieving revision 1.19 diff -u -d -u -r1.18 -r1.19 --- STATUS1999/11/23 08:30:21 1.18 +++ STATUS1999/11/30 00:10:23 1.19 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/11/23 08:30:21 $] +Last modified at [$Date: 1999/11/30 00:10:23 $] Release: @@ -81,6 +81,17 @@ Should be much cleaner with APR in 2.0. * Switch to autoconf, et al. for configuration. +Status: Manoj has placed an initial version into the 2.0 +repository. Todos include: + + - Support selection of modules (including MPMs; only Dexter +works right now). Probably involved creating a directory for +each module + - make install and make depend should work + - a tool to simplfy third-party module building should be +written. Something like apxs + - Add a lot more checks to satisfy the various platforms, e.g. +for threading * Use APR to get rid of more platform dependancies. Status: Ryan Bloom [EMAIL PROTECTED] is working on this.
cvs commit: apache-2.0 STATUS
dgaudet 99/11/29 16:52:32 Modified:.STATUS Log: more accurate reflection of reality Revision ChangesPath 1.20 +4 -2 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.19 retrieving revision 1.20 diff -u -r1.19 -r1.20 --- STATUS1999/11/30 00:10:23 1.19 +++ STATUS1999/11/30 00:52:30 1.20 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/11/30 00:10:23 $] +Last modified at [$Date: 1999/11/30 00:52:30 $] Release: @@ -125,7 +125,9 @@ * Work on other MPMs. Possible MPMs discussed or in progress include: - Dean Gaudet's async MPM - Status: ? + Status: dean is unlikely to begin work on this any time soon... + has lots of ideas he'll share with folks willing to begin + tackling the issues. this is apache 2.1 material. - Zach Brown's [EMAIL PROTECTED] Linux siginfo MPM Status: ?
cvs commit: apache-2.0 STATUS
manoj 99/11/23 00:30:23 Modified:.STATUS Log: Hmmm, turns out that the nice event library for network events on PalmOS only applies to the wireless transceiver. Oh well, I guess select() will have to do... Revision ChangesPath 1.18 +3 -1 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.17 retrieving revision 1.18 diff -u -d -u -r1.17 -r1.18 --- STATUS1999/11/19 18:59:48 1.17 +++ STATUS1999/11/23 08:30:21 1.18 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/11/19 18:59:48 $] +Last modified at [$Date: 1999/11/23 08:30:21 $] Release: @@ -19,6 +19,8 @@ * Pipes to CGI scripts are not being timed out Status: code has been added to APR to support timing out pipes. This needs to be used in Apache now. + +* Put back resource limit code * Port mod_rewrite Status: Paul Reder [EMAIL PROTECTED] is working on this.
cvs commit: apache-2.0 STATUS
rbb 99/11/16 07:11:51 Modified:.STATUS Log: Update the STATUS file a bit. Revision ChangesPath 1.14 +3 -6 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.13 retrieving revision 1.14 diff -u -r1.13 -r1.14 --- STATUS1999/11/11 13:21:10 1.13 +++ STATUS1999/11/16 15:11:51 1.14 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/11/11 13:21:10 $] +Last modified at [$Date: 1999/11/16 15:11:51 $] Release: @@ -17,7 +17,8 @@ Status: Bill [EMAIL PROTECTED] is working on this * Pipes to CGI scripts are not being timed out -Status: +Status: code has been added to APR to support timing out pipes. +This needs to be used in Apache now. * Port mod_rewrite Status: Paul Reder [EMAIL PROTECTED] is working on this. @@ -41,7 +42,6 @@ * Current 2.0 code is not tested on many Unix platforms. Make 2.0 work on most, if not all the systems 1.3 did - Status: Autoconf and APR will get us most of the way there. * The module API is a weird combination of the old table and the new @@ -72,9 +72,6 @@ Status: Bill [EMAIL PROTECTED] is working on this. * Move I/O layering into APR. - -* APR'ize I/O layering. iol is using native sockets now, which is the source of - the double close problem Manoj ran into. * There are still a number of places in the code where we are loosing error status (i.e. throwing away the error returned by a system call and replacing
cvs commit: apache-2.0 STATUS
rbb 99/11/11 05:21:11 Modified:.STATUS Log: Update the STATUS file with three new items. Revision ChangesPath 1.13 +11 -1 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.12 retrieving revision 1.13 diff -u -r1.12 -r1.13 --- STATUS1999/11/04 18:59:39 1.12 +++ STATUS1999/11/11 13:21:10 1.13 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/11/04 18:59:39 $] +Last modified at [$Date: 1999/11/11 13:21:10 $] Release: @@ -48,7 +48,17 @@ hook system. Switch completely to the new hook system Status: Ben Laurie is working on this. +* Modules. Which modules do not work yet, can we get a list? + RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP: + +* Clean the code. There are a lot of places we used APR but didn't + remove the hacks that were required for the cross-platform code in + 1.3. We need to make the code look like APR was supposed to be there. + +* Go throught the Bug DB and research the bugs marked suspended. People + were told these would be considered for inclusion in Apache 2.0, it + would be nice to actually do so. * Win32: Redirect stderr to the error log. This is tougher than it appears because Apache for Windows now uses native file I/O libraries rather than
cvs commit: apache-2.0 STATUS
stoddard99/11/04 07:16:15 Modified:.STATUS Log: Some updates to STATUS... Revision ChangesPath 1.10 +36 -3 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.9 retrieving revision 1.10 diff -u -r1.9 -r1.10 --- STATUS1999/11/03 07:29:02 1.9 +++ STATUS1999/11/04 15:16:15 1.10 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/11/03 07:29:02 $] +Last modified at [$Date: 1999/11/04 15:16:15 $] Release: @@ -11,14 +11,24 @@ is active. RELEASE SHOWSTOPPERS: +* Port mod_rewrite +Status: Paul Reder is working on this. +* Port mod_mime_magic +Status: Paul Reder is working on this. + * suEXEC doesn't work Status: Manoj has posted an patch to fix this. [EMAIL PROTECTED] * Windows NT port isn't done -Status: Bill is working on this through his MPM work, and Ryan -is with APR. +Status: Bill is working on MPM and APR. Remaining work: +1. Add back ability to run Apache as a service +2. Fix Win9* specific code in the winnt MPM +3. Get the MPM working in multi process mode (one parent one child) +4. Test access logging with multiple threads. Will the native file I/O + calls serialize automagically like the CRT calls or do we need to + add region locking each time we write to the access/error logs? * Current 2.0 code is not tested on many Unix platforms. Make 2.0 work on most, if not all the systems 1.3 did @@ -30,6 +40,29 @@ Status: Ben Laurie is working on this. RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP: + +* Win32: Redirect stderr to the error log. This is tougher than it appears + because Apache for Windows now uses native file I/O libraries rather than + the CRT library. No problem redirecting native I/O STDERR to the error log. + However, we need to replace all the fprintf(stderr...) calls with something + that will write to the native stderr handle instead of the CRT stderr handle. + Bleh. + +* Win32: Migrate the MPM over to use APR thread/process calls. + +* Move I/O layering into APR. + +* There are still a number of places in the code where we are loosing error + status (i.e. throwing away the error returned by a system call and replacing + it with a generic error code) + +* APRize the stat function. CRT stat() is about 30% slower on Windows than + the equivalent native Windows call. The APR call should return the Unix + style stat structure, just to keep it familier. + +* Implement reliable piped logs on Windows + Status: Bill has prototype code (not reliable) in 1.3. Should be much + cleaner with APR in 2.0. * Switch to autoconf, et al. for configuration.
cvs commit: apache-2.0 STATUS
stoddard99/11/04 10:25:26 Modified:.STATUS Log: ToDo: timeout pipes to CGIs Revision ChangesPath 1.11 +5 -1 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.10 retrieving revision 1.11 diff -u -r1.10 -r1.11 --- STATUS1999/11/04 15:16:15 1.10 +++ STATUS1999/11/04 18:25:23 1.11 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/11/04 15:16:15 $] +Last modified at [$Date: 1999/11/04 18:25:23 $] Release: @@ -11,6 +11,10 @@ is active. RELEASE SHOWSTOPPERS: + +* Pipes to CGI scripts are not being timed out +Status: + * Port mod_rewrite Status: Paul Reder is working on this.
cvs commit: apache-2.0 STATUS
stoddard99/11/04 10:59:42 Modified:.STATUS Log: More stuff to do Revision ChangesPath 1.12 +17 -7 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.11 retrieving revision 1.12 diff -u -r1.11 -r1.12 --- STATUS1999/11/04 18:25:23 1.11 +++ STATUS1999/11/04 18:59:39 1.12 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/11/04 18:25:23 $] +Last modified at [$Date: 1999/11/04 18:59:39 $] Release: @@ -12,21 +12,26 @@ RELEASE SHOWSTOPPERS: +* Complete HAVE_SENDFILE work. Need to add iovec head and tail pointers to + the iol sendfile API and make it work under Unix. +Status: Bill [EMAIL PROTECTED] is working on this + * Pipes to CGI scripts are not being timed out Status: * Port mod_rewrite -Status: Paul Reder is working on this. +Status: Paul Reder [EMAIL PROTECTED] is working on this. * Port mod_mime_magic -Status: Paul Reder is working on this. +Status: Paul Reder [EMAIL PROTECTED] is working on this. * suEXEC doesn't work Status: Manoj has posted an patch to fix this. [EMAIL PROTECTED] * Windows NT port isn't done -Status: Bill is working on MPM and APR. Remaining work: +Status: Bill [EMAIL PROTECTED] is working on MPM and APR. + Remaining work: 1. Add back ability to run Apache as a service 2. Fix Win9* specific code in the winnt MPM 3. Get the MPM working in multi process mode (one parent one child) @@ -52,10 +57,15 @@ that will write to the native stderr handle instead of the CRT stderr handle. Bleh. -* Win32: Migrate the MPM over to use APR thread/process calls. +* Win32: Migrate the MPM over to use APR thread/process calls. This would eliminate + some code in the Win32 branch that essentially duplicates what is in APR. + Status: Bill [EMAIL PROTECTED] is working on this. * Move I/O layering into APR. +* APR'ize I/O layering. iol is using native sockets now, which is the source of + the double close problem Manoj ran into. + * There are still a number of places in the code where we are loosing error status (i.e. throwing away the error returned by a system call and replacing it with a generic error code) @@ -65,8 +75,8 @@ style stat structure, just to keep it familier. * Implement reliable piped logs on Windows - Status: Bill has prototype code (not reliable) in 1.3. Should be much - cleaner with APR in 2.0. +Status: Bill [EMAIL PROTECTED] has prototype code (not reliable) in 1.3. +Should be much cleaner with APR in 2.0. * Switch to autoconf, et al. for configuration.
cvs commit: apache-2.0 STATUS
manoj 99/10/31 20:30:54 Modified:.STATUS Log: Still no progress on the PalmPilot MPM, though. Revision ChangesPath 1.7 +1 -8 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.6 retrieving revision 1.7 diff -u -d -u -r1.6 -r1.7 --- STATUS1999/10/13 05:24:11 1.6 +++ STATUS1999/11/01 04:30:52 1.7 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/10/13 05:24:11 $] +Last modified at [$Date: 1999/11/01 04:30:52 $] Release: @@ -12,11 +12,7 @@ RELEASE SHOWSTOPPERS: -* CGI doesn't work -Status: Ben Laurie is working on this. - * suEXEC doesn't work -Ben Laurie's work to fix CGIs will also fix this. * Windows NT port isn't done Status: Bill is working on this through his MPM work, and Ryan @@ -67,9 +63,6 @@ * Odd comments in source (egrep for ZZZ) need to be cleaned-up * Work on other MPMs. Possible MPMs discussed or in progress include: - - - David Reid's [EMAIL PROTECTED] Beos MPM - Status: in progress - Dean Gaudet's async MPM Status: ?
cvs commit: apache-2.0 STATUS
fielding99/08/27 19:44:10 Modified:.STATUS Log: no shortage of things TODO Revision ChangesPath 1.3 +3 -1 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.2 retrieving revision 1.3 diff -u -r1.2 -r1.3 --- STATUS1999/08/27 19:02:51 1.2 +++ STATUS1999/08/28 02:44:09 1.3 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/08/27 19:02:51 $] +Last modified at [$Date: 1999/08/28 02:44:09 $] Release: @@ -57,6 +57,8 @@ * Finish infrastructure in core for async MPMs Status: ? + +* TODO in source -- just do an egrep on TODO and see what's there * Work on other MPMs. Possible MPMs discussed or in progress include:
cvs commit: apache-2.0 STATUS
fielding99/08/27 20:25:05 Modified:.STATUS Log: Enough for one week. Revision ChangesPath 1.4 +17 -1 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.3 retrieving revision 1.4 diff -u -r1.3 -r1.4 --- STATUS1999/08/28 02:44:09 1.3 +++ STATUS1999/08/28 03:25:04 1.4 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1999/08/28 02:44:09 $] +Last modified at [$Date: 1999/08/28 03:25:04 $] Release: @@ -53,6 +53,11 @@ behavior. Dean has suggested getting rid of regex containers completely. +* Areas where APRFile is being used need to be cleaned-up + +* mod_mmap_static hides ap_block_alarms() and ap_unblock_alarms() + This should really be done in some header file elsewhere. + Other features that need writing: * Finish infrastructure in core for async MPMs @@ -60,6 +65,8 @@ * TODO in source -- just do an egrep on TODO and see what's there +* Odd comments in source (egrep for ZZZ) need to be cleaned-up + * Work on other MPMs. Possible MPMs discussed or in progress include: - David Reid's [EMAIL PROTECTED] Beos MPM @@ -82,6 +89,12 @@ Status: Ben Laurie has written some hooks documentation (apache-2.0/htdocs/hooks.html) +* Changes since 1.3.9 can be more easily seen in the commitlog file + dev.apache.org:/home/cvs/CVSROOT/commitlogs/apache-2.0 + which includes some of Roy's comments when the changes were + committed in rough change-sets by purpose. Note that the commitlog + does not show the contents of new files until later. + Available Patches: * Mike Abbott's [EMAIL PROTECTED] patches to improve @@ -94,3 +107,6 @@ * What do we do about mod_proxy? * Which MPMs will be included with Apache 2.0? + + * Is conf/highperformance.conf-dist obsolete? It looks obsolete. +
cvs commit: apache-2.0 STATUS
manoj 99/08/27 11:57:20 Added: .STATUS Log: First crack at a status file for Apache 2.0. Please fix mistakes and add/rearrange your pet projects Revision ChangesPath 1.1 apache-2.0/STATUS Index: STATUS === Apache 2.0 STATUS: Last modified at [$Date: 1999/08/27 18:57:17 $] Release: 2.0: In pre-alpha development Plan: * Vague goal of an alpha or beta release in 1999. Commit-then-review is active. RELEASE SHOWSTOPPERS: * CGI doesn't work Status: Ben Laurie is working on this. * mod_status doesn't work Status: Manoj Kasichainula [EMAIL PROTECTED] is working on this. A patch has been posted that starts to fix this: ([EMAIL PROTECTED] and [EMAIL PROTECTED]) * suEXEC doesn't work Ben Laurie's work to fix CGIs will also fix this. * Windows NT port isn't done Status: Bill is working on this through his MPM work, and Ryan is with APR. * Current 2.0 code is not tested on many Unix platforms. Make 2.0 work on most, if not all the systems 1.3 did Status: Autoconf and APR will get us most of the way there. * The module API is a weird combination of the old table and the new hook system. Switch completely to the new hook system Status: Ben Laurie is working on this. RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP: (none listed) Other bugs that need fixing: * MaxRequestsPerChild measures connections, not requests. * Regex containers don't work in an intutive way Status: No one has come up with an efficient way to fix this behavior. Dean has suggested getting rid of regex containers completely. Other features that need writing: * Switch to autoconf, et al. for configuration. * Use APR to get rid of more platform dependancies. Status: Ryan Bloom [EMAIL PROTECTED] is working on this. * Finish infrastructure in core for async MPMs Status: ? * Work on other MPMs. Possible MPMs discussed or in progress include: - David Reid's [EMAIL PROTECTED] Beos MPM Status: in progress - Dean Gaudet's async MPM Status: ? - Zach Brown's [EMAIL PROTECTED] Linux siginfo MPM Status: ? Documentation that needs writing: * The concept of MPMs, especially if we ship more than one MPM for a given platform * New directives in the various MPMs * API documentation Status: Ben Laurie has written some hooks documentation (apache-2.0/htdocs/hooks.html) Available Patches: * Mike Abbott's [EMAIL PROTECTED] patches to improve performance Status: These were written for 1.3, and are awaiting a port to 2.0 Open issues: * What do we do about mod_proxy? * Which MPMs will be included with Apache 2.0?
cvs commit: apache-2.0 STATUS
fielding99/01/07 14:55:43 Modified:.STATUS Log: Add more info on sfio option Revision ChangesPath 1.47 +14 -3 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.46 retrieving revision 1.47 diff -u -r1.46 -r1.47 --- STATUS1998/12/29 22:21:42 1.46 +++ STATUS1999/01/07 22:55:42 1.47 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1998/12/29 22:21:42 $] +Last modified at [$Date: 1999/01/07 22:55:42 $] Release: @@ -151,7 +151,8 @@ Roy, Lars Status: Ken has volunteered. - o sfio + o sfio_1998 http://www.research.att.com/sw/tools/sfio/ + Dean says it's not threadsafe. Specifically these need to be fixed: - sfprints.c has a static Sfio_t *f which would need to be @@ -169,7 +170,17 @@ There could be more... that was just a quick look around. But the perl folks seem to be using sfio, so they may have already - looked into the threading issues. + looked into the threading issues. Also, the layering only provides +for four primitives: read, write, seek, except (an exception +mechanism). I kind of like NSPR's approach which has all the +socket primitives in the list as well. It's also lacking the +timeout parameters, because it wasn't specifically designed to +deal with sockets like NSPR was. + +Roy talked to the authors (Glenn Fowler, David Korn, and +Kiem-Phong Vo) at ATT Research when he gave an Apache +talk there in October. We might be able to get one of them +to participate, or at least get answers to tough questions. o bstdio This was written by Chris Provenzano as part of his implementation
cvs commit: apache-2.0 STATUS
dgaudet 98/12/29 14:21:43 Modified:.STATUS Log: more details on sfio Revision ChangesPath 1.46 +19 -2 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.45 retrieving revision 1.46 diff -u -r1.45 -r1.46 --- STATUS1998/12/17 15:14:06 1.45 +++ STATUS1998/12/29 22:21:42 1.46 @@ -1,5 +1,5 @@ Apache 2.0 STATUS: -Last modified at [$Date: 1998/12/17 15:14:06 $] +Last modified at [$Date: 1998/12/29 22:21:42 $] Release: @@ -152,7 +152,24 @@ Status: Ken has volunteered. o sfio --1: Dean [until it's shown to be thread safe (RST claims it isn't)] +Dean says it's not threadsafe. Specifically these need + to be fixed: + - sfprints.c has a static Sfio_t *f which would need to be + made an auto for thread safety + - sfcvt.c has a static char *Buf... needs to have similar + change that we did to the cvt foo in apache + - sftmp.c has a few statics... probably best to just lock in here + - sfpopen.c has a couple statics that need initializing once + - sfexit.c needs a few locks + - sfmode.c Sfrsrv pool of buffers needs to be locked + + These can just be avoided: + - the Stdio_b interface has a bunch of statics + - the sfdcdos example discipline has a static + + There could be more... that was just a quick look around. But + the perl folks seem to be using sfio, so they may have already + looked into the threading issues. o bstdio This was written by Chris Provenzano as part of his implementation
cvs commit: apache-2.0 STATUS
coar98/10/22 13:44:37 Modified:.STATUS Log: Add Ryan Bloom's name, since he told me he'd like to be involved in the config language rework. Revision ChangesPath 1.43 +1 -1 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.42 retrieving revision 1.43 diff -u -r1.42 -r1.43 --- STATUS1998/09/28 17:33:14 1.42 +++ STATUS1998/10/22 20:44:36 1.43 @@ -221,7 +221,7 @@ * new configuration language +1: Dean, Marc, Ben, Martin, Roy, Lars +0: Ralf, Paul -Status: Ken has volunteered +Status: Ken has volunteered, as has Ryan Bloom [EMAIL PROTECTED] o use XML +0: Roy, Ralf
cvs commit: apache-2.0 STATUS
lars98/09/28 10:33:15 Modified:.STATUS Log: Revision ChangesPath 1.42 +28 -21apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.41 retrieving revision 1.42 diff -u -r1.41 -r1.42 --- STATUS1998/09/25 19:44:59 1.41 +++ STATUS1998/09/28 17:33:14 1.42 @@ -25,17 +25,19 @@ segments should persist until the refcount drops to zero. It would be cool if pools could be created in such segments to allow things like shared tables and arrays. - +1: Roy, Paul, Ken, Ralf, Martin + +1: Roy, Paul, Ken, Ralf, Martin, Lars * Apache ports project - simple-to-install (a la CPAN) one-off tools, scripts (such as counters, guest books, et cetera) +1: Roy, Paul, Ken, Ralf, Martin David Southwell has volunteered to work on this. + -0: Lars [If we do this for modules only I'm +1, but + I'm not very happy with scripts, tools or other stuff.] * Embed API documentation in the source, a la Java docs, so a script can extract the parameters and description and build a dictionary. If we do this, it needs to be *mandatory* in the style guide. - +1: Ken + +1: Ken, Lars * Abstract the '+/-' capabilities of the Options keyword so other directives can use it without re-inventing the square peg. @@ -51,7 +53,7 @@ formats that bear a faint resemblance to other usages (such as allowing =~ and !~) in other languages, and provide for caseless matches. - +1: Roy, Ken, Martin + +1: Roy, Ken, Martin, Lars Martin sez: I've been thinking about replacing the parser by a recursive descent parser which would reduce complexity tremendously. @@ -62,7 +64,7 @@ to have things like the arrays, tables, pools, and related primitives moved into a library of which httpd is just a client and other things can be too. - +1: Roy, Paul, Jim, Ken, Ralf, Martin + +1: Roy, Paul, Jim, Ken, Ralf, Martin, Lars * Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset. @@ -73,11 +75,12 @@ varies from the above such that if it's shown that the autoconf toolset can do what we want, with less headache than what we have, then we go for it) - +1: Jim, Ken, Marc, MarkC, Ben, Paul, Martin + +1: Jim, Ken, Marc, MarkC, Ben, Paul, Martin, Lars * The autoconf toolset should include all three: autoconf, automake, and libtool. +1: Brian, Jim, Roy, Dean, Ken, Ralf, Martin + +0: Lars * Whatever we do regarding autoconf, we should be able to configure to build objects other than in the source tree. autoconf allows for this... you @@ -85,7 +88,7 @@ platforms... or even on a single platform, one copy with profiling another without. (There are a lot of possibilities: creating shadow trees, VPATH-style Makefile settings, etc.) - +1: Dean, Roy, Paul, Ralf, Martin + +1: Dean, Roy, Paul, Ralf, Martin, Lars * One of the main restrictions on Apache has been that we must assume a very low-level common denominator for the OSs out there. For example, @@ -116,7 +119,8 @@ if they can help parallelize development or simplify the core code.] * multithreading. -+1: Brian, Ken, Jim, Paul, Sameer, Marc, Ralf, MarkC, Ben, Martin, Roy ++1: Brian, Ken, Jim, Paul, Sameer, Marc, Ralf, MarkC, Ben, Martin, +Roy, Lars o Thread Abstraction +1: Sameer, Marc, MarkC, Ben, Dean, Paul, Martin, Roy, Ralf Status: nobody has volunteered yet @@ -124,12 +128,13 @@ * revamped process model (Dean's proposal) Dean says: it's hard to do the multithreading work cleanly without considering a bunch of this -+1: MarkC, Paul, Dean, Martin, Roy, Ralf ++1: MarkC, Paul, Dean, Martin, Roy, Ralf, Lars Marc (+1 on much of it; threads aren't enough for perf.), Status: nobody has volunteered yet * new layered I/O. -+1: Brian, Ken, Dean, Jim, Paul, Sameer, Marc, Ralf, MarkC, Ben, Roy ++1: Brian, Ken, Dean, Jim, Paul, Sameer, Marc, Ralf, MarkC, Ben, +Roy, Lars Status: Ken has volunteered. o sfio @@ -152,7 +157,7 @@ The only problem is reclaiming buffers so that a large SSI won't suck up all available memory just sending it out the pipe.] +1: Dean, Marc, Ben, Paul, Martin, Roy -+0: Jim [what about examples? Portability concerns?], Ralf ++0: Jim [what about
cvs commit: apache-2.0 STATUS
coar98/09/25 12:45:01 Modified:.STATUS Log: Add another item for 2.0, that Great White Hope for Whirled Peas. Revision ChangesPath 1.41 +7 -1 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.40 retrieving revision 1.41 diff -u -r1.40 -r1.41 --- STATUS1998/07/12 18:48:19 1.40 +++ STATUS1998/09/25 19:44:59 1.41 @@ -10,7 +10,7 @@ should add them to the repository now. Use a descriptive filename. -Other code will be copied over when 1.3.0 is finished. +Other code will be copied over when 1.3.? is finished. Showstoppers: Committed Code Changes: @@ -30,11 +30,17 @@ * Apache ports project - simple-to-install (a la CPAN) one-off tools, scripts (such as counters, guest books, et cetera) +1: Roy, Paul, Ken, Ralf, Martin + David Southwell has volunteered to work on this. * Embed API documentation in the source, a la Java docs, so a script can extract the parameters and description and build a dictionary. If we do this, it needs to be *mandatory* in the style guide. +1: Ken + + * Abstract the '+/-' capabilities of the Options keyword so other +directives can use it without re-inventing the square peg. + +1: Ken + Status: Ken has volunteered to work on this. * Really, really comply with 2068 (or whatever Roy says is the right document). Like handling Vary response fields properly and reliably.
cvs commit: apache-2.0 STATUS
coar98/07/12 11:48:20 Modified:.STATUS Log: Another issue for future work.. Revision ChangesPath 1.40 +4 -0 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.39 retrieving revision 1.40 diff -u -r1.39 -r1.40 --- STATUS1998/06/11 10:01:30 1.39 +++ STATUS1998/07/12 18:48:19 1.40 @@ -36,6 +36,10 @@ If we do this, it needs to be *mandatory* in the style guide. +1: Ken + * Really, really comply with 2068 (or whatever Roy says is the right +document). Like handling Vary response fields properly and reliably. + +1: Ken + * Improve mod_include SSI handling, perhaps by caching offsets to directives. Also, clean up the conditional syntax to allow formats that bear a faint resemblance to other usages (such
cvs commit: apache-2.0 STATUS
coar98/06/11 03:01:30 Modified:.STATUS Log: Another SSI enhancement tweak. Revision ChangesPath 1.39 +2 -1 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.38 retrieving revision 1.39 diff -u -r1.38 -r1.39 --- STATUS1998/06/02 11:43:41 1.38 +++ STATUS1998/06/11 10:01:30 1.39 @@ -39,7 +39,8 @@ * Improve mod_include SSI handling, perhaps by caching offsets to directives. Also, clean up the conditional syntax to allow formats that bear a faint resemblance to other usages (such -as allowing =~ and !~) in other languages. +as allowing =~ and !~) in other languages, and provide +for caseless matches. +1: Roy, Ken, Martin Martin sez: I've been thinking about replacing the parser by a recursive descent parser which would
cvs commit: apache-2.0 STATUS
coar98/06/02 04:43:42 Modified:.STATUS Log: Your local documentation fiend, at your service.. Revision ChangesPath 1.38 +5 -0 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.37 retrieving revision 1.38 diff -u -r1.37 -r1.38 --- STATUS1998/05/06 21:30:04 1.37 +++ STATUS1998/06/02 11:43:41 1.38 @@ -31,6 +31,11 @@ tools, scripts (such as counters, guest books, et cetera) +1: Roy, Paul, Ken, Ralf, Martin + * Embed API documentation in the source, a la Java docs, so a script +can extract the parameters and description and build a dictionary. +If we do this, it needs to be *mandatory* in the style guide. + +1: Ken + * Improve mod_include SSI handling, perhaps by caching offsets to directives. Also, clean up the conditional syntax to allow formats that bear a faint resemblance to other usages (such
cvs commit: apache-2.0 STATUS
martin 98/05/06 14:30:05 Modified:.STATUS Log: Cool temperatures, but next weekend will be *warm* Revision ChangesPath 1.37 +4 -2 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.36 retrieving revision 1.37 diff -u -u -r1.36 -r1.37 --- STATUS1998/04/27 22:50:17 1.36 +++ STATUS1998/05/06 21:30:04 1.37 @@ -209,12 +209,14 @@ +0: Marc, Ralf -1: MarkC, Paul, Roy, Ken -* make everything C++ friendly (fix that damned pool decl) -+1: Roy, Paul, Ken, Ralf +* make everything C++ friendly ++1: Roy, Paul, Ken, Ralf, Martin Status: nobody has volunteered yet +(that damned pool decl is fixed now) * Proxy enhancements (or drop proxy altogether?) for HTTP/1.1 and better ftp proxy Auth handling + +1: Martin Closed issues:
cvs commit: apache-2.0 STATUS
martin 98/04/27 15:50:18 Modified:.STATUS Log: vote (as long as I still can) Revision ChangesPath 1.36 +15 -9 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.35 retrieving revision 1.36 diff -u -u -r1.35 -r1.36 --- STATUS1998/04/03 19:48:44 1.35 +++ STATUS1998/04/27 22:50:17 1.36 @@ -25,17 +25,20 @@ segments should persist until the refcount drops to zero. It would be cool if pools could be created in such segments to allow things like shared tables and arrays. - +1: Roy, Paul, Ken, Ralf + +1: Roy, Paul, Ken, Ralf, Martin * Apache ports project - simple-to-install (a la CPAN) one-off tools, scripts (such as counters, guest books, et cetera) - +1: Roy, Paul, Ken, Ralf + +1: Roy, Paul, Ken, Ralf, Martin * Improve mod_include SSI handling, perhaps by caching offsets to directives. Also, clean up the conditional syntax to allow formats that bear a faint resemblance to other usages (such as allowing =~ and !~) in other languages. - +1: Roy, Ken + +1: Roy, Ken, Martin +Martin sez: I've been thinking about replacing the parser + by a recursive descent parser which would + reduce complexity tremendously. * Apache reusable code library, wherein we can put some of the stuff developed during the HTTP project that would be useful elsewhere. @@ -43,11 +46,11 @@ to have things like the arrays, tables, pools, and related primitives moved into a library of which httpd is just a client and other things can be too. - +1: Roy, Paul, Jim, Ken, Ralf + +1: Roy, Paul, Jim, Ken, Ralf, Martin * Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset. - +1: Brian, Roy, Dean, Ralf + +1: Brian, Roy, Dean, Ralf, Martin * Investigate replacing the current Unix compilation model (Configuration.tmpl home-brew Configure script) with the autoconf toolset. (this @@ -66,7 +69,7 @@ platforms... or even on a single platform, one copy with profiling another without. (There are a lot of possibilities: creating shadow trees, VPATH-style Makefile settings, etc.) - +1: Dean, Roy, Paul, Ralf + +1: Dean, Roy, Paul, Ralf, Martin * One of the main restrictions on Apache has been that we must assume a very low-level common denominator for the OSs out there. For example, @@ -164,7 +167,7 @@ o change API 'phase' model to use module-registered hooks rather than a fixed static structure -+1: Ken, Ralf, MarkC, Paul, Dean, Roy ++1: Ken, Ralf, MarkC, Paul, Dean, Roy, Martin Status: Ken has volunteered o use virtual functions for module hooks @@ -172,14 +175,14 @@ -1: Paul, Roy [would require two APIs] o clearly identify API functions by renaming them -+1: Ken, Ralf, Ben, Paul (plus back compat.), Dean, Jim, Roy ++1: Ken, Ralf, Ben, Paul (plus back compat.), Dean, Jim, Roy, Martin Status: Ken has volunteered [Roy: looks like it is already in 1.3] o backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers +1: Paul, Sameer, Marc, MarkC --1: Roy, Ralf +-1: Roy, Ralf, Martin o make API call syntax rational (e.g., all r*() routines list r as their first argument, et cetera) @@ -209,6 +212,9 @@ * make everything C++ friendly (fix that damned pool decl) +1: Roy, Paul, Ken, Ralf Status: nobody has volunteered yet + +* Proxy enhancements (or drop proxy altogether?) for HTTP/1.1 + and better ftp proxy Auth handling Closed issues:
cvs commit: apache-2.0 STATUS
sameer 98/04/03 11:48:45 Modified:.STATUS Log: evaluate nspr Revision ChangesPath 1.35 +10 -0 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.34 retrieving revision 1.35 diff -c -u -r1.34 -r1.35 /usr/bin/diff: conflicting specifications of output style --- STATUS1998/03/26 14:48:17 1.34 +++ STATUS1998/04/03 19:48:44 1.35 @@ -135,6 +135,16 @@ +1: Dean, Marc, Ben, Paul, Martin, Roy +0: Jim [what about examples? Portability concerns?], Ralf + o NSPR: Sameer thinks that we should evaluate NSPR (ns/nspr in the +Mozilla source tree) and determine whether or not it sucks and +decide if we can integrate it. It appears to me, from reading +the NPL, that including NSPR in Apache 2.0 is legit. Vote + +1 if you like NSPR, and want to use it. + + +1: + +0: Sameer + + * API work o radically revamped API
cvs commit: apache-2.0 STATUS
fielding98/03/25 23:13:44 Modified:.STATUS Log: Add my comments/votes. Fixed the formatting and removed the kindler, gentler form of saying piss off to a proposal. Revision ChangesPath 1.30 +108 -92 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.29 retrieving revision 1.30 diff -u -r1.29 -r1.30 --- STATUS1998/03/17 15:00:24 1.29 +++ STATUS1998/03/26 07:13:43 1.30 @@ -3,9 +3,6 @@ Release: 2.0 : In pre-alpha development - see: http://www.arctic.org/~dgaudet/apache/2.0/process-model -Dean says: This sounds like I'm working on implementing this -proposal. I'm not. Nobody is as far as I know. Plan: @@ -28,14 +25,17 @@ segments should persist until the refcount drops to zero. It would be cool if pools could be created in such segments to allow things like shared tables and arrays. + +1: Roy * Apache ports project - simple-to-install (a la CPAN) one-off tools, scripts (such as counters, guest books, et cetera) + +1: Roy * Improve mod_include SSI handling, perhaps by caching offsets to directives. Also, clean up the conditional syntax to allow formats that bear a faint resemblance to other usages (such as allowing =~ and !~) in other languages. + +1: Roy * Apache reusable code library, wherein we can put some of the stuff developed during the HTTP project that would be useful elsewhere. @@ -43,28 +43,29 @@ to have things like the arrays, tables, pools, and related primitives moved into a library of which httpd is just a client and other things can be too. + +1: Roy * Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset. - Status: Brian +1, Roy +1, Dean +1, Ralf +1 + +1: Brian, Roy, Dean, Ralf * Investigate replacing the current Unix compilation model (Configuration.tmpl home-brew Configure script) with the autoconf toolset. (this varies from the above such that if it's shown that the autoconf toolset can do what we want, with less headache than what we have, then we go for it) - Status: Jim +1, Ken +1, Marc +1, MarkC +1, Ben +1, Paul +1, Martin +1 + +1: Jim, Ken, Marc, MarkC, Ben, Paul, Martin * The autoconf toolset should include all three: autoconf, automake, and libtool. - Status: Brian +1, Jim +1, Roy +1, Dean +1, Ken +1, Ralf +1, Martin +1 + +1: Brian, Jim, Roy, Dean, Ken, Ralf, Martin * Whatever we do regarding autoconf, we should be able to configure to build objects other than in the source tree. autoconf allows for this... you can do mkdir obj; cd obj; ../configure. This is great for multiple platforms... or even on a single platform, one copy with profiling another without. -Status: Dean +1 + +1: Dean, Roy * One of the main restrictions on Apache has been that we must assume a very low-level common denominator for the OSs out there. For example, @@ -76,6 +77,11 @@ compile and run. Many of the ideas floating around for 2.0 would, by default, (drastically) change this. Is this a good idea? + [Roy: That overstates most of the changes proposed so far. In any +case, older systems will continue to be supported by older +versions of Apache -- it is not desirable to dull the cutting edge +just so people nowhere near the cutting edge won't get cut.] + FEATURE SET FOR 2.0 Here, we decide how many of the following feature ideas we will set for ourselves as work items for 2.0. We can't do everything we would want @@ -83,105 +89,115 @@ conservative with your votes. Items in no particular order. Feel free to add more, but try not to duplicate earlier items too much. + [Roy: The amount of time it will take to complete 2.0 will have +very little to do with how much change is attempted --- it depends +only on how many people can change things simultaneously, and thus +many suggested changes will actually speed-up the overall schedule +if they can help parallelize development or simplify the core code.] + * multithreading. - Status: Brian +1, Ken +1, Jim +1, Paul +1, Sameer +1, Marc +1, - Ralf +1, MarkC +1, Ben +1, Martin +1 - - Thread Abstraction - Status: Sameer +1, Marc +1, MarkC +1, Ben +1, Dean +1, Paul +1, - Martin +1 - Volunteers: ++1: Brian, Ken, Jim, Paul, Sameer, Marc, Ralf, MarkC, Ben, Martin, Roy + o Thread Abstraction ++1: Sameer,
cvs commit: apache-2.0 STATUS
pcs 98/03/26 00:57:58 Modified:.STATUS Log: Votes! Dull and rainy again. 11-14. Revision ChangesPath 1.31 +5 -5 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.30 retrieving revision 1.31 diff -u -r1.30 -r1.31 --- STATUS1998/03/26 07:13:43 1.30 +++ STATUS1998/03/26 08:57:58 1.31 @@ -25,11 +25,11 @@ segments should persist until the refcount drops to zero. It would be cool if pools could be created in such segments to allow things like shared tables and arrays. - +1: Roy + +1: Roy, Paul * Apache ports project - simple-to-install (a la CPAN) one-off tools, scripts (such as counters, guest books, et cetera) - +1: Roy + +1: Roy, Paul * Improve mod_include SSI handling, perhaps by caching offsets to directives. Also, clean up the conditional syntax to allow @@ -43,7 +43,7 @@ to have things like the arrays, tables, pools, and related primitives moved into a library of which httpd is just a client and other things can be too. - +1: Roy + +1: Roy, Paul * Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset. @@ -65,7 +65,7 @@ can do mkdir obj; cd obj; ../configure. This is great for multiple platforms... or even on a single platform, one copy with profiling another without. - +1: Dean, Roy + +1: Dean, Roy, Paul * One of the main restrictions on Apache has been that we must assume a very low-level common denominator for the OSs out there. For example, @@ -196,7 +196,7 @@ -1: MarkC, Paul, Roy, Ken * make everything C++ friendly (fix that damned pool decl) -+1: Roy ++1: Roy, Paul Status: nobody has volunteered yet Closed issues:
cvs commit: apache-2.0 STATUS
jim 98/03/26 01:36:10 Modified:.STATUS Log: Some votes Revision ChangesPath 1.32 +1 -1 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.31 retrieving revision 1.32 diff -u -r1.31 -r1.32 --- STATUS1998/03/26 08:57:58 1.31 +++ STATUS1998/03/26 09:36:08 1.32 @@ -43,7 +43,7 @@ to have things like the arrays, tables, pools, and related primitives moved into a library of which httpd is just a client and other things can be too. - +1: Roy, Paul + +1: Roy, Paul, Jim * Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset.
cvs commit: apache-2.0 STATUS
coar98/03/26 03:45:11 Modified:.STATUS Log: Votes, votes, votes. Haven't looked outside yet - and what has the weather to do with it, anyway? Revision ChangesPath 1.33 +5 -5 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.32 retrieving revision 1.33 diff -u -r1.32 -r1.33 --- STATUS1998/03/26 09:36:08 1.32 +++ STATUS1998/03/26 11:45:10 1.33 @@ -25,17 +25,17 @@ segments should persist until the refcount drops to zero. It would be cool if pools could be created in such segments to allow things like shared tables and arrays. - +1: Roy, Paul + +1: Roy, Paul, Ken * Apache ports project - simple-to-install (a la CPAN) one-off tools, scripts (such as counters, guest books, et cetera) - +1: Roy, Paul + +1: Roy, Paul, Ken * Improve mod_include SSI handling, perhaps by caching offsets to directives. Also, clean up the conditional syntax to allow formats that bear a faint resemblance to other usages (such as allowing =~ and !~) in other languages. - +1: Roy + +1: Roy, Ken * Apache reusable code library, wherein we can put some of the stuff developed during the HTTP project that would be useful elsewhere. @@ -43,7 +43,7 @@ to have things like the arrays, tables, pools, and related primitives moved into a library of which httpd is just a client and other things can be too. - +1: Roy, Paul, Jim + +1: Roy, Paul, Jim, Ken * Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset. @@ -196,7 +196,7 @@ -1: MarkC, Paul, Roy, Ken * make everything C++ friendly (fix that damned pool decl) -+1: Roy, Paul ++1: Roy, Paul, Ken Status: nobody has volunteered yet Closed issues:
cvs commit: apache-2.0 STATUS
rse 98/03/26 06:48:18 Modified:.STATUS Log: votes... Revision ChangesPath 1.34 +14 -13apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.33 retrieving revision 1.34 diff -u -r1.33 -r1.34 --- STATUS1998/03/26 11:45:10 1.33 +++ STATUS1998/03/26 14:48:17 1.34 @@ -25,11 +25,11 @@ segments should persist until the refcount drops to zero. It would be cool if pools could be created in such segments to allow things like shared tables and arrays. - +1: Roy, Paul, Ken + +1: Roy, Paul, Ken, Ralf * Apache ports project - simple-to-install (a la CPAN) one-off tools, scripts (such as counters, guest books, et cetera) - +1: Roy, Paul, Ken + +1: Roy, Paul, Ken, Ralf * Improve mod_include SSI handling, perhaps by caching offsets to directives. Also, clean up the conditional syntax to allow @@ -43,7 +43,7 @@ to have things like the arrays, tables, pools, and related primitives moved into a library of which httpd is just a client and other things can be too. - +1: Roy, Paul, Jim, Ken + +1: Roy, Paul, Jim, Ken, Ralf * Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset. @@ -64,8 +64,9 @@ objects other than in the source tree. autoconf allows for this... you can do mkdir obj; cd obj; ../configure. This is great for multiple platforms... or even on a single platform, one copy with profiling another -without. - +1: Dean, Roy, Paul +without. (There are a lot of possibilities: creating shadow trees, +VPATH-style Makefile settings, etc.) + +1: Dean, Roy, Paul, Ralf * One of the main restrictions on Apache has been that we must assume a very low-level common denominator for the OSs out there. For example, @@ -98,13 +99,13 @@ * multithreading. +1: Brian, Ken, Jim, Paul, Sameer, Marc, Ralf, MarkC, Ben, Martin, Roy o Thread Abstraction -+1: Sameer, Marc, MarkC, Ben, Dean, Paul, Martin, Roy ++1: Sameer, Marc, MarkC, Ben, Dean, Paul, Martin, Roy, Ralf Status: nobody has volunteered yet * revamped process model (Dean's proposal) Dean says: it's hard to do the multithreading work cleanly without considering a bunch of this -+1: MarkC, Paul, Dean, Martin, Roy, ++1: MarkC, Paul, Dean, Martin, Roy, Ralf Marc (+1 on much of it; threads aren't enough for perf.), Status: nobody has volunteered yet @@ -132,7 +133,7 @@ The only problem is reclaiming buffers so that a large SSI won't suck up all available memory just sending it out the pipe.] +1: Dean, Marc, Ben, Paul, Martin, Roy -+0: Jim [what about examples? Portability concerns?] ++0: Jim [what about examples? Portability concerns?], Ralf * API work @@ -167,8 +168,8 @@ o backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers -+1: Paul, Sameer, Marc, Ralf, MarkC --1: Roy ++1: Paul, Sameer, Marc, MarkC +-1: Roy, Ralf o make API call syntax rational (e.g., all r*() routines list r as their first argument, et cetera) @@ -178,7 +179,7 @@ o abstract module layering for plugins (e.g., a mod_auth interface into which mod_auth_mumble modules can be plugged) -+1: Ken, Martin ++1: Ken, Martin, Ralf +0: Roy [needs more detailed proposal] Status: Ken has volunteered @@ -188,7 +189,7 @@ Status: Ken has volunteered o use XML -+0: Roy ++0: Roy, Ralf * rewrite in C++ +1: Ben, Martin @@ -196,7 +197,7 @@ -1: MarkC, Paul, Roy, Ken * make everything C++ friendly (fix that damned pool decl) -+1: Roy, Paul, Ken ++1: Roy, Paul, Ken, Ralf Status: nobody has volunteered yet Closed issues:
cvs commit: apache-2.0 STATUS
jim 98/03/17 07:00:25 Modified:.STATUS Log: Ahhh... At this rate, Apache 2.0 should be called Apache 2000 :) Revision ChangesPath 1.29 +16 -7 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.28 retrieving revision 1.29 diff -u -r1.28 -r1.29 --- STATUS1998/03/14 17:12:51 1.28 +++ STATUS1998/03/17 15:00:24 1.29 @@ -66,6 +66,16 @@ without. Status: Dean +1 + * One of the main restrictions on Apache has been that we must assume +a very low-level common denominator for the OSs out there. For example, +Configure is always being derided as crappy, but it is restricted, +by long tradition and common sense, to only use those capabilities +that existed in the System 7 'sh' (eg: no function, etc...). One +possible key to Apache's success is that it does not require any +more than basic UNIX tools (and an ANSI C compiler) to build, +compile and run. Many of the ideas floating around for 2.0 would, +by default, (drastically) change this. Is this a good idea? + FEATURE SET FOR 2.0 Here, we decide how many of the following feature ideas we will set for ourselves as work items for 2.0. We can't do everything we would want @@ -99,10 +109,7 @@ . bstdio This was written by Chris Provenzano as part of his implementation - of Posix threads... Jim can place a copy of the RST-hacked - version on dev.apache.org if needed and possible. --- RST never donated his hacks to the Group. Don't put it up for - download unless you've cleared it with him. --Brian + of Posix threads. . page flipping friendly, page-sized buffer oriented, zero copy I/O (In this model there are functions like readbuf() which return a @@ -113,7 +120,8 @@ style of stdio, which adds an extra user space copy.) Martin asks: Is there some software flying around where such a model has been tried? Or is it a totally new technique? - Status: Dean +1, Marc +1, Ben +1, Paul +1, Martin +1 + Jim asks: examples? Portability concerns? + Status: Dean +1, Marc +1, Ben +1, Paul +1, Martin +1, Jim 0 Volunteers: * API work @@ -124,7 +132,7 @@ . documented API Status: Ken +1, Sameer +1, Marc +1, Ralf +1, Paul +1, Dean +1, - Martin +1 + Martin +1, Jim +1 Volunteers: Ken . just new API phases @@ -143,7 +151,8 @@ Volunteers: . clearly identify API functions by renaming them - Status: Ken +1, Ralf +1, Ben +1, Paul +1 (plus back compat.), Dean +1 + Status: Ken +1, Ralf +1, Ben +1, Paul +1 (plus back compat.), Dean +1, + Jim +1 Volunteers: Ken . backward compatibility with 1.3 (just require a recompile)
cvs commit: apache-2.0 STATUS
dgaudet 98/03/14 02:13:17 Modified:.STATUS Log: Revision ChangesPath 1.27 +7 -0 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.26 retrieving revision 1.27 diff -u -r1.26 -r1.27 --- STATUS1998/03/08 15:07:21 1.26 +++ STATUS1998/03/14 10:13:15 1.27 @@ -53,6 +53,13 @@ * The autoconf toolset should include all three: autoconf, automake, and libtool. Status: Brian +1, Jim +1, Roy +1, Dean +1, Ken +1, Ralf +1, Martin +1 + + * Whatever we do regarding autoconf, we should be able to configure to build +objects other than in the source tree. autoconf allows for this... you +can do mkdir obj; cd obj; ../configure. This is great for multiple +platforms... or even on a single platform, one copy with profiling another +without. +Status: Dean +1 FEATURE SET FOR 2.0 Here, we decide how many of the following feature ideas we will set for
cvs commit: apache-2.0 STATUS
coar98/03/08 07:07:21 Modified:.STATUS Log: Another blue-sky idea for a primitive-set. Revision ChangesPath 1.26 +6 -0 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.25 retrieving revision 1.26 diff -u -r1.25 -r1.26 --- STATUS1998/03/07 21:26:17 1.25 +++ STATUS1998/03/08 15:07:21 1.26 @@ -23,6 +23,12 @@ Open issues: + * Library of routines to allow access to memory shared between +processes, for things like per-module caches. Such shared +segments should persist until the refcount drops to zero. +It would be cool if pools could be created in such segments +to allow things like shared tables and arrays. + * Apache ports project - simple-to-install (a la CPAN) one-off tools, scripts (such as counters, guest books, et cetera)
cvs commit: apache-2.0 STATUS
coar98/03/07 13:26:18 Modified:.STATUS Log: Just recording a couple of other things that have been blue-skied. Revision ChangesPath 1.25 +10 -0 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.24 retrieving revision 1.25 diff -u -r1.24 -r1.25 --- STATUS1998/02/18 20:26:22 1.24 +++ STATUS1998/03/07 21:26:17 1.25 @@ -23,6 +23,16 @@ Open issues: + * Apache ports project - simple-to-install (a la CPAN) one-off +tools, scripts (such as counters, guest books, et cetera) + + * Apache reusable code library, wherein we can put some of the stuff +developed during the HTTP project that would be useful elsewhere. +We've got a start on this with libap, but it would be really cool +to have things like the arrays, tables, pools, and related primitives +moved into a library of which httpd is just a client and other things +can be too. + * Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset. Status: Brian +1, Roy +1, Dean +1, Ralf +1
cvs commit: apache-2.0 STATUS
martin 98/02/18 12:26:23 Modified:.STATUS Log: Vote. It's been clear blue sky all day long Revision ChangesPath 1.24 +17 -11apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.23 retrieving revision 1.24 diff -u -u -r1.23 -r1.24 --- STATUS1998/02/11 15:59:44 1.23 +++ STATUS1998/02/18 20:26:22 1.24 @@ -32,11 +32,11 @@ varies from the above such that if it's shown that the autoconf toolset can do what we want, with less headache than what we have, then we go for it) - Status: Jim +1, Ken +1, Marc +1, MarkC +1, Ben +1, Paul +1 + Status: Jim +1, Ken +1, Marc +1, MarkC +1, Ben +1, Paul +1, Martin +1 * The autoconf toolset should include all three: autoconf, automake, and libtool. - Status: Brian +1, Jim +1, Roy +1, Dean +1, Ken +1, Ralf +1 + Status: Brian +1, Jim +1, Roy +1, Dean +1, Ken +1, Ralf +1, Martin +1 FEATURE SET FOR 2.0 Here, we decide how many of the following feature ideas we will set for @@ -47,16 +47,17 @@ * multithreading. Status: Brian +1, Ken +1, Jim +1, Paul +1, Sameer +1, Marc +1, - Ralf +1, MarkC +1, Ben +1 + Ralf +1, MarkC +1, Ben +1, Martin +1 - Thread Abstraction - Status: Sameer +1, Marc +1, MarkC +1, Ben +1, Dean +1, Paul +1 + Status: Sameer +1, Marc +1, MarkC +1, Ben +1, Dean +1, Paul +1, + Martin +1 Volunteers: * revamped process model (Dean's proposal) Dean says: it's hard to do the multithreading work cleanly without considering a bunch of this Status: Marc +1 on much of it; threads aren't enough for perf. - MarkC +1, Paul +1, Dean +1 + MarkC +1, Paul +1, Dean +1, Martin +1 Volunteers: * new layered I/O. @@ -82,7 +83,9 @@ can get zero-copy in the user space, which is a big win for caching modules of all sorts. You can also support the traditional slow style of stdio, which adds an extra user space copy.) - Status: Dean +1, Marc +1, Ben +1, Paul +1 + Martin asks: Is there some software flying around where such + a model has been tried? Or is it a totally new technique? + Status: Dean +1, Marc +1, Ben +1, Paul +1, Martin +1 Volunteers: * API work @@ -92,7 +95,8 @@ Volunteers: Ken . documented API - Status: Ken +1, Sameer +1, Marc +1, Ralf +1, Paul +1, Dean +1 + Status: Ken +1, Sameer +1, Marc +1, Ralf +1, Paul +1, Dean +1, + Martin +1 Volunteers: Ken . just new API phases @@ -121,20 +125,22 @@ . make API call syntax rational (e.g., all r*() routines list r as their first argument, et cetera) - Status: Ken +1, Ralf +0, Paul +0, Dean +0 + Status: Ken +1, Ralf +0, Paul +0, Dean +0, Martin +0 Volunteers: Ken . abstract module layering for plugins (e.g., a mod_auth interface into which mod_auth_mumble modules can be plugged) - Status: Ken +1 + Status: Ken +1, Martin +1 Volunteers: Ken * new configuration language - Status: Dean +1, Marc +1, Ralf +0, Ben +1, Paul +0 + Martin notes: There have been proposals to maybe make the config + language XML-based. + Status: Dean +1, Marc +1, Ralf +0, Ben +1, Paul +0, Martin +1 Volunteers: Ken * rewrite in C++ - . Yes: Ben +1 + . Yes: Ben +1, Martin +1 . doesn't like the idea, but is open to it: Marc +1, Ralf +1 . No way ever: MarkC +1, Paul +1 . Not for 2.0, but maybe later: Ken +1
cvs commit: apache-2.0 STATUS
pcs 98/02/11 07:59:45 Modified:.STATUS Log: Hey, how'd that vote get on the wrong line? Revision ChangesPath 1.23 +2 -2 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.22 retrieving revision 1.23 diff -u -r1.22 -r1.23 --- STATUS1998/02/10 18:19:27 1.22 +++ STATUS1998/02/11 15:59:44 1.23 @@ -49,7 +49,7 @@ Status: Brian +1, Ken +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Ben +1 - Thread Abstraction - Status: Sameer +1, Marc +1, MarkC +1, Ben +1, Dean +1 + Status: Sameer +1, Marc +1, MarkC +1, Ben +1, Dean +1, Paul +1 Volunteers: * revamped process model (Dean's proposal) @@ -116,7 +116,7 @@ . backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers - Status: Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Paul +1 + Status: Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1 Volunteers: . make API call syntax rational (e.g., all r*() routines list r
cvs commit: apache-2.0 STATUS
dgaudet 98/02/09 22:49:35 Modified:.STATUS Log: Revision ChangesPath 1.21 +10 -8 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.20 retrieving revision 1.21 diff -u -r1.20 -r1.21 --- STATUS1998/02/09 11:10:31 1.20 +++ STATUS1998/02/10 06:49:34 1.21 @@ -4,6 +4,8 @@ 2.0 : In pre-alpha development see: http://www.arctic.org/~dgaudet/apache/2.0/process-model +Dean says: This sounds like I'm working on implementing this +proposal. I'm not. Nobody is as far as I know. Plan: @@ -44,14 +46,14 @@ to add more, but try not to duplicate earlier items too much. * multithreading. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Ben +1 + Status: Brian +1, Ken +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Ben +1 - Thread Abstraction - Status: Sameer +1, Marc +1, MarkC +1, Ben +1 + Status: Sameer +1, Marc +1, MarkC +1, Ben +1, Dean +1 * revamped process model (Dean's proposal) Dean says: it's hard to do the multithreading work cleanly without considering a bunch of this - Status: Marc +1 on much of it; threads aren't enough for perf. MarkC +1, Paul +1 + Status: Marc +1 on much of it; threads aren't enough for perf. MarkC +1, Paul +1, Dean +1 * new layered I/O. Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Ben +1 @@ -81,7 +83,7 @@ Status: Ken +1 . documented API - Status: Ken +1, Sameer +1, Marc +1, Ralf +1, Paul +1 + Status: Ken +1, Sameer +1, Marc +1, Ralf +1, Paul +1, Dean +1 . just new API phases Status: Brian +1, Jim +1, Sameer +1 (just the gaping holes), @@ -89,13 +91,13 @@ . change API 'phase' model to use module-registered hooks rather than a fixed static structure - Status: Ken +1, Ralf +1, MarkC +1, Paul +1 + Status: Ken +1, Ralf +1, MarkC +1, Paul +1, Dean +1 - . use virtual functions for module hooks + . use virtual functions for module hooks Status: Ben +1, Paul -1 . clearly identify API functions by renaming them - Status: Ken +1, Ralf +1, Ben +1, Paul +1 (plus back compat.) + Status: Ken +1, Ralf +1, Ben +1, Paul +1 (plus back compat.), Dean +1 . backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers @@ -103,7 +105,7 @@ . make API call syntax rational (e.g., all r*() routines list r as their first argument, et cetera) - Status: Ken +1, Ralf +0, Paul +0 + Status: Ken +1, Ralf +0, Paul +0, Dean +0 * new configuration language Status: Dean +1, Marc +1, Ralf +0, Ben +1, Paul +0
cvs commit: apache-2.0 STATUS
coar98/02/10 10:19:27 Modified:.STATUS Log: Add another work area discussed in the past. Also, things are starting to get potentially overcommitted here; add an opportunity to check sanity by letting people sign up for different pieces. Items with lots of votes but no volunteers should maybe be struck from the plan, or moved to 'nice-to-have' status. Revision ChangesPath 1.22 +39 -15apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.21 retrieving revision 1.22 diff -u -r1.21 -r1.22 --- STATUS1998/02/10 06:49:34 1.21 +++ STATUS1998/02/10 18:19:27 1.22 @@ -46,75 +46,99 @@ to add more, but try not to duplicate earlier items too much. * multithreading. - Status: Brian +1, Ken +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Ben +1 - - Thread Abstraction - Status: Sameer +1, Marc +1, MarkC +1, Ben +1, Dean +1 - + Status: Brian +1, Ken +1, Jim +1, Paul +1, Sameer +1, Marc +1, + Ralf +1, MarkC +1, Ben +1 + - Thread Abstraction + Status: Sameer +1, Marc +1, MarkC +1, Ben +1, Dean +1 + Volunteers: + * revamped process model (Dean's proposal) Dean says: it's hard to do the multithreading work cleanly without considering a bunch of this - Status: Marc +1 on much of it; threads aren't enough for perf. MarkC +1, Paul +1, Dean +1 + Status: Marc +1 on much of it; threads aren't enough for perf. + MarkC +1, Paul +1, Dean +1 + Volunteers: * new layered I/O. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Ben +1 - + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, + Marc +1, Ralf +1, MarkC +1, Ben +1 + Volunteers: Ken + . sfio Status: Dean -1 until it's shown to be thread safe (RST claims it isn't) + Volunteers: . bstdio - This was written by Chris Provenzano as part of his implementation - of Posix threads... Jim can place a copy of the RST-hacked - version on dev.apache.org if needed and possible. + This was written by Chris Provenzano as part of his implementation + of Posix threads... Jim can place a copy of the RST-hacked + version on dev.apache.org if needed and possible. -- RST never donated his hacks to the Group. Don't put it up for - download unless you've cleared it with him. --Brian + download unless you've cleared it with him. --Brian . page flipping friendly, page-sized buffer oriented, zero copy I/O -(In this model there are functions like readbuf() which return a + (In this model there are functions like readbuf() which return a pointer to a buffer, rather than taking a pointer to a buffer. This is a lot like how kernels actually work. The advantage is that you can get zero-copy in the user space, which is a big win for caching modules of all sorts. You can also support the traditional slow style of stdio, which adds an extra user space copy.) Status: Dean +1, Marc +1, Ben +1, Paul +1 + Volunteers: * API work . radically revamped API Status: Ken +1 + Volunteers: Ken . documented API Status: Ken +1, Sameer +1, Marc +1, Ralf +1, Paul +1, Dean +1 - + Volunteers: Ken + . just new API phases Status: Brian +1, Jim +1, Sameer +1 (just the gaping holes), - Ralf +1 (especially url2url and file2file in addition to url2file) + Ralf +1 (especially url2url and file2file in addition to + url2file) + Volunteers: Ken . change API 'phase' model to use module-registered hooks rather than a fixed static structure Status: Ken +1, Ralf +1, MarkC +1, Paul +1, Dean +1 + Volunteers: Ken . use virtual functions for module hooks Status: Ben +1, Paul -1 + Volunteers: . clearly identify API functions by renaming them Status: Ken +1, Ralf +1, Ben +1, Paul +1 (plus back compat.), Dean +1 + Volunteers: Ken . backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers Status: Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Paul +1 + Volunteers: . make API call syntax rational (e.g., all r*() routines list r as their first argument, et cetera) Status: Ken +1, Ralf +0, Paul +0, Dean +0 + Volunteers: Ken + + . abstract module layering for plugins (e.g., a mod_auth interface
cvs commit: apache-2.0 STATUS
pcs 98/02/09 03:10:32 Modified:.STATUS Log: What a surprise more votes! Revision ChangesPath 1.20 +11 -11apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.19 retrieving revision 1.20 diff -u -r1.19 -r1.20 --- STATUS1998/02/08 12:05:54 1.19 +++ STATUS1998/02/09 11:10:31 1.20 @@ -30,7 +30,7 @@ varies from the above such that if it's shown that the autoconf toolset can do what we want, with less headache than what we have, then we go for it) - Status: Jim +1, Ken +1, Marc +1, MarkC +1, Ben +1 + Status: Jim +1, Ken +1, Marc +1, MarkC +1, Ben +1, Paul +1 * The autoconf toolset should include all three: autoconf, automake, and libtool. @@ -51,7 +51,7 @@ * revamped process model (Dean's proposal) Dean says: it's hard to do the multithreading work cleanly without considering a bunch of this - Status: Marc +1 on much of it; threads aren't enough for perf. MarkC +1 + Status: Marc +1 on much of it; threads aren't enough for perf. MarkC +1, Paul +1 * new layered I/O. Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Ben +1 @@ -73,7 +73,7 @@ can get zero-copy in the user space, which is a big win for caching modules of all sorts. You can also support the traditional slow style of stdio, which adds an extra user space copy.) - Status: Dean +1, Marc +1, Ben +1 + Status: Dean +1, Marc +1, Ben +1, Paul +1 * API work @@ -81,7 +81,7 @@ Status: Ken +1 . documented API - Status: Ken +1, Sameer +1, Marc +1, Ralf +1 + Status: Ken +1, Sameer +1, Marc +1, Ralf +1, Paul +1 . just new API phases Status: Brian +1, Jim +1, Sameer +1 (just the gaping holes), @@ -89,29 +89,29 @@ . change API 'phase' model to use module-registered hooks rather than a fixed static structure - Status: Ken +1, Ralf +1, MarkC +1 + Status: Ken +1, Ralf +1, MarkC +1, Paul +1 . use virtual functions for module hooks - Status: Ben +1 + Status: Ben +1, Paul -1 . clearly identify API functions by renaming them - Status: Ken +1, Ralf +1, Ben +1 + Status: Ken +1, Ralf +1, Ben +1, Paul +1 (plus back compat.) . backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers - Status: Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1 + Status: Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Paul +1 . make API call syntax rational (e.g., all r*() routines list r as their first argument, et cetera) - Status: Ken +1, Ralf +0 + Status: Ken +1, Ralf +0, Paul +0 * new configuration language - Status: Dean +1, Marc +1, Ralf +0, Ben +1 + Status: Dean +1, Marc +1, Ralf +0, Ben +1, Paul +0 * rewrite in C++ . Yes: Ben +1 . doesn't like the idea, but is open to it: Marc +1, Ralf +1 - . No way ever: MarkC +1 + . No way ever: MarkC +1, Paul +1 Closed issues:
cvs commit: apache-2.0 STATUS
mjc 98/02/08 02:08:18 Modified:.STATUS Log: My feelings. Multi-fibre options would be good imho. Revision ChangesPath 1.18 +8 -8 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.17 retrieving revision 1.18 diff -u -r1.17 -r1.18 --- STATUS1998/02/07 20:12:45 1.17 +++ STATUS1998/02/08 10:08:17 1.18 @@ -30,7 +30,7 @@ varies from the above such that if it's shown that the autoconf toolset can do what we want, with less headache than what we have, then we go for it) - Status: Jim +1, Ken +1, Marc +1 + Status: Jim +1, Ken +1, Marc +1, MarkC +1 * The autoconf toolset should include all three: autoconf, automake, and libtool. @@ -44,17 +44,17 @@ to add more, but try not to duplicate earlier items too much. * multithreading. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1 + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1 - Thread Abstraction - Status: Sameer +1, Marc +1 + Status: Sameer +1, Marc +1, MarkC +1 * revamped process model (Dean's proposal) Dean says: it's hard to do the multithreading work cleanly without considering a bunch of this - Status: Marc +1 on much of it; threads aren't enough for perf. + Status: Marc +1 on much of it; threads aren't enough for perf. MarkC +1 * new layered I/O. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1 + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1 . sfio Status: Dean -1 until it's shown to be thread safe (RST claims it isn't) @@ -89,14 +89,14 @@ . change API 'phase' model to use module-registered hooks rather than a fixed static structure - Status: Ken +1, Ralf +1 + Status: Ken +1, Ralf +1, MarkC +1 . clearly identify API functions by renaming them Status: Ken +1, Ralf +1 . backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers - Status: Paul +1, Sameer +1, Marc +1, Ralf +1 + Status: Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1 . make API call syntax rational (e.g., all r*() routines list r as their first argument, et cetera) @@ -108,7 +108,7 @@ * rewrite in C++ . Yes: . doesn't like the idea, but is open to it: Marc +1, Ralf +1 - . No way ever: + . No way ever: MarkC +1 Closed issues:
cvs commit: apache-2.0 STATUS
ben 98/02/08 04:05:55 Modified:.STATUS Log: Revision ChangesPath 1.19 +11 -8 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.18 retrieving revision 1.19 diff -u -r1.18 -r1.19 --- STATUS1998/02/08 10:08:17 1.18 +++ STATUS1998/02/08 12:05:54 1.19 @@ -30,7 +30,7 @@ varies from the above such that if it's shown that the autoconf toolset can do what we want, with less headache than what we have, then we go for it) - Status: Jim +1, Ken +1, Marc +1, MarkC +1 + Status: Jim +1, Ken +1, Marc +1, MarkC +1, Ben +1 * The autoconf toolset should include all three: autoconf, automake, and libtool. @@ -44,9 +44,9 @@ to add more, but try not to duplicate earlier items too much. * multithreading. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1 + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Ben +1 - Thread Abstraction - Status: Sameer +1, Marc +1, MarkC +1 + Status: Sameer +1, Marc +1, MarkC +1, Ben +1 * revamped process model (Dean's proposal) Dean says: it's hard to do the multithreading work cleanly without @@ -54,7 +54,7 @@ Status: Marc +1 on much of it; threads aren't enough for perf. MarkC +1 * new layered I/O. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1 + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1, MarkC +1, Ben +1 . sfio Status: Dean -1 until it's shown to be thread safe (RST claims it isn't) @@ -73,7 +73,7 @@ can get zero-copy in the user space, which is a big win for caching modules of all sorts. You can also support the traditional slow style of stdio, which adds an extra user space copy.) - Status: Dean +1, Marc +1 + Status: Dean +1, Marc +1, Ben +1 * API work @@ -91,8 +91,11 @@ than a fixed static structure Status: Ken +1, Ralf +1, MarkC +1 + . use virtual functions for module hooks + Status: Ben +1 + . clearly identify API functions by renaming them - Status: Ken +1, Ralf +1 + Status: Ken +1, Ralf +1, Ben +1 . backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers @@ -103,10 +106,10 @@ Status: Ken +1, Ralf +0 * new configuration language - Status: Dean +1, Marc +1, Ralf +0 + Status: Dean +1, Marc +1, Ralf +0, Ben +1 * rewrite in C++ - . Yes: + . Yes: Ben +1 . doesn't like the idea, but is open to it: Marc +1, Ralf +1 . No way ever: MarkC +1
cvs commit: apache-2.0 STATUS
marc98/02/06 17:17:03 Modified:.STATUS Log: I'm glad I'm not in California. Here El Nino gives nice weather. Revision ChangesPath 1.14 +13 -10apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.13 retrieving revision 1.14 diff -u -r1.13 -r1.14 --- STATUS1998/02/06 19:59:47 1.13 +++ STATUS1998/02/07 01:17:03 1.14 @@ -30,7 +30,7 @@ varies from the above such that if it's shown that the autoconf toolset can do what we want, with less headache than what we have, then we go for it) - Status: Jim +1, Ken +1 + Status: Jim +1, Ken +1, Marc +1 * The autoconf toolset should include all three: autoconf, automake, and libtool. @@ -44,17 +44,17 @@ to add more, but try not to duplicate earlier items too much. * multithreading. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1 + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1 - Thread Abstraction - Status: Sameer +1 + Status: Sameer +1, Marc +1 * revamped process model (Dean's proposal) Dean says: it's hard to do the multithreading work cleanly without considering a bunch of this - Status: + Status: Marc +1 on much of it; threads aren't enough for perf. * new layered I/O. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1 + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1 . sfio Status: Dean -1 until it's shown to be thread safe (RST claims it isn't) @@ -69,7 +69,7 @@ can get zero-copy in the user space, which is a big win for caching modules of all sorts. You can also support the traditional slow style of stdio, which adds an extra user space copy.) - Status: Dean +1 + Status: Dean +1, Marc +1 * API work @@ -77,7 +77,7 @@ Status: Ken +1 . documented API - Status: Ken +1, Sameer +1 + Status: Ken +1, Sameer +1, Marc +1 . just new API phases Status: Brian +1, Jim +1, Sameer +1 (just the gaping holes) @@ -91,17 +91,20 @@ . backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers - Status: Paul +1, Sameer +1 + Status: Paul +1, Sameer +1, Marc +1 . make API call syntax rational (e.g., all r*() routines list r as their first argument, et cetera) Status: Ken +1 * new configuration language - Status: Dean +1 + Status: Dean +1, Marc +1 * rewrite in C++ - Status: + . Yes: + . doesn't like the idea, but is open to it: Marc +1 + . No way ever: + Closed issues:
cvs commit: apache-2.0 STATUS
jim 98/02/06 19:30:25 Modified:.STATUS Log: Just some bstdio stuff Revision ChangesPath 1.15 +3 -1 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.14 retrieving revision 1.15 diff -u -r1.14 -r1.15 --- STATUS1998/02/07 01:17:03 1.14 +++ STATUS1998/02/07 03:30:24 1.15 @@ -60,7 +60,9 @@ Status: Dean -1 until it's shown to be thread safe (RST claims it isn't) . bstdio - url please? + This was written by Chris Provenzano as part of his implementation + of Posix threads... Jim can place a copy of the RST-hacked + version on dev.apache.org if needed and possible. . page flipping friendly, page-sized buffer oriented, zero copy I/O (In this model there are functions like readbuf() which return a
cvs commit: apache-2.0 STATUS
brian 98/02/06 19:34:38 Modified:.STATUS Log: Revision ChangesPath 1.16 +2 -0 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.15 retrieving revision 1.16 diff -u -r1.15 -r1.16 --- STATUS1998/02/07 03:30:24 1.15 +++ STATUS1998/02/07 03:34:37 1.16 @@ -63,6 +63,8 @@ This was written by Chris Provenzano as part of his implementation of Posix threads... Jim can place a copy of the RST-hacked version on dev.apache.org if needed and possible. +-- RST never donated his hacks to the Group. Don't put it up for + download unless you've cleared it with him. --Brian . page flipping friendly, page-sized buffer oriented, zero copy I/O (In this model there are functions like readbuf() which return a
cvs commit: apache-2.0 STATUS
rse 98/02/07 12:12:46 Modified:.STATUS Log: just my $0.02 according to Apache 2.0 Revision ChangesPath 1.17 +13 -12apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.16 retrieving revision 1.17 diff -u -r1.16 -r1.17 --- STATUS1998/02/07 03:34:37 1.16 +++ STATUS1998/02/07 20:12:45 1.17 @@ -23,7 +23,7 @@ * Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset. - Status: Brian +1, Roy +1, Dean +1 + Status: Brian +1, Roy +1, Dean +1, Ralf +1 * Investigate replacing the current Unix compilation model (Configuration.tmpl home-brew Configure script) with the autoconf toolset. (this @@ -34,7 +34,7 @@ * The autoconf toolset should include all three: autoconf, automake, and libtool. - Status: Brian +1, Jim +1, Roy +1, Dean +1, Ken +1 + Status: Brian +1, Jim +1, Roy +1, Dean +1, Ken +1, Ralf +1 FEATURE SET FOR 2.0 Here, we decide how many of the following feature ideas we will set for @@ -44,7 +44,7 @@ to add more, but try not to duplicate earlier items too much. * multithreading. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1 + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1 - Thread Abstraction Status: Sameer +1, Marc +1 @@ -54,7 +54,7 @@ Status: Marc +1 on much of it; threads aren't enough for perf. * new layered I/O. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1 + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1, Marc +1, Ralf +1 . sfio Status: Dean -1 until it's shown to be thread safe (RST claims it isn't) @@ -81,32 +81,33 @@ Status: Ken +1 . documented API - Status: Ken +1, Sameer +1, Marc +1 + Status: Ken +1, Sameer +1, Marc +1, Ralf +1 . just new API phases - Status: Brian +1, Jim +1, Sameer +1 (just the gaping holes) + Status: Brian +1, Jim +1, Sameer +1 (just the gaping holes), + Ralf +1 (especially url2url and file2file in addition to url2file) . change API 'phase' model to use module-registered hooks rather than a fixed static structure - Status: Ken +1 + Status: Ken +1, Ralf +1 . clearly identify API functions by renaming them - Status: Ken +1 + Status: Ken +1, Ralf +1 . backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers - Status: Paul +1, Sameer +1, Marc +1 + Status: Paul +1, Sameer +1, Marc +1, Ralf +1 . make API call syntax rational (e.g., all r*() routines list r as their first argument, et cetera) - Status: Ken +1 + Status: Ken +1, Ralf +0 * new configuration language - Status: Dean +1, Marc +1 + Status: Dean +1, Marc +1, Ralf +0 * rewrite in C++ . Yes: - . doesn't like the idea, but is open to it: Marc +1 + . doesn't like the idea, but is open to it: Marc +1, Ralf +1 . No way ever:
cvs commit: apache-2.0 STATUS
coar98/02/05 18:58:40 Modified:.STATUS Log: Tra la la.. Revision ChangesPath 1.7 +10 -9 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.6 retrieving revision 1.7 diff -u -r1.6 -r1.7 --- STATUS1998/02/06 02:35:12 1.6 +++ STATUS1998/02/06 02:58:40 1.7 @@ -21,35 +21,36 @@ Open issues: - Replace the current Unix compilation model (Configuration.tmpl, home-brew + * Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset. - Status: Brian +1, Roy +1, Dean +1 + Status: Brian +1, Roy +1, Dean +1 - Investigate replacing the current Unix compilation model (Configuration.tmpl + * Investigate replacing the current Unix compilation model (Configuration.tmpl home-brew Configure script) with the autoconf toolset. (this varies from the above such that if it's shown that the autoconf toolset can do what we want, with less headache than what we have, then we go for it) - Status: Jim +1 + Status: Jim +1, Ken +1 - The autoconf toolset should include all three: autoconf, automake, and + * The autoconf toolset should include all three: autoconf, automake, and libtool. - Status: Brian +1, Jim +1, Roy +1, Dean +1 + Status: Brian +1, Jim +1, Roy +1, Dean +1, Ken +1 FEATURE SET FOR 2.0 - Here, we decide how many of the following feature ideas we will set for +Here, we decide how many of the following feature ideas we will set for ourselves as work items for 2.0. We can't do everything we would want to, otherwise 2.0 will never be released. So please try and be conservative with your votes. Items in no particular order. Feel free to add more, but try not to duplicate earlier items too much. a) multithreading. - Brian +1 + Brian +1, Ken +1 b) new layered I/O. - Brian +1 + Brian +1, Ken +1 c) radically revamped API + Ken +1 d) just new API phases Brian +1
cvs commit: apache-2.0 STATUS
dgaudet 98/02/06 02:10:25 Modified:.STATUS Log: Revision ChangesPath 1.8 +4 -2 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.7 retrieving revision 1.8 diff -u -r1.7 -r1.8 --- STATUS1998/02/06 02:58:40 1.7 +++ STATUS1998/02/06 10:10:23 1.8 @@ -44,10 +44,10 @@ to add more, but try not to duplicate earlier items too much. a) multithreading. - Brian +1, Ken +1 + Brian +1, Ken +1, Dean +1 b) new layered I/O. - Brian +1, Ken +1 + Brian +1, Ken +1, Dean +1 c) radically revamped API Ken +1 @@ -60,6 +60,8 @@ f) rewrite in C++ g) revamped process model (Dean's proposal) + Dean says: it's hard to do (a) cleanly without considering a + bunch of this Closed issues:
cvs commit: apache-2.0 STATUS
jim 98/02/06 05:20:09 Modified:.STATUS Log: Hurump Revision ChangesPath 1.9 +6 -3 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.8 retrieving revision 1.9 diff -u -r1.8 -r1.9 --- STATUS1998/02/06 10:10:23 1.8 +++ STATUS1998/02/06 13:20:08 1.9 @@ -44,16 +44,19 @@ to add more, but try not to duplicate earlier items too much. a) multithreading. - Brian +1, Ken +1, Dean +1 + Brian +1, Ken +1, Dean +1, Jim +1 b) new layered I/O. - Brian +1, Ken +1, Dean +1 + Brian +1, Ken +1, Dean +1, Jim +1 + + b1) sfio + b2) bstdio c) radically revamped API Ken +1 d) just new API phases - Brian +1 + Brian +1, Jim +1 (just the gaping holes) e) new configuration language
cvs commit: apache-2.0 STATUS
pcs 98/02/06 05:49:16 Modified:.STATUS Log: Votes Revision ChangesPath 1.10 +2 -2 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.9 retrieving revision 1.10 diff -u -r1.9 -r1.10 --- STATUS1998/02/06 13:20:08 1.9 +++ STATUS1998/02/06 13:49:16 1.10 @@ -44,10 +44,10 @@ to add more, but try not to duplicate earlier items too much. a) multithreading. - Brian +1, Ken +1, Dean +1, Jim +1 + Brian +1, Ken +1, Dean +1, Jim +1, Paul +1 b) new layered I/O. - Brian +1, Ken +1, Dean +1, Jim +1 + Brian +1, Ken +1, Dean +1, Jim +1, Paul +1 b1) sfio b2) bstdio
Re: cvs commit: apache-2.0 STATUS
On 6 Feb 1998 [EMAIL PROTECTED] wrote: +c) radically revamped API + +d) just new API phases + Brian +1 Um, I'm not sure how to vote on this. Personally I'd like to see (a) the function API left pretty much as it is, (b) the phase API radically updated (a la Alexei et al's proposal -- i.e. remove the module structure from modules etc), and (c) backward compatibility with modules written for 1.3. Paul
cvs commit: apache-2.0 STATUS
coar98/02/06 07:03:22 Modified:.STATUS Log: Woo-hoo, on a roll this morning. Sunny, cloudless, and bitterly cold. Revision ChangesPath 1.11 +38 -15apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.10 retrieving revision 1.11 diff -u -r1.10 -r1.11 --- STATUS1998/02/06 13:49:16 1.10 +++ STATUS1998/02/06 15:03:21 1.11 @@ -43,28 +43,51 @@ conservative with your votes. Items in no particular order. Feel free to add more, but try not to duplicate earlier items too much. -a) multithreading. - Brian +1, Ken +1, Dean +1, Jim +1, Paul +1 +* multithreading. + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1 -b) new layered I/O. - Brian +1, Ken +1, Dean +1, Jim +1, Paul +1 +* revamped process model (Dean's proposal) + Dean says: it's hard to do the multithreading work cleanly without + considering a bunch of this + Status: + +* new layered I/O. + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1 - b1) sfio - b2) bstdio + . sfio + . bstdio + +* API work + + . radically revamped API + Status: Ken +1 -c) radically revamped API - Ken +1 + . documented API + Status: Ken +1 -d) just new API phases - Brian +1, Jim +1 (just the gaping holes) + . just new API phases + Status: Brian +1, Jim +1 (just the gaping holes) + + . change API 'phase' model to use module-registered hooks rather + than a fixed static structure + Status: Ken +1 + + . clearly identify API functions by renaming them + Status: Ken +1 + + . backward compatibility with 1.3 (just require a recompile) + if functions get renamed, old names retained as wrappers + Status: Paul +1 -e) new configuration language + . make API call syntax rational (e.g., all r*() routines list r + as their first argument, et cetera) + Status: Ken +1 -f) rewrite in C++ +* new configuration language + Status: -g) revamped process model (Dean's proposal) - Dean says: it's hard to do (a) cleanly without considering a - bunch of this +* rewrite in C++ + Status: Closed issues:
Re: cvs commit: apache-2.0 STATUS
On Fri, 6 Feb 1998, Paul Sutton wrote: On 6 Feb 1998 [EMAIL PROTECTED] wrote: +c) radically revamped API + +d) just new API phases + Brian +1 Um, I'm not sure how to vote on this. Personally I'd like to see (a) the function API left pretty much as it is, (b) the phase API radically updated (a la Alexei et al's proposal -- i.e. remove the module structure from modules etc), and (c) backward compatibility with modules written for 1.3. Um... if you remove the module structure, it is physically impossible to retain bacwards compatibility with modules written for 1.3. And I don't think it's a good idea to try. Especially not if we need to rename all our functions (see other threads). And there are a number of other parts of the Apache API that really need to be rethought (can we say configuration?) Backwards compatibility with 1.3, either for the API or the config language or anything else, has never (in the year and a half we've been discussing it) been a high priority for 2.0. Also, if, as certain threads here might be leaning, Apache 2.0 is C++-based (I'm +0 personally - I don't mind C++, but the lack of legacy support is a bit troubling), you pretty much have to throw the 1.x API completely out the window. -- Alexei Kosut [EMAIL PROTECTED] http://www.stanford.edu/~akosut/ Stanford University, Class of 2001 * Apache http://www.apache.org *
Re: cvs commit: apache-2.0 STATUS
On Fri, 6 Feb 1998, Alexei Kosut wrote: Especially not if we need to rename all our functions (see other threads). And there are a number of other parts of the Apache API that really need to be rethought (can we say configuration?) Backwards compatibility with 1.3, either for the API or the config language or anything else, has never (in the year and a half we've been discussing it) been a high priority for 2.0. Well, I think now is probably the time to at least think about how we are going to support current modules in the 2.0 framework. I'm not suggesting we should bend over backwards to make everything 100% backward compatible, but that we should value the work that has been done by third-party module authors and not disregard their efforts. Forcing radical changes on module authors could devalue them and maybe hurt the prospects of them continuing to develop Apache API modules (and move them towards, say, ISAPI where no-compilation is necessary to use later versions of the server). Paul
cvs commit: apache-2.0 STATUS
sameer 98/02/06 11:59:48 Modified:.STATUS Log: some rain Revision ChangesPath 1.13 +7 -5 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.12 retrieving revision 1.13 diff -c -u -r1.12 -r1.13 /usr/bin/diff: conflicting specifications of output style --- STATUS1998/02/06 19:54:19 1.12 +++ STATUS1998/02/06 19:59:47 1.13 @@ -44,7 +44,9 @@ to add more, but try not to duplicate earlier items too much. * multithreading. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1 + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1 + - Thread Abstraction + Status: Sameer +1 * revamped process model (Dean's proposal) Dean says: it's hard to do the multithreading work cleanly without @@ -52,7 +54,7 @@ Status: * new layered I/O. - Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1 + Status: Brian +1, Ken +1, Dean +1, Jim +1, Paul +1, Sameer +1 . sfio Status: Dean -1 until it's shown to be thread safe (RST claims it isn't) @@ -75,10 +77,10 @@ Status: Ken +1 . documented API - Status: Ken +1 + Status: Ken +1, Sameer +1 . just new API phases - Status: Brian +1, Jim +1 (just the gaping holes) + Status: Brian +1, Jim +1, Sameer +1 (just the gaping holes) . change API 'phase' model to use module-registered hooks rather than a fixed static structure @@ -89,7 +91,7 @@ . backward compatibility with 1.3 (just require a recompile) if functions get renamed, old names retained as wrappers - Status: Paul +1 + Status: Paul +1, Sameer +1 . make API call syntax rational (e.g., all r*() routines list r as their first argument, et cetera)
cvs commit: apache-2.0 STATUS
fielding98/02/05 11:23:29 Modified:.STATUS Log: 61 degress and partly sunny Revision ChangesPath 1.5 +2 -2 apache-2.0/STATUS Index: STATUS === RCS file: /home/cvs/apache-2.0/STATUS,v retrieving revision 1.4 retrieving revision 1.5 diff -u -r1.4 -r1.5 --- STATUS1998/02/04 03:26:29 1.4 +++ STATUS1998/02/05 19:23:28 1.5 @@ -23,7 +23,7 @@ Replace the current Unix compilation model (Configuration.tmpl, home-brew Configure script) with the autoconf toolset. - Status: Brian +1 + Status: Brian +1, Roy +1 Investigate replacing the current Unix compilation model (Configuration.tmpl home-brew Configure script) with the autoconf toolset. (this @@ -34,7 +34,7 @@ The autoconf toolset should include all three: autoconf, automake, and libtool. - Status: Brian +1, Jim +1 + Status: Brian +1, Jim +1, Roy +1 Closed issues:
cvs commit: apache-2.0 STATUS
brian 98/02/03 19:01:54 Modified:.STATUS Log: The quiet before the storm Revision ChangesPath 1.2 +9 -0 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.1 retrieving revision 1.2 diff -u -r1.1 -r1.2 --- STATUS1998/02/01 02:11:45 1.1 +++ STATUS1998/02/04 03:01:53 1.2 @@ -20,5 +20,14 @@ Needs patch: Open issues: + + Replace the current Unix compilation model (Configuration.tmpl, home-brew +Configure script) with the autoconf toolset. + Status: Brian +1 + + The autoconf toolset should include all three: autoconf, automake, and +libtool. + Status: Brian +1 + Closed issues:
cvs commit: apache-2.0 STATUS
jim 98/02/03 19:26:30 Modified:.STATUS Log: explain my pov Revision ChangesPath 1.4 +5 -2 apache-2.0/STATUS Index: STATUS === RCS file: /export/home/cvs/apache-2.0/STATUS,v retrieving revision 1.3 retrieving revision 1.4 diff -u -r1.3 -r1.4 --- STATUS1998/02/04 03:23:34 1.3 +++ STATUS1998/02/04 03:26:29 1.4 @@ -26,12 +26,15 @@ Status: Brian +1 Investigate replacing the current Unix compilation model (Configuration.tmpl -home-brew Configure script) with the autoconf toolset. +home-brew Configure script) with the autoconf toolset. (this +varies from the above such that if it's shown that the autoconf +toolset can do what we want, with less headache than what we +have, then we go for it) Status: Jim +1 The autoconf toolset should include all three: autoconf, automake, and libtool. - Status: Brian +1 + Status: Brian +1, Jim +1 Closed issues:
cvs commit: apache-2.0 STATUS
fielding98/01/31 18:11:46 Added: .STATUS Log: A new beginning. Revision ChangesPath 1.1 apache-2.0/STATUS Index: STATUS === Apache 2.0 STATUS: Release: 2.0 : In pre-alpha development see: http://www.arctic.org/~dgaudet/apache/2.0/process-model Plan: Everyone with plans on things they want to do for 2.0 should add them to the repository now. Use a descriptive filename. Other code will be copied over when 1.3.0 is finished. Showstoppers: Committed Code Changes: Available Patches: In progress: Needs patch: Open issues: Closed issues:
