2 Last modified at [$Date: 2000/03/11 05:42:50 $]
6 Vague plan, assuming more features aren't demanded for the 2.0 release
7 2.0a1 : released March 10, 2000
16 * MPMs should have responsibility for closing sockets and
17 registering the socket cleanup. Stop registering multiple
18 cleanups for each socket. Move socket close code out of
19 http_connection.c and into the MPMs.
22 * Reliable piped logs look broken everywhere. Each MPM includes essentially
23 identical code to ap_register_other_child(), etc. Most of this code can
24 be moved out of the MPMs and into some common file (http_core.c?).
25 Dean says presumably you mean an os-specific file?
27 * DSO hooks registered during the preflight call to ap_read_config
28 are not 'unregistered' during the preflight clean-up. Thus, when
29 Apache comes up, the hook tables contain two entries for each DSO hook,
30 one of which is bogus, leftover from the preflight.
32 * Pipes to CGI scripts are not being timed out
33 Status: code has been added to APR to support timing out pipes.
34 This needs to be used in Apache now.
36 * Put back resource limit code
39 Status: Manoj has posted an patch to fix this.
40 <19991103003605.A20612@samosa.mindspring.com>
42 * Win32: Enable the Windows MPM to honor max_requests_per_child
43 Status: Bill will fix this.
45 * Win32: Fix Win9* specific code in the winnt MPM
48 * Win32: Test access logging with multiple threads. Will the
49 native file I/O calls serialize automagically like the
50 CRT calls or do we need to add region locking each time
54 * Win32: Smoke test all ported modules to make sure they work
55 correctly under Windows.
58 * Current 2.0 code is not tested on many Unix platforms. Make 2.0
59 work on most, if not all the systems 1.3 did
60 Status: Autoconf and APR will get us most of the way there.
62 * The handler hook needs to use the new registration system
64 * Modules. Which modules do not work yet, can we get a list?
66 * We need a thread-safe resolver, at least on Unix.
67 Status: The best known candidate would be something from BIND
68 (v8 or v9?) The only other option would be to mutex all the
71 RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
73 * "make distclean" leaves (at least these) turds:
76 lib/apr/shmem/unix/mm/config.log
77 lib/apr/shmem/unix/mm/mm_conf.h
78 lib/apr/shmem/unix/mm/libtool
79 lib/apr/shmem/unix/mm/config.status
80 lib/apr/shmem/unix/mm/Makefile
81 lib/apr/shmem/unix/mm/mm-config
87 * apaci permissions problem:
88 cvs checkout apache-2.0
93 now apaci is no longer executable and future makes fail
95 * the top-level configure.in appears to use $USE_MAINTAINER_MODE = yes
96 to set -Wall/etc. flags, and the APR configure.in appears to use
97 --with-debug. both seem like more of a hardwired pain compaired
98 to the pre-2.0 method of setting EXTRA_CFLAGS.
100 * Win32: Reuse accept socket after transmitfile/close
101 This is not a bug, but would be nice to get this feature in
104 * Win32: Enable the winnt MPM to use the new scoreboard API
106 * Win32: Non-blocking CGI pipes
108 * Win32: Graceful restart of a service is broken (code does shutdown then start)
110 * Dynamically loadable MPMs
111 Enable users of the binary distributions of Apache to be able to
112 dynamically load the MPM.
114 Dean says this probably isn't a good idea. For example consider
115 -lpthread, you couldn't have the same core run with pthreads,
116 prefork, and gnu pth without a lot of headaches is my guess.
118 * Platforms that do not support fork (primarily Win32 and AS/400)
119 Consider introducing HAVE_FORK feature macro. Architect start-up code
120 that avoids initializing all the modules in the parent process on
121 platforms that do not support fork.
123 * Clean the code. There are a lot of places we used APR but didn't
124 remove the hacks that were required for the cross-platform code in
125 1.3. We need to make the code look like APR was supposed to be there.
127 * Go throught the 1.3 Bug DB and research the bugs marked "suspended".
128 People were told these would be considered for inclusion in Apache 2.0,
129 it would be nice to actually do so.
131 * Win32: Migrate the MPM over to use APR thread/process calls. This
132 would eliminate some code in the Win32 branch that essentially
133 duplicates what is in APR.
134 Status: Bill <stoddard@raleigh.ibm.com> is working on this.
136 * Move I/O layering into APR.
138 * There are still a number of places in the code where we are
139 loosing error status (i.e. throwing away the error returned by a
140 system call and replacing it with a generic error code)
142 * Win32: Implement reliable piped logs on Windows
143 Status: Bill <stoddard@raleigh.ibm.com> has prototype code (not
144 reliable) in 1.3. Should be much cleaner with APR in 2.0.
146 * Switch to autoconf, et al. for configuration.
147 Status: Manoj has placed an initial version into the 2.0
148 repository. Todos include:
150 - "make install" should work
151 - a tool to simplfy third-party module building should be
152 written. Something like apxs
153 - Add a lot more checks to satisfy the various platforms, e.g.
156 * Use APR to get rid of more platform dependancies.
157 Status: Ryan Bloom <rbb@raleigh.ibm.com> is working on this.
159 * The connection status table is not very efficient. Also, very few stats
160 are exported to the connection status table (easy to fix), and mod_status
163 Other bugs that need fixing:
165 * MaxRequestsPerChild measures connections, not requests.
166 Until someone has a better way, we'll probably just rename it
167 "MaxConnectionsPerChild".
169 * Regex containers don't work in an intutive way
170 Status: No one has come up with an efficient way to fix this
171 behavior. Dean has suggested getting rid of regex containers
174 * SIGSEGV on Linux (glibc 2.1.2) isn't caught properly by a
175 sigwaiting thread. We need to work around this, perhaps unless
176 there is hope soon for a fixed glibc.
178 Other features that need writing:
180 * Finish infrastructure in core for async MPMs
183 * TODO in source -- just do an egrep on "TODO" and see what's there
185 Documentation that needs writing:
187 * The concept of MPMs, especially if we ship more than one MPM for a
190 * New directives in the various MPMs
193 Status: Ben Laurie has written some hooks documentation
194 (apache-2.0/htdocs/hooks.html)
196 * Changes since 1.3.9 can be more easily seen in the commitlog file
197 dev.apache.org:/home/cvs/CVSROOT/commitlogs/apache-2.0
198 which includes some of Roy's comments when the changes were
199 committed in rough change-sets by purpose. Note that the commitlog
200 does not show the contents of new files until later.
204 * Mike Abbott's <mja@trudge.engr.sgi.com> patches to improve
206 Status: These were written for 1.3, and are awaiting a port to
211 * What do we do about mod_proxy?
213 * Which MPMs will be included with Apache 2.0?
215 * Is conf/highperformance.conf-dist obsolete? It looks obsolete.