2 Last modified at [$Date: 2000/03/08 21:36:39 $]
6 Vague plan, assuming more features aren't demanded for the 2.0 release
21 * MPMs should have responsibility for closing sockets and
22 registering the socket cleanup. Stop registering multiple
23 cleanups for each socket. Move socket close code out of
24 http_connection.c and into the MPMs.
27 * Reliable piped logs look broken everywhere. Each MPM includes essentially
28 identical code to ap_register_other_child(), etc. Most of this code can
29 be moved out of the MPMs and into some common file (http_core.c?).
30 Dean says presumably you mean an os-specific file?
32 * DSO hooks registered during the preflight call to ap_read_config
33 are not 'unregistered' during the preflight clean-up. Thus, when
34 Apache comes up, the hook tables contain two entries for each DSO hook,
35 one of which is bogus, leftover from the preflight.
37 * Pipes to CGI scripts are not being timed out
38 Status: code has been added to APR to support timing out pipes.
39 This needs to be used in Apache now.
41 * Put back resource limit code
44 Status: Manoj has posted an patch to fix this.
45 <19991103003605.A20612@samosa.mindspring.com>
47 * Win32: Enable the Windows MPM to honor max_requests_per_child
48 Status: Bill will fix this.
50 * Win32: Fix Win9* specific code in the winnt MPM
53 * Win32: Test access logging with multiple threads. Will the
54 native file I/O calls serialize automagically like the
55 CRT calls or do we need to add region locking each time
59 * Win32: Smoke test all ported modules to make sure they work
60 correctly under Windows.
63 * Current 2.0 code is not tested on many Unix platforms. Make 2.0
64 work on most, if not all the systems 1.3 did
65 Status: Autoconf and APR will get us most of the way there.
67 * The module API is a weird combination of the old table and the new
68 hook system. Switch completely to the new hook system
69 Status: Ben Laurie is working on this.
71 * Modules. Which modules do not work yet, can we get a list?
73 * We need a thread-safe resolver, at least on Unix.
74 Status: The best known candidate would be something from BIND
75 (v8 or v9?) The only other option would be to mutex all the
78 RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
80 * "make distclean" leaves (at least these) turds:
83 lib/apr/shmem/unix/mm/config.log
84 lib/apr/shmem/unix/mm/mm_conf.h
85 lib/apr/shmem/unix/mm/libtool
86 lib/apr/shmem/unix/mm/config.status
87 lib/apr/shmem/unix/mm/Makefile
88 lib/apr/shmem/unix/mm/mm-config
94 * apaci permissions problem:
95 cvs checkout apache-2.0
100 now apaci is no longer executable and future makes fail
102 * the top-level configure.in appears to use $USE_MAINTAINER_MODE = yes
103 to set -Wall/etc. flags, and the APR configure.in appears to use
104 --with-debug. both seem like more of a hardwired pain compaired
105 to the pre-2.0 method of setting EXTRA_CFLAGS.
107 * Win32: Reuse accept socket after transmitfile/close
108 This is not a bug, but would be nice to get this feature in
111 * Win32: Enable the winnt MPM to use the new scoreboard API
113 * Win32: Non-blocking CGI pipes
115 * Win32: Graceful restart of a service is broken (code does shutdown then start)
117 * Dynamically loadable MPMs
118 Enable users of the binary distributions of Apache to be able to
119 dynamically load the MPM.
121 Dean says this probably isn't a good idea. For example consider
122 -lpthread, you couldn't have the same core run with pthreads,
123 prefork, and gnu pth without a lot of headaches is my guess.
125 * Platforms that do not support fork (primarily Win32 and AS/400)
126 Consider introducing HAVE_FORK feature macro. Architect start-up code
127 that avoids initializing all the modules in the parent process on
128 platforms that do not support fork.
130 * Clean the code. There are a lot of places we used APR but didn't
131 remove the hacks that were required for the cross-platform code in
132 1.3. We need to make the code look like APR was supposed to be there.
134 * Go throught the Bug DB and research the bugs marked "suspended". People
135 were told these would be considered for inclusion in Apache 2.0, it
136 would be nice to actually do so.
138 * Win32: Migrate the MPM over to use APR thread/process calls. This
139 would eliminate some code in the Win32 branch that essentially
140 duplicates what is in APR.
141 Status: Bill <stoddard@raleigh.ibm.com> is working on this.
143 * Move I/O layering into APR.
145 * There are still a number of places in the code where we are
146 loosing error status (i.e. throwing away the error returned by a
147 system call and replacing it with a generic error code)
149 * Win32: Implement reliable piped logs on Windows
150 Status: Bill <stoddard@raleigh.ibm.com> has prototype code (not
151 reliable) in 1.3. Should be much cleaner with APR in 2.0.
153 * Switch to autoconf, et al. for configuration.
154 Status: Manoj has placed an initial version into the 2.0
155 repository. Todos include:
157 - "make install" should work
158 - a tool to simplfy third-party module building should be
159 written. Something like apxs
160 - Add a lot more checks to satisfy the various platforms, e.g.
163 * Use APR to get rid of more platform dependancies.
164 Status: Ryan Bloom <rbb@raleigh.ibm.com> is working on this.
166 * The connection status table is not very efficient. Also, very few stats
167 are exported to the connection status table (easy to fix), and mod_status
170 Other bugs that need fixing:
172 * MaxRequestsPerChild measures connections, not requests.
173 Until someone has a better way, we'll probably just rename it
174 "MaxConnectionsPerChild".
176 * Regex containers don't work in an intutive way
177 Status: No one has come up with an efficient way to fix this
178 behavior. Dean has suggested getting rid of regex containers
181 * SIGSEGV on Linux (glibc 2.1.2) isn't caught properly by a
182 sigwaiting thread. We need to work around this, perhaps unless
183 there is hope soon for a fixed glibc.
185 Other features that need writing:
187 * Finish infrastructure in core for async MPMs
190 * TODO in source -- just do an egrep on "TODO" and see what's there
192 Documentation that needs writing:
194 * The concept of MPMs, especially if we ship more than one MPM for a
197 * New directives in the various MPMs
200 Status: Ben Laurie has written some hooks documentation
201 (apache-2.0/htdocs/hooks.html)
203 * Changes since 1.3.9 can be more easily seen in the commitlog file
204 dev.apache.org:/home/cvs/CVSROOT/commitlogs/apache-2.0
205 which includes some of Roy's comments when the changes were
206 committed in rough change-sets by purpose. Note that the commitlog
207 does not show the contents of new files until later.
211 * Mike Abbott's <mja@trudge.engr.sgi.com> patches to improve
213 Status: These were written for 1.3, and are awaiting a port to
218 * What do we do about mod_proxy?
220 * Which MPMs will be included with Apache 2.0?
222 * Is conf/highperformance.conf-dist obsolete? It looks obsolete.