2 Last modified at [$Date: 2000/02/16 16:48:32 $]
6 Vague plan, assuming more features aren't demanded for the 2.0 release
11 * DSO hooks registered during the preflight call to ap_read_config
12 are not 'unregistered' during the preflight clean-up. Thus, when
13 Apache comes up, the hook tables contain two entries for each DSO hook,
14 one of which is bogus, leftover from the preflight.
16 * Pipes to CGI scripts are not being timed out
17 Status: code has been added to APR to support timing out pipes.
18 This needs to be used in Apache now.
20 * Put back resource limit code
23 Status: Paul Reder <rederpj@raleigh.ibm.com> is working on this.
26 Status: Manoj has posted an patch to fix this.
27 <19991103003605.A20612@samosa.mindspring.com>
29 * Win32: Fix Win9* specific code in the winnt MPM
32 * Win32: Test access logging with multiple threads. Will the
33 native file I/O calls serialize automagically like the
34 CRT calls or do we need to add region locking each time
38 * Win32: Smoke test all ported modules to make sure they work
39 correctly under Windows.
42 * Current 2.0 code is not tested on many Unix platforms. Make 2.0
43 work on most, if not all the systems 1.3 did
44 Status: Autoconf and APR will get us most of the way there.
46 * The module API is a weird combination of the old table and the new
47 hook system. Switch completely to the new hook system
48 Status: Ben Laurie is working on this.
50 * Modules. Which modules do not work yet, can we get a list?
52 * We need a thread-safe resolver, at least on Unix.
53 Status: The best known candidate would be something from BIND
54 (v8 or v9?) The only other option would be to mutex all the
57 RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
59 * Win32: Non-blocking CGI pipes
61 * Win32: Graceful restart of a service is broken (code does shutdown then start)
63 * Dynamically loadable MPMs
64 Enable users of the binary distributions of Apache to be able to
65 dynamically load the MPM.
67 * Platforms that do not support fork (primarily Win32 and AS/400)
68 Consider introducing HAVE_FORK feature macro. Architect start-up code
69 that avoids initializing all the modules in the parent process on
70 platforms that do not support fork.
72 * Clean the code. There are a lot of places we used APR but didn't
73 remove the hacks that were required for the cross-platform code in
74 1.3. We need to make the code look like APR was supposed to be there.
76 * Go throught the Bug DB and research the bugs marked "suspended". People
77 were told these would be considered for inclusion in Apache 2.0, it
78 would be nice to actually do so.
80 * Win32: Migrate the MPM over to use APR thread/process calls. This
81 would eliminate some code in the Win32 branch that essentially
82 duplicates what is in APR.
83 Status: Bill <stoddard@raleigh.ibm.com> is working on this.
85 * Move I/O layering into APR.
87 * There are still a number of places in the code where we are
88 loosing error status (i.e. throwing away the error returned by a
89 system call and replacing it with a generic error code)
91 * Win32: Implement reliable piped logs on Windows
92 Status: Bill <stoddard@raleigh.ibm.com> has prototype code (not
93 reliable) in 1.3. Should be much cleaner with APR in 2.0.
95 * Switch to autoconf, et al. for configuration.
96 Status: Manoj has placed an initial version into the 2.0
97 repository. Todos include:
99 - "make install" should work
100 - a tool to simplfy third-party module building should be
101 written. Something like apxs
102 - Add a lot more checks to satisfy the various platforms, e.g.
105 * Use APR to get rid of more platform dependancies.
106 Status: Ryan Bloom <rbb@raleigh.ibm.com> is working on this.
108 * The connection status table is not very efficient. Also, very few stats
109 are exported to the connection status table (easy to fix), and mod_status
112 Other bugs that need fixing:
114 * MaxRequestsPerChild measures connections, not requests.
115 Until someone has a better way, we'll probably just rename it
116 "MaxConnectionsPerChild".
118 * Regex containers don't work in an intutive way
119 Status: No one has come up with an efficient way to fix this
120 behavior. Dean has suggested getting rid of regex containers
123 * SIGSEGV on Linux (glibc 2.1.2) isn't caught properly by a
124 sigwaiting thread. We need to work around this, perhaps unless
125 there is hope soon for a fixed glibc.
127 Other features that need writing:
129 * Finish infrastructure in core for async MPMs
132 * TODO in source -- just do an egrep on "TODO" and see what's there
134 * Work on other MPMs. Possible MPMs discussed or in progress include:
136 - Dean Gaudet's async MPM
137 Status: dean is unlikely to begin work on this any time soon...
138 has lots of ideas he'll share with folks willing to begin
139 tackling the issues. this is apache 2.1 material.
141 - Zach Brown's <zab@zabbo.net> Linux siginfo MPM
144 Documentation that needs writing:
146 * The concept of MPMs, especially if we ship more than one MPM for a
149 * New directives in the various MPMs
152 Status: Ben Laurie has written some hooks documentation
153 (apache-2.0/htdocs/hooks.html)
155 * Changes since 1.3.9 can be more easily seen in the commitlog file
156 dev.apache.org:/home/cvs/CVSROOT/commitlogs/apache-2.0
157 which includes some of Roy's comments when the changes were
158 committed in rough change-sets by purpose. Note that the commitlog
159 does not show the contents of new files until later.
163 * Mike Abbott's <mja@trudge.engr.sgi.com> patches to improve
165 Status: These were written for 1.3, and are awaiting a port to
170 * What do we do about mod_proxy?
172 * Which MPMs will be included with Apache 2.0?
174 * Is conf/highperformance.conf-dist obsolete? It looks obsolete.