1 APACHE 2.3 STATUS: -*-text-*-
2 Last modified at [$Date$]
4 The current version of this file can be found at:
6 * http://svn.apache.org/repos/asf/httpd/httpd/trunk/STATUS
8 Documentation status is maintained seperately and can be found at:
10 * docs/STATUS in this source tree, or
11 * http://svn.apache.org/repos/asf/httpd/httpd/trunk/docs/STATUS
13 Consult the following STATUS files for information on related projects:
15 * http://svn.apache.org/repos/asf/apr/apr/trunk/STATUS
16 * http://svn.apache.org/repos/asf/apr/apr-util/trunk/STATUS
18 Patches considered for backport are noted in their branches' STATUS:
20 * http://svn.apache.org/repos/asf/httpd/httpd/branches/2.0.x/STATUS
21 * http://svn.apache.org/repos/asf/httpd/httpd/branches/2.2.x/STATUS
26 [NOTE that x.{odd}.z versions are strictly Alpha/Beta releases,
27 while x.{even}.z versions are Stable/GA releases.]
29 2.3.15 : In development.
30 2.3.14 : Tagged on August 1, 2011.
31 2.3.13 : Tagged on June 28, 2011, not released.
32 2.3.12 : Tagged on May 11, 2011. Released May 23, 2011.
33 2.3.11 : Released as Beta on March 7, 2011.
34 2.3.10 : Tagged on December 13, 2010. Released Dec 21, 2010.
35 2.3.9 : Tagged on November 23, 2010, not released.
36 2.3.8 : Tagged on August 24, 2010.
37 2.3.7 : Tagged on August 19, 2010, not released.
38 2.3.6 : Released on June 21, 2010.
39 2.3.5 : Released on January 26, 2010.
40 2.3.4 : Released on December 8, 2009.
41 2.3.3 : Tagged on November 11, 2009, not released.
42 2.3.2 : Tagged on March 23, 2009, not released.
43 2.3.1 : Tagged on January 2, 2009, not released.
44 2.3.0 : Tagged on December 6, 2008, not released.
46 Contributors looking for a mission:
48 * Just do an egrep on "TODO" or "XXX" in the source.
50 * Review the bug database at: http://issues.apache.org/bugzilla/
52 * Review the "PatchAvailable" bugs in the bug database:
54 https://issues.apache.org/bugzilla/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&product=Apache+httpd-2&keywords=PatchAvailable
56 After testing, you can append a comment saying "Reviewed and tested".
58 * Open bugs in the bug database.
60 * See also the STATUS file in the docs/ directory, which lists documentation-specific TODO items.
63 CURRENT RELEASE NOTES:
68 Jim proposes final beta release the mid/end of Sept, with a
69 GA in October (at the latest). Jim volunteers to RM these.
75 * Modules that are not ready for production use must be removed.
76 The same for modules without documentation.
79 - mod_serf (which is optimal for async httpd anyways)
81 * The mod_session* modules need to be checked that their hooks respect
82 the returning of int (HTTP status codes) and apr_status_t as appropriate,
83 and any anomolies fixed.
84 jim sez: from what I can see, mod_session* is no worse that other
85 modules that mix these 2 types... clean up is
86 forthcoming but should not be considered a blocker, imo
88 wrowe asks: what's the API change required?
89 wrowe asks; why are we shipping this if it requires apr_ssl
91 * mod_ssl's proxy support only allows one proxy client certificate per
92 frontend virtual host. Lift this restriction.
93 jim sez: Why a blocker?, pgollucci +1 jim
94 wrowe asks: what's the API change required?
96 * INCLUDE mod_fcgid with 2.4.0, esp to help php users etc to enjoy
97 a painless event mpm experience.
99 * Investigate Windows AcceptFilter issues. Thread starts at
100 http://mail-archives.apache.org/mod_mbox/httpd-dev/201107.mbox/%3C84552AF751934DA6AD70C93AAEA2B553@mother%3E
102 * Get a recent beta (2.3.14?) installed on apache.org machines for testing.
107 OLD ISSUES THAT WERE THOUGHT TO BE SHOWSTOPPERS FOR 2.2 BUT OBVIOUSLY WEREN'T:
109 * Handling of non-trailing / config by non-default handler is broken
110 http://marc.theaimsgroup.com/?l=apache-httpd-dev&m=105451701628081&w=2
111 jerenkrantz asks: Why should this block a release?
112 wsanchez agrees: this may be a change in behavior, but isn't
113 clearly wrong, and even if so, it doesn't seem like a
116 * the edge connection filter cannot be removed
117 http://marc.theaimsgroup.com/?l=apache-httpd-dev&m=105366252619530&w=2
119 jerenkrantz asks: Why should this block a release?
121 stas replies: because it requires a rewrite of the filters stack
122 implementation (you have suggested that) and once 2.2 is
123 released you can't do that anymore.
125 pgollucci: this affects mod_perl I'm pretty sure.
129 * Name the Server (version 2.4 or 3.0, depending on the final call)
130 Recent discussion indicates we should designate a (short name).
131 This is not yet a [Vote] - Your nominations please:
132 * Apache HTTP Server (httpd)
133 +1: sctemme (why mess with it?), pgollucci
135 RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
137 * Clean up all the kruft and *extremely* outdated stuff below...
139 * Maybe remove Limit/LimitExcept or at least make it log warnings when
142 * Patches submitted to the bug database:
143 http://issues.apache.org/bugzilla/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&product=Apache+httpd-2&keywords=PatchAvailable
145 * Filter stacks and subrequests, redirects and fast redirects.
146 There's at least one PR that suffers from the current unclean behaviour
147 (which lets the server send garbage): PR 17629
148 nd says: Every subrequest should get its own filter stack with the
149 subreq_core filter as bottom-most. That filter does two things:
150 - swallow EOS buckets
151 - redirect the data stream to the upper request's (rr->main)
152 filter chain directly after the subrequest's starting
154 Once we have a clean solution, we can try to optimize
155 it, so that the server won't be slow down too much.
157 * RFC 2616 violations.
158 Closed PRs: 15852, 15857, 15859, 15861, 15864, 15869, 15870, 16120,
159 16125, 16135, 16136, 16137, 16138, 16139, 16140, 16518,
161 Open PRs: 15865, 15866, 15868, 16126, 16133, 16142, 16521, 42978
162 jerenkrantz says: need to decide how many we need to backport and/or
163 if these rise to showstopper status.
164 wrowe suggests: it would be nice to see "MUST" v.s. "SHOULD" v.s. "MAY"
165 out of this list, without reviewing them individually.
167 * pipes deadlock on all platforms with limited pipe buffers (e.g. both
168 Linux and Win32, as opposed to only Win32 on 1.3). The right solution
169 is either GStein's proposal for a "CGI Brigade", or OtherBill's proposal
170 for "Poll Buckets" for "Polling Filter Chains". Or maybe both :-)
172 * All handlers should always send content down even if r->header_only
173 is set. If not, it means that the HEAD requests don't generate the
174 same headers as a GET which is wrong.
176 * exec cmd and suexec arg-passing enhancements
177 Status: Patches proposed
178 Message-ID: <20020526041748.A29148@prodigy.Redbrick.DCU.IE>
179 (see the "proc.patch" and "suexec-shell.patch" links in this message)
181 * The 2.0.36 worker MPM graceless shutdown changes work but are
182 a bit clunky on some platforms; eg, on Linux, the loop to
183 join each worker thread seems to hang, and the parent ends up
184 killing off the child with SIGKILL. But at least it shuts down.
186 chrisd: Has this been fixed by the changes for PR 38737?
188 * We do not properly substitute the prefix-variables in the configuration
189 scripts or generated-configs. (i.e. if sysconfdir is etc,
190 httpd-std.conf points to conf.)
192 * If any request gets through ap_process_request_internal() and is
193 scheduled to be served by the core handler, without a flag that this
194 r->filename was tested by dir/file_walk, we need to 500 at the very
195 end of the ap_process_request_internal() processing so sub_req-esters
196 know this request cannot be run. This provides authors of older
197 modules better compatibility, while still improving the security and
200 Status: still need to decide where this goes, OtherBill comments...
201 Message-ID: <065701c14526$495203b0$96c0b0d0@roweclan.net>
202 [Deleted comments regarding the ap_run_handler phase, as irrelevant
203 as BillS points out that "common case will be caught in
204 default_handler already (with the r->finfo.filetype == 0 check)"
205 and the issue is detecting this -before- we try to run the req.]
207 gregames says: can this happen somehow without a broken module
208 being involved? If not, why waste cycles trying to defend against
209 potential broken modules? It seems futile.
210 wrowe counters: no, it shouldn't happen unless the module is broken.
211 But the right answer is to fail the request up-front in dir/file
212 walk if the path was entirely invalid; and we can't do that either
213 UNTIL 2.1 or we break modules that haven't hooked map_to_storage.
215 * Can a static httpd be built reliably?
216 Message-ID: <20020207142751.T31582@clove.org>
218 * Usage of APR_BRIGADE_NORMALIZE in core_input_filter should be
221 <Pine.LNX.4.33.0201202232430.318-100000@deepthought.cs.virginia.edu>
222 Jeff wonders if we still care about this. It is no longer an
223 API issue but simply an extra trip through the brigade.
225 * Try to get libtool inter-library dependency code working on AIX.
226 Message-ID: <cm3n10lx555.fsf@rdu163-40-092.nc.rr.com>
228 Justin says: If we get it working on AIX, we can enable this
229 on all platforms and clean up our build system somewhat.
230 Jeff says: I thought I tested a patch for you sometime in
231 January that you were going to commit within a few days.
233 * Handling of %2f in URIs. Currently both 1.3 and 2.0
234 completely disallow %2f in the request URI path (see
235 ap_unescape_url() in util.c). It's permitted and passed
236 through in the query string, however. Roy says the
237 original reason for disallowing it, from five years ago,
238 was to protect CGI scripts that applied PATH_INFO to
239 a filesystem location and which might be tricked by
240 ..%2f..%2f(...). We *should* allow path-info of the
241 form 'http://foo.com/index.cgi/path/to/path%2finfo'.
242 Since we've revamped a lot of our processing of path
243 segments, it would be nice to allow this, or at least
244 allow it conditionally with a directive.
246 OtherBill adds that %2f as the SECOND character of a multibyte
247 sequence causes the request to fail! This happens notably in
250 * There is increasing demand from module writers for an API
251 that will allow them to control the server à la apachectl.
252 Reasons include sole-function servers that need to die if
253 an external dependency (e.g., a database) fails, et cetera.
254 Perhaps something in the (ever more abused) scoreboard?
256 On the other hand, we already have a pipe that goes between parent
257 and child for graceful shutdown events, along with an API that
258 can be used to send a message down that pipe. In threaded MPMs,
259 it is easy enough to make that one pipe be used for graceful
260 and graceless events, and it is also easy to open that pipe
261 to both parent and child for writing. Then we just need to
262 figure out how to do graceless on non-threaded MPMs.
264 * Allow the DocumentRoot directive within <Location > scopes? This
265 allows the beloved (crusty) Alias /foo/ /somepath/foo/ followed
266 by a <Directory /somepath/foo> to become simply
267 <Location /foo/> DocumentRoot /somefile/foo (IMHO a bit more legible
268 and in-your-face.) DocumentRoot unset would be accepted [and would
269 not permit content to be served, only virtual resources such as
270 server-info or server-status.
271 This proposed change would _not_ depricate Alias.
272 striker: See the thread starting with Message-ID:
273 JLEGKKNELMHCJPNMOKHOGEEJFBAA.striker@apache.org.
275 * Win32: Rotatelogs sometimes is not terminated when Apache
276 goes down hard. FirstBill was looking at possibly tracking the
277 child's-child processes in the parent process.
278 stoddard: Shared scoreboard might offer a good way for the parent
279 to keep track of 'other child' processes and whack them if the child
281 Other thoughts on walking the process chain using the NT kernel
282 have also been proposed on APR.
284 * Eliminate unnecessary creation of pipes in mod_cgid
286 * Combine log_child and piped_log_spawn. Clean up http_log.c.
289 * Platforms that do not support fork (primarily Win32 and AS/400)
290 Architect start-up code that avoids initializing all the modules
291 in the parent process on platforms that do not support fork.
293 * There are still a number of places in the code where we are
294 losing error status (i.e. throwing away the error returned by a
295 system call and replacing it with a generic error code)
297 * Mass vhosting version of suEXEC.
299 * All DBMs suffer from confusion in support/dbmmanage (perl script) since
300 the dbmmanage employs the first-matched dbm format. This is not
301 necessarily the library that Apache was built with. Aught to
302 rewrite dbmmanage upon installation to bin/ with the proper library
303 for predictable mod_auth_dbm administration.
304 Questions; htdbm exists, time to kill dbmmanage, or does it remain
305 useful as a perl dbm management example? If we keep it,
306 do we address the issue above?
309 Some additional items remaining:
310 - case_preserved_filename stuff
311 (use the new canonical name stuff?)
312 - find a new home for ap_text(_header)
313 - is it possible to remove the DAV: namespace stuff from util_xml?
315 * ap_core_translate() and its use by mod_mmap_static and mod_file_cache
316 are a bit wonky. The function should probably be exposed as a utility
317 function (such as ap_translate_url2fs() or ap_validate_fs_url() or
318 something). Another approach would be a new hook phase after
319 "translate" which would allow the module to munge what the
320 translation has decided to do.
321 Status: Greg +1 (volunteers)
323 * Explore use of a post-config hook for the code in http_main.c which
324 calls ap_fixup_virutal_hosts(), ap_fini_vhost_config(), and
325 ap_sort_hooks() [to reduce the logic in main()]
327 * read the config tree just once, and process N times (as necessary)
329 * (possibly) use UUIDs in mod_unique_id and/or mod_usertrack
331 * (possibly) port the bug fix for PR 6942 (segv when LoadModule is put
332 into a VirtualHost container) to 2.0.
334 * shift stuff to mod_core.h
336 * callers of ap_run_create_request() should check the return value
337 for failure (Doug volunteers)
339 * Fix the worker MPM to use POD to kill child processes instead
340 of ap_os_killpg, regardless of how they should die.
342 chrisd: Is this done, by any chance? See r92598 and r93358.
344 * Scoreboard structures could be changed in the future such that
345 proper alignment is not maintained, leading to segfaults on
346 some systems. Cliff posted a patch to deal with this issue but
347 later recanted. See this message to dev@apr.apache.org:
349 <Pine.LNX.4.44.0203011354090.16457-200000@deepthought.cs.virginia.edu>
351 * APXS either needs to be fixed completely for use when apr is out of tree,
352 or it should drop query mode altogether, and we just grow an
353 httpd-config or similar arrangement.
354 To quote a discussion in STATUS earlier:
356 thommay: this doesn't fix all the problems with apxs and out of
357 tree apr/apr-util, but it's a good start. There's still the
358 query cases; but I'm beginning to think that in these cases
359 the app should be querying ap{r,u}-config directly
360 deprecate -q: add htpd-config: gstein, pquerna, minfrin, pgollucci
363 TODO ISSUES REMAINING IN MOD_SSL:
365 * Do we need SSL_set_read_ahead()?
367 * SSLRequire directive (parsing of) leaks memory
369 * Diffie-Hellman-Parameters for temporary keys are hardcoded in
370 ssl_engine_dh.c, while the comment in ssl_engine_kernel.c says:
371 "it is suggested that keys be changed daily or every 500
372 transactions, and more often if possible."
374 * ssl_var_lookup could be rewritten to be MUCH faster
376 * CRL callback should be pluggable
378 * session cache store should be pluggable
380 * init functions should return status code rather than ssl_die()
382 * ssl_engine_pphrase.c needs to be reworked so it is generic enough
383 to also decrypt proxy keys
385 * output warning when allowing SSL v2.0 ? its so old
388 * mod_proxy: Ability to run SSL over proxy gateway connections,
389 encrypting (or reencrypting) at the proxy.
391 * mod_cache: Handle ESI tags.
393 * mod_cache: Resolve issue of how to cache page fragments (or perhaps
394 -if- we want to cache page fragments). Today, mod_cache/mod_mem_cache
395 will cache #include 'virtual' requests (but not #include 'file'
396 requests). This was accomplished by making CACHE_IN a
397 CONTENT_SET-1 filter to force it to run before the SUBREQ_CORE
398 filter. But now responses cannot be cached that include the
399 effects of having been run through CONTENT_SET filters
400 (mod_deflate, mod_expires, etc). We could rerun all the
401 CONTENT_SET filters on the cached response, but this will not
402 work in all cases. For example, mod_expires relies on installing
403 the EXPIRATION filter during fixups. Contents served out of
404 mod_cache (out of the quick_handler) bypass -all- the request
405 line server hooks (Ryan really hated this. It is great for
406 performance, but bad because of the complications listed above).
408 mod_cache/mod_mem_cache/mod_cache_disk:
410 * mod_mem_cache: Consider adding a RevalidateTimeout directive to
411 specify time at which local cached content is to be revalidated
412 (ie, underlying file stat'ed to see if it has changed).
414 * mod_cache: CacheEnable/CacheDisable should accept regular expressions.
415 jerenkrantz says: Too slow. Get regexs away from speedy caches by
416 default. Introduce a new CacheEnableRegex if you want.
418 * mod_mem_cache/mod_cache_disk: Need to be able to query cache
419 status (num of entries, cache object properties, etc.).
420 mod_status could be extended to query optional hooks defined
421 by modules for the purpose of reporting module status.
422 mod_cache (et. al.) could define optional hooks that are called
423 to collect status. Status should be queryable by
425 jerenkrantz says: Yawn. Who cares.
427 * Regex containers don't work in an intutive way
428 Status: No one has come up with an efficient way to fix this
429 behavior. Dean has suggested getting rid of regex containers
431 OtherBill suggests: We at least seem to agree on eliminating
432 the <Container ~ foo> forms, and using only
433 <ContainerMatch foo> semantics.
435 * orig_ct in the byterange/multipart handling may not be
436 needed. Apache 1.3 just never stashed "multipart" into
437 r->content_type. We should probably follow suit since the
438 byterange stuff doesn't want the rest of the code to see the
439 multipart content-type; the other code should still think it is
440 dealing with the <orig_ct> stuff.
441 Status: Greg volunteers to investigate (esp. since he was most
442 likely the one to break it :-)
444 EXPERIMENTAL MODULES:
446 Experimental modules should eventually be be promoted to fully supported
447 status or removed from the repository entirely (ie, the
448 'experiment' failed). This section tracks what needs to happen to
449 get the modules promoted to fully supported status.