STATUS revision 5c9a6bc911baa7cef34d64de59af1e8d6e09c13a
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan ZelenyAPACHE 2.0 STATUS: -*-text-*-
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan ZelenyLast modified at [$Date: 2002/10/28 07:41:09 $]
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.44 : in development
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.43 : rolled October 2, 2002
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.42 : released September 24, 2002 as GA.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.41 : rolled September 16, 2002. not released.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.40 : released August 9, 2002 as GA.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.39 : released June 17, 2002 as GA.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.38 : rolled June 16, 2002. not released.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.37 : rolled June 11, 2002. not released.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.36 : released May 6, 2002 as GA.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.35 : released April 5, 2002 as GA.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.34 : tagged March 26, 2002.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.33 : tagged March 6, 2002. not released.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.32 : released Feburary 16, 2002 as beta.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.31 : rolled Feburary 1, 2002. not released.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.30 : tagged January 8, 2002. not rolled.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.29 : tagged November 27, 2001. not rolled.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.28 : released November 13, 2001 as beta.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.27 : rolled November 6, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.26 : tagged October 16, 2001. not rolled.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.25 : rolled August 29, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.24 : rolled August 18, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.23 : rolled August 9, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.22 : rolled July 29, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.21 : rolled July 20, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.20 : rolled July 8, 2001
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek 2.0.19 : rolled June 27, 2001
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek 2.0.18 : rolled May 18, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.17 : rolled April 17, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.16 : rolled April 4, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.15 : rolled March 21, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0.14 : rolled March 7, 2001
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 2.0a9 : released December 12, 2000
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek 2.0a8 : released November 20, 2000
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek 2.0a7 : released October 8, 2000
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek 2.0a6 : released August 18, 2000
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek 2.0a5 : released August 4, 2000
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek 2.0a4 : released June 7, 2000
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek 2.0a3 : released April 28, 2000
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek 2.0a2 : released March 31, 2000
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek 2.0a1 : released March 10, 2000
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub HrozekPlease consult the following STATUS files for information
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zelenyon related projects:
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan ZelenyContributors looking for a mission:
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * just do an egrep on "TODO" and see what's there
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan ZelenyCURRENT RELEASE NOTES:
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub HrozekRELEASE SHOWSTOPPERS:
da0b829a2002987339aadaa01b85adbb5ddba20dJakub HrozekCURRENT VOTES:
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek * Adopt backwards compatibility for future Apache 2.0 releases
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek such that MMN major number changes and eliminating non-experimental
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek modules are deferred for the next minor version bump (e.g. 2.1, 2.2
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek +1: wrowe, jerenkrantz, aaron, brianp
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Defer the Auth module overhaul to the next minor version bump
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny (e.g. 2.1, 2.2, 3.0) on the condition that forward compatibility
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny resolution is adopted.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny +1: wrowe, aaron
3f94d6718d44185137e13b6d326dfd63e8dc61c6Simo Sorce 0: jerenkrantz
1e6f2180724de4722a5218826c9401181168d9d4Simo Sorce * Adopt an even/odd release paradigm (see VERSIONING) such that
1e6f2180724de4722a5218826c9401181168d9d4Simo Sorce even numbered releases are stable, and odd numbered releases
1e6f2180724de4722a5218826c9401181168d9d4Simo Sorce are development efforts, keeping in the tradition of Linux,
480195bd70467bf05e2fd92cf25cec822a670cedSimo Sorce Perl, etc. In pratical terms, this implies C-T-R-T-C, where
480195bd70467bf05e2fd92cf25cec822a670cedSimo Sorce patches are (generally) first applied to the development branch,
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek tested, and then (after vote) applied to the stable branch.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek +1: wrowe, jerenkrantz, aaron
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Branch APACHE_2_0_BRANCH today, changing the version in CVS HEAD
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny to 2.1.0-dev.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny +1 [from APACHE_2_0_43]: wrowe, aaron
69f6f2ed116d1c987851bfcc410cf2bdd1b0cc97Michal Zidek +1 [from HEAD]:
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek 0: jerenkrantz
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek a) httpd-std.conf should be tailored by install (from src or
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny binbuild) even if user has existing httpd.conf
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny +1: trawick, slive, gregames, ianh, Ken, wrowe
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny wrowe - prefer httpd.default.conf to avoid ambiguity with cvs
72aa8e7b1d234b6b68446d42efa1cff22b70c81bSimo Sorce b) tailored httpd-std.conf should be copied by install to
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek c) tailored httpd-std.conf should be installed to
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek +1: slive, trawick, Ken
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek d) Installing a set of default config files when upgrading a server
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny doesn't make ANY sense at all.
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek ianh - medium/big sites don't use 'standard config' anyway, as it
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny usually needs major customizations
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny -1: Ken, wrowe
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny wrowe - diff is wonderful when comparing old/new default configs,
1e6f2180724de4722a5218826c9401181168d9d4Simo Sorce even for customized sites that ianh mentions
1e6f2180724de4722a5218826c9401181168d9d4Simo Sorce * If the parent process dies, should the remaining child processes
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek "gracefully" self-terminate. Or maybe we should make it a runtime
1e6f2180724de4722a5218826c9401181168d9d4Simo Sorce option, or have a concept of 2 parent processes (one being a
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek "hot spare").
1e6f2180724de4722a5218826c9401181168d9d4Simo Sorce See: Message-ID: <3C58232C.FE91F19F@Golux.Com>
1e6f2180724de4722a5218826c9401181168d9d4Simo Sorce Self-destruct: Ken, Martin
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek Not self-destruct: BrianP, Ian, Cliff, BillS
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny Make it runtime configurable: Aaron, Jim, Justin, wrowe
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny Have 2 parents: +1: Jim
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny -1: Justin, wrowe
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek +0: Martin (while standing by, could it do
21d485184df986e1a123f70c689517386e51a5ceMichal Zidek something useful?)
21d485184df986e1a123f70c689517386e51a5ceMichal Zidek * Make the worker MPM the default MPM for threaded Unix boxes.
21d485184df986e1a123f70c689517386e51a5ceMichal Zidek +1: Justin, Ian, Cliff, BillS, striker, wrowe
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny +0: BrianP, Aaron (mutex contention is looking better with the
69f6f2ed116d1c987851bfcc410cf2bdd1b0cc97Michal Zidek latest code, let's continue tuning and testing)
69f6f2ed116d1c987851bfcc410cf2bdd1b0cc97Michal ZidekRELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
69f6f2ed116d1c987851bfcc410cf2bdd1b0cc97Michal Zidek * There is a bug in how we sort some hooks, at least the pre-config
69f6f2ed116d1c987851bfcc410cf2bdd1b0cc97Michal Zidek hook. The first time we call the hooks, they are in the correct
a290ace39af184d878568e17588b6b2210aea63fMichal Zidek order, but the second time, we don't sort them correctly. Currently,
33cbb789ff71be5dccbb4a0acd68814b0d53da34Michal Zidek the modules/http/config.m4 file has been renamed to
a290ace39af184d878568e17588b6b2210aea63fMichal Zidek modules/http/config2.m4 to work around this problem, it should moved
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek back when this is fixed.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny OtherBill offers that this is a SERIOUS problem. We do not sort
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny correctly by the ordering arguments passed to the register hook
a290ace39af184d878568e17588b6b2210aea63fMichal Zidek functions. This was proven when I reordered the open_logs hook
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny to attempt to open the error logs prior to the access logs. Possibly
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny the entire sorting code needs to be refactored.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * pipes deadlock on all platforms with limited pipe buffers (e.g. both
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny Linux and Win32, as opposed to only Win32 on 1.3). The right solution
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek is either GStein's proposal for a "CGI Brigade", or OtherBill's proposal
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek for "Poll Buckets" for "Polling Filter Chains".
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * server pushed CGI's not working.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek This might be an interaction with the above pipes deadlock issue.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Message-ID: <3CE15B85.2FF45121@apache.org>
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * All handlers should always send content down even if r->header_only
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek is set. If not, it means that the HEAD requests don't generate the
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek same headers as a GET which is wrong.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * HP/UX 10.20: compile breakage in APR. Looks like it should be easy
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek to fix, probably just some extraneous #include's that are fouling
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Jeff: See my reply and patch in the PR (and previous commit to
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek stop using "pipe" as a field name). If patch is committed, we
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek should be okay. I'll wait to see if the user tests the patch.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Update by Jeff 20020722: I got an account on HP 10.20. It looks
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek like some of the APR thread detection is screwed up. If we find
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek pthread.h but we can't compile the pthread test program we still
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek think we can use threads. For that reason, the patch I posted
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek to the PR won't work as-is since a failed compile of the test
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek program means nothing.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * exec cmd and suexec arg-passing enhancements
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Status: Patches proposed
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Message-ID: <20020526041748.A29148@prodigy.Redbrick.DCU.IE>
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek (see the "proc.patch" and "suexec-shell.patch" links in this message)
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * The 2.0.36 worker MPM graceless shutdown changes work but are
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek a bit clunky on some platforms; eg, on Linux, the loop to
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek join each worker thread seems to hang, and the parent ends up
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek killing off the child with SIGKILL. But at least it shuts down.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * --enable-mods-shared="foo1 foo2" is busted on Darwin. Pier
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek posted a patch (Message-ID: <B8DBBE8D.575A%pier@betaversion.org>).
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * We do not properly substitute the prefix-variables in the configuration
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek scripts or generated-configs. (i.e. if sysconfdir is etc,
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek httpd-std.conf points to conf.)
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * If any request gets through ap_process_request_internal() and is
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek scheduled to be served by the core handler, without a flag that this
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek r->filename was tested by dir/file_walk, we need to 500 at the very
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek end of the ap_process_request_internal() processing so sub_req-esters
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek know this request cannot be run. This provides authors of older
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek modules better compatibility, while still improving the security and
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek robustness of 2.0.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Status: still need to decide where this goes, OtherBill comments...
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Message-ID: <065701c14526$495203b0$96c0b0d0@roweclan.net>
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek [Deleted comments regarding the ap_run_handler phase, as irrelevant
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek as BillS points out that "common case will be caught in
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek default_handler already (with the r->finfo.filetype == 0 check)"
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek and the issue is detecting this -before- we try to run the req.]
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek gregames says: can this happen somehow without a broken module
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek being involved? If not, why waste cycles trying to defend against
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek potential broken modules? It seems futile.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek wrowe counters: no, it shouldn't happen unless the module is broken.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek But the right answer is to fail the request up-front in dir/file
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek walk if the path was entirely invalid; and we can't do that either
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek UNTIL 2.1 or we break modules that haven't hooked map_to_storage.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * Rewrite core_output_filter. It is nearly impossible to support
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek it with predictable results as it is implemented now.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * With AP_MODE_EXHAUSTIVE in the core, it is finally clear to me
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek how the Perchild MPM should be re-written. It hasn't worked
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek correctly since filters were added because it wasn't possible to
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek get the content that had already been written and the socket at
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek the same time. This mode lets us do that, so the MPM can be
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * htpasswd blindly processes the file you give it, and does no
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek sanity checking before totally corrupting whatever file it was
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek you thought you had. It should check the input file and bail
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek if it finds non-comment lines that do not contain exactly 1
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek ':' character.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Message-ID: <20020217150457.A31632@clove.org>
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * Can a static httpd be built reliably?
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Message-ID: <20020207142751.T31582@clove.org>
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * [Ken] Test suite failures:
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek o worker is also failing some of the 'cgi' subtests
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek (see <URL:http://Source-Zone.Org/Apache/regression/>):
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Justin says: "Worker should be fine and passes httpd-test here.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek I think it's a perl or a httpd-test problem."
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * Usage of APR_BRIGADE_NORMALIZE in core_input_filter should be
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek removed if possible.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Message-ID: <Pine.LNX.4.33.0201202232430.318-100000@deepthought.cs.virginia.edu>
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Jeff wonders if we still care about this. It is no longer an
e3ce042be3a6c66aa720fc139f557b065ae6dc5eSimo Sorce API issue but simply an extra trip through the brigade.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * The Add...Filter and Set...Filter directives do not allow the
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek administrator to order filters, beyond the order of filename (mime)
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek extensions. It isn't clear if Set...Filter(s) should be inserted
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek before or after the Add...Filter(s) which are ordered by sequence of
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek filename extensions. At minimum, some sort of +-[0-10] syntax seems
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek like a nice solution. See ROADMAP.
e3ce042be3a6c66aa720fc139f557b065ae6dc5eSimo Sorce * Get perchild to work on platforms other than Linux. This
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek will require a portable mechanism to pass data and file/socket
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek descriptors between vhost child groups. An API was proposed
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Message-ID: <20020111115006.K1529@clove.org>
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek * Try to get libtool inter-library dependency code working on AIX.
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek Message-ID: <cm3n10lx555.fsf@rdu163-40-092.nc.rr.com>
1e6f2180724de4722a5218826c9401181168d9d4Simo Sorce Justin says: If we get it working on AIX, we can enable this
1e6f2180724de4722a5218826c9401181168d9d4Simo Sorce on all platforms and clean up our build system
480195bd70467bf05e2fd92cf25cec822a670cedSimo Sorce Jeff says: I thought I tested a patch for you sometime in
480195bd70467bf05e2fd92cf25cec822a670cedSimo Sorce January that you were going to commit within a few
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Handling of %2f in URIs. Currently both 1.3 and 2.0
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny completely disallow %2f in the request URI path (see
480195bd70467bf05e2fd92cf25cec822a670cedSimo Sorce ap_unescape_url() in util.c). It's permitted and passed
480195bd70467bf05e2fd92cf25cec822a670cedSimo Sorce through in the query string, however. Roy says the
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny original reason for disallowing it, from five years ago,
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny was to protect CGI scripts that applied PATH_INFO to
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek a filesystem location and which might be tricked by
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek ..%2f..%2f(...). We *should* allow path-info of the
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek form 'http://foo.com/index.cgi/path/to/path%2finfo'.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny Since we've revamped a lot of our processing of path
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce segments, it would be nice to allow this, or at least
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce allow it conditionally with a directive.
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce OtherBill adds that %2f as the SECOND character of a multibyte
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce sequence causes the request to fail! This happens notably in
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce the ja-jis encoding.
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce * FreeBSD, threads, and worker MPM. All seems to work fine
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce if you only have one worker process with many threads. Add
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce a second worker process and the accept lock seems to be
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce lost. This might be an APR issue with how it deals with
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce the child_init hook (i.e. the fcntl lock needs to be resynced).
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce More examination and analysis is required.
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce Status: This has also been reported on Cygwin.
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce Message-ID: <3C2CC514.8EF3BED1@wapme-systems.de> (cygnus)
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce Aaron says: I spent some time disecting this and have come to
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce the conclusion that it is not a problem in the worker MPM
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce (or at least, it is not isolated to a problem in worker).
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce I'll list some of the problems I'm seeing in case someone
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce else wants to pick up where I've left off:
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce - Delivery of just about any signal to one of the child
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce processes will send it into an infinite loop as well.
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek - Even though the parent is spinning out of control,
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek at first the child or children will appear to work
69f6f2ed116d1c987851bfcc410cf2bdd1b0cc97Michal Zidek properly. At times it is possible to get it into a state,
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce however, where a request will hang until another concurrent
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek request "kicks" the first, at which point the second will
69f6f2ed116d1c987851bfcc410cf2bdd1b0cc97Michal Zidek hang. My theory is that this has to do with the
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce pthread_cond_*() implementation in FreeBSD, but it's still
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek possible that it is in APR.
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce Justin adds: Oh, FreeBSD threads are implemented entirely with
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny select()/poll()/longjmp(). Welcome to the nightmare.
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek So, that means a ktrace output also has the thread
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek scheduling internals in it (since it is all the same to
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny the kernel). Which makes it hard to distinguish between
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek our select() calls and their select() calls.
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek *bangs head on wall repeatedly* But, some of the libc_r
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek files have a DBG_MSG #define. This is moderately helpful
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek when used with -DNO_DETACH. The kernel scheduler isn't
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek waking up the threads on a select(). Yum. And, I bet
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek those decrementing select calls have to do with the
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce scheduler. Time to brush up on our OS fundamentals.
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek * There is increasing demand from module writers for an API
3f94d6718d44185137e13b6d326dfd63e8dc61c6Simo Sorce that will allow them to control the server � la apachectl.
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek Reasons include sole-function servers that need to die if
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek an external dependency (e.g., a database) fails, et cetera.
2eaf2045b5b55cdaff6ae5704225cf9a75b16950Simo Sorce Perhaps something in the (ever more abused) scoreboard?
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek On the other hand, we already have a pipe that goes between parent
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny and child for graceful shutdown events, along with an API that
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny can be used to send a message down that pipe. In threaded MPMs,
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny it is easy enough to make that one pipe be used for graceful
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek and graceless events, and it is also easy to open that pipe
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek to both parent and child for writing. Then we just need to
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny figure out how to do graceless on non-threaded MPMs.
3f94d6718d44185137e13b6d326dfd63e8dc61c6Simo Sorce * Allow the DocumentRoot directive within <Location > scopes? This
3f94d6718d44185137e13b6d326dfd63e8dc61c6Simo Sorce allows the beloved (crusty) Alias /foo/ /somepath/foo/ followed
3f94d6718d44185137e13b6d326dfd63e8dc61c6Simo Sorce by a <Directory /somepath/foo> to become simply
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny <Location /foo/> DocumentRoot /somefile/foo (IMHO a bit more legible
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny and in-your-face.) DocumentRoot unset would be accepted [and would
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny not permit content to be served, only virtual resources such as
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny server-info or server-status.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny This proposed change would _not_ depricate Alias.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny striker: See the thread starting with Message-ID:
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny JLEGKKNELMHCJPNMOKHOGEEJFBAA.striker@apache.org.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Win32: Rotatelogs sometimes is not terminated when Apache
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny goes down hard. FirstBill was looking at possibly tracking the
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny child's-child processes in the parent process.
3f94d6718d44185137e13b6d326dfd63e8dc61c6Simo Sorce stoddard: Shared scoreboard might offer a good way for the parent
e79d23932ef9d52cf4eb32ddec2d0a9b3af9a9ebJan Zeleny to keep track of 'other child' processes and whack them if the child
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny Other thoughts on walking the process chain using the NT kernel
e79d23932ef9d52cf4eb32ddec2d0a9b3af9a9ebJan Zeleny have also been proposed on APR.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Eliminate unnecessary creation of pipes in mod_cgid
8a1738f9379a1b8fb5c95c3df649e014ff5a1434Jan Zeleny * Combine log_child and piped_log_spawn. Clean up http_log.c.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny Common logging API.
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * Document mod_file_cache.
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * Platforms that do not support fork (primarily Win32 and AS/400)
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek Architect start-up code that avoids initializing all the modules
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek in the parent process on platforms that do not support fork.
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * Win32: Migrate the MPM over to use APR thread/process calls. This
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek would eliminate some code in the Win32 branch that essentially
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny duplicates what is in APR.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * There are still a number of places in the code where we are
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny losing error status (i.e. throwing away the error returned by a
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny system call and replacing it with a generic error code)
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Mass vhosting version of suEXEC.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * All DBMs suffer from confusion in support/dbmmanage (perl script) since
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny the dbmmanage employs the first-matched dbm format. This is not
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny necessarily the library that Apache was built with. Aught to
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny rewrite dbmmanage upon installation to bin/ with the proper library
d35c8317ed23aff66fde8147bfa2ea91f43a12e4Jan Zeleny for predictable mod_auth_dbm administration.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny Questions; htdbm exists, time to kill dbmmanage, or does it remain
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny useful as a perl dbm management example? If we keep it,
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny do we address the issue above?
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * Integrate mod_dav.
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek Some additional items remaining:
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny - case_preserved_filename stuff
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny (use the new canonical name stuff?)
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny - find a new home for ap_text(_header)
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny - is it possible to remove the DAV: namespace stuff from util_xml?
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * ap_core_translate() and its use by mod_mmap_static and mod_file_cache
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny are a bit wonky. The function should probably be exposed as a utility
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny function (such as ap_translate_url2fs() or ap_validate_fs_url() or
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny something). Another approach would be a new hook phase after
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny "translate" which would allow the module to munge what the
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny translation has decided to do.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny Status: Greg +1 (volunteers)
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Explore use of a post-config hook for the code in http_main.c which
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny calls ap_fixup_virutal_hosts(), ap_fini_vhost_config(), and
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny ap_sort_hooks() [to reduce the logic in main()]
234958be042980242fff6da936af674da877c5efSimo Sorce * read the config tree just once, and process N times (as necessary)
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * (possibly) use UUIDs in mod_unique_id and/or mod_usertrack
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * (possibly) port the bug fix for PR 6942 (segv when LoadModule is put
72aa8e7b1d234b6b68446d42efa1cff22b70c81bSimo Sorce into a VirtualHost container) to 2.0.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * shift stuff to mod_core.h
72aa8e7b1d234b6b68446d42efa1cff22b70c81bSimo Sorce * callers of ap_run_create_request() should check the return value
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny for failure (Doug volunteers)
72aa8e7b1d234b6b68446d42efa1cff22b70c81bSimo Sorce * Win32: Get Apache working on Windows 95/98. The following work
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny (at least) needs to be done:
72aa8e7b1d234b6b68446d42efa1cff22b70c81bSimo Sorce - Document warning that OSR2 is required (for Crypt functions, in
72aa8e7b1d234b6b68446d42efa1cff22b70c81bSimo Sorce rand.c, at least.) This could be resolved with an SSL library, or
4c9a85ab708ec7debecad51e4240e04d8bc6ca4eOndrej Kos randomization in APR itself.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny - Bring the Win9xConHook.dll from 1.3 into 2.0 (no sense till it
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny actually works) and add in a splash of Win9x service code.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Fix the worker MPM to use POD to kill child processes instead
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny of ap_os_killpg, regardless of how they should die.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Scoreboard structures could be changed in the future such that
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny proper alignment is not maintained, leading to segfaults on
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny some systems. Cliff posted a patch to deal with this issue but
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny later recanted. See this message to dev@apr.apache.org:
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny Message-ID: <Pine.LNX.4.44.0203011354090.16457-200000@deepthought
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan ZelenyTODO ISSUES REMAINING IN MOD_SSL:
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * In order to use a DSO version of mod_ssl we have to link with
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek -lssl and -lcrypto. A workaround is in place right now where the
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek entire EXTRA_LIBS macro is being appended to the objects list, but
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny this is a hack. We should either revamp the APACHE_CHECK_SSL_TOOLKIT
89caf5edcc99f5731e89bd51e6ffaad3ec11c304Pavel Březina autoconf function or come up with some other autoconf checks to
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny search for libssl and libcrypto and properly add them to mod_ssl's
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * SSL renegotiations in combination with POST request
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Port or dispose all code inside #if 0...#endif blocks that remain
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny from the porting effort.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Do we need SSL_set_read_ahead()?
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * the ssl_expr api is NOT THREAD SAFE. race conditions exist:
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny -in ssl_expr_comp() if SSLRequire is used in .htaccess
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny (ssl_expr_info is global)
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny -is ssl_expr_eval() if there is an error
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny (ssl_expr_error is global)
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * SSLRequire directive (parsing of) leaks memory
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Diffie-Hellman-Parameters for temporary keys are hardcoded in
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek ssl_engine_dh.c, while the comment in ssl_engine_kernel.c says:
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek "it is suggested that keys be changed daily or every 500
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek transactions, and more often if possible."
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * ssl_var_lookup could be rewritten to be MUCH faster
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * CRL callback should be pluggable
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * session cache store should be pluggable
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * init functions should return status code rather than ssl_die()
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * ssl_engine_pphrase.c needs to be reworked so it is generic enough
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny to also decrypt proxy keys
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * the shmcb code should just align its memory segment rather than
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny jumping through all the "safe" memcpy and memset hoops
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan ZelenyEXPERIMENTAL MODULES:
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny Experimental modules should eventually be be promoted to fully supported
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny status or removed from the repository entirely (ie, the
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny 'experiment' failed). This section tracks what needs to happen to
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny get the modules promoted to fully supported status.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * mod_cache: handle cache_control: no_cache "field_name" to enable
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny cacheing the response w/o header "field_name"
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny See RFC2616 section 14.9.1
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * mod_cache: CacheEnable/CacheDisable should accept regular expressions.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * mod_cache: Fix dependency on ATOMIC operators. Need
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek APR_HAS_ATOMIC_* feature macros.
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * mod_disk_cache: Implement garbage collection
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * mod_mem_cache/mod_disk_cache: Need to be able to query cache
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek status (num of entries, cache object properties, etc.).
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny mod_status could be extended to query optional hooks defined
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny by modules for the purpose of reporting module status.
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek mod_cache (et. al.) could define optional hooks that are called
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny to collect status. Status should be queryable by
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny HTTP or SNMP?
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * Enable mod_cache/mod_mem_cache/mod_disk_cache to handle
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek * mod_mem_cache fd caching is broken on Unix. The cached fd cannot
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek be used across multiple threads except for use on sendfile().
9e2c64c6d4f5560e27207193efea6536a566865eMichal Zidek * implement cross-platform/cross-library TLS support
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * General stabilization and testing
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan ZelenyPRs that have been suspended forever waiting for someone to
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zelenyput them into 'the next release':
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * documentation and Q&A
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#2221: Make online documentation search link back to my installation
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek PR#2906: Propose that Apache recommend $UNIQUE_ID for all "session id"
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#2793: When will Apache support P3P? Any Plans?
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#2113: HTTP Server Rebuild Line Needs Changing for the better
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#2421: problem specifying ndbm library for build ?with autoconfigure
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#76: missing call to "setlocale();"
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#628: Request of "Options SymLinksIfGroupMatch"
e3ce042be3a6c66aa720fc139f557b065ae6dc5eSimo Sorce PR#793: RLimitCPU and RLimitMEM don't apply to all children like they should
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek PR#922: it is useful to allow specifiction that root-owned symlinks
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek should always be followed
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek PR#1028: DoS attacks involving memory consumption
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek PR#1191: setlogin() is not called, causing problems with e.g. identd
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek PR#1204: regerror() exists, use it
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek PR#2284: Can not POST to ErrorDocument - Apache/1.3b6
d27d11b360b92f1728206802005bb3da359e1ea4Michal Zidek PR#2396: Proposal for TimeZone directive
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek PR#2446: AllowOverride FileInfo is too coarse
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#2760: [PATCH] User/Group for <Directory> and <Location> i.e. not only
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny in global and <Virtual>.
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#2907: suggestion: power up your Include directive :)
95cc95749a5e783f2b5d2124d783f85820baf937Stephen Gallagher PR#3018: cannot limit some HTTP methods
da0b829a2002987339aadaa01b85adbb5ddba20dJakub Hrozek PR#3677: New ErrorDocumentMatch directive
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#4244: "Files" and "FilesMatch" regexp does not recognize bang as
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny negation operator
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#5993: AllowOverride should have a 'CheckNone' and 'AllowNone' argument
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny instead of only 'None'
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny * mod_access
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#537: mod_access syntax allows hosts that should be restricted
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#1287: add allow,deny/deny,allow warning to mod_access
c737e1444fb186e349e59bfa9dac4995b720b4b1Jan Zeleny PR#2512: <IfDenied> directive wanted
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * mod_auth-any
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek PR#557: ~UserHome directories are not honored in absolute pathname
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek requests (.htaccess)
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek PR#1117: Using NIS passwd.byname dbm files with AuthDBMUserFile
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek PR#1809: Suggestion for improving authentication modules and core source
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek code, problem with 401 and ErrorDocument
82b5429d1438392c45e70a0f84dd4d0f5fa1a171Jakub Hrozek * mod_autoindex
PR#2873: Feedback/Comment on APACI
PR#2431: A small addition to rotatelogs.c to improve program functionality.
PR#2889: Inclusion of RPM spec file in CVS/distributions
* orig_ct in the byterange/multipart handling may not be