STATUS revision dba571e5c5a2586c7e2b382789f3250136ff7577
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceAPACHE 2.0 STATUS: -*-text-*-
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceLast modified at [$Date: 2002/08/12 19:30:12 $]
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.41 : in development.
027e89d47af308db4b41761ca9f847c026b63ec8Andreas Gustafsson 2.0.40 : released August 9, 2002 as GA.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.39 : released June 17, 2002 as GA.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.38 : rolled June 16, 2002. not released.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.37 : rolled June 11, 2002. not released.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.36 : released May 6, 2002 as GA.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.35 : released April 5, 2002 as GA.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.34 : tagged March 26, 2002.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.33 : tagged March 6, 2002. not released.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.32 : released Feburary 16, 2002 as beta.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.31 : rolled Feburary 1, 2002. not released.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.30 : tagged January 8, 2002. not rolled.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.29 : tagged November 27, 2001. not rolled.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.28 : released November 13, 2001 as beta.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.27 : rolled November 6, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.26 : tagged October 16, 2001. not rolled.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.25 : rolled August 29, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.24 : rolled August 18, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.23 : rolled August 9, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.22 : rolled July 29, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.21 : rolled July 20, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.20 : rolled July 8, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.19 : rolled June 27, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.18 : rolled May 18, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.17 : rolled April 17, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.16 : rolled April 4, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.15 : rolled March 21, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0.14 : rolled March 7, 2001
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0a9 : released December 12, 2000
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0a8 : released November 20, 2000
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0a7 : released October 8, 2000
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0a6 : released August 18, 2000
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0a5 : released August 4, 2000
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0a4 : released June 7, 2000
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0a3 : released April 28, 2000
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0a2 : released March 31, 2000
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 2.0a1 : released March 10, 2000
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LucePlease consult the following STATUS files for information
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luceon related projects:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceCURRENT RELEASE NOTES:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceRELEASE SHOWSTOPPERS:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceCURRENT VOTES:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce a) httpd-std.conf should be tailored by install (from src or
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce binbuild) even if user has existing httpd.conf
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce +1: trawick, slive, gregames, ianh, Ken
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce b) tailored httpd-std.conf should be copied by install to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce c) tailored httpd-std.conf should be installed to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce +1: slive, trawick, Ken
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce d) Installing a set of default config files when upgrading a server
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce doesn't make ANY sense at all.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce +1: rbb, striker
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce ianh - medium/big sites don't use 'standard config' anyway, as it
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce usually needs major customizations
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Should we always build [support*] binaries statically unless otherwise
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <20020129210006.B23512@Lithium.MeepZor.Com>
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce +1: Ken, *wrowe [they are PITAs on OSX]
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce -1: Justin, Ian
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * If the parent process dies, should the remaining child processes
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce "gracefully" self-terminate. Or maybe we should make it a runtime
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce option, or have a concept of 2 parent processes (one being a
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce See: Message-ID: <3C58232C.FE91F19F@Golux.Com>
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Self-destruct: Ken, Martin
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Not self-destruct: BrianP, Ian, Cliff, BillS
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Make it runtime configurable: Aaron, Jim, Justin
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Have 2 parents: +1: Jim
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce -1: Justin, wrowe [for 2.0]
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce +0: Martin (while standing by, could it do
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce something useful?)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Make the worker MPM the default MPM for threaded Unix boxes.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce +1: Justin, Ian, Cliff, BillS, striker
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce +0: BrianP, Aaron (mutex contention is looking better with the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce latest code, let's continue tuning and testing)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceRELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Performance problems in the content-length filter:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce - In many situations, the C-L filter sets aside buckets
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce until it sees EOS. Setting aside file buckets is bad
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce because it requires an mmap+memcpy+munmap.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce - In addition, the C-L filter reads and buffers all the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce content from a pipe bucket.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * All handlers should always send content down even if r->header_only
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce is set. If not, it means that the HEAD requests don't generate the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce same headers as a GET which is wrong.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Is this a showstopper?
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce +1: Justin, Ken
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * server pushed CGI's not working. (Is this a showstopper??)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <3CE15B85.2FF45121@apache.org>
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * HP/UX 10.20: compile breakage in APR. Looks like it should be easy
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce to fix, probably just some extraneous #include's that are fouling
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Jeff: See my reply and patch in the PR (and previous commit to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce stop using "pipe" as a field name). If patch is committed, we
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce should be okay. I'll wait to see if the user tests the patch.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Update by Jeff 20020722: I got an account on HP 10.20. It looks
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce like some of the APR thread detection is screwed up. If we find
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce pthread.h but we can't compile the pthread test program we still
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce think we can use threads. For that reason, the patch I posted
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce to the PR won't work as-is since a failed compile of the test
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce program means nothing.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * exec cmd and suexec arg-passing enhancements
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Status: Patches proposed
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <20020526041748.A29148@prodigy.Redbrick.DCU.IE>
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce (see the "proc.patch" and "suexec-shell.patch" links in this message)
aeb8fffc841865c3336383eadfd9987332a03286Andreas Gustafsson * Get mod_cache/mod_mem_cache out of experimental (still some
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce work items left to complete)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * The 2.0.36 worker MPM graceless shutdown changes work but are
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce a bit clunky on some platforms; eg, on Linux, the loop to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce join each worker thread seems to hang, and the parent ends up
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce killing off the child with SIGKILL. But at least it shuts down.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * --enable-mods-shared="foo1 foo2" is busted on Darwin. Pier
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce posted a patch (Message-ID: <B8DBBE8D.575A%pier@betaversion.org>).
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * We do not properly substitute the prefix-variables in the configuration
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce scripts or generated-configs. (i.e. if sysconfdir is etc,
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce httpd-std.conf points to conf.)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * If any request gets through ap_process_request_internal() and is
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce scheduled to be served by the core handler, without a flag that this
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce r->filename was tested by dir/file_walk, we need to 500 at the very
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce end of the ap_process_request_internal() processing so sub_req-esters
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce know this request cannot be run. This provides authors of older
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce modules better compatibility, while still improving the security and
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce robustness of 2.0.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Status: still need to decide where this goes, OtherBill comments...
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <065701c14526$495203b0$96c0b0d0@roweclan.net>
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce [Deleted comments regarding the ap_run_handler phase, as irrelevant
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce as BillS points out that "common case will be caught in
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce default_handler already (with the r->finfo.filetype == 0 check)"
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce and the issue is detecting this -before- we try to run the req.]
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce gregames says: can this happen somehow without a broken module
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce being involved? If not, why waste cycles trying to defend against
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce potential broken modules? It seems futile.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce wrowe counters: no, it shouldn't happen unless the module is broken.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce But the right answer is to fail the request up-front in dir/file
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce walk if the path was entirely invalid; and we can't do that either
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce or we break modules that are unwilling to hook map_to_storage.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Rewrite core_output_filter. It is nearly impossible to support
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce it with predictable results as it is implemented now.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * With AP_MODE_EXHAUSTIVE in the core, it is finally clear to me
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce how the Perchild MPM should be re-written. It hasn't worked
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce correctly since filters were added because it wasn't possible to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce get the content that had already been written and the socket at
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce the same time. This mode lets us do that, so the MPM can be
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson * htpasswd blindly processes the file you give it, and does no
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce sanity checking before totally corrupting whatever file it was
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce you thought you had. It should check the input file and bail
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson if it finds non-comment lines that do not contain exactly 1
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce ':' character.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <20020217150457.A31632@clove.org>
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Can a static httpd be built reliably?
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <20020207142751.T31582@clove.org>
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * [Ken] Test suite failures:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce o worker is also failing some of the 'cgi' subtests
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce (see <URL:http://Source-Zone.Org/Apache/regression/>):
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Justin says: "Worker should be fine and passes httpd-test here.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce If you can provide evidence that it can be reproduced
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce outside of httpd-test, then it's a showstopper. I
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce think it's a perl or a httpd-test problem."
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Not a showstopper: Justin
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Usage of APR_BRIGADE_NORMALIZE in core_input_filter should be
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce removed if possible.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <Pine.LNX.4.33.0201202232430.318-100000@deepthought.cs.virginia.edu>
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Jeff wonders if we still care about this. It is no longer an
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce API issue but simply an extra trip through the brigade.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * There is a bug in how we sort some hooks, at least the pre-config
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce hook. The first time we call the hooks, they are in the correct
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce order, but the second time, we don't sort them correctly. Currently,
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce the modules/http/config.m4 file has been renamed to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce modules/http/config2.m4 to work around this problem, it should moved
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce back when this is fixed. rbb
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Justin says: "Is this really a showstopper? This has been here
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce forever. What's wrong? Does this have to do with
b9c96971964d87c2705c8dc29300ff8103479ee6Andreas Gustafsson autoconf or m4?"
b9c96971964d87c2705c8dc29300ff8103479ee6Andreas Gustafsson Not a showstopper: Justin, BrianP, trawick, gregames
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * The Add...Filter and Set...Filter directives do not allow the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce administrator to order filters, beyond the order of filename (mime)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce extensions. It isn't clear if Set...Filter(s) should be inserted
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce before or after the Add...Filter(s) which are ordered by sequence of
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce filename extensions. At minimum, some sort of +-[0-10] syntax seems
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce like the quickest fix for a 2.0 gold release.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Justin says: "Could we delay this for a point release or 2.1?"
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Not a showstopper: justin, wrowe, trawick, stoddard, Jim, Ian, Aaron,
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Get perchild to work on platforms other than Linux. This
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce will require a portable mechanism to pass data and file/socket
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce descriptors between vhost child groups. An API was proposed
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <20020111115006.K1529@clove.org>
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Try to get libtool inter-library dependency code working on AIX.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <cm3n10lx555.fsf@rdu163-40-092.nc.rr.com>
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Justin says: If we get it working on AIX, we can enable this
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce on all platforms and clean up our build system
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Jeff says: I thought I tested a patch for you sometime in
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce January that you were going to commit within a few
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Handling of %2f in URIs. Currently both 1.3 and 2.0
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce completely disallow %2f in the request URI path (see
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce ap_unescape_url() in util.c). It's permitted and passed
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce through in the query string, however. Roy says the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce original reason for disallowing it, from five years ago,
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce was to protect CGI scripts that applied PATH_INFO to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce a filesystem location and which might be tricked by
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce ..%2f..%2f(...). We *should* allow path-info of the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce form 'http://foo.com/index.cgi/path/to/path%2finfo'.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Since we've revamped a lot of our processing of path
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce segments, it would be nice to allow this, or at least
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce allow it conditionally with a directive.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * FreeBSD, threads, and worker MPM. All seems to work fine
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce if you only have one worker process with many threads. Add
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce a second worker process and the accept lock seems to be
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce lost. This might be an APR issue with how it deals with
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce the child_init hook (i.e. the fcntl lock needs to be resynced).
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce More examination and analysis is required.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Status: This has also been reported on Cygwin.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <3C2CC514.8EF3BED1@wapme-systems.de> (cygnus)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Justin says: So, FreeBSD-CURRENT and Cywin have the same
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce problem. Yum. If another platform has this
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce with worker, this becomes a showstopper.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Aaron says: I spent some time disecting this and have come to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce the conclusion that it is not a problem in the worker MPM
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce (or at least, it is not isolated to a problem in worker).
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce I'll list some of the problems I'm seeing in case someone
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce else wants to pick up where I've left off:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce - Delivery of just about any signal to one of the child
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce processes will send it into an infinite loop as well.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce - Even though the parent is spinning out of control,
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce at first the child or children will appear to work
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce properly. At times it is possible to get it into a state,
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce however, where a request will hang until another concurrent
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce request "kicks" the first, at which point the second will
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce hang. My theory is that this has to do with the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce pthread_cond_*() implementation in FreeBSD, but it's still
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce possible that it is in APR.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Justin adds: Oh, FreeBSD threads are implemented entirely with
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce select()/poll()/longjmp(). Welcome to the nightmare.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce So, that means a ktrace output also has the thread
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce scheduling internals in it (since it is all the same to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce the kernel). Which makes it hard to distinguish between
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce our select() calls and their select() calls.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce *bangs head on wall repeatedly* But, some of the libc_r
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce files have a DBG_MSG #define. This is moderately helpful
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce when used with -DNO_DETACH. The kernel scheduler isn't
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce waking up the threads on a select(). Yum. And, I bet
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce those decrementing select calls have to do with the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce scheduler. Time to brush up on our OS fundamentals.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * There is increasing demand from module writers for an API
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce that will allow them to control the server � la apachectl.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Reasons include sole-function servers that need to die if
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce an external dependency (e.g., a database) fails, et cetera.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Perhaps something in the (ever more abused) scoreboard?
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce rbb: I don't believe the scoreboard is the correct mechanism
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce for this. We already have a pipe that goes between parent
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce and child for graceful shutdown events, along with an API that
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce can be used to send a message down that pipe. In threaded MPMs,
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce it is easy enough to make that one pipe be used for graceful
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce and graceless events, and it is also easy to open that pipe
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce to both parent and child for writing. Then we just need to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce figure out how to do graceless on non-threaded MPMs.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Allow the DocumentRoot directive within <Location > scopes? This
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce allows the beloved (crusty) Alias /foo/ /somepath/foo/ followed
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson by a <Directory /somepath/foo> to become simply
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce <Location /foo/> DocumentRoot /somefile/foo (IMHO a bit more legible
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce and in-your-face.) DocumentRoot unset would be accepted [and would
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce not permit content to be served, only virtual resources such as
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce server-info or server-status.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce This proposed change would _not_ depricate Alias.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce striker: See the thread starting with Message-ID:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce JLEGKKNELMHCJPNMOKHOGEEJFBAA.striker@apache.org.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Win32: Rotatelogs sometimes is not terminated when Apache
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce goes down hard. FirstBill was looking at possibly tracking the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce child's-child processes in the parent process.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce OtherBill asks, wasn't this fixed?
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce stoddard: Not fixed. Shared scoreboard might offer a good
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce way for the parent to keep track of 'other child' processes
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce and whack them if the child goes down.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Other thoughts on walking the process chain using the NT kernel
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce have also been proposed on APR.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Win32: Add a simple hold console open patch (wait for close or
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce the ESC key, with a nice message) if the server died a bad
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce death (non-zero exit code) in console mode.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Resolution: bring forward same ugly hacks from 1.3.13-.20
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce This is not so simple. Any exit() from APR or other libraries
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce can't be caught unless we add some sort of apr_exit(rv) with
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce registered apr_atexit() fn's that have the return code as an
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce argument to the registered fn.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Eliminate unnecessary creation of pipes in mod_cgid
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Combine log_child and piped_log_spawn. Clean up http_log.c.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Common logging API.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Document mod_file_cache.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Platforms that do not support fork (primarily Win32 and AS/400)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Architect start-up code that avoids initializing all the modules
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce in the parent process on platforms that do not support fork.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Win32: Migrate the MPM over to use APR thread/process calls. This
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce would eliminate some code in the Win32 branch that essentially
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce duplicates what is in APR.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * There are still a number of places in the code where we are
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce losing error status (i.e. throwing away the error returned by a
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce system call and replacing it with a generic error code)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Mass vhosting version of suEXEC.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * All DBMs suffer from confusion in support/dbmmanage (perl script) since
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce the dbmmanage employs the first-matched dbm format. This is not
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce necessarily the library that Apache was built with. Aught to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce rewrite dbmmanage upon installation to bin/ with the proper library
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce for predictable mod_auth_dbm administration.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Questions; htdbm exists, time to kill dbmmanage, or does it remain
ab19d688255b3a333a41b4ebe6f4213538e89c2aEric Luce useful as a perl dbm management example? If we keep it,
ab19d688255b3a333a41b4ebe6f4213538e89c2aEric Luce do we address the issue above?
ab19d688255b3a333a41b4ebe6f4213538e89c2aEric Luce * Integrate mod_dav.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Some additional items remaining:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce - case_preserved_filename stuff
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce (use the new canonical name stuff?)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce - find a new home for ap_text(_header)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce - is it possible to remove the DAV: namespace stuff from util_xml?
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * ap_core_translate() and its use by mod_mmap_static and mod_file_cache
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce are a bit wonky. The function should probably be exposed as a utility
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce function (such as ap_translate_url2fs() or ap_validate_fs_url() or
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce something). Another approach would be a new hook phase after
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce "translate" which would allow the module to munge what the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce translation has decided to do.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Status: Greg +1 (volunteers), Ryan +1
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Explore use of a post-config hook for the code in http_main.c which
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce calls ap_fixup_virutal_hosts(), ap_fini_vhost_config(), and
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce ap_sort_hooks() [to reduce the logic in main()]
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * read the config tree just once, and process N times (as necessary)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * (possibly) use UUIDs in mod_unique_id and/or mod_usertrack
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * (possibly) port the bug fix for PR 6942 (segv when LoadModule is put
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce into a VirtualHost container) to 2.0.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * shift stuff to mod_core.h
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * callers of ap_run_create_request() should check the return value
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce for failure (Doug volunteers)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Win32: Get Apache working on Windows 95/98. The following work
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce (at least) needs to be done:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce - Document warning that OSR2 is required (for Crypt functions, in
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce rand.c, at least.) This could be resolved with an SSL library, or
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce randomization in APR itself.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce - Bring the Win9xConHook.dll from 1.3 into 2.0 (no sense till it
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce actually works) and add in a splash of Win9x service code.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Fix the worker MPM to use POD to kill child processes instead
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce of ap_os_killpg, regardless of how they should die. (Ryan Bloom)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Scoreboard structures could be changed in the future such that
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce proper alignment is not maintained, leading to segfaults on
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce some systems. Cliff posted a patch to deal with this issue but
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce later recanted. See this message to dev@apr.apache.org:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Message-ID: <Pine.LNX.4.44.0203011354090.16457-200000@deepthought
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceTODO ISSUES REMAINING IN MOD_SSL:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * In order to use a DSO version of mod_ssl we have to link with
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce -lssl and -lcrypto. A workaround is in place right now where the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce entire EXTRA_LIBS macro is being appended to the objects list, but
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce this is a hack. We should either revamp the APACHE_CHECK_SSL_TOOLKIT
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce autoconf function or come up with some other autoconf checks to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce search for libssl and libcrypto and properly add them to mod_ssl's
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * SSL renegotiations in combination with POST request
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Port or dispose all code inside #if 0...#endif blocks that remain
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce from the porting effort.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Do we need SSL_set_read_ahead()?
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * the ssl_expr api is NOT THREAD SAFE. race conditions exist:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce -in ssl_expr_comp() if SSLRequire is used in .htaccess
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce (ssl_expr_info is global)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce -is ssl_expr_eval() if there is an error
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce (ssl_expr_error is global)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * SSLRequire directive (parsing of) leaks memory
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Diffie-Hellman-Parameters for temporary keys are hardcoded in
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce ssl_engine_dh.c, while the comment in ssl_engine_kernel.c says:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce "it is suggested that keys be changed daily or every 500
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce transactions, and more often if possible."
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * ssl_var_lookup could be rewritten to be MUCH faster
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * CRL callback should be pluggable
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * session cache store should be pluggable
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * init functions should return status code rather than ssl_die()
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * ssl_engine_pphrase.c needs to be reworked so it is generic enough
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce to also decrypt proxy keys
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * the shmcb code should just align its memory segment rather than
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce jumping through all the "safe" memcpy and memset hoops
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceEXPERIMENTAL MODULES:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Experimental modules should eventually be be promoted to fully supported
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce status or removed from the repository entirely (ie, the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce 'experiment' failed). This section tracks what needs to happen to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce get the modules promoted to fully supported status.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_cache: handle cache_control: no_cache "field_name" to enable
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce cacheing the response w/o header "field_name"
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce See RFC2616 section 14.9.1
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_cache: CacheEnable/CacheDisable should accept regular expressions.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_cache: Fix dependency on ATOMIC operators. Need
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce APR_HAS_ATOMIC_* feature macros.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_disk_cache: Implement garbage collection
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_mem_cache/mod_disk_cache: Need to be able to query cache
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce status (num of entries, cache object properties, etc.).
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce mod_status could be extended to query optional hooks defined
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce by modules for the purpose of reporting module status.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce mod_cache (et. al.) could define optional hooks that are called
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce to collect status. Status should be queryable by
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce HTTP or SNMP?
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_mem_cache: garbage collection. One strategy is to simply
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce remove stale entries as we attempt to serve them. Another
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce strategy is to kick off a GC thread that traverses the cache
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce and preemptively remove stale entries. How to manage a
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce cache that is full? Do LRU GC? Other? Bueller?
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_mem_cache/mod_disk_cache: Complete implementing config
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Sample config for mod_cache/mod_mem_cache/mod_disk_cache for
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce inclusion into httpd.conf.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_cache/mod_mem_cache/mod_disk_cache: Documentation.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LucePRs that have been suspended forever waiting for someone to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luceput them into 'the next release':
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * documentation and Q&A
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2221: Make online documentation search link back to my installation
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2906: Propose that Apache recommend $UNIQUE_ID for all "session id"
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2793: When will Apache support P3P? Any Plans?
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2113: HTTP Server Rebuild Line Needs Changing for the better
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2421: problem specifying ndbm library for build ?with autoconfigure
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#76: missing call to "setlocale();"
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#628: Request of "Options SymLinksIfGroupMatch"
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#793: RLimitCPU and RLimitMEM don't apply to all children like they should
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#922: it is useful to allow specifiction that root-owned symlinks
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce should always be followed
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1028: DoS attacks involving memory consumption
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1191: setlogin() is not called, causing problems with e.g. identd
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1204: regerror() exists, use it
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2284: Can not POST to ErrorDocument - Apache/1.3b6
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2396: Proposal for TimeZone directive
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2446: AllowOverride FileInfo is too coarse
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2760: [PATCH] User/Group for <Directory> and <Location> i.e. not only
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce in global and <Virtual>.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2907: suggestion: power up your Include directive :)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#3018: cannot limit some HTTP methods
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#3677: New ErrorDocumentMatch directive
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#4244: "Files" and "FilesMatch" regexp does not recognize bang as
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce negation operator
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#5993: AllowOverride should have a 'CheckNone' and 'AllowNone' argument
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce instead of only 'None'
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_access
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#537: mod_access syntax allows hosts that should be restricted
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1287: add allow,deny/deny,allow warning to mod_access
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2512: <IfDenied> directive wanted
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_auth-any
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#557: ~UserHome directories are not honored in absolute pathname
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce requests (.htaccess)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1117: Using NIS passwd.byname dbm files with AuthDBMUserFile
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1809: Suggestion for improving authentication modules and core source
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce code, problem with 401 and ErrorDocument
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_autoindex
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1263: Add frame-safe anchor attribute to mod_autoindex links
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_cgi (and suexec)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#921: suexec Uses cwd before filling it in, doesn't use syslog
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1176: Apache cannot handle continuation line in headers
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1120: suexec does not parse arguments to #exec cmd
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1268: CGI scripts running as Apache user: security (suexec etc.)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1285: Error messages could be easier to spot in cgi.log file for suexec.c
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1905: suexec - Allow modules to set user:group for execution.
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#2360: suexec for general access of user content?
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#2460: TimeOut applies to output of CGI scripts
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#2573: CGI's for general use still have to be run as another user
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#4241: Need to be able to override shebang line to make CGI scripts
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson more portable.
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#4490: mod_cgi prevents handling of OPTIONS requests
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#370: Modified PATH environemnt variable is not passed, instead
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson system's is used
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#1383: I make mod_headers to modify request headers as well as
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson response ones.
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#1677: mod_headers should allow mod_log_config-style formats in
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#759: imap should read <MAP><AREA>*</MAP> too!
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#78: Additional status for XBitHack directive
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#623: A smarter "Last Modified" value for SSI documents (see PR number 600)
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#1145: mod_include
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson Allow for Last-Modified: without resorting to XBitHack
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#1803: patches to mod_include to allow for file tests
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#4459: Suggestion for better handling of Last-modified headers
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#2415: /server-info doesn't check for the virtual host to list the info
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#1050: Logging of virtual server to error_log as well
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson PR#1358: Selective url-encode of log fields (or maybe a pseudo
5f09ce124cad9712a9675f17f83ddc915e734909Andreas Gustafsson log_rewrite module?)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2073: pipelined connections are not logged correctly
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#4448: Please allow CGI env variables (QUERY_STRING, ...) to be logged
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_negotiation
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#3191: no way to set global quality-of-source (qs) coneg values
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce with multiviews
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#362: Mod_proxy doesn't allow change of error pages
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#440: Proxy doesn't deliver documents if not connected
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#534: proxy converts ~name to %7Ename when name starts with a dot (.)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#612: Proxy FTP Authentication Fails
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#700: Proxy doesn't do links right for OpenVMS files through ftp:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#980: Controlling Access to Remote Proxies would be nice...
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#994: Adding authentication "on the fly" through the proxy module
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1085: ProxyRemote make a dead cycle.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1166: ``nph-'' not honored (no buffering) for ProxyRemote mapping
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1290: Need to know "hit-rate" on proxy cache
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1532: Proxy transfer logging
b9c96971964d87c2705c8dc29300ff8103479ee6Andreas Gustafsson PR#1547: No HTTP_X_FORWARDED_FOR set...
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1567: ProxyRemote proxy requests fail authentication by firewall
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1702: mod_proxy to support persistent conns?
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1878: listing of proxy cache content
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2314: patterns in ProxyRemote
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2648: Cache file names in Proxy module
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#3568: Accessing URL through proxy server corrupts data.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#3605: Some anonymous FTP URLs ask for authentication
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_rewrite
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1582: mod_rewrite forms REQUEST_URI different than mod_cgi does
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2074: mod_rewrite doesn't pass Proxy Throughput on internal subrequests
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * mod_status
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2138: mod_status always displays 256 possible connection slots
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2343: Status module averages are for entire uptime
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * apache-api
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1004: request_config field in request_rec is moderately bogus
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1158: improvements to child spawning API
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#1233: there is no way to keep per-connection per-module state
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2024: adding auth_why to conn_rec
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2873: Feedback/Comment on APACI
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#3143: No module specific data hook for per-connection data
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * generally odds and ends
f1fd37f759991616d454ce371a2390da45141593Andreas Gustafsson PR#2431: A small addition to rotatelogs.c to improve program functionality.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2763: mailto tags and bundling bug report script
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2785: os-aix Support for System Resource Controller
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#2889: Inclusion of RPM spec file in CVS/distributions
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce PR#5713: os-windows [PATCH] install as win32 service with domain account
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Status: Cannot accept password-as-arg, we should prompt the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce user when -k install/-k config with a user argument.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceOther bugs that need fixing:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * ap_discard_request should be converted to use the bucket API
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce directly rather than waste cycles copying buffers with the old API.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * MaxRequestsPerChild measures connections, not requests.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Until someone has a better way, we'll probably just rename it
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce "MaxConnectionsPerChild".
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Regex containers don't work in an intutive way
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Status: No one has come up with an efficient way to fix this
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce behavior. Dean has suggested getting rid of regex containers
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce OtherBill suggests: We at least seem to agree on eliminating
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce the <Container ~ foo> forms, and using only
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce <ContainerMatch foo> semantics.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * SIGSEGV on Linux (glibc 2.1.2) isn't caught properly by a
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce sigwaiting thread. We need to work around this, perhaps unless
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce there is hope soon for a fixed glibc.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * orig_ct in the byterange/multipart handling may not be
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce needed. Apache 1.3 just never stashed "multipart" into
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce r->content_type. We should probably follow suit since the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce byterange stuff doesn't want the rest of the code to see the
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce multipart content-type; the other code should still think it is
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce dealing with the <orig_ct> stuff.
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Status: Greg volunteers to investigate (esp. since he was most
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce likely the one to break it :-)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceBinaries (2.0.40):
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Platform Avail. Volunteer
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce ------------------------------------------------------------------
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce AIX 4.3.3 no Bill Stoddard
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Mandrake 8.1 no Ryan Bloom
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce FreeBSD 4.1 no Ryan Bloom
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce hppa2.0w-hp-hpux11.00 yes Cliff Woolley
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce i386-pc-solaris2.8 no Aaron Bannert
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce i386-unknown-freebsd4.5 no
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce i386-unknown-freebsd4.6 yes Cliff Woolley
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce i686-pc-linux-gnu-slackware81 yes Cliff Woolley
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce i686-pc-linux-gnu-rh70 no Aaron Bannert
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce i686-pc-linux-gnu-rh73 yes Cliff Woolley
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce ia64-hp-hpux11.20 no
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce powerpc-apple-darwin5.5 no Aaron Bannert
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce powerpc-unknown-linux-gnu no Graham Leggett
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce sparc-sun-solaris2.8 no Jim Jagielski
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce NetWare no Brad Nicholes
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce OS/2 no Brian Havard
b9c96971964d87c2705c8dc29300ff8103479ee6Andreas Gustafsson Win32-x86 no William Rowe
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceOther features that need writing:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Finish infrastructure in core for async MPMs
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Status: post 2.0
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * TODO in source -- just do an egrep on "TODO" and see what's there
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric LuceAvailable Patches:
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Martin Sojka <msojka@gmx.de>'s patch to add error reporting for failed
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce htpasswd actions due to a full /tmp volume (other programs may have
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce similar problems?)
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Mike Abbott's <mja@trudge.engr.sgi.com> patches to improve
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce Status: These were written for 1.3, and are awaiting a port to
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Jim Winstead's <jimw@trainedmonkey.com> patch to add CookieDomain and
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce other small mod_usertrack features
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce * Dan Rench's <drench@xnet.com> patch to add allow the errmsg and timefmt
f293a69bcd1c1dd7bdac8f4102fc2398b9e475c8Eric Luce of SSI's to be modified in the config file. Patch is available in