STATUS revision 18fcb3eaebbdbc0c0a47e1d6b7c3a6c6819cdafa
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrmAPACHE 2.0 STATUS: -*-text-*-
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrmLast modified at [$Date: 2002/04/08 15:09:16 $]
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.35 : released April 5, 2002 as GA.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.34 : tagged March 26, 2002.
9dc68827cbd515262ecb8d5ae8547d9e82c72e00Jon A. Cruz 2.0.33 : tagged March 6, 2002. not released.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.32 : released Feburary 16, 2002 as beta.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.31 : rolled Feburary 1, 2002. not released.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.30 : tagged January 8, 2002. not rolled.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.29 : tagged November 27, 2001. not rolled.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.28 : released November 13, 2001 as beta.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.27 : rolled November 6, 2001
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.26 : tagged October 16, 2001. not rolled.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.25 : rolled August 29, 2001
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.24 : rolled August 18, 2001
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.23 : rolled August 9, 2001
eb4caa8f4cdc2955b58dcd2de06fe770533414c8Jon A. Cruz 2.0.22 : rolled July 29, 2001
eb4caa8f4cdc2955b58dcd2de06fe770533414c8Jon A. Cruz 2.0.21 : rolled July 20, 2001
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.20 : rolled July 8, 2001
92d1c774b263648abaaf65699c867f37d6a43644cilix 2.0.19 : rolled June 27, 2001
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.18 : rolled May 18, 2001
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.17 : rolled April 17, 2001
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.16 : rolled April 4, 2001
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.15 : rolled March 21, 2001
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0.14 : rolled March 7, 2001
4ddc5ae5a40c90607ca4ea4a9beab40f9c2715dapjrm 2.0a9 : released December 12, 2000
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0a8 : released November 20, 2000
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0a7 : released October 8, 2000
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0a6 : released August 18, 2000
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0a5 : released August 4, 2000
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm 2.0a4 : released June 7, 2000
b802808a0226a87371021393c4f1da776aa6a6adjohanengelen 2.0a3 : released April 28, 2000
b802808a0226a87371021393c4f1da776aa6a6adjohanengelen 2.0a2 : released March 31, 2000
b802808a0226a87371021393c4f1da776aa6a6adjohanengelen 2.0a1 : released March 10, 2000
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrmPlease consult the following STATUS files for information
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelenon related projects:
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrmCURRENT RELEASE NOTES:
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * 34 status: Let's get all API changes and showstoppers in this one.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrmFINAL RELEASE SHOWSTOPPERS:
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * We do not properly substitute the prefix-variables in the configuration
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm scripts or generated-configs. (i.e. if sysconfdir is etc,
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm httpd-std.conf points to conf.)
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Aaron says: This is not a showstopper, these problems have existed
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen for as long as I can remember. It would be nice to fix
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen them but they are not new.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Not a showstopper: Jeff, Aaron, BillS, gregames, Jim
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrmCURRENT VOTES:
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * Should we always build [support*] binaries statically unless otherwise
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Message-ID: <20020129210006.B23512@Lithium.MeepZor.Com>
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm +1: Ken, *wrowe [they are PITAs on OSX]
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm -1: Justin, Ian
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * If the parent process dies, should the remaining child processes
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm "gracefully" self-terminate. Or maybe we should make it a runtime
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm option, or have a concept of 2 parent processes (one being a
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm "hot spare").
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm See: Message-ID: <3C58232C.FE91F19F@Golux.Com>
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Self-destruct: Ken, Martin
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Not self-destruct: BrianP, Ian, Cliff, BillS
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Make it runtime configurable: Aaron, Jim, Justin
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Have 2 parents: +1: Jim
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm -1: Justin, wrowe [for 2.0]
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm +0: Martin (while standing by, could it do
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm something useful?)
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * Make the worker MPM the default MPM for threaded Unix boxes.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm +1: Justin, Ian, Cliff
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm -0: Aaron (premature decision, needs more discussion), Lars
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrmRELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * ap_cache_hex2msec() and ap_cache_msec2hex() are completely
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm hosed. They do not handle apr_time_t (64 bit ints) properly.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * If any request gets through ap_process_request_internal() and is
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm scheduled to be served by the core handler, without a flag that this
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm r->filename was tested by dir/file_walk, we need to 500 at the very
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm end of the ap_process_request_internal() processing so sub_req-esters
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm know this request cannot be run. This provides authors of older
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm modules better compatibility, while still improving the security and
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm robustness of 2.0.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Status: still need to decide where this goes, OtherBill comments...
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Message-ID: <065701c14526$495203b0$96c0b0d0@roweclan.net>
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm [Deleted comments regarding the ap_run_handler phase, as irrelevant
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm as BillS points out that "common case will be caught in
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm default_handler already (with the r->finfo.filetype == 0 check)"
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm and the issue is detecting this -before- we try to run the req.]
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm gregames says: can this happen somehow without a broken module
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm being involved? If not, why waste cycles trying to defend against
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm potential broken modules? It seems futile.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen wrowe counters: no, it shouldn't happen unless the module is broken.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm But the right answer is to fail the request up-front in dir/file
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm walk if the path was entirely invalid; and we can't do that either
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm or we break modules that are unwilling to hook map_to_storage.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Rewrite core_output_filter. It is nearly impossible to support
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel it with predictable results as it is implemented now.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Convert all instances of the old apr_lock_t type to the new
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * With AP_MODE_EXHAUSTIVE in the core, it is finally clear to me
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm how the Perchild MPM should be re-written. It hasn't worked
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm correctly since filters were added because it wasn't possible to
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm get the content that had already been written and the socket at
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel the same time. This mode lets us do that, so the MPM can be
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * htpasswd blindly processes the file you give it, and does no
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen sanity checking before totally corrupting whatever file it was
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen you thought you had. It should check the input file and bail
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen if it finds non-comment lines that do not contain exactly 1
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen ':' character.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Message-ID: <20020217150457.A31632@clove.org>
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * Can a static httpd be built reliably?
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Message-ID: <20020207142751.T31582@clove.org>
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * [Ken] Test suite failures:
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm o worker is also failing some of the 'cgi' subtests
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Justin says: "Worker should be fine and passes httpd-test here.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm If you can provide evidence that it can be reproduced
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm outside of httpd-test, then it's a showstopper. I
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm think it's a perl or a httpd-test problem."
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Not a showstopper: Justin
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Usage of APR_BRIGADE_NORMALIZE in core_input_filter should be
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel removed if possible.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Message-ID: <Pine.LNX.4.33.0201202232430.318-100000@deepthought.cs.virginia.edu>
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * There is a bug in how we sort some hooks, at least the pre-config
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel hook. The first time we call the hooks, they are in the correct
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel order, but the second time, we don't sort them correctly. Currently,
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm the modules/http/config.m4 file has been renamed to
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm modules/http/config2.m4 to work around this problem, it should moved
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm back when this is fixed. rbb
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Justin says: "Is this really a showstopper? This has been here
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm forever. What's wrong? Does this have to do with
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm autoconf or m4?"
09ba3247163582bf2e30e17c4c154aa259ce038acilix Not a showstopper: Justin, BrianP, trawick, gregames
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * The Add...Filter and Set...Filter directives do not allow the
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm administrator to order filters, beyond the order of filename (mime)
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm extensions. It isn't clear if Set...Filter(s) should be inserted
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel before or after the Add...Filter(s) which are ordered by sequence of
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel filename extensions. At minimum, some sort of +-[0-10] syntax seems
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm like the quickest fix for a 2.0 gold release.
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen Justin says: "Could we delay this for a point release or 2.1?"
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen Not a showstopper: justin, wrowe, trawick, stoddard, Jim, Ian, Aaron,
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen * Get perchild to work on platforms other than Linux. This
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen will require a portable mechanism to pass data and file/socket
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen descriptors between vhost child groups. An API was proposed
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Message-ID: <20020111115006.K1529@clove.org>
96d7f74ab56c7ce6cf9df0108ce925f59bcf0035buliabyak * Try to get libtool inter-library dependency code working on AIX.
96d7f74ab56c7ce6cf9df0108ce925f59bcf0035buliabyak Message-ID: <cm3n10lx555.fsf@rdu163-40-092.nc.rr.com>
96d7f74ab56c7ce6cf9df0108ce925f59bcf0035buliabyak Justin says: If we get it working on AIX, we can enable this
96d7f74ab56c7ce6cf9df0108ce925f59bcf0035buliabyak on all platforms and clean up our build system
96d7f74ab56c7ce6cf9df0108ce925f59bcf0035buliabyak Jeff says: I thought I tested a patch for you sometime in
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen January that you were going to commit within a few
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Handling of %2f in URIs. Currently both 1.3 and 2.0
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm completely disallow %2f in the request URI path (see
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm ap_unescape_url() in util.c). It's permitted and passed
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel through in the query string, however. Roy says the
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm original reason for disallowing it, from five years ago,
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel was to protect CGI scripts that applied PATH_INFO to
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm a filesystem location and which might be tricked by
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm ..%2f..%2f(...). We *should* allow path-info of the
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel form 'http://foo.com/index.cgi/path/to/path%2finfo'.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Since we've revamped a lot of our processing of path
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel segments, it would be nice to allow this, or at least
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel allow it conditionally with a directive.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * FreeBSD, threads, and worker MPM. All seems to work fine
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel if you only have one worker process with many threads. Add
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel a second worker process and the accept lock seems to be
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm lost. This might be an APR issue with how it deals with
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm the child_init hook (i.e. the fcntl lock needs to be resynced).
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm More examination and analysis is required.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Status: This has also been reported on Cygwin.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Message-ID: <3C2CC514.8EF3BED1@wapme-systems.de> (cygnus)
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Justin says: So, FreeBSD-CURRENT and Cywin have the same
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen problem. Yum. If another platform has this
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen with worker, this becomes a showstopper.
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen Aaron says: I spent some time disecting this and have come to
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen the conclusion that it is not a problem in the worker MPM
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen (or at least, it is not isolated to a problem in worker).
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen I'll list some of the problems I'm seeing in case someone
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen else wants to pick up where I've left off:
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen - Delivery of just about any signal to one of the child
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen processes will send it into an infinite loop as well.
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen - Even though the parent is spinning out of control,
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen at first the child or children will appear to work
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen properly. At times it is possible to get it into a state,
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen however, where a request will hang until another concurrent
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen request "kicks" the first, at which point the second will
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen hang. My theory is that this has to do with the
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen pthread_cond_*() implementation in FreeBSD, but it's still
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen possible that it is in APR.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Justin adds: Oh, FreeBSD threads are implemented entirely with
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen select()/poll()/longjmp(). Welcome to the nightmare.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen So, that means a ktrace output also has the thread
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen scheduling internals in it (since it is all the same to
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen the kernel). Which makes it hard to distinguish between
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen our select() calls and their select() calls.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen *bangs head on wall repeatedly* But, some of the libc_r
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen files have a DBG_MSG #define. This is moderately helpful
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen when used with -DNO_DETACH. The kernel scheduler isn't
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen waking up the threads on a select(). Yum. And, I bet
e27ae51fe3131016d977bce585cce82cdb1a89aejohanengelen those decrementing select calls have to do with the
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould scheduler. Time to brush up on our OS fundamentals.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * There is increasing demand from module writers for an API
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen that will allow them to control the server � la apachectl.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Reasons include sole-function servers that need to die if
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen an external dependency (e.g., a database) fails, et cetera.
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould Perhaps something in the (ever more abused) scoreboard?
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen rbb: I don't believe the scoreboard is the correct mechanism
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen for this. We already have a pipe that goes between parent
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen and child for graceful shutdown events, along with an API that
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen can be used to send a message down that pipe. In threaded MPMs,
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen it is easy enough to make that one pipe be used for graceful
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould and graceless events, and it is also easy to open that pipe
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould to both parent and child for writing. Then we just need to
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould figure out how to do graceless on non-threaded MPMs.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * Allow the DocumentRoot directive within <Location > scopes? This
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm allows the beloved (crusty) Alias /foo/ /somepath/foo/ followed
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen by a <Directory /somepath/foo> to become simply
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen <Location /foo/> DocumentRoot /somefile/foo (IMHO a bit more legible
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm and in-your-face.) DocumentRoot unset would be accepted [and would
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm not permit content to be served, only virtual resources such as
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm server-info or server-status.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel This proposed change would _not_ depricate Alias.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * Win32: Rotatelogs sometimes is not terminated when Apache
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm goes down hard. FirstBill was looking at possibly tracking the
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm child's-child processes in the parent process.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel OtherBill asks, wasn't this fixed?
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm stoddard: Not fixed. Shared scoreboard might offer a good
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel way for the parent to keep track of 'other child' processes
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm and whack them if the child goes down.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Other thoughts on walking the process chain using the NT kernel
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel have also been proposed on APR.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Win32: Add a simple hold console open patch (wait for close or
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel the ESC key, with a nice message) if the server died a bad
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel death (non-zero exit code) in console mode.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Resolution: bring forward same ugly hacks from 1.3.13-.20
b2b2d102f63f1f7eeb8353146d0b2bc75938acbccilix * Port of mod_ssl to Apache 2.0:
16a8c7d5e433b176636a4a1260c42ea43932110bKrzysztof Kosiński The current porting state is summarized in modules/ssl/README. The
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm remaining work includes:
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm (1) stablizing/optimizing the SSL filter logic
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel (2) Enabling SSL extentions
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm (3) Trying to seperate the https filter logic from mod_ssl -
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel This is to facilitate other modules that wish to use the https
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm filter or the mod_ssl logic or both as required.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Eliminate unnecessary creation of pipes in mod_cgid
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Combine log_child and piped_log_spawn. Clean up http_log.c.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Common logging API.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Document mod_file_cache.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Platforms that do not support fork (primarily Win32 and AS/400)
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Architect start-up code that avoids initializing all the modules
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen in the parent process on platforms that do not support fork.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * Win32: Migrate the MPM over to use APR thread/process calls. This
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen would eliminate some code in the Win32 branch that essentially
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel duplicates what is in APR.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * There are still a number of places in the code where we are
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel losing error status (i.e. throwing away the error returned by a
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel system call and replacing it with a generic error code)
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * Mass vhosting version of suEXEC.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * All DBMs suffer from confusion in support/dbmmanage (perl script) since
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm the dbmmanage employs the first-matched dbm format. This is not
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel necessarily the library that Apache was built with. Aught to
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm rewrite dbmmanage upon installation to bin/ with the proper library
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel for predictable mod_auth_dbm administration.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Questions; htdbm exists, time to kill dbmmanage, or does it remain
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm useful as a perl dbm management example? If we keep it,
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel do we address the issue above?
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Integrate mod_dav.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Some additional items remaining:
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel - case_preserved_filename stuff
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm (use the new canonical name stuff?)
b2275bb1684a0e6775bf32414ebd25a04a1e705fjohanengelen - find a new home for ap_text(_header)
b2275bb1684a0e6775bf32414ebd25a04a1e705fjohanengelen - is it possible to remove the DAV: namespace stuff from util_xml?
b2275bb1684a0e6775bf32414ebd25a04a1e705fjohanengelen * ap_core_translate() and its use by mod_mmap_static and mod_file_cache
b2275bb1684a0e6775bf32414ebd25a04a1e705fjohanengelen are a bit wonky. The function should probably be exposed as a utility
b2275bb1684a0e6775bf32414ebd25a04a1e705fjohanengelen function (such as ap_translate_url2fs() or ap_validate_fs_url() or
b2275bb1684a0e6775bf32414ebd25a04a1e705fjohanengelen something). Another approach would be a new hook phase after
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm "translate" which would allow the module to munge what the
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm translation has decided to do.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Status: Greg +1 (volunteers), Ryan +1
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * Explore use of a post-config hook for the code in http_main.c which
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm calls ap_fixup_virutal_hosts(), ap_fini_vhost_config(), and
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm ap_sort_hooks() [to reduce the logic in main()]
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * read the config tree just once, and process N times (as necessary)
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * (possibly) use UUIDs in mod_unique_id and/or mod_usertrack
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * (possibly) port the bug fix for PR 6942 (segv when LoadModule is put
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel into a VirtualHost container) to 2.0.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * shift stuff to mod_core.h
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould * callers of ap_run_create_request() should check the return value
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould for failure (Doug volunteers)
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould * Win32: Get Apache working on Windows 95/98. The following work
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould (at least) needs to be done:
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould - Document warning that OSR2 is required (for Crypt functions, in
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm rand.c, at least.) This could be resolved with an SSL library, or
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen randomization in APR itself.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen - Bring the Win9xConHook.dll from 1.3 into 2.0 (no sense till it
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen actually works) and add in a splash of Win9x service code.
e2dcdc01dfe5ed43500f0d16eea9c0c022a5f1a4johanengelen * In order to use a DSO version of mod_ssl we have to link with
e9b6af083e34e2397a8ddbe9781920733d09d151Ted Gould -lssl and -lcrypto. A workaround is in place right now where the
e9b6af083e34e2397a8ddbe9781920733d09d151Ted Gould entire EXTRA_LIBS macro is being appended to the objects list, but
e9b6af083e34e2397a8ddbe9781920733d09d151Ted Gould this is a hack. We should either revamp the APACHE_CHECK_SSL_TOOLKIT
e2dcdc01dfe5ed43500f0d16eea9c0c022a5f1a4johanengelen autoconf function or come up with some other autoconf checks to
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould search for libssl and libcrypto and properly add them to mod_ssl's
90a3966dd44e306d23febc15ebd65cde07d7a4ddTed Gould * Fix the worker MPM to use POD to kill child processes instead
e9b6af083e34e2397a8ddbe9781920733d09d151Ted Gould of ap_os_killpg, regardless of how they should die. (Ryan Bloom)
e9b6af083e34e2397a8ddbe9781920733d09d151Ted Gould * mod_cache: handle cache_control: no_cache "field_name" to enable
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould cacheing the response w/o header "field_name"
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould See RFC2616 section 14.9.1
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * Scoreboard structures could be changed in the future such that
77a4a003111bd5cfb771d4849801c898aeb889b0cilix proper alignment is not maintained, leading to segfaults on
494c671e141564431d7d05f141c885d9a2789db5cilix some systems. Cliff posted a patch to deal with this issue but
77a4a003111bd5cfb771d4849801c898aeb889b0cilix later recanted. See this message to dev@apr.apache.org:
77a4a003111bd5cfb771d4849801c898aeb889b0cilix Message-ID: <Pine.LNX.4.44.0203011354090.16457-200000@deepthought
77a4a003111bd5cfb771d4849801c898aeb889b0cilixPRs that have been suspended forever waiting for someone to
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelenput them into 'the next release':
77a4a003111bd5cfb771d4849801c898aeb889b0cilix * PR#76: general
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen missing call to "setlocale();"
b802808a0226a87371021393c4f1da776aa6a6adjohanengelen * PR#78: mod_include
b802808a0226a87371021393c4f1da776aa6a6adjohanengelen Additional status for XBitHack directive
e9b6af083e34e2397a8ddbe9781920733d09d151Ted Gould * PR#362: mod_proxy
e9b6af083e34e2397a8ddbe9781920733d09d151Ted Gould Mod_proxy doesn't allow change of error pages
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould * PR#370: mod_env
b802808a0226a87371021393c4f1da776aa6a6adjohanengelen Modified PATH environemnt variable is not passed, instead
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen system's is used
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould * PR#440: mod_proxy
6c3e745a94ef6b25a4ef9f018d350a7535aa45afTed Gould Proxy doesn't deliver documents if not connected
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * PR#534: mod_proxy
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel proxy converts ~name to %7Ename when name starts with a dot (.)
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * PR#537: mod_access
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm mod_access syntax allows hosts that should be restricted
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#557: mod_auth-any
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen ~UserHome directories are not honored in absolute pathname
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm requests (.htaccess)
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#612: mod_proxy
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Proxy FTP Authentication Fails
c3cfd82f2e8dd61022150a027d529613c38289c6bgk * PR#623: mod_include
c3cfd82f2e8dd61022150a027d529613c38289c6bgk A smarter "Last Modified" value for SSI documents (see PR number 600)
92d1c774b263648abaaf65699c867f37d6a43644cilix * PR#628: config
92d1c774b263648abaaf65699c867f37d6a43644cilix Request of "Options SymLinksIfGroupMatch"
e9b6af083e34e2397a8ddbe9781920733d09d151Ted Gould * PR#700: mod_proxy
e9b6af083e34e2397a8ddbe9781920733d09d151Ted Gould Proxy doesn't do links right for OpenVMS files through ftp:
16a8c7d5e433b176636a4a1260c42ea43932110bKrzysztof Kosiński * PR#759: mod_imap
16a8c7d5e433b176636a4a1260c42ea43932110bKrzysztof Kosiński imap should read <MAP><AREA>*</MAP> too!
92d1c774b263648abaaf65699c867f37d6a43644cilix * PR#793: general
92d1c774b263648abaaf65699c867f37d6a43644cilix RLimitCPU and RLimitMEM don't apply to all children like they should
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#921: suexec
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Uses cwd before filling it in, doesn't use syslog
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#922: config
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen it is useful to allow specifiction that root-owned symlinks
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen should always be followed
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#980: mod_proxy
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Controlling Access to Remote Proxies would be nice...
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#994: mod_proxy
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Adding authentication "on the fly" through the proxy module
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1004: apache-api
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm request_config field in request_rec is moderately bogus
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1028: other
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm DoS attacks involving memory consumption
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen * PR#1050: mod_log-any
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen Logging of virtual server to error_log as well
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1085: mod_proxy
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm ProxyRemote make a dead cycle.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1117: mod_auth-any
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Using NIS passwd.byname dbm files with AuthDBMUserFile
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1120: suexec
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm suexec does not parse arguments to #exec cmd
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1145: mod_include
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Allow for Last-Modified: without resorting to XBitHack
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1158: apache-api
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm improvements to child spawning API
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1166: mod_proxy
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm ``nph-'' not honored (no buffering) for ProxyRemote mapping
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1176: mod_cgi
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Apache cannot handle continuation line in headers
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1191: general
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm setlogin() is not called, causing problems with e.g. identd
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1204: general
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm regerror() exists, use it
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1233: apache-api
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm there is no way to keep per-connection per-module state
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1263: mod_autoexec
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Add frame-safe anchor attribute to mod_autoindex links
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1268: suexec
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm CGI scripts running as Apache user: security (suexec etc.)
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1285: suexec
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Error messages could be easier to spot in cgi.log file for suexec.c
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1287: mod_access
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen add allow,deny/deny,allow warning to mod_access
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1290: mod_proxy
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Need to know "hit-rate" on proxy cache
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1358: mod_log-any
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Selective url-encode of log fields (or maybe a pseudo
6dee13f1978edf532e28b55de684694f0b2e9449bgk log_rewrite module?)
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1383: mod_headers
60282e8335d7b6ae7020613bb22c7c69a6909fbbJon A. Cruz I make mod_headers to modify request headers as well as
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen response ones.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1532: mod_proxy
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Proxy transfer logging
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1547: mod_proxy
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen No HTTP_X_FORWARDED_FOR set...
77a4a003111bd5cfb771d4849801c898aeb889b0cilix * PR#1567: mod_proxy
77a4a003111bd5cfb771d4849801c898aeb889b0cilix ProxyRemote proxy requests fail authentication by firewall
77a4a003111bd5cfb771d4849801c898aeb889b0cilix * PR#1582: mod_rewrite
c27006137a3ad49e074b97b3297db753e1ac8eefcilix mod_rewrite forms REQUEST_URI different than mod_cgi does
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1677: mod_headers
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen mod_headers should allow mod_log_config-style formats in
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen header values
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1702: mod_proxy
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm mod_proxy to support persistent conns?
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1803: mod_include
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm patches to mod_include to allow for file tests
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1809: mod_auth-any
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Suggestion for improving authentication modules and core source
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm code, problem with 401 and ErrorDocument
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#1878: mod_proxy
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen listing of proxy cache content
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#1905: suexec
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Allow modules to set user:group for execution.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2024: apache-api
07f872bd2a4fee3c346819cbaf5af6627e698499scislac adding auth_why to conn_rec
60282e8335d7b6ae7020613bb22c7c69a6909fbbJon A. Cruz * PR#2073: mod_log-any
07f872bd2a4fee3c346819cbaf5af6627e698499scislac pipelined connections are not logged correctly
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#2074: mod_rewrite
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm mod_rewrite doesn't pass Proxy Throughput on internal subrequests
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2113: config
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen HTTP Server Rebuild Line Needs Changing for the better
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2138: mod_status
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen mod_status always displays 256 possible connection slots
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2221: documentation
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Make online documentation search link back to my installation
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2284: general
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Can not POST to ErrorDocument - Apache/1.3b6
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2314: mod_proxy
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen patterns in ProxyRemote
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2343: mod_status
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Status module averages are for entire uptime
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2360: suexec
60282e8335d7b6ae7020613bb22c7c69a6909fbbJon A. Cruz suexec for general access of user content?
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2396: general
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Proposal for TimeZone directive
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2415: mod_info
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen /server-info doesn't check for the virtual host to list the info
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2421: config
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen problem specifying ndbm library for build ?with autoconfigure
84d6d1f7365e49f2936df9df890ce49d2c000ce2Kris * PR#2431: general
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen A small addition to rotatelogs.c to improve program functionality.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2446: config
60282e8335d7b6ae7020613bb22c7c69a6909fbbJon A. Cruz AllowOverride FileInfo is too coarse
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#2460: mod_cgi
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen TimeOut applies to output of CGI scripts
1e26c94d38d46957b039b5ce059d093b5821ea18buliabyak * PR#2512: mod_access
1e26c94d38d46957b039b5ce059d093b5821ea18buliabyak <IfDenied> directive wanted
1e26c94d38d46957b039b5ce059d093b5821ea18buliabyak * PR#2573: suexec
620d9f76e90641a5a91f0c45fcb9a2f07b3881dbJohan Engelen CGI's for general use still have to be run as another user
1e26c94d38d46957b039b5ce059d093b5821ea18buliabyak * PR#2648: general
1e26c94d38d46957b039b5ce059d093b5821ea18buliabyak Cache file names in Proxy module
1e26c94d38d46957b039b5ce059d093b5821ea18buliabyak * PR#2760: config
1e26c94d38d46957b039b5ce059d093b5821ea18buliabyak [PATCH] User/Group for <Directory> and <Location> i.e. not only
1e26c94d38d46957b039b5ce059d093b5821ea18buliabyak in global and <Virtual>.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#2763: general
ba387fec25883c72319bf6107eab9e5def453e3ebuliabyak mailto tags and bundling bug report script
ba387fec25883c72319bf6107eab9e5def453e3ebuliabyak * PR#2785: os-aix
620d9f76e90641a5a91f0c45fcb9a2f07b3881dbJohan Engelen Support for System Resource Controller
ba387fec25883c72319bf6107eab9e5def453e3ebuliabyak * PR#2793: protocol
ba387fec25883c72319bf6107eab9e5def453e3ebuliabyak When will Apache support P3P? Any Plans?
ba387fec25883c72319bf6107eab9e5def453e3ebuliabyak * PR#2873: config
620d9f76e90641a5a91f0c45fcb9a2f07b3881dbJohan Engelen * PR#2889: general
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Inclusion of RPM spec file in CVS/distributions
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#2906: general
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Propose that Apache recommend $UNIQUE_ID for all "session id"
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm algorithms
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#2907: config
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm suggestion: power up your Include directive :)
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix * PR#3018: general
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix cannot limit some HTTP methods
90a3966dd44e306d23febc15ebd65cde07d7a4ddTed Gould * PR#3143: apache-api
90a3966dd44e306d23febc15ebd65cde07d7a4ddTed Gould No module specific data hook for per-connection data
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix * PR#3191: mod_negotiation
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix no way to set global quality-of-source (qs) coneg values
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix with multiviews
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix * PR#3568: mod_proxy
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix Accessing URL through proxy server corrupts data.
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix * PR#3605: mod_proxy
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix Some anonymous FTP URLs ask for authentication
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix * PR#3677: general
29f9623ba77fc735b89765ae3a13e0c06aabafcecilix New ErrorDocumentMatch directive
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#4241: config
90a3966dd44e306d23febc15ebd65cde07d7a4ddTed Gould Need to be able to override shebang line to make CGI scripts
8a0a34615f080ecd1c09986d8963d6afe5937a25Jon A. Cruz more portable.
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#4244: config
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel "Files" and "FilesMatch" regexp does not recognize bang as
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel negation operator
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * PR#4448: mod_log-any
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Please allow CGI env variables (QUERY_STRING, ...) to be logged
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * PR#4459: mod_include
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm Suggestion for better handling of Last-modified headers
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * PR#4490: mod_cgi
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm mod_cgi prevents handling of OPTIONS requests
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * PR#5713: os-windows
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel [PATCH] install as win32 service with domain account
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Status: Cannot accept password-as-arg, we should prompt the
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel user when -k install/-k config with a user argument.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * PR#5993: general
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm AllowOverride should have a 'CheckNone' and 'AllowNone' argument
ab17729e09d54ccfcb65645e6dae62a343e9c6b7Abhishek Sharma Public instead of only 'None'
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrmOther bugs that need fixing:
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm * MaxRequestsPerChild measures connections, not requests.
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Until someone has a better way, we'll probably just rename it
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel "MaxConnectionsPerChild".
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel * Regex containers don't work in an intutive way
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel Status: No one has come up with an efficient way to fix this
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel behavior. Dean has suggested getting rid of regex containers
b33e66f7da8a9371bc4e89373b075d27e2d1ec20Markus Engel OtherBill suggests: We at least seem to agree on eliminating
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm the <Container ~ foo> forms, and using only
019dcd82834b28a0204655ae1f3766a4c4d7cb7cpjrm <ContainerMatch foo> semantics.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * SIGSEGV on Linux (glibc 2.1.2) isn't caught properly by a
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen sigwaiting thread. We need to work around this, perhaps unless
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen there is hope soon for a fixed glibc.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * orig_ct in the byterange/multipart handling may not be
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen needed. Apache 1.3 just never stashed "multipart" into
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen r->content_type. We should probably follow suit since the
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen byterange stuff doesn't want the rest of the code to see the
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen multipart content-type; the other code should still think it is
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen dealing with the <orig_ct> stuff.
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Status: Greg volunteers to investigate (esp. since he was most
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen likely the one to break it :-)
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelenBinaries (2.0.35):
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Platform Avail. Volunteer
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen ------------------------------------------------------------------
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Mandrake 8.1 no Ryan Bloom
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen FreeBSD 4.1 yes Ryan Bloom
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen OS X 10.1.3/Darwin 5.3 no Jim Jagielski
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Solaris 8.x/sparc yes Jim Jagielski
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen RedHat 7.2 no Jim Jagielski
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen i686-pc-linux-gnu no Aaron Bannert
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Solaris 8.x/i386 no Aaron Bannert
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen powerpc-unknown-linux-gnu no Graham Leggett
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelenOther features that need writing:
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * Finish infrastructure in core for async MPMs
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Status: post 2.0
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * TODO in source -- just do an egrep on "TODO" and see what's there
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelenAvailable Patches:
620d9f76e90641a5a91f0c45fcb9a2f07b3881dbJohan Engelen * Martin Sojka <msojka@gmx.de>'s patch to add error reporting for failed
620d9f76e90641a5a91f0c45fcb9a2f07b3881dbJohan Engelen htpasswd actions due to a full /tmp volume (other programs may have
620d9f76e90641a5a91f0c45fcb9a2f07b3881dbJohan Engelen similar problems?)
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * Mike Abbott's <mja@trudge.engr.sgi.com> patches to improve
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen Status: These were written for 1.3, and are awaiting a port to
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * Jim Winstead's <jimw@trainedmonkey.com> patch to add CookieDomain and
f0adf2766388b21de0a6c429e39496cb4f95992cjohanengelen other small mod_usertrack features
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * Dan Rench's <drench@xnet.com> patch to add allow the errmsg and timefmt
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen of SSI's to be modified in the config file. Patch is available in
0563fd55cbad59e8a878e6d4cbbdd8e47f74488djohanengelen * Which MPMs will be included with Apache 2.0?