STATUS revision c517528fe4c49a6c5a1e1c66e79544092d39e15f
c330021bf3f45cbf187fa644781e67f7e470a58awroweAPACHE 2.0 STATUS: -*-text-*-
a3f7a6fb4aedec8585efd990cf99d3917b99a0d3jimLast modified at [$Date: 2002/02/06 18:32:03 $]
c330021bf3f45cbf187fa644781e67f7e470a58awrowe 2.0.32 : in development
2a6c49cfaef5979a5a06098f3ce987cd76769409manoj 2.0.31 : rolled Feburary 1, 2002.
af4c982a7cf4515f124935f99a329744035fc699slive 2.0.30 : tagged January 8, 2002. not rolled.
af4c982a7cf4515f124935f99a329744035fc699slive 2.0.29 : tagged November 27, 2001. not rolled.
af4c982a7cf4515f124935f99a329744035fc699slive 2.0.28 : released November 13, 2001
af4c982a7cf4515f124935f99a329744035fc699slive 2.0.27 : rolled November 6, 2001
af4c982a7cf4515f124935f99a329744035fc699slive 2.0.26 : tagged October 16, 2001. not rolled.
af4c982a7cf4515f124935f99a329744035fc699slive 2.0.25 : rolled August 29, 2001
af4c982a7cf4515f124935f99a329744035fc699slive 2.0.24 : rolled August 18, 2001
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe 2.0.23 : rolled August 9, 2001
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe 2.0.22 : rolled July 29, 2001
abdf88f96a56cf5c242d3195c493dbb8a7355d03wrowe 2.0.21 : rolled July 20, 2001
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe 2.0.20 : rolled July 8, 2001
10270f6f94b2069d0d357805c140a9897449b9ccianh 2.0.19 : rolled June 27, 2001
4b62424416882687387923b3130b96241503cbe0jerenkrantz 2.0.18 : rolled May 18, 2001
5ca8e11fadb6f7a8d9d0367c1800205c99d4bcd6jerenkrantz 2.0.17 : rolled April 17, 2001
5ca8e11fadb6f7a8d9d0367c1800205c99d4bcd6jerenkrantz 2.0.16 : rolled April 4, 2001
45b0e1c775c1cfed6473c9e5304179ccb9609f53stoddard 2.0.15 : rolled March 21, 2001
dbec4658981e4f9127e8676457c28d42932be7cdtrawick 2.0.14 : rolled March 7, 2001
1b7e9aeb9867a1c94e24e8ae491d2dfcd0d225e3jerenkrantz 2.0a9 : released December 12, 2000
f4cb04eb78da02a38fcdd87489dc7b660107d55fjerenkrantz 2.0a8 : released November 20, 2000
f4cb04eb78da02a38fcdd87489dc7b660107d55fjerenkrantz 2.0a7 : released October 8, 2000
a3f7a6fb4aedec8585efd990cf99d3917b99a0d3jim 2.0a6 : released August 18, 2000
a3f7a6fb4aedec8585efd990cf99d3917b99a0d3jim 2.0a5 : released August 4, 2000
a3f7a6fb4aedec8585efd990cf99d3917b99a0d3jim 2.0a4 : released June 7, 2000
a3f7a6fb4aedec8585efd990cf99d3917b99a0d3jim 2.0a3 : released April 28, 2000
a3f7a6fb4aedec8585efd990cf99d3917b99a0d3jim 2.0a2 : released March 31, 2000
f95a0b59eb24c631f15bd83e20c6cf823c432d83trawick 2.0a1 : released March 10, 2000
f95a0b59eb24c631f15bd83e20c6cf823c432d83trawickPlease consult the following STATUS files for information
c95a8116bde20ab75f2e54d0e867bdf2e13643c3jimon related projects:
f95a0b59eb24c631f15bd83e20c6cf823c432d83trawickRELEASE SHOWSTOPPERS:
ddafc111b94558ef4e2d7357ceda623315566ce3slive * 31 BETA STATUS:
ddafc111b94558ef4e2d7357ceda623315566ce3slive running on Daedalus since 02-Feb-2002 7:58 PST (need 3 days)
7718f3d5b4da70eb063877f5300ee361435910f4nd Compiles on : AIX 4.3, Solaris, FreeBSD 3.4 & 4.5, Win32,
ddafc111b94558ef4e2d7357ceda623315566ce3slive Linux 2.2 & 2.4, HPUX
93f189f1198f539d3cfa75a15b23dcde60ee35ffrbb Broken on: Win32 [no error logging within service, other bugs]
bca5b27d271b6e1690134a83963424b9825d93bdstriker +1 : Aaron, Jim
bca5b27d271b6e1690134a83963424b9825d93bdstriker +0 : Lars, Justin, trawick
7718f3d5b4da70eb063877f5300ee361435910f4nd -0 : OtherBill
13402b2193f52031b2acfbee2b0965e02f3f29b4wrowe -1 : BillS, Ian, gregames, BrianP
13402b2193f52031b2acfbee2b0965e02f3f29b4wrowe bumps since original tag:
38d2c5d41cdb5eb28668d0290b59f8c76ae2a4bfjim * mod-dir patch
38d2c5d41cdb5eb28668d0290b59f8c76ae2a4bfjim * scoreboard x2 : 1 to fix gracefull restarts
38d2c5d41cdb5eb28668d0290b59f8c76ae2a4bfjim 1 to fix netware
93f189f1198f539d3cfa75a15b23dcde60ee35ffrbb * win32/locks.c : to fix mod_rewrite on win32
f4cb04eb78da02a38fcdd87489dc7b660107d55fjerenkrantz problems with v31:
f4cb04eb78da02a38fcdd87489dc7b660107d55fjerenkrantz * libtool/binbuild on AIX -- possible addition of patched
f4cb04eb78da02a38fcdd87489dc7b660107d55fjerenkrantz binbuild.sh to 31-beta roll
f4cb04eb78da02a38fcdd87489dc7b660107d55fjerenkrantz +1: Ian, Justin
f4cb04eb78da02a38fcdd87489dc7b660107d55fjerenkrantz Jeff says: We can't do anything about libtool since AIX
7718f3d5b4da70eb063877f5300ee361435910f4nd needs a version that won't work on some
38d2c5d41cdb5eb28668d0290b59f8c76ae2a4bfjim platforms. Handle this in the README.
38d2c5d41cdb5eb28668d0290b59f8c76ae2a4bfjim The binbuild issue isn't AIX. It is
38d2c5d41cdb5eb28668d0290b59f8c76ae2a4bfjim something that could happen anywhere that
7718f3d5b4da70eb063877f5300ee361435910f4nd the binbuild-er has their own expat.
b9b0788ea79de3ee230ccb0abf93a0c2c08f6287lars Handle this by patching binbuild.sh for a
b9b0788ea79de3ee230ccb0abf93a0c2c08f6287lars beta roll or putting a patch in the README for
f4cb04eb78da02a38fcdd87489dc7b660107d55fjerenkrantz use by people who want to do binbuild but
f4cb04eb78da02a38fcdd87489dc7b660107d55fjerenkrantz have expat installed locally.
7718f3d5b4da70eb063877f5300ee361435910f4nd gregames: why can't we roll a second tarball with the
1af5c0e25a649bb298e25ddfa5418fa18bdcb107aaron appropriate libtool version? doesn't
38d2c5d41cdb5eb28668d0290b59f8c76ae2a4bfjim Darwin need it as well as AIX?
f4cb04eb78da02a38fcdd87489dc7b660107d55fjerenkrantz * erroneous check in an AP_DEBUG_ASSERT() call. Only happens
1b3f48fd6b1ccb8745f908e40156c5a85ca3c347jerenkrantz in maintainer mode. Fixed in modules/http/http_protocol.c
b05930e6008f69bd323abe0c10f81f40ffd27983brianp revision 1.391. Add item in release notes:
016f2545c9375ec7fc5e9cb70aa1ae0cace83c98jerenkrantz +1: Justin, Cliff, Aaron, BillS, Jim, trawick
016f2545c9375ec7fc5e9cb70aa1ae0cace83c98jerenkrantz * seg faults in core_input_filter when the client goes away
016f2545c9375ec7fc5e9cb70aa1ae0cace83c98jerenkrantz before any POST body bytes are received.
016f2545c9375ec7fc5e9cb70aa1ae0cace83c98jerenkrantz Jeff committed a fix with server/core.c revision 1.144.
016f2545c9375ec7fc5e9cb70aa1ae0cace83c98jerenkrantz Justin committed a fix with server/protocol.c revision 1.78
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe (server/protocol.c revision 1.81 demotes a potentially
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe annoying error message)
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe * FirstBill reports problem [re]starting as-a-service, shared
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe score is suspect. OtherBill is investigating... has found
b3b4e853e4958357ee2d50e2fe41effecfde9eedwrowe . created restart and shutdown events, only restart
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe survived initialization on XP, although breaking into
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe the debugger interferes with reproducing the bug.
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe Perhaps in FirstBill's example _restart didn't survive.
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe . Scoreboard appears irrelevant to the problem.
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe . This looks like handle corruption in NT/XP
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe . Args are not initialized correctly when the -k install
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe OtherBill will not create Win32 binaries due to this bug.
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe * mod_auth_dbm can't open a Berkeley DB password file on Unix
c330021bf3f45cbf187fa644781e67f7e470a58awrowe Justin postulates that this might be related to the fact
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe that mod_auth_dbm wasn't using apr-util. See
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe modules/aaa/mod_auth_dbm.c revision 1.42 and
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe module/aaa/config.m4 revision 1.54. Also, FreeBSD's DB
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe variant was not properly detected. See
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe apr-util/build/apu-conf.m4 revision 1.31 and
56b2562cb4041a21bc313fde3204beabdb758800jwoolley * Madhu reports that the worker MPM doesn't shutdown nicely
56b2562cb4041a21bc313fde3204beabdb758800jwoolley after a stress test.
56b2562cb4041a21bc313fde3204beabdb758800jwoolley * mod_autoindex displays the wrong icon for subdirectories
f881b8be216de36c6b964324c772450bca38a4e6trawick * anon shared memory not allocating enough
f881b8be216de36c6b964324c772450bca38a4e6trawick * other BETA possibilities
f881b8be216de36c6b964324c772450bca38a4e6trawick 31 + fixes for all non-cosmetic problems seen on daedalus
56b2562cb4041a21bc313fde3204beabdb758800jwoolley See: http://www.apache.org/~jerenkrantz/httpd-2.0.jre.patch
b05930e6008f69bd323abe0c10f81f40ffd27983brianp +1: trawick, Justin
b05930e6008f69bd323abe0c10f81f40ffd27983brianp * ap_directory_walk skips some per-dir config merge functions
4c7dab038d90d7feb67ef8ddbfacc77be8c9dbf0jwoolley if there is no "<Directory />" block in the configuration
4c7dab038d90d7feb67ef8ddbfacc77be8c9dbf0jwoolley Message-ID: <m3itbdiijq.fsf@rdu163-40-092.nc.rr.com>
4c7dab038d90d7feb67ef8ddbfacc77be8c9dbf0jwoolley * That is very unlikely, merges are additive. Much more likely,
4c7dab038d90d7feb67ef8ddbfacc77be8c9dbf0jwoolley the default SetOutputFilter default or merge is borked.
bb2a72a6e2327ae4f177c9d26e9d433033cfc27eaaron Unless it's the code that permits 'null' merges, per module.
8c8fbb8546af54582539898be704411a60058d85trawick Still, it's probably in SetOutputFilter's behavior.
8c8fbb8546af54582539898be704411a60058d85trawick * BrianP notes: directory_walk's handling of trailing slashes
8c8fbb8546af54582539898be704411a60058d85trawick appears to be the cause: http://www.apachelabs.org/apache-mbox/200112.mbox/%3c3C1CF721.1090300@pacbell.net%3e
f9b8e29cfca92cf0a996e8ab17fa1a1f447cecc7stoddard * Test suite failures:
f9b8e29cfca92cf0a996e8ab17fa1a1f447cecc7stoddard o worker is also failing some of the 'cgi' subtests
a23b1c5a74208b03884c09a6f9dd5d6c97fa6415trawick (see <URL:http://Source-Zone.Org/Apache/regression/>):
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe Justin says: "Worker should be fine and passes httpd-test here.
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe If you can provide evidence that it can be reproduced
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe outside of httpd-test, then it's a showstopper. I
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe think it's a perl or a httpd-test problem."
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe Not a showstopper: Justin
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe * If any request gets to the core handler, without a flag that this
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe r->filename was tested by dir/file_walk, we need to 500 at the very
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe end of the ap_process_request_internal() processing. This provides
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe authors of older modules better compatibility, while still improving
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe the security and robustness of 2.0.
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe Status: still need to decide where this goes, OtherBill comments...
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe Message-ID: <065701c14526$495203b0$96c0b0d0@roweclan.net>
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe we need to look at halting this in the 'default handler' case,
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe and that implies pushing the 'handler election' into the request
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe internal processing phase from the run request phase.
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe * Convert all instances of the old apr_lock_t type to the new
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe types (once they are fully supported in APR).
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe Status: Aaron is working on converting INTRAPROCESS
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe to apr_thread_mutex_t types. Full replacements for
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe LOCKALL and CROSS_PROCESS are not yet complete on all
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe platforms, and should only be used in MPMs like worker
db2d668e6233d8949b35ee7f9f42f444758f9ce9rbb with limited OS exposure.
db2d668e6233d8949b35ee7f9f42f444758f9ce9rbb BrianP asks: "Is this really a showstopper?"
db2d668e6233d8949b35ee7f9f42f444758f9ce9rbb IanH says: "If we don't do it before we go live we will have 2
db2d668e6233d8949b35ee7f9f42f444758f9ce9rbb different API's to support in APR"
db2d668e6233d8949b35ee7f9f42f444758f9ce9rbb * A binbuild installation picks up the right libraries when
db2d668e6233d8949b35ee7f9f42f444758f9ce9rbb running apachectl because we set the appropriate environment
5d12baef135b5d3cb94745e007a1575398469724jerenkrantz variable, but ab, htpasswd, etc. don't know how to pick up apr,
5d12baef135b5d3cb94745e007a1575398469724jerenkrantz apr-util, etc.
5d12baef135b5d3cb94745e007a1575398469724jerenkrantz Message-ID: <20020116000226.GA15991@ebuilt.com>
5d12baef135b5d3cb94745e007a1575398469724jerenkrantz Justin says: "You could always build all of the binaries
5d12baef135b5d3cb94745e007a1575398469724jerenkrantz statically if you don't want to let libtool
5d12baef135b5d3cb94745e007a1575398469724jerenkrantz handle it. At most, I don't consider this a
5d12baef135b5d3cb94745e007a1575398469724jerenkrantz a release showstopper as it is a packaging issue."
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe Ken sez: "Showstopper because, until it's the default, it
5d12baef135b5d3cb94745e007a1575398469724jerenkrantz violates the P of LA. Binaries will give strange
b865daaa4ef731a7066ee6d97e2aae36c7743939jerenkrantz errors and not run if moved."
b865daaa4ef731a7066ee6d97e2aae36c7743939jerenkrantz Status: Aaron volunteers: I'll commit a patch to allow all
b865daaa4ef731a7066ee6d97e2aae36c7743939jerenkrantz the support binaries to be built statically against
f881b8be216de36c6b964324c772450bca38a4e6trawick libapr/libaprutil/etc and then enable it in binbuild.
f881b8be216de36c6b964324c772450bca38a4e6trawick gregames says: what about httpd? yeah, it works at the moment
f881b8be216de36c6b964324c772450bca38a4e6trawick if you use apachectl. But I've already heard
dc098c7ce5d36179c504d09fc722d190683d0262aaron complaints about not being able to execute
dc098c7ce5d36179c504d09fc722d190683d0262aaron httpd directly.
dc098c7ce5d36179c504d09fc722d190683d0262aaronRELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe * Usage of APR_BRIGADE_NORMALIZE in core_input_filter should be
dc098c7ce5d36179c504d09fc722d190683d0262aaron removed if possible.
0bcb1fe39dfaacf9745b6633f5cc9ebc8e2596caaaron Message-ID: <Pine.LNX.4.33.0201202232430.318-100000@deepthought.cs.virginia.edu>
0bcb1fe39dfaacf9745b6633f5cc9ebc8e2596caaaron * There is a bug in how we sort some hooks, at least the pre-config
0bcb1fe39dfaacf9745b6633f5cc9ebc8e2596caaaron hook. The first time we call the hooks, they are in the correct
0bcb1fe39dfaacf9745b6633f5cc9ebc8e2596caaaron order, but the second time, we don't sort them correctly. Currently,
0bcb1fe39dfaacf9745b6633f5cc9ebc8e2596caaaron the modules/http/config.m4 file has been renamed to
33f5961d34a8b5390cebad0543b3ebe67830e5d7jerenkrantz modules/http/config2.m4 to work around this problem, it should moved
33f5961d34a8b5390cebad0543b3ebe67830e5d7jerenkrantz back when this is fixed. rbb
33f5961d34a8b5390cebad0543b3ebe67830e5d7jerenkrantz Justin says: "Is this really a showstopper? This has been here
33f5961d34a8b5390cebad0543b3ebe67830e5d7jerenkrantz forever. What's wrong? Does this have to do with
33f5961d34a8b5390cebad0543b3ebe67830e5d7jerenkrantz autoconf or m4?"
33f5961d34a8b5390cebad0543b3ebe67830e5d7jerenkrantz Not a showstopper: Justin, BrianP, trawick, gregames
ff42f83cbf31893bcde9712332a8e5ee970f6a74trawick * The Add...Filter and Set...Filter directives do not allow the
ff42f83cbf31893bcde9712332a8e5ee970f6a74trawick administrator to order filters, beyond the order of filename (mime)
33f5961d34a8b5390cebad0543b3ebe67830e5d7jerenkrantz extensions. It isn't clear if Set...Filter(s) should be inserted
54e1babd5a5a56c576eeeace54110150769cc916coar before or after the Add...Filter(s) which are ordered by sequence of
54e1babd5a5a56c576eeeace54110150769cc916coar filename extensions. At minimum, some sort of +-[0-10] syntax seems
54e1babd5a5a56c576eeeace54110150769cc916coar like the quickest fix for a 2.0 gold release.
54e1babd5a5a56c576eeeace54110150769cc916coar Justin says: "Could we delay this for a point release or 2.1?"
54e1babd5a5a56c576eeeace54110150769cc916coar Not a showstopper: justin, wrowe, trawick, stoddard, Jim, Ian, Aaron,
54e1babd5a5a56c576eeeace54110150769cc916coar * Should we always build binaries statically unless otherwise
54e1babd5a5a56c576eeeace54110150769cc916coar indicated?
54e1babd5a5a56c576eeeace54110150769cc916coar Message-ID: <20020129210006.B23512@Lithium.MeepZor.Com>
54e1babd5a5a56c576eeeace54110150769cc916coar -1: Justin, Ian
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe * If the parent process dies, should the remaining child processes
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe "gracefully" self-terminate. Or maybe we should make it a runtime
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe option, or have a concept of 2 parent processes (one being a
949aa7bba7f804faa8e6b08cad42a98fc0255d85jerenkrantz "hot spare").
949aa7bba7f804faa8e6b08cad42a98fc0255d85jerenkrantz See: Message-ID: <3C58232C.FE91F19F@Golux.Com>
949aa7bba7f804faa8e6b08cad42a98fc0255d85jerenkrantz Self-destruct: Ken, Martin
949aa7bba7f804faa8e6b08cad42a98fc0255d85jerenkrantz Not self-destruct: BrianP, Ian, Cliff, BillS
949aa7bba7f804faa8e6b08cad42a98fc0255d85jerenkrantz Make it runtime configurable: Aaron, Jim, Justin
e0427bf8e52a8fb920cb8b6adb5cdb3b6535b7fecoar Have 2 parents: +1: Jim
e0427bf8e52a8fb920cb8b6adb5cdb3b6535b7fecoar +0: Martin (while standing by, could it do
07021d9f405849228b859d9fb4b877f20e4fbba3jerenkrantz something useful?)
f126ee03179eb54308118f1ec3de5a7b461685d8aaron look at accept() error handling first: gregames
f126ee03179eb54308118f1ec3de5a7b461685d8aaron * Make some modifications to the scoreboard creation routines,
f126ee03179eb54308118f1ec3de5a7b461685d8aaron prefer anonymous shared memory, and allow a configuration
f126ee03179eb54308118f1ec3de5a7b461685d8aaron directive to override the defaults.
f126ee03179eb54308118f1ec3de5a7b461685d8aaron Message-ID: <20020130080804.C16977@clove.org>
f126ee03179eb54308118f1ec3de5a7b461685d8aaron Status: Aaron volunteers
f126ee03179eb54308118f1ec3de5a7b461685d8aaron * Get perchild to work on platforms other than Linux. This
f126ee03179eb54308118f1ec3de5a7b461685d8aaron will require a portable mechanism to pass data and file/socket
f126ee03179eb54308118f1ec3de5a7b461685d8aaron descriptors between vhost child groups. An API was proposed
f126ee03179eb54308118f1ec3de5a7b461685d8aaron on dev@apr:
f126ee03179eb54308118f1ec3de5a7b461685d8aaron Message-ID: <20020111115006.K1529@clove.org>
f126ee03179eb54308118f1ec3de5a7b461685d8aaron Vote: Is a non-portable perchild going to hold up a GA release?
109faf633e12ab0bbdd602c7addc795cce59e8addreid No: Aaron, Justin, trawick, stoddard, Jim, Ian, BrianP, gregames
109faf633e12ab0bbdd602c7addc795cce59e8addreid * Recent changes to ap_rgetline may have broken EBCDIC boxes.
109faf633e12ab0bbdd602c7addc795cce59e8addreid Message-ID: <20020122072605.GF28051@ebuilt.com>
109faf633e12ab0bbdd602c7addc795cce59e8addreid Justin says: "I don't have an EBCDIC box to test on. A potential
109faf633e12ab0bbdd602c7addc795cce59e8addreid solution is to split out ap_rgetline into two
109faf633e12ab0bbdd602c7addc795cce59e8addreid functions as described in this message."
109faf633e12ab0bbdd602c7addc795cce59e8addreid gregames says: I see the breakage now, and volunteer to fix it
109faf633e12ab0bbdd602c7addc795cce59e8addreid when things calm down a little. It looks OK when
109faf633e12ab0bbdd602c7addc795cce59e8addreid there are complete lines and no mime continuations.
109faf633e12ab0bbdd602c7addc795cce59e8addreid * Modify the worker MPM so that it doesn't need to create and
109faf633e12ab0bbdd602c7addc795cce59e8addreid destroy a pool for each request--possibly by adopting a
109faf633e12ab0bbdd602c7addc795cce59e8addreid leader/follower model in which each worker owns a persistent
4ca13a5e126946272f02637e268a8e09193c553ecoar ptrans pool (like the prefork MPM) and the workers take
4ca13a5e126946272f02637e268a8e09193c553ecoar turns acting as listeners...this approach might also help
4ca13a5e126946272f02637e268a8e09193c553ecoar reduce context-switching
4ca13a5e126946272f02637e268a8e09193c553ecoar * CGI single-byte reads
b3b4e853e4958357ee2d50e2fe41effecfde9eedwrowe BrianP suggests that this is caused by the ap_scan_script_header_err()
b3b4e853e4958357ee2d50e2fe41effecfde9eedwrowe routine, which will do single-byte reads until it finds the end
48c0c81cd6fabac9d3386406d97633780365b839coar of the header, at which point it constructs a pipe-bucket (buffered)
48c0c81cd6fabac9d3386406d97633780365b839coar to read from.
48c0c81cd6fabac9d3386406d97633780365b839coar Proposed solution in:
48c0c81cd6fabac9d3386406d97633780365b839coar Message-ID: <3C36ADAF.60601@cnet.com>
4ca13a5e126946272f02637e268a8e09193c553ecoar * Try to get libtool inter-library dependency code working on AIX.
b84f66c93f820824b1d5455181f55598b766319cwrowe Message-ID: <cm3n10lx555.fsf@rdu163-40-092.nc.rr.com>
b84f66c93f820824b1d5455181f55598b766319cwrowe Justin says: If we get it working on AIX, we can enable this
7fe18c15b669db9d191859695901dc4fcf3829dawrowe on all platforms and clean up our build system
7fe18c15b669db9d191859695901dc4fcf3829dawrowe Jeff says: I thought I tested a patch for you sometime in
cc22a72861c58dda7f3768613aec864e4c4e0353striker January that you were going to commit within a few
cc22a72861c58dda7f3768613aec864e4c4e0353striker * Handling of %2f in URIs. Currently both 1.3 and 2.0
60d567a0c2aae815ee6fc20c0d65032bea52c92cwrowe completely disallow %2f in the request URI path (see
60d567a0c2aae815ee6fc20c0d65032bea52c92cwrowe ap_unescape_url() in util.c). It's permitted and passed
60d567a0c2aae815ee6fc20c0d65032bea52c92cwrowe through in the query string, however. Roy says the
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe original reason for disallowing it, from five years ago,
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe was to protect CGI scripts that applied PATH_INFO to
f610c7c704235bc327dbe9b62982f5b3f8e30a77wrowe a filesystem location and which might be tricked by
2fb49a1d25f38421a68d31b4cbb5d9293fdeafbewrowe ..%2f..%2f(...). We *should* allow path-info of the
60d567a0c2aae815ee6fc20c0d65032bea52c92cwrowe Since we've revamped a lot of our processing of path
19cbe4d7b7c931723e7249de6829bf965a1fee72stoddard segments, it would be nice to allow this, or at least
19cbe4d7b7c931723e7249de6829bf965a1fee72stoddard allow it conditionally with a directive.
b187d568e1507d75139ebc13ca945b38fc05d55cstoddard * FreeBSD, threads, and worker MPM. All seems to work fine
b187d568e1507d75139ebc13ca945b38fc05d55cstoddard if you only have one worker process with many threads. Add
a5ed555df952c85bc1b179f5981e8a6c54ba16e6stoddard a second worker process and the accept lock seems to be
d2f8b010487ffa990a9c268df5a25579e7291bcdrbb lost. This might be an APR issue with how it deals with
d2f8b010487ffa990a9c268df5a25579e7291bcdrbb the child_init hook (i.e. the fcntl lock needs to be resynced).
a5ed555df952c85bc1b179f5981e8a6c54ba16e6stoddard More examination and analysis is required.
0bff2f28ef945280c17099c142126178a78e1e54manoj Status: This has also been reported on Cygwin.
0bff2f28ef945280c17099c142126178a78e1e54manoj Message-ID: <3C2CC514.8EF3BED1@wapme-systems.de> (cygnus)
ff849e4163ed879288f0df15f78b6c9d278ec804fanf Justin says: So, FreeBSD-CURRENT and Cywin have the same
ff849e4163ed879288f0df15f78b6c9d278ec804fanf problem. Yum. If another platform has this
447c6ce3ff08073c44f6785d5256271fcb877512wrowe with worker, this becomes a showstopper.
e0427bf8e52a8fb920cb8b6adb5cdb3b6535b7fecoar Aaron says: I spent some time disecting this and have come to
e0427bf8e52a8fb920cb8b6adb5cdb3b6535b7fecoar the conclusion that it is not a problem in the worker MPM
e0427bf8e52a8fb920cb8b6adb5cdb3b6535b7fecoar (or at least, it is not isolated to a problem in worker).
e0427bf8e52a8fb920cb8b6adb5cdb3b6535b7fecoar I'll list some of the problems I'm seeing in case someone
6758b07b4b79f898b0f56375016cea7da0bfb495wrowe else wants to pick up where I've left off:
6758b07b4b79f898b0f56375016cea7da0bfb495wrowe - Delivery of just about any signal to one of the child
6758b07b4b79f898b0f56375016cea7da0bfb495wrowe processes will send it into an infinite loop as well.
7fe18c15b669db9d191859695901dc4fcf3829dawrowe - Even though the parent is spinning out of control,
db3ccce11afac4fc1d4f51a65424412f7480c46cgstein at first the child or children will appear to work
dd4713dc5b186f4d1be7b88f86608fdb84cbe5d5gstein properly. At times it is possible to get it into a state,
48c0c81cd6fabac9d3386406d97633780365b839coar however, where a request will hang until another concurrent
48c0c81cd6fabac9d3386406d97633780365b839coar request "kicks" the first, at which point the second will
48c0c81cd6fabac9d3386406d97633780365b839coar hang. My theory is that this has to do with the
48c0c81cd6fabac9d3386406d97633780365b839coar pthread_cond_*() implementation in FreeBSD, but it's still
db3ccce11afac4fc1d4f51a65424412f7480c46cgstein possible that it is in APR.
79d5106a9b65b956d646f5daae4b94bc79e315b8trawick Justin adds: Oh, FreeBSD threads are implemented entirely with
79d5106a9b65b956d646f5daae4b94bc79e315b8trawick select()/poll()/longjmp(). Welcome to the nightmare.
cf6bf6c34c936e6a6fe731dbce4a5c3c8bf8e9a3gstein So, that means a ktrace output also has the thread
cf6bf6c34c936e6a6fe731dbce4a5c3c8bf8e9a3gstein scheduling internals in it (since it is all the same to
cf6bf6c34c936e6a6fe731dbce4a5c3c8bf8e9a3gstein the kernel). Which makes it hard to distinguish between
b3b4e853e4958357ee2d50e2fe41effecfde9eedwrowe our select() calls and their select() calls.
93c5cba06b623ebe8e4372e886eece12d9a80c3egstein *bangs head on wall repeatedly* But, some of the libc_r
14cccaddba3a9263cf0d0ddc311e18f3e3dc9b0fgstein files have a DBG_MSG #define. This is moderately helpful
14cccaddba3a9263cf0d0ddc311e18f3e3dc9b0fgstein when used with -DNO_DETACH. The kernel scheduler isn't
14cccaddba3a9263cf0d0ddc311e18f3e3dc9b0fgstein waking up the threads on a select(). Yum. And, I bet
14cccaddba3a9263cf0d0ddc311e18f3e3dc9b0fgstein those decrementing select calls have to do with the
14cccaddba3a9263cf0d0ddc311e18f3e3dc9b0fgstein scheduler. Time to brush up on our OS fundamentals.
823c303d33c9e637a83d82208bcbafaf5f532d7bgstein * There is increasing demand from module writers for an API
823c303d33c9e637a83d82208bcbafaf5f532d7bgstein that will allow them to control the server � la apachectl.
e636eba7474e0010b5c7198af1c2fe5ad8652dbbmanoj Reasons include sole-function servers that need to die if
e636eba7474e0010b5c7198af1c2fe5ad8652dbbmanoj an external dependency (e.g., a database) fails, et cetera.
e636eba7474e0010b5c7198af1c2fe5ad8652dbbmanoj Perhaps something in the (ever more abused) scoreboard?
281da4c02cf40c663298ded7e4e5b913a8f8b814gstein rbb: I don't believe the scoreboard is the correct mechanism
281da4c02cf40c663298ded7e4e5b913a8f8b814gstein for this. We already have a pipe that goes between parent
2f728b2e8555fee1b7cc11e886488692f2575fbddougm and child for graceful shutdown events, along with an API that
2f728b2e8555fee1b7cc11e886488692f2575fbddougm can be used to send a message down that pipe. In threaded MPMs,
2f728b2e8555fee1b7cc11e886488692f2575fbddougm it is easy enough to make that one pipe be used for graceful
60d567a0c2aae815ee6fc20c0d65032bea52c92cwrowe and graceless events, and it is also easy to open that pipe
48c0c81cd6fabac9d3386406d97633780365b839coar to both parent and child for writing. Then we just need to figure
60d567a0c2aae815ee6fc20c0d65032bea52c92cwrowe out how to do graceless on non-threaded MPMs.
60d567a0c2aae815ee6fc20c0d65032bea52c92cwrowe * revamp the input filter behavior, per discussions since
60d567a0c2aae815ee6fc20c0d65032bea52c92cwrowe February (and especially at the hackathon last
60d567a0c2aae815ee6fc20c0d65032bea52c92cwrowe April). Specifically, ap_get_brigade will return a brigade with
60d567a0c2aae815ee6fc20c0d65032bea52c92cwrowe *up to* a specific number of bytes, or a "line" of data. The
27757f6699a924d4b493a1b6cceb27df27a43287dreid read may be blocking or nonblocking. ap_getline() will be
b3b4e853e4958357ee2d50e2fe41effecfde9eedwrowe refactored into apr_brigade_getline(), and then DECHUNK can use
27757f6699a924d4b493a1b6cceb27df27a43287dreid f->next (ap_getline will always read "top of input stack"). Also
21e01f13f717faeca3e498d7d9c9b4d3af98ae27trawick fix the bug where request body content will end up closing the
21e01f13f717faeca3e498d7d9c9b4d3af98ae27trawick connection (buggering up persistent conns).
21e01f13f717faeca3e498d7d9c9b4d3af98ae27trawick Status: Justin is working on this as fast as he can.
21e01f13f717faeca3e498d7d9c9b4d3af98ae27trawick The core input filters, HTTP-related filters, mod_ssl,
21e01f13f717faeca3e498d7d9c9b4d3af98ae27trawick mod_proxy, and ap_[r]getline are switched to the new logic.
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard - socket bucket and core input filter changes. see end of
0a5d24f1bcf38baf13b60c1a021057a9979ce4fbcoar message ID (Feb 27): <20010227075326.S2297@lyra.org>
0a5d24f1bcf38baf13b60c1a021057a9979ce4fbcoar - fix up ap_get_brigade() semantics, fix bug in DECHUNK /
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe ap_getline. many messages (plus their threads) (Apr/May):
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe Message-ID: <20010402101207.J27539@lyra.org>
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe Message-ID: <3AF7F921.D2EEC41A@algroup.co.uk>
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe Message-ID: <20010508190029.E18404@lyra.org>
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe - further work with combining/tweaking the builtin filters:
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe Message-ID: <20010509115445.D1374@lyra.org>
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe - thoughts on filter modes:
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe Message-ID: <021b01c14dee$09782af0$93c0b0d0@roweclan.net>
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe - proposal for getline rewrite:
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe Message-ID: <20011231104019.GD3904@ebuilt.com>
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe - patch for getline rewrite:
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe Message-ID: <20020122072605.GF28051@ebuilt.com>
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe - Change ap_get_brigade prototype:
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe Message-ID: <20020120115136.GI17601@ebuilt.com>
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe * Allow the DocumentRoot directive within <Location > scopes? This
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe allows the beloved (crusty) Alias /foo/ /somepath/foo/ followed
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe by a <Directory /somepath/foo> to become simply
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe <Location /foo/> DocumentRoot /somefile/foo (IMHO a bit more legible
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe and in-your-face.) DocumentRoot unset would be accepted [and would
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe not permit content to be served, only virtual resources such as
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe server-info or server-status.
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe This proposed change would _not_ depricate Alias.
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe * Win32: Rotatelogs sometimes is not terminated when Apache
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe goes down hard. FirstBill was looking at possibly tracking the
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe child's-child processes in the parent process.
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe OtherBill asks, wasn't this fixed?
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe stoddard: Not fixed. Shared scoreboard might offer a good
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe way for the parent to keep track of 'other child' processes
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe and whack them if the child goes down.
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe * Win32: Add a simple hold console open patch (wait for close or
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe the ESC key, with a nice message) if the server died a bad
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe death (non-zero exit code) in console mode.
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe Resolution: bring forward same ugly hacks from 1.3.13-.20
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe * Port of mod_ssl to Apache 2.0:
afbd720d176856630fed7c6576cdd3ae25a407edstoddard The current porting state is summarized in modules/ssl/README. The
afbd720d176856630fed7c6576cdd3ae25a407edstoddard remaining work includes:
afbd720d176856630fed7c6576cdd3ae25a407edstoddard (1) stablizing/optimizing the SSL filter logic
afbd720d176856630fed7c6576cdd3ae25a407edstoddard (2) Enabling the various SSL caching mechanisms (shmcb, shmht)
afbd720d176856630fed7c6576cdd3ae25a407edstoddard (3) Enabling SSL extentions
afbd720d176856630fed7c6576cdd3ae25a407edstoddard (4) Trying to seperate the https filter logic from mod_ssl -
afbd720d176856630fed7c6576cdd3ae25a407edstoddard This is to facilitate other modules that wish to use the https
afbd720d176856630fed7c6576cdd3ae25a407edstoddard filter or the mod_ssl logic or both as required.
afbd720d176856630fed7c6576cdd3ae25a407edstoddard Justin: mod_ssl filter logic is redone, so that should be fine.
afbd720d176856630fed7c6576cdd3ae25a407edstoddard Madhu has submitted a patch for SSL caching - however, I
afbd720d176856630fed7c6576cdd3ae25a407edstoddard am -0 on that patch as I *think* we could implement the
afbd720d176856630fed7c6576cdd3ae25a407edstoddard shared memory another way that is much cleaner (i.e.
afbd720d176856630fed7c6576cdd3ae25a407edstoddard treat shmem directly as a dbm via APR routines). Justin
afbd720d176856630fed7c6576cdd3ae25a407edstoddard also thinks that the https filter logic may be sufficiently
28c4fe67d75f8f26504d75b7aa8dc5d868032888wrowe decoupled now, but isn't really sure.
f881b8be216de36c6b964324c772450bca38a4e6trawick * Performance & Debug: Eliminate most (and perhaps all) of the
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard malloc/free calls in the bucket brigade code. Need some
6f1e2a1eb9944358dc96ee52f2048377c57f1cfaaaron light weight memory management functions that allow freeing
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard memory (putting it back into a memory pool) when it is no
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard longer needed. Enabling simple debugging features like guard
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard bands, double free detection, etc. would be cool but certainly
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard not a hard requirement.
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard Status: Cliff started to implement this using SMS as has
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard been discussed at length for months, but since
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard SMS is not being used anywhere else in the server,
afbd720d176856630fed7c6576cdd3ae25a407edstoddard several people expressed the opinion that we should
afbd720d176856630fed7c6576cdd3ae25a407edstoddard get rid of it entirely, meaning that the buckets
afbd720d176856630fed7c6576cdd3ae25a407edstoddard need their own memory management (free list) functions.
afbd720d176856630fed7c6576cdd3ae25a407edstoddard Cliff will implement that this weekend so we at least
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard have something to look at/compare with.
c408915e2a85fedc520e97e96abda8f61aafc804stoddard * Eliminate unnecessary creation of pipes in mod_cgid
c408915e2a85fedc520e97e96abda8f61aafc804stoddard * the autoconf setup should be fixed to default to using the
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard "Apache" layout from config.layout, and each variable settable
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard in a layout should be overridable on the command line. Plus,
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard what we do right now just doesn't seem to fully fit into how autoconf
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard works, eg. AC_PREFIX_DEFAULT issues.
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard Message-ID: <Pine.BSF.4.20.0104031557420.20876-100000@alive.znep.com>
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard * Combine log_child and piped_log_spawn. Clean up http_log.c.
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard Common logging API.
ddd44b06b04507cae083c52451e28f54f0bdb5afstoddard * Document mod_file_cache.
a32ddb6cabb635ca29ccff87f8bf5c6af3b78ecfstoddard * Platforms that do not support fork (primarily Win32 and AS/400)
a32ddb6cabb635ca29ccff87f8bf5c6af3b78ecfstoddard Architect start-up code that avoids initializing all the modules
783e9d9e55a1ecb3c0eedb14493667ed642191c5rederpj in the parent process on platforms that do not support fork.
75f8e1cae5ca3a16a7400cdddf604815ab06b5a8rederpj * Win32: Migrate the MPM over to use APR thread/process calls. This
75f8e1cae5ca3a16a7400cdddf604815ab06b5a8rederpj would eliminate some code in the Win32 branch that essentially
75f8e1cae5ca3a16a7400cdddf604815ab06b5a8rederpj duplicates what is in APR.
951342b69609b2b45303f89f91275676d5325217bnicholes * There are still a number of places in the code where we are
951342b69609b2b45303f89f91275676d5325217bnicholes losing error status (i.e. throwing away the error returned by a
951342b69609b2b45303f89f91275676d5325217bnicholes system call and replacing it with a generic error code)
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * Mass vhosting version of suEXEC.
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * All DBMs suffer from confusion in support/dbmmanage (perl script) since
7787343a549c7309035b3b7836e00edb6e3a915ewrowe the dbmmanage employs the first-matched dbm format. This is not
64ad864fa0f4493eebb181e393b40a8a90beccb9coar necessarily the library that Apache was built with. Aught to
7787343a549c7309035b3b7836e00edb6e3a915ewrowe rewrite dbmmanage upon installation to bin/ with the proper library
48c0c81cd6fabac9d3386406d97633780365b839coar for predictable mod_auth_dbm administration.
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Questions; htdbm exists, time to kill dbmmanage, or does it remain
7787343a549c7309035b3b7836e00edb6e3a915ewrowe useful as a perl dbm management example? If we keep it,
7787343a549c7309035b3b7836e00edb6e3a915ewrowe do we address the issue above?
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * use apu_dbm in mod_auth_dbm
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Status: Greg +1 (low-priority volunteer)
48c0c81cd6fabac9d3386406d97633780365b839coar Justin says: "Seems like this is already there, so should we just
64ad864fa0f4493eebb181e393b40a8a90beccb9coar remove the other DBM code in that file? If you want
7787343a549c7309035b3b7836e00edb6e3a915ewrowe to use gdbm, or dbm, etc, you should tell apr-util."
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Will says: "bs - I may choose the fastest - most efficient native
7787343a549c7309035b3b7836e00edb6e3a915ewrowe dbm implementation, for shared proc caches, ssl session
48c0c81cd6fabac9d3386406d97633780365b839coar caching, etc, but that has nothing to do with maintaining
64ad864fa0f4493eebb181e393b40a8a90beccb9coar a userlist via dbm, which has to remain readable between
7787343a549c7309035b3b7836e00edb6e3a915ewrowe builds/machines, etc. The use-multiple database schema
48c0c81cd6fabac9d3386406d97633780365b839coar for apr-util would let us do this with just apr, though."
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Ian says: "multi-dbm is in, but it still has ndbm support hardcoded
7787343a549c7309035b3b7836e00edb6e3a915ewrowe is this still required? isn't ndbm supported via gdbm?"
48c0c81cd6fabac9d3386406d97633780365b839coar * Integrate mod_dav.
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Some additional items remaining:
7787343a549c7309035b3b7836e00edb6e3a915ewrowe - case_preserved_filename stuff
48c0c81cd6fabac9d3386406d97633780365b839coar (use the new canonical name stuff?)
64ad864fa0f4493eebb181e393b40a8a90beccb9coar - find a new home for ap_text(_header)
7787343a549c7309035b3b7836e00edb6e3a915ewrowe - is it possible to remove the DAV: namespace stuff from util_xml?
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * ap_core_translate() and its use by mod_mmap_static and mod_file_cache
7787343a549c7309035b3b7836e00edb6e3a915ewrowe are a bit wonky. The function should probably be exposed as a utility
7787343a549c7309035b3b7836e00edb6e3a915ewrowe function (such as ap_translate_url2fs() or ap_validate_fs_url() or
48c0c81cd6fabac9d3386406d97633780365b839coar something). Another approach would be a new hook phase after
64ad864fa0f4493eebb181e393b40a8a90beccb9coar "translate" which would allow the module to munge what the
7787343a549c7309035b3b7836e00edb6e3a915ewrowe translation has decided to do.
48c0c81cd6fabac9d3386406d97633780365b839coar Status: Greg +1 (volunteers), Ryan +1
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * Explore use of a post-config hook for the code in http_main.c which
48c0c81cd6fabac9d3386406d97633780365b839coar calls ap_fixup_virutal_hosts(), ap_fini_vhost_config(), and
64ad864fa0f4493eebb181e393b40a8a90beccb9coar ap_sort_hooks() [to reduce the logic in main()]
48c0c81cd6fabac9d3386406d97633780365b839coar * read the config tree just once, and process N times (as necessary)
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * (possibly) use UUIDs in mod_unique_id and/or mod_usertrack
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * (possibly) port the bug fix for PR 6942 (segv when LoadModule is put
7787343a549c7309035b3b7836e00edb6e3a915ewrowe into a VirtualHost container) to 2.0.
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * shift stuff to mod_core.h
48c0c81cd6fabac9d3386406d97633780365b839coar * callers of ap_run_create_request() should check the return value
64ad864fa0f4493eebb181e393b40a8a90beccb9coar for failure (Doug volunteers)
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * Win32: Get Apache working on Windows 95/98. The following work
48c0c81cd6fabac9d3386406d97633780365b839coar (at least) needs to be done:
64ad864fa0f4493eebb181e393b40a8a90beccb9coar - winnt MPM: Fix 95/98 code paths in the winnt MPM. There is some NT
7787343a549c7309035b3b7836e00edb6e3a915ewrowe specific code that is still not in NT only code paths
48c0c81cd6fabac9d3386406d97633780365b839coar - IOL binds to APR sendfile, implemented with TransmitFile, which
64ad864fa0f4493eebb181e393b40a8a90beccb9coar is not available on 95/98.
7787343a549c7309035b3b7836e00edb6e3a915ewrowe - Document warning that OSR2 is required (for Crypt functions, in
48c0c81cd6fabac9d3386406d97633780365b839coar rand.c, at least.) This could be resolved with an SSL library, or
64ad864fa0f4493eebb181e393b40a8a90beccb9coar randomization in APR itself.
7787343a549c7309035b3b7836e00edb6e3a915ewrowe - Bring the Win9xConHook.dll from 1.3 into 2.0 (no sense till it
48c0c81cd6fabac9d3386406d97633780365b839coar actually works) and add in a splash of Win9x service code.
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * In order to use a DSO version of mod_ssl we have to link with
7787343a549c7309035b3b7836e00edb6e3a915ewrowe -lssl and -lcrypto. A workaround is in place right now where the
48c0c81cd6fabac9d3386406d97633780365b839coar entire EXTRA_LIBS macro is being appended to the objects list, but
64ad864fa0f4493eebb181e393b40a8a90beccb9coar this is a hack. We should either revamp the APACHE_CHECK_SSL_TOOLKIT
7787343a549c7309035b3b7836e00edb6e3a915ewrowe autoconf function or come up with some other autoconf checks to
7787343a549c7309035b3b7836e00edb6e3a915ewrowe search for libssl and libcrypto and properly add them to mod_ssl's
48c0c81cd6fabac9d3386406d97633780365b839coar link flags.
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * Make the worker MPM the default MPM for threaded Unix boxes.
7787343a549c7309035b3b7836e00edb6e3a915ewrowe +1: Justin, Jeff, Ian
7787343a549c7309035b3b7836e00edb6e3a915ewrowe -0: Aaron (premature decision, needs more discussion), Lars
48c0c81cd6fabac9d3386406d97633780365b839coar -0: Cliff (I think the default config should be the safest possible)
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * Fix the worker MPM to use POD to kill child processes instead
48c0c81cd6fabac9d3386406d97633780365b839coar of ap_os_killpg, regardless of how they should die. (Ryan Bloom)
7787343a549c7309035b3b7836e00edb6e3a915ewrowePRs that have been suspended forever waiting for someone to
48c0c81cd6fabac9d3386406d97633780365b839coarput them into 'the next release':
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#76: general
7787343a549c7309035b3b7836e00edb6e3a915ewrowe missing call to "setlocale();"
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#78: mod_include
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Additional status for XBitHack directive
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#362: mod_proxy
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Mod_proxy doesn't allow change of error pages
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#370: mod_env
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Modified PATH environemnt variable is not passed, instead
7787343a549c7309035b3b7836e00edb6e3a915ewrowe system's is used
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#440: mod_proxy
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Proxy doesn't deliver documents if not connected
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#534: mod_proxy
64ad864fa0f4493eebb181e393b40a8a90beccb9coar proxy converts ~name to %7Ename when name starts with a dot (.)
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#537: mod_access
7787343a549c7309035b3b7836e00edb6e3a915ewrowe mod_access syntax allows hosts that should be restricted
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#557: mod_auth-any
48c0c81cd6fabac9d3386406d97633780365b839coar ~UserHome directories are not honored in absolute pathname
64ad864fa0f4493eebb181e393b40a8a90beccb9coar requests (.htaccess)
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#612: mod_proxy
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Proxy FTP Authentication Fails
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#623: mod_include
48c0c81cd6fabac9d3386406d97633780365b839coar A smarter "Last Modified" value for SSI documents (see PR number 600)
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#628: config
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Request of "Options SymLinksIfGroupMatch"
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#700: mod_proxy
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Proxy doesn't do links right for OpenVMS files through ftp:
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#759: mod_imap
64ad864fa0f4493eebb181e393b40a8a90beccb9coar imap should read <MAP><AREA>*</MAP> too!
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#793: general
7787343a549c7309035b3b7836e00edb6e3a915ewrowe RLimitCPU and RLimitMEM don't apply to all children like they should
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#921: suexec
48c0c81cd6fabac9d3386406d97633780365b839coar Uses cwd before filling it in, doesn't use syslog
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#922: config
7787343a549c7309035b3b7836e00edb6e3a915ewrowe it is useful to allow specifiction that root-owned symlinks
7787343a549c7309035b3b7836e00edb6e3a915ewrowe should always be followed
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#980: mod_proxy
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Controlling Access to Remote Proxies would be nice...
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#994: mod_proxy
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Adding authentication "on the fly" through the proxy module
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1004: apache-api
7787343a549c7309035b3b7836e00edb6e3a915ewrowe request_config field in request_rec is moderately bogus
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#1028: other
64ad864fa0f4493eebb181e393b40a8a90beccb9coar DoS attacks involving memory consumption
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1050: mod_log-any
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Logging of virtual server to error_log as well
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1085: mod_proxy
7787343a549c7309035b3b7836e00edb6e3a915ewrowe ProxyRemote make a dead cycle.
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#1117: mod_auth-any
48c0c81cd6fabac9d3386406d97633780365b839coar Using NIS passwd.byname dbm files with AuthDBMUserFile
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#1120: suexec
7787343a549c7309035b3b7836e00edb6e3a915ewrowe suexec does not parse arguments to #exec cmd
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#1145: mod_include
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Allow for Last-Modified: without resorting to XBitHack
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1158: apache-api
7787343a549c7309035b3b7836e00edb6e3a915ewrowe improvements to child spawning API
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1166: mod_proxy
7787343a549c7309035b3b7836e00edb6e3a915ewrowe ``nph-'' not honored (no buffering) for ProxyRemote mapping
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#1176: mod_cgi
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Apache cannot handle continuation line in headers
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#1191: general
7787343a549c7309035b3b7836e00edb6e3a915ewrowe setlogin() is not called, causing problems with e.g. identd
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#1204: general
64ad864fa0f4493eebb181e393b40a8a90beccb9coar regerror() exists, use it
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#1233: apache-api
48c0c81cd6fabac9d3386406d97633780365b839coar there is no way to keep per-connection per-module state
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#1263: mod_autoexec
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Add frame-safe anchor attribute to mod_autoindex links
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1268: suexec
7787343a549c7309035b3b7836e00edb6e3a915ewrowe CGI scripts running as Apache user: security (suexec etc.)
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#1285: suexec
48c0c81cd6fabac9d3386406d97633780365b839coar Error messages could be easier to spot in cgi.log file for suexec.c
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#1287: mod_access
64ad864fa0f4493eebb181e393b40a8a90beccb9coar add allow,deny/deny,allow warning to mod_access
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1290: mod_proxy
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Need to know "hit-rate" on proxy cache
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#1358: mod_log-any
48c0c81cd6fabac9d3386406d97633780365b839coar Selective url-encode of log fields (or maybe a pseudo
64ad864fa0f4493eebb181e393b40a8a90beccb9coar log_rewrite module?)
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1383: mod_headers
7787343a549c7309035b3b7836e00edb6e3a915ewrowe I make mod_headers to modify request headers as well as
48c0c81cd6fabac9d3386406d97633780365b839coar response ones.
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#1532: mod_proxy
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Proxy transfer logging
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1547: mod_proxy
7787343a549c7309035b3b7836e00edb6e3a915ewrowe No HTTP_X_FORWARDED_FOR set...
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#1567: mod_proxy
48c0c81cd6fabac9d3386406d97633780365b839coar ProxyRemote proxy requests fail authentication by firewall
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#1582: mod_rewrite
64ad864fa0f4493eebb181e393b40a8a90beccb9coar mod_rewrite forms REQUEST_URI different than mod_cgi does
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1677: mod_headers
7787343a549c7309035b3b7836e00edb6e3a915ewrowe mod_headers should allow mod_log_config-style formats in
48c0c81cd6fabac9d3386406d97633780365b839coar header values
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#1702: mod_proxy
64ad864fa0f4493eebb181e393b40a8a90beccb9coar mod_proxy to support persistent conns?
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#1803: mod_include
48c0c81cd6fabac9d3386406d97633780365b839coar patches to mod_include to allow for file tests
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#1809: mod_auth-any
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Suggestion for improving authentication modules and core source
7787343a549c7309035b3b7836e00edb6e3a915ewrowe code, problem with 401 and ErrorDocument
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#1878: mod_proxy
64ad864fa0f4493eebb181e393b40a8a90beccb9coar listing of proxy cache content
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#1905: suexec
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Allow modules to set user:group for execution.
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#2024: apache-api
64ad864fa0f4493eebb181e393b40a8a90beccb9coar adding auth_why to conn_rec
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#2073: mod_log-any
7787343a549c7309035b3b7836e00edb6e3a915ewrowe pipelined connections are not logged correctly
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#2074: mod_rewrite
48c0c81cd6fabac9d3386406d97633780365b839coar mod_rewrite doesn't pass Proxy Throughput on internal subrequests
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#2113: config
64ad864fa0f4493eebb181e393b40a8a90beccb9coar HTTP Server Rebuild Line Needs Changing for the better
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#2138: mod_status
7787343a549c7309035b3b7836e00edb6e3a915ewrowe mod_status always displays 256 possible connection slots
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#2221: documentation
64ad864fa0f4493eebb181e393b40a8a90beccb9coar Make online documentation search link back to my installation
64ad864fa0f4493eebb181e393b40a8a90beccb9coar * PR#2284: general
7787343a549c7309035b3b7836e00edb6e3a915ewrowe Can not POST to ErrorDocument - Apache/1.3b6
7787343a549c7309035b3b7836e00edb6e3a915ewrowe * PR#2314: mod_proxy
7787343a549c7309035b3b7836e00edb6e3a915ewrowe patterns in ProxyRemote
48c0c81cd6fabac9d3386406d97633780365b839coar * PR#2343: mod_status
6758b07b4b79f898b0f56375016cea7da0bfb495wrowe Status module averages are for entire uptime
28d1da9ca818f831ea491f110dafcc10f7f07050coar * PR#2360: suexec
2a6c49cfaef5979a5a06098f3ce987cd76769409manoj suexec for general access of user content?
b865b179e150aae40ff75e580dce6ec4b5294ebdfielding * PR#2396: general
b865b179e150aae40ff75e580dce6ec4b5294ebdfielding Proposal for TimeZone directive
d5defd5a0c5cdbaf74b85939484dc2b6c8317d19manoj * PR#2415: mod_info
2a6c49cfaef5979a5a06098f3ce987cd76769409manoj /server-info doesn't check for the virtual host to list the info
2a6c49cfaef5979a5a06098f3ce987cd76769409manoj * PR#2421: config
2a6c49cfaef5979a5a06098f3ce987cd76769409manoj problem specifying ndbm library for build ?with autoconfigure
6758b07b4b79f898b0f56375016cea7da0bfb495wrowe * PR#2431: general
2a6c49cfaef5979a5a06098f3ce987cd76769409manoj A small addition to rotatelogs.c to improve program functionality.
0bff2f28ef945280c17099c142126178a78e1e54manoj * PR#2446: config
d6b3cb141f0667101c1bca883ad15b383402c93bfielding AllowOverride FileInfo is too coarse
ec0315cdf832eac2b78e50ad636af84fe4c9118cgstein * PR#2460: mod_cgi
ec0315cdf832eac2b78e50ad636af84fe4c9118cgstein TimeOut applies to output of CGI scripts
ec0315cdf832eac2b78e50ad636af84fe4c9118cgstein * PR#2512: mod_access
ec0315cdf832eac2b78e50ad636af84fe4c9118cgstein <IfDenied> directive wanted
92d75991039fc1155961044d642816b4c0323687aaron * PR#2573: suexec
92d75991039fc1155961044d642816b4c0323687aaron CGI's for general use still have to be run as another user
92d75991039fc1155961044d642816b4c0323687aaron with suExec
b3b4e853e4958357ee2d50e2fe41effecfde9eedwrowe * PR#2648: general
9a93784d34c322d1996e8a7b794a75230eac81a4wrowe Cache file names in Proxy module
9a93784d34c322d1996e8a7b794a75230eac81a4wrowe * PR#2760: config
9a93784d34c322d1996e8a7b794a75230eac81a4wrowe [PATCH] User/Group for <Directory> and <Location> i.e. not only
dba571e5c5a2586c7e2b382789f3250136ff7577jwoolley in global and <Virtual>.
dba571e5c5a2586c7e2b382789f3250136ff7577jwoolley * PR#2763: general
dba571e5c5a2586c7e2b382789f3250136ff7577jwoolley mailto tags and bundling bug report script
9a93784d34c322d1996e8a7b794a75230eac81a4wrowe * PR#2785: os-aix
dba571e5c5a2586c7e2b382789f3250136ff7577jwoolley Support for System Resource Controller
Feedback/Comment on APACI
Inclusion of RPM spec file in CVS/distributions
* orig_ct in the byterange/multipart handling may not be