STATUS revision e04911e1869585acf01ee0876c2752afbc91af71
f743002678eb67b99bbc29fee116b65d9530fec0wroweAPACHE 2.0 STATUS: -*-text-*-
80833bb9a1bf25dcf19e814438a4b311d2e1f4cffuankgLast modified at [$Date: 2001/11/13 05:04:24 $]
1337c7673efc1f80f634139fbad7cbb98a0dc657ylavic 2.0.28 : rolled November 12, 2001
4da61833a1cbbca94094f9653fd970582b97a72etrawick 2.0.27 : rolled November 6, 2001
4da61833a1cbbca94094f9653fd970582b97a72etrawick 2.0.26 : tagged October 16, 2001. not rolled.
4da61833a1cbbca94094f9653fd970582b97a72etrawick 2.0.25 : rolled August 29, 2001
4da61833a1cbbca94094f9653fd970582b97a72etrawick 2.0.24 : rolled August 18, 2001
4da61833a1cbbca94094f9653fd970582b97a72etrawick 2.0.23 : rolled August 9, 2001
4789804be088bcd86ae637a29cdb7fda25169521jailletc 2.0.22 : rolled July 29, 2001
4789804be088bcd86ae637a29cdb7fda25169521jailletc 2.0.21 : rolled July 20, 2001
4789804be088bcd86ae637a29cdb7fda25169521jailletc 2.0.20 : rolled July 8, 2001
4789804be088bcd86ae637a29cdb7fda25169521jailletc 2.0.19 : rolled June 27, 2001
e50c3026198fd496f183cda4c32a202925476778covener 2.0.18 : rolled May 18, 2001
e50c3026198fd496f183cda4c32a202925476778covener 2.0.17 : rolled April 17, 2001
e50c3026198fd496f183cda4c32a202925476778covener 2.0.16 : rolled April 4, 2001
5b88c8507d5ef6d0c4cfbc78230294968175b638minfrin 2.0.15 : rolled March 21, 2001
5b88c8507d5ef6d0c4cfbc78230294968175b638minfrin 2.0.14 : rolled March 7, 2001
6c3b9cebb551140fbb25d58bae08b539b3802133ylavic 2.0a9 : released December 12, 2000
6c3b9cebb551140fbb25d58bae08b539b3802133ylavic 2.0a8 : released November 20, 2000
6c3b9cebb551140fbb25d58bae08b539b3802133ylavic 2.0a7 : released October 8, 2000
4f29b65ab4b547ad5dbe506e2d0ff5d12ead9247ylavic 2.0a6 : released August 18, 2000
4f29b65ab4b547ad5dbe506e2d0ff5d12ead9247ylavic 2.0a5 : released August 4, 2000
0a0df13b7f1f4f1a74fe295253d89ca3911b301aylavic 2.0a4 : released June 7, 2000
0a0df13b7f1f4f1a74fe295253d89ca3911b301aylavic 2.0a3 : released April 28, 2000
0a0df13b7f1f4f1a74fe295253d89ca3911b301aylavic 2.0a2 : released March 31, 2000
0a0df13b7f1f4f1a74fe295253d89ca3911b301aylavic 2.0a1 : released March 10, 2000
69301145375a889e7e37caf7cc7321ac0f91801erpluemPlease consult the following STATUS files for information
69301145375a889e7e37caf7cc7321ac0f91801erpluemon related projects:
d58a848a016d401b965111e50ef829e1641f7834minfrinRELEASE SHOWSTOPPERS:
2e6f4d654c96c98b761fb012fd25c5d5b1558c44sf * If any request gets to the core handler, without a flag that this
2e6f4d654c96c98b761fb012fd25c5d5b1558c44sf r->filename was tested by dir/file_walk, we need to 500 at the very
2e6f4d654c96c98b761fb012fd25c5d5b1558c44sf end of the ap_process_request_internal() processing. This provides
17e6c95f3b22d18acdf8380fb26a8d0e10c80767ylavic authors of older modules better compatibility, while still improving
17e6c95f3b22d18acdf8380fb26a8d0e10c80767ylavic the security and robustness of 2.0.
17e6c95f3b22d18acdf8380fb26a8d0e10c80767ylavic Status: still need to decide where this goes, OtherBill comments...
17e6c95f3b22d18acdf8380fb26a8d0e10c80767ylavic Message-ID: <065701c14526$495203b0$96c0b0d0@roweclan.net>
17e6c95f3b22d18acdf8380fb26a8d0e10c80767ylavic we need to look at halting this in the 'default handler' case,
e8bd80a4bb88199d2f9a24a50345688e52d9c116ylavic and that implies pushing the 'handler election' into the request
e8bd80a4bb88199d2f9a24a50345688e52d9c116ylavic internal processing phase from the run request phase.
330e16bea8fe9cace4de90c349750c03dfb1fe64ylavic * There is a bug in how we sort some hooks, at least the pre-config
330e16bea8fe9cace4de90c349750c03dfb1fe64ylavic hook. The first time we call the hooks, they are in the correct
330e16bea8fe9cace4de90c349750c03dfb1fe64ylavic order, but the second time, we don't sort them correctly. Currently,
330e16bea8fe9cace4de90c349750c03dfb1fe64ylavic the modules/http/config.m4 file has been renamed to
330e16bea8fe9cace4de90c349750c03dfb1fe64ylavic modules/http/config2.m4 to work around this problem, it should moved
330e16bea8fe9cace4de90c349750c03dfb1fe64ylavic back when this is fixed. rbb
d7205b1a86c51c27b71a2c458dc453fd53a261c1covener * The Add...Filter and Set...Filter directives do not allow the
d7205b1a86c51c27b71a2c458dc453fd53a261c1covener administrator to order filters, beyond the order of filename (mime)
d7205b1a86c51c27b71a2c458dc453fd53a261c1covener extensions. It isn't clear if Set...Filter(s) should be inserted
d7205b1a86c51c27b71a2c458dc453fd53a261c1covener before or after the Add...Filter(s) which are ordered by sequence of
d7205b1a86c51c27b71a2c458dc453fd53a261c1covener filename extensions. Add...FilterByType will add to this quandry.
44ff304057225e944e220e981d434a046d14cf06covener Some sort of resolution needs to be proposed,
44ff304057225e944e220e981d434a046d14cf06covener * mod_dir should normally redirect ALL directory requests which do
44ff304057225e944e220e981d434a046d14cf06covener not include a trailing slash on the URI. However, if a "notes"
5d1ba75b8794925e67591c209085a49279791de9covener flag is set (say, via BrowserMatch), this behavior will be
5d1ba75b8794925e67591c209085a49279791de9covener disabled for non-GET requests.
5d1ba75b8794925e67591c209085a49279791de9covener Status: Greg volunteers
032982212dbcc7c3cce95bf89c503bb56e185ac7kbrand MsgId: <20010227104646.E2297@lyra.org>
032982212dbcc7c3cce95bf89c503bb56e185ac7kbrand MsgId: <3A9C0097.9C83F07C@Golux.Com>
032982212dbcc7c3cce95bf89c503bb56e185ac7kbrand * mod_negotiation needs a new option or directive, something like
caad2986f81ab263f7af41467dd622dc9add17f3ylavic ForceLanguagePriority, to fall back to the LanguagePriority
caad2986f81ab263f7af41467dd622dc9add17f3ylavic directive instead of returning a "no acceptable variant" error.
caad2986f81ab263f7af41467dd622dc9add17f3ylavic Status: Bill has some code in his tree that accomplishes
caad2986f81ab263f7af41467dd622dc9add17f3ylavic this, and will commit it Friday after it's tested.
f7317ff316c2b141feea31bddb74d5d3fa1584edjorton * Usability: Sanitize the MPM config directives. MaxClients in
f7317ff316c2b141feea31bddb74d5d3fa1584edjorton the threaded MPM is totally misleading now as it has little to
2165214331e4afafca4048f66f303d0253d7b001covener do with limiting the number of clients (it limits the number
a34684a59b60a4173c25035d0c627ef17e6dc215rpluem of child processes). Bill proposed nomenclature change to
a34684a59b60a4173c25035d0c627ef17e6dc215rpluem something like "StartWorkers, MaxWorkers, etc." that could
1e2d421a36999d292042a5539971070d54aa6c63ylavic apply to most all the MPMs (with some notable exceptions).
1e2d421a36999d292042a5539971070d54aa6c63ylavic Bill would be happy with changing MaxClients to MaxServers
1e2d421a36999d292042a5539971070d54aa6c63ylavic to make it agree with the operation of the StartServers
fa7ed98b9dc94c5845cf845aea0a44ecacd290c9humbedooh * revamp the input filter semantics, per discussions since
0b67eb8568cd58bb77082703951679b42cf098actrawick February (and especially at the hackathon last
0b67eb8568cd58bb77082703951679b42cf098actrawick April). Specifically, ap_get_brigade will return a brigade with
0b67eb8568cd58bb77082703951679b42cf098actrawick *up to* a specific number of bytes, or a "line" of data. The
0b67eb8568cd58bb77082703951679b42cf098actrawick read may be blocking or nonblocking. ap_getline() will be
5ef3c61605a3a021ff71f488983cb0065f8e1a79covener refactored into apr_brigade_getline(), and then DECHUNK can use
fb1985a97912b25ec6564c73e610a31e5fc6e25fcovener f->next (ap_getline will always read "top of input stack"). Also
09c87c777bed1655621bb20e1c46cb6b1a63279dcovener fix the bug where request body content will end up closing the
6502b7b32f980cc2093bb3ebce37e5e4dc68fba4ylavic connection (buggering up persistent conns).
6502b7b32f980cc2093bb3ebce37e5e4dc68fba4ylavic Status: Justin is working on this as fast as he can.
3060ce7f798fbda7999cd4ddf89b525d2b294185covener The core input filters, HTTP-related filters, mod_ssl, and
c1a63b8fad09c419c1a64f75993feb8a343a6801ylavic mod_proxy are switched to the new logic.
c1a63b8fad09c419c1a64f75993feb8a343a6801ylavic However, ap_getline() still needs to be refactored out. But,
c1a63b8fad09c419c1a64f75993feb8a343a6801ylavic there's a problem there: ap_getline() peeks ahead for MIME
e6b4bd1113567627ab6bb6c6a7105e1e01a7d889jailletc continuation (first character on line is space or \t) and
e6b4bd1113567627ab6bb6c6a7105e1e01a7d889jailletc stores unused data in core_request_config which violates the
e466c40e1801982602ee0200c9e8b61cc148742djailletc abstraction. That's cheating. So, we may not be able to
e466c40e1801982602ee0200c9e8b61cc148742djailletc implement this without setting some data aside (yuck!).
457468b82e59d01eba00dd9d0817309c8f5e414ejim I believe this is OtherBill's main complaint with the current
457468b82e59d01eba00dd9d0817309c8f5e414ejim filtering.
457468b82e59d01eba00dd9d0817309c8f5e414ejim AIUI (correct me if I'm wrong!), OtherBill believes we
04983e3bd1754764eec7d6bb772fe3b0bf391771jorton should have a pushback option so that we can return unread
04983e3bd1754764eec7d6bb772fe3b0bf391771jorton data - this would solve this case. However, my question to
15890c9306ba98f6fc243e15a3c4778ddc7d773erpluem him is how do we handle stuff like mod_ssl - we can't "unread"
15660979a30d251681463de2e0584853890082accovener data. So, do we have two brigades for each filter? An in
15660979a30d251681463de2e0584853890082accovener brigade and a returned brigade? That seems messy. To
49dacedb6c387b786b7911082ff35121a45f414bcovener everyone else, can we refactor ap_getline() without pushback
cfd9415521847b2f9394fad04fb701cfb955f503rjung - socket bucket and core input filter changes. see end of
cfd9415521847b2f9394fad04fb701cfb955f503rjung message ID (Feb 27): <20010227075326.S2297@lyra.org>
28c31fb73c1264bd1d0ff932573677030b024c7dwrowe - fix up ap_get_brigade() semantics, fix bug in DECHUNK /
28c31fb73c1264bd1d0ff932573677030b024c7dwrowe ap_getline. many messages (plus their threads) (Apr/May):
28c31fb73c1264bd1d0ff932573677030b024c7dwrowe Message-ID: <20010402101207.J27539@lyra.org>
28c31fb73c1264bd1d0ff932573677030b024c7dwrowe Message-ID: <3AF7F921.D2EEC41A@algroup.co.uk>
8491e0600f69b0405e156ea8a419653c065c645bcovener Message-ID: <20010508190029.E18404@lyra.org>
63b9f1f5880391261705f696d7d65507bbe9ace3covener - further work with combining/tweaking the builtin filters:
63b9f1f5880391261705f696d7d65507bbe9ace3covener Message-ID: <20010509115445.D1374@lyra.org>
49dacedb6c387b786b7911082ff35121a45f414bcovener - thoughts on filter modes:
49dacedb6c387b786b7911082ff35121a45f414bcovener Message-ID: <021b01c14dee$09782af0$93c0b0d0@roweclan.net>
3c990331fc6702119e4f5b8ba9eae3021aea5265jim * Fold mod_auth_db features back into mod_auth_dbm, and depricate it.
3c990331fc6702119e4f5b8ba9eae3021aea5265jim This can't wait until we have a 2.0-gold release, if folks need
3c990331fc6702119e4f5b8ba9eae3021aea5265jim to move over to auth_dbm, we can't do that to them after 2.0 gold.
fc42512879dd0504532f52fe5d0d0383dda96a1eniqRELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
fc42512879dd0504532f52fe5d0d0383dda96a1eniq * Source code should follow style guidelines.
0451df5dc50fa5d8b3e07d92ee6a92e36a1181a5niq This shouldn't wait until we have a 2.0-gold release because
0451df5dc50fa5d8b3e07d92ee6a92e36a1181a5niq then style corrections will conflict with bug fixes found after
0451df5dc50fa5d8b3e07d92ee6a92e36a1181a5niq release which is not nice.
983528026996668ea295be95aedb9c7a346af470ylavic * Allow the DocumentRoot directive within <Location > scopes? This
da0442c0440caef34706e2c2f3af05cb65921cc0jailletc allows the beloved (crusty) Alias /foo/ /somepath/foo/ followed
da0442c0440caef34706e2c2f3af05cb65921cc0jailletc by a <Directory /somepath/foo> to become simply
06b8f183140c8e02e0974e938a05078b511d1603covener <Location /foo/> DocumentRoot /somefile/foo (IMHO a bit more legible
06b8f183140c8e02e0974e938a05078b511d1603covener and in-your-face.) DocumentRoot unset would be accepted [and would
06b8f183140c8e02e0974e938a05078b511d1603covener not permit content to be served, only virtual resources such as
15890c9306ba98f6fc243e15a3c4778ddc7d773erpluem server-info or server-status.
259878293a997ff49f5ddfc53d3739cbdc25444ecovener This proposed change would _not_ depricate Alias.
259878293a997ff49f5ddfc53d3739cbdc25444ecovener * add mod_gz to httpd-2.0 (in modules/experimental/)
259878293a997ff49f5ddfc53d3739cbdc25444ecovener +1: Greg, Justin, ben, Ken, Jeff, Bill, Ian
15890c9306ba98f6fc243e15a3c4778ddc7d773erpluem 0: Cliff (there's now another candidate to be evaluated)
b54b024c06a19926832d77d40ba35ad8c41e4d3dminfrin 0: Jim, Aaron (premature decision at present, IMO)
b54b024c06a19926832d77d40ba35ad8c41e4d3dminfrin -0: Doug, Ryan
65967d05f839dbf27cf91d91fa79585eeae19660minfrin * daedalus: mod_cgid and suexec have a problem co-existing. suexec
65967d05f839dbf27cf91d91fa79585eeae19660minfrin sees a null command string sometimes. The problem happens when
65967d05f839dbf27cf91d91fa79585eeae19660minfrin you access bugs.apache.org, then click on the "search the bug db"
8152945ae46857b170cb227e79bb799f4fc7710dminfrin * Win32: Rotatelogs sometimes is not terminated when Apache
8152945ae46857b170cb227e79bb799f4fc7710dminfrin goes down hard. FirstBill was looking at possibly tracking the
8152945ae46857b170cb227e79bb799f4fc7710dminfrin child's-child processes in the parent process.
75f5c2db254c0167a0e396254460de09b775d203trawick OtherBill asks, wasn't this fixed?
75f5c2db254c0167a0e396254460de09b775d203trawick * Win32: Add a simple hold console open patch (wait for close or
4f0358189bfa57b8e75bd6b94db264302a8f336amrumph the ESC key, with a nice message) if the server died a bad
4f0358189bfa57b8e75bd6b94db264302a8f336amrumph death (non-zero exit code) in console mode.
4f0358189bfa57b8e75bd6b94db264302a8f336amrumph Resolution: bring forward same ugly hacks from 1.3.13-.20
5716f9c6daa92dde5f2f9d11ed63f7c9549c223atrawick * Port of mod_ssl to Apache 2.0:
5716f9c6daa92dde5f2f9d11ed63f7c9549c223atrawick The current porting state is summarized in modules/ssl/README. The
54d750a84a175d8e338880514d440773eb986b50covener remaining work includes:
54d750a84a175d8e338880514d440773eb986b50covener (1) stablizing/optimizing the SSL filter logic
54d750a84a175d8e338880514d440773eb986b50covener (2) Enabling the various SSL caching mechanisms (shmcb, shmht)
54d750a84a175d8e338880514d440773eb986b50covener (3) Enabling SSL extentions
54d750a84a175d8e338880514d440773eb986b50covener (4) Trying to seperate the https filter logic from mod_ssl -
54d750a84a175d8e338880514d440773eb986b50covener This is to facilitate other modules that wish to use the https
54d750a84a175d8e338880514d440773eb986b50covener filter or the mod_ssl logic or both as required.
54d750a84a175d8e338880514d440773eb986b50covener Justin: mod_ssl filter logic is redone, so that should be fine.
7a3aa12f0eda24793ee26d6a179bd53132e9dae8covener Madhu has submitted a patch for SSL caching - however, I
54d750a84a175d8e338880514d440773eb986b50covener am -0 on that patch as I *think* we could implement the
54d750a84a175d8e338880514d440773eb986b50covener shared memory another way that is much cleaner (i.e.
83b50288fa7d306324bba68832011ea08f5c7832covener treat shmem directly as a dbm via APR routines). Justin
4e30ef014533a7e93c92d88306291f5e49c9692ftrawick also thinks that the https filter logic may be sufficiently
83b50288fa7d306324bba68832011ea08f5c7832covener decoupled now, but isn't really sure.
5f066f496cd9f20a2a701255bc67d44e7cb46daetrawick * Performance: Get the SINGLE_LISTEN_UNSERIALIZED_ACCEPT
5f066f496cd9f20a2a701255bc67d44e7cb46daetrawick optimization working in threaded. prefork's new design for how
2e15620d724fb8e3a5be183b917359a2fd6e9468covener to notice data on the pod should be sufficient.
2e15620d724fb8e3a5be183b917359a2fd6e9468covener * Performance & Debug: Eliminate most (and perhaps all) of the
2e15620d724fb8e3a5be183b917359a2fd6e9468covener malloc/free calls in the bucket brigade code. Need some
1b988c41ee505962781d110a3e4c2c90f1ea0aa4covener light weight memory management functions that allow freeing
1b988c41ee505962781d110a3e4c2c90f1ea0aa4covener memory (putting it back into a memory pool) when it is no
1b988c41ee505962781d110a3e4c2c90f1ea0aa4covener longer needed. Enabling simple debugging features like guard
1b988c41ee505962781d110a3e4c2c90f1ea0aa4covener bands, double free detection, etc. would be cool but certainly
b8efdc95bec9cf089aa1be0bfd07d46aa1137a7acovener not a hard requirement.
b8efdc95bec9cf089aa1be0bfd07d46aa1137a7acovener Status: Cliff started to implement this using SMS as has
f06e7c4b1bce6b6491e5de0b7998d3f5696b293dchrisd been discussed at length for months, but since
f06e7c4b1bce6b6491e5de0b7998d3f5696b293dchrisd SMS is not being used anywhere else in the server,
f06e7c4b1bce6b6491e5de0b7998d3f5696b293dchrisd several people expressed the opinion that we should
179565be4043d7e5f9161aa75271fa0a001866d9covener get rid of it entirely, meaning that the buckets
179565be4043d7e5f9161aa75271fa0a001866d9covener need their own memory management (free list) functions.
179565be4043d7e5f9161aa75271fa0a001866d9covener Cliff will implement that this weekend so we at least
111436a32ba1254291e4883292fb116d15fe8f64covener have something to look at/compare with.
fce4949fb0b309a5744afcd503c6ed2d35621ee2covener * Eliminate unnecessary creation of pipes in mod_cgid
fce4949fb0b309a5744afcd503c6ed2d35621ee2covener * the autoconf setup should be fixed to default to using the
7b7430e701e9a31ce809da7c220bb8dfcf68c86etrawick "Apache" layout from config.layout, and each variable settable
7b7430e701e9a31ce809da7c220bb8dfcf68c86etrawick in a layout should be overridable on the command line. Plus,
7b7430e701e9a31ce809da7c220bb8dfcf68c86etrawick what we do right now just doesn't seem to fully fit into how autoconf
ccc20788c1e5fc973f36df634399c89acb70deaejerenkrantz works, eg. AC_PREFIX_DEFAULT issues.
ccc20788c1e5fc973f36df634399c89acb70deaejerenkrantz Message-ID: <Pine.BSF.4.20.0104031557420.20876-100000@alive.znep.com>
273e512f20f262e5e2aa8e0e83371d1929fb76adjkaluza * Combine log_child and piped_log_spawn. Clean up http_log.c.
273e512f20f262e5e2aa8e0e83371d1929fb76adjkaluza Common logging API.
efe780dcf13b2b95effabf897d694d8f23feac74trawick * Document mod_file_cache.
fe83f60b41477b14a37edcfcd1f7f5c5a1ebfe44minfrin * OS/2: Make mod_status work for spmt_os2 MPM.
993d1261a278d7322bccef219101220b7b4fb8c5jkaluza * Platforms that do not support fork (primarily Win32 and AS/400)
993d1261a278d7322bccef219101220b7b4fb8c5jkaluza Architect start-up code that avoids initializing all the modules
993d1261a278d7322bccef219101220b7b4fb8c5jkaluza in the parent process on platforms that do not support fork.
ba050a6f942b9fa0e81ed73437588005c569655ccovener * Win32: Migrate the MPM over to use APR thread/process calls. This
ba050a6f942b9fa0e81ed73437588005c569655ccovener would eliminate some code in the Win32 branch that essentially
ba050a6f942b9fa0e81ed73437588005c569655ccovener duplicates what is in APR.
135ddda3a989215d2bedbcf1529bfb269c3eda23niq * There are still a number of places in the code where we are
135ddda3a989215d2bedbcf1529bfb269c3eda23niq losing error status (i.e. throwing away the error returned by a
001a44c352f89c9ec332ffd3e0a6927dcd19432chumbedooh system call and replacing it with a generic error code)
001a44c352f89c9ec332ffd3e0a6927dcd19432chumbedooh * Mass vhosting version of suEXEC.
793214f67dede32edfd9ee96c664ead04d175cbbjfclere * All DBMs suffer from confusion in support/dbmmanage (perl script) since
cc5a4a08dc9783fcbc52ce86f11e01c281a43810minfrin the dbmmanage employs the first-matched dbm format. This is not
9b0076ddd1103e5fa9c1f9bafde4b06ce244fbaecovener necessarily the library that Apache was built with. Aught to
9b0076ddd1103e5fa9c1f9bafde4b06ce244fbaecovener rewrite dbmmanage upon installation to bin/ with the proper library
9b0076ddd1103e5fa9c1f9bafde4b06ce244fbaecovener for predictable mod_auth_db/dbm administration.
249d09d51808cb7981af99762c3b3736ca126cd5jkaluza Status: Mladen Turk has posted several patches and ideas.
249d09d51808cb7981af99762c3b3736ca126cd5jkaluza Key question, part of htpasswd, or a seperate utility?
249d09d51808cb7981af99762c3b3736ca126cd5jkaluza prefer htpasswd:
249d09d51808cb7981af99762c3b3736ca126cd5jkaluza prefer seperate: OtherBill
56589be3d7a3e9343370df240010c6928cc78b39jkaluza * use apu_dbm in mod_auth_dbm
77ca16c5676da23155311e13cee61e7eaba9fa3ejailletc Status: Greg +1 (low-priority volunteer)
77ca16c5676da23155311e13cee61e7eaba9fa3ejailletc Justin says: "Seems like this is already there, so should we just
77ca16c5676da23155311e13cee61e7eaba9fa3ejailletc remove the other DBM code in that file? If you want
77ca16c5676da23155311e13cee61e7eaba9fa3ejailletc to use gdbm, or dbm, etc, you should tell apr-util."
f87299dab99bc04b51a6b8cad51b6795db862c0atrawick Will says: "bs - I may choose the fastest - most efficient native
f87299dab99bc04b51a6b8cad51b6795db862c0atrawick dbm implementation, for shared proc caches, ssl session
f87299dab99bc04b51a6b8cad51b6795db862c0atrawick caching, etc, but that has nothing to do with maintaining
4d12805e6c18253040223ea637acd6b3b3c18f60jorton a userlist via dbm, which has to remain readable between
4d12805e6c18253040223ea637acd6b3b3c18f60jorton builds/machines, etc. The use-multiple database schema
4d12805e6c18253040223ea637acd6b3b3c18f60jorton for apr-util would let us do this with just apr, though."
85eacfc96a04547ef25aabbc06440039715084c2jorton * Integrate mod_dav.
e5d909f2b06bd880fb3675cd49363df981caa631trawick Some additional items remaining:
a4df2cd1e1391575a327c2a90ba4315f805a0a78covener - case_preserved_filename stuff
a4df2cd1e1391575a327c2a90ba4315f805a0a78covener (use the new canonical name stuff?)
a4df2cd1e1391575a327c2a90ba4315f805a0a78covener - find a new home for ap_text(_header)
cb666b29f81df1d11d65002250153353568021fccovener - is it possible to remove the DAV: namespace stuff from util_xml?
cb666b29f81df1d11d65002250153353568021fccovener * ap_core_translate() and its use by mod_mmap_static and mod_file_cache
6a80c3c6f4b8ea7ba5e89402b8b779b09ce020e0covener are a bit wonky. The function should probably be exposed as a utility
1c2cab00d988fc48cbe59032cf76cc0bab20d6f7covener function (such as ap_translate_url2fs() or ap_validate_fs_url() or
6a80c3c6f4b8ea7ba5e89402b8b779b09ce020e0covener something). Another approach would be a new hook phase after
75a230a728338d84dcfe81edd375352f34de22d0covener "translate" which would allow the module to munge what the
75a230a728338d84dcfe81edd375352f34de22d0covener translation has decided to do.
75a230a728338d84dcfe81edd375352f34de22d0covener Status: Greg +1 (volunteers), Ryan +1
1f50dc34ae069adeed20b2986e5ffdefa5c410e0covener * Explore use of a post-config hook for the code in http_main.c which
1f50dc34ae069adeed20b2986e5ffdefa5c410e0covener calls ap_fixup_virutal_hosts(), ap_fini_vhost_config(), and
63a5ea80bddcc84a462e40f402b4f330e0e05411covener ap_sort_hooks() [to reduce the logic in main()]
63a5ea80bddcc84a462e40f402b4f330e0e05411covener * read the config tree just once, and process N times (as necessary)
65a4e663b82f8bce28ac22ab2edfd7502de36998sf * (possibly) use UUIDs in mod_unique_id and/or mod_usertrack
65a4e663b82f8bce28ac22ab2edfd7502de36998sf * (possibly) port the bug fix for PR 6942 (segv when LoadModule is put
65a4e663b82f8bce28ac22ab2edfd7502de36998sf into a VirtualHost container) to 2.0.
74e7f6c55fd67b10cb400b3f6d1dc718a303d944minfrin * shift stuff to mod_core.h
74e7f6c55fd67b10cb400b3f6d1dc718a303d944minfrin * APR-ize resolver stuff in mod_unique_id (Jeff volunteers)
a511a29faf2ff7ead3b67680154a624effb31aafminfrin * callers of ap_run_create_request() should check the return value
a511a29faf2ff7ead3b67680154a624effb31aafminfrin for failure (Doug volunteers)
a511a29faf2ff7ead3b67680154a624effb31aafminfrin * when prefork bails out due to an initialization error in the detached
a511a29faf2ff7ead3b67680154a624effb31aafminfrin process (e.g., mutex init failure), other children (cgid, at least)
63921358ef93fcb41bc71d9894221ba3d7fbb87bminfrin are left hanging around
63921358ef93fcb41bc71d9894221ba3d7fbb87bminfrin * Win32: Get Apache working on Windows 95/98. The following work
deec48c67d4786bc77112ffbf3a4e70b931097edminfrin (at least) needs to be done:
6d601599d3d65df0410eae6e573e75b2dbfb1fb4minfrin - winnt MPM: Fix 95/98 code paths in the winnt MPM. There is some NT
6d601599d3d65df0410eae6e573e75b2dbfb1fb4minfrin specific code that is still not in NT only code paths
6d601599d3d65df0410eae6e573e75b2dbfb1fb4minfrin - IOL binds to APR sendfile, implemented with TransmitFile, which
6d601599d3d65df0410eae6e573e75b2dbfb1fb4minfrin is not available on 95/98.
684e0cfc200f66287a93bbd1708d1dd8a92a7eefcovener - Document warning that OSR2 is required (for Crypt functions, in
684e0cfc200f66287a93bbd1708d1dd8a92a7eefcovener rand.c, at least.) This could be resolved with an SSL library, or
5c43d2fb853f84497b5ece2d414ef9484aa87e5fsf randomization in APR itself.
05a5a9c3e16f21566e1b61f4bd68025ce1b741ccjoes - Bring the Win9xConHook.dll from 1.3 into 2.0 (no sense till it
05a5a9c3e16f21566e1b61f4bd68025ce1b741ccjoes actually works) and add in a splash of Win9x service code.
26c5829347f6a355c00f1ba0301d575056b69536niq * In order to use a DSO version of mod_ssl we have to link with
ef82e8fa164e0a1f8b813f7deb6b7ead96018c94niq -lssl and -lcrypto. A workaround is in place right now where the
ef82e8fa164e0a1f8b813f7deb6b7ead96018c94niq entire EXTRA_LIBS macro is being appended to the objects list, but
ef82e8fa164e0a1f8b813f7deb6b7ead96018c94niq this is a hack. We should either revamp the APACHE_CHECK_SSL_TOOLKIT
ef82e8fa164e0a1f8b813f7deb6b7ead96018c94niq autoconf function or come up with some other autoconf checks to
ef82e8fa164e0a1f8b813f7deb6b7ead96018c94niq search for libssl and libcrypto and properly add them to mod_ssl's
ef82e8fa164e0a1f8b813f7deb6b7ead96018c94niq link flags.
c12917da693bae4028a1d5a5e8224bceed8c739dsfPRs that have been suspended forever waiting for someone to
c12917da693bae4028a1d5a5e8224bceed8c739dsfput them into 'the next release':
eafcc0ebf263d0ba69855b6e10958c4c1a2361bdsf * PR#76: general
eafcc0ebf263d0ba69855b6e10958c4c1a2361bdsf missing call to "setlocale();"
d7ffd2da16d58b1a0de212e4d56f7aebb72bef26sf * PR#78: mod_include
d7ffd2da16d58b1a0de212e4d56f7aebb72bef26sf Additional status for XBitHack directive
4576c1a9ef54cd1e5555ee07d016a7f559f80338sf * PR#362: mod_proxy
4576c1a9ef54cd1e5555ee07d016a7f559f80338sf Mod_proxy doesn't allow change of error pages
9811aed12bbc71783d2e544ccb5fecd193843eadsf * PR#370: mod_env
88fac54d9d64f85bbdab5d7010816f4377f95bd7rjung Modified PATH environemnt variable is not passed, instead
88fac54d9d64f85bbdab5d7010816f4377f95bd7rjung system's is used
bd3f5647b96d378d9c75c954e3f13582af32c643sf * PR#440: mod_proxy
bd3f5647b96d378d9c75c954e3f13582af32c643sf Proxy doesn't deliver documents if not connected
2a7beea91d46beb41f043a84eaad060047ee04aafabien * PR#534: mod_proxy
2a7beea91d46beb41f043a84eaad060047ee04aafabien proxy converts ~name to %7Ename when name starts with a dot (.)
584a85dd4047e38d3ed3a29b6662fcc9d100ae4csf * PR#537: mod_access
584a85dd4047e38d3ed3a29b6662fcc9d100ae4csf mod_access syntax allows hosts that should be restricted
f21e9e3d0bfb7a507ecc5bc963f2159d693503d1sf * PR#557: mod_auth-any
f6b9c755a0b793e8a3a3aebd327ca20a86478117sf ~UserHome directories are not honored in absolute pathname
f6b9c755a0b793e8a3a3aebd327ca20a86478117sf requests (.htaccess)
132ee6ac1c26d6e8953836316ba50734eefab47bsf * PR#612: mod_proxy
132ee6ac1c26d6e8953836316ba50734eefab47bsf Proxy FTP Authentication Fails
85eacfc96a04547ef25aabbc06440039715084c2jorton * PR#623: mod_include
536d2e7cd1fdec1255b8c3bdf41fdc714c506a54trawick A smarter "Last Modified" value for SSI documents (see PR number 600)
536d2e7cd1fdec1255b8c3bdf41fdc714c506a54trawick * PR#628: config
79c5787b92ac5f0e1cc82393816c77a006399316trawick Request of "Options SymLinksIfGroupMatch"
79c5787b92ac5f0e1cc82393816c77a006399316trawick * PR#700: mod_proxy
c967bf3bc89e8aa60dbd30d9da388e448ddc1cc4trawick Proxy doesn't do links right for OpenVMS files through ftp:
79c5787b92ac5f0e1cc82393816c77a006399316trawick * PR#759: mod_imap
79c5787b92ac5f0e1cc82393816c77a006399316trawick imap should read <MAP><AREA>*</MAP> too!
7b395e4e878c28a4784919cfd2e704ddd14a3390jorton * PR#793: general
7b395e4e878c28a4784919cfd2e704ddd14a3390jorton RLimitCPU and RLimitMEM don't apply to all children like they should
536e48c08d674acac5d44929318f2ad928edc361jorton * PR#921: suexec
e81785da447b469da66f218b3f0244aab507958djorton Uses cwd before filling it in, doesn't use syslog
3e4e54d4e3fc0123c63d57aa84ac7ad7a8c73ff8jorton * PR#922: config
3e4e54d4e3fc0123c63d57aa84ac7ad7a8c73ff8jorton it is useful to allow specifiction that root-owned symlinks
53e9b27aba029b18be814df40bcf6f0428771d1efuankg should always be followed
53e9b27aba029b18be814df40bcf6f0428771d1efuankg * PR#980: mod_proxy
53e9b27aba029b18be814df40bcf6f0428771d1efuankg Controlling Access to Remote Proxies would be nice...
ca61ccd0c306c2c72df153688ba1b49f3eceed80sf * PR#994: mod_proxy
6bb524f1895f30265a1431afc460977d391cb36bsf Adding authentication "on the fly" through the proxy module
e6dd71992459d05a676b98b7963423dc5dc1e24aminfrin * PR#1004: apache-api
e6dd71992459d05a676b98b7963423dc5dc1e24aminfrin request_config field in request_rec is moderately bogus
23f1535d6a60817d2846bac0aea230ea475d7dccminfrin * PR#1028: other
23f1535d6a60817d2846bac0aea230ea475d7dccminfrin DoS attacks involving memory consumption
ec7520b24cd80d34d82bbcaca153cbb23cc04bc0rjung * PR#1050: mod_log-any
ec7520b24cd80d34d82bbcaca153cbb23cc04bc0rjung Logging of virtual server to error_log as well
ec7520b24cd80d34d82bbcaca153cbb23cc04bc0rjung * PR#1085: mod_proxy
ec7520b24cd80d34d82bbcaca153cbb23cc04bc0rjung ProxyRemote make a dead cycle.
0827cb14e550f6f65018431c22c2c913631c8f25kbrand * PR#1117: mod_auth-any
6249dfa569d3b4f1f539665b979a80c6e335d93etrawick Using NIS passwd.byname dbm files with AuthDBMUserFile
cfa64348224b66dd1c9979b809406c4d15b1c137fielding * PR#1120: suexec
74499a117b3b2cd9666715a14f90c0e5d1a4ee8ajim suexec does not parse arguments to #exec cmd
cfa64348224b66dd1c9979b809406c4d15b1c137fielding * PR#1145: mod_include
74499a117b3b2cd9666715a14f90c0e5d1a4ee8ajim Allow for Last-Modified: without resorting to XBitHack
cfa64348224b66dd1c9979b809406c4d15b1c137fielding * PR#1158: apache-api
setlogin() is not called, causing problems with e.g. identd
add allow,deny/deny,allow warning to mod_access
A small addition to rotatelogs.c to improve program functionality.
Feedback/Comment on APACI
Inclusion of RPM spec file in CVS/distributions
* orig_ct in the byterange/multipart handling may not be