STATUS revision 7ac1ffe4121328055c55cc1db7fc4aeae0fcc3d6
02c335c23bf5fa225a467c19f2c063fb0dc7b8c3Timo SirainenAPACHE 2.3 STATUS: -*-text-*-
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan BoschLast modified at [$Date$]
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan BoschThe current version of this file can be found at:
bdd36cfdba3ff66d25570a9ff568d69e1eb543cfTimo Sirainen * http://svn.apache.org/repos/asf/httpd/httpd/trunk/STATUS
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan BoschDocumentation status is maintained seperately and can be found at:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * docs/STATUS in this source tree, or
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * http://svn.apache.org/repos/asf/httpd/httpd/trunk/docs/STATUS
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan BoschConsult the following STATUS files for information on related projects:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * http://svn.apache.org/repos/asf/apr/apr/trunk/STATUS
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * http://svn.apache.org/repos/asf/apr/apr-util/trunk/STATUS
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan BoschPatches considered for backport are noted in their branches' STATUS:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * http://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x/STATUS
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * http://svn.apache.org/repos/asf/httpd/httpd/branches/2.0.x/STATUS
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * http://svn.apache.org/repos/asf/httpd/httpd/branches/2.2.x/STATUS
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan BoschRelease history:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch [NOTE that x.{odd}.z versions are strictly Alpha/Beta releases,
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch while x.{even}.z versions are Stable/GA releases.]
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch 2.3.9 : In Development.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch 2.3.8 : Tagged on August 24, 2010.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch 2.3.7 : Tagged on August 19, 2010, not released.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch 2.3.6 : Released on June 21, 2010.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch 2.3.5 : Released on January 26, 2010.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch 2.3.4 : Released on December 8, 2009.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch 2.3.3 : Tagged on November 11, 2009, not released.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch 2.3.2 : Tagged on March 23, 2009, not released.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch 2.3.1 : Tagged on January 2, 2009, not released.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch 2.3.0 : Tagged on December 6, 2008, not released.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan BoschContributors looking for a mission:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Just do an egrep on "TODO" or "XXX" in the source.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Review the bug database at: http://issues.apache.org/bugzilla/
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Review the "PatchAvailable" bugs in the bug database:
0dffa25d211be541ee3c953b23566a1a990789dfTimo Sirainen https://issues.apache.org/bugzilla/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&product=Apache+httpd-2&keywords=PatchAvailable
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch After testing, you can append a comment saying "Reviewed and tested".
0dffa25d211be541ee3c953b23566a1a990789dfTimo Sirainen * Open bugs in the bug database.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan BoschCURRENT RELEASE NOTES:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan BoschRELEASE SHOWSTOPPERS:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Modules that are not ready for production use should be marked as
96fbbe9c6fd04bca25770020bf94eec50888bc9bStephan Bosch experimental. Candidates:
84a31ddebcf8c3aeaef7b9e54171a48ee1374c45Timo Sirainen * Review the example configuration. It should be based on current best
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch practices and not use deprecated features.
84a31ddebcf8c3aeaef7b9e54171a48ee1374c45Timo Sirainen * Modules without documentation need to be moved to experimental or be
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Not all MPMs are updated to set conn_rec::current_thread correctly.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch (Prefork, Worker, Event, Simple are updated).
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch jim sez: Then we just ship with those... mark any others as
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch FOR NEXT ALPHA:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * The mod_session* modules need to be checked that their hooks respect
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch the returning of int (HTTP status codes) and apr_status_t as appropriate,
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch and any anomolies fixed.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch OLD ISSUES THAT WERE THOUGHT TO BE SHOWSTOPPERS FOR 2.2 BUT OBVIOUSLY WEREN'T:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Handling of non-trailing / config by non-default handler is broken
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch http://marc.theaimsgroup.com/?l=apache-httpd-dev&m=105451701628081&w=2
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch jerenkrantz asks: Why should this block a release?
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch wsanchez agrees: this may be a change in behavior, but isn't
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch clearly wrong, and even if so, it doesn't seem like a
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * the edge connection filter cannot be removed
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch http://marc.theaimsgroup.com/?l=apache-httpd-dev&m=105366252619530&w=2
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch jerenkrantz asks: Why should this block a release?
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch stas replies: because it requires a rewrite of the filters stack
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch implementation (you have suggested that) and once 2.2 is
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch released you can't do that anymore.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch pgollucci: this affects mod_perl I'm pretty sure.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan BoschCURRENT VOTES:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Name the Server (version 2.4 or 3.0, depending on the final call)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Recent discussion indicates we should designate a (short name).
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch This is not yet a [Vote] - Your nominations please:
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch * Apache HTTP Server (httpd)
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch +1: sctemme (why mess with it?)
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan BoschRELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch * Add mod_allow_method or some other (usable) functionality to replace
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Sort out modules selections for most/all/reallyall. Maybe rename
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch all -> most, reallyall -> all, and remove the old 'most'.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Add mod_define.
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch * Patches submitted to the bug database:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch http://issues.apache.org/bugzilla/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&product=Apache+httpd-2&keywords=PatchAvailable
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Filter stacks and subrequests, redirects and fast redirects.
a103139eb1c241c2aff0c001758eb2da9d8abf44Stephan Bosch There's at least one PR that suffers from the current unclean behaviour
a103139eb1c241c2aff0c001758eb2da9d8abf44Stephan Bosch (which lets the server send garbage): PR 17629
a103139eb1c241c2aff0c001758eb2da9d8abf44Stephan Bosch nd says: Every subrequest should get its own filter stack with the
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch subreq_core filter as bottom-most. That filter does two things:
df97665f0f471d21566b38cc856253a0adadc3b7Stephan Bosch - swallow EOS buckets
df97665f0f471d21566b38cc856253a0adadc3b7Stephan Bosch - redirect the data stream to the upper request's (rr->main)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch filter chain directly after the subrequest's starting
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Once we have a clean solution, we can try to optimize
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch it, so that the server won't be slow down too much.
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch * RFC 2616 violations.
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch Closed PRs: 15852, 15857, 15859, 15861, 15864, 15869, 15870, 16120,
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch 16125, 16135, 16136, 16137, 16138, 16139, 16140, 16518,
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Open PRs: 15865, 15866, 15868, 16126, 16133, 16142, 16521
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch jerenkrantz says: need to decide how many we need to backport and/or
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch if these rise to showstopper status.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch wrowe suggests: it would be nice to see "MUST" v.s. "SHOULD" v.s. "MAY"
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch out of this list, without reviewing them individually.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * There is a bug in how we sort some hooks, at least the pre-config
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch hook. The first time we call the hooks, they are in the correct
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch order, but the second time, we don't sort them correctly. Currently,
efe78d3ba24fc866af1c79b9223dc0809ba26cadStephan Bosch the modules/http/config.m4 file has been renamed to
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch modules/http/config2.m4 to work around this problem, it should moved
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch back when this is fixed.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch OtherBill offers that this is a SERIOUS problem. We do not sort
84a31ddebcf8c3aeaef7b9e54171a48ee1374c45Timo Sirainen correctly by the ordering arguments passed to the register hook
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch functions. This was proven when I reordered the open_logs hook
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch to attempt to open the error logs prior to the access logs. Possibly
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch the entire sorting code needs to be refactored.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * pipes deadlock on all platforms with limited pipe buffers (e.g. both
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch Linux and Win32, as opposed to only Win32 on 1.3). The right solution
df97665f0f471d21566b38cc856253a0adadc3b7Stephan Bosch is either GStein's proposal for a "CGI Brigade", or OtherBill's proposal
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch for "Poll Buckets" for "Polling Filter Chains". Or maybe both :-)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * All handlers should always send content down even if r->header_only
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch is set. If not, it means that the HEAD requests don't generate the
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch same headers as a GET which is wrong.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * exec cmd and suexec arg-passing enhancements
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Status: Patches proposed
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Message-ID: <20020526041748.A29148@prodigy.Redbrick.DCU.IE>
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch (see the "proc.patch" and "suexec-shell.patch" links in this message)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * The 2.0.36 worker MPM graceless shutdown changes work but are
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch a bit clunky on some platforms; eg, on Linux, the loop to
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch join each worker thread seems to hang, and the parent ends up
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch killing off the child with SIGKILL. But at least it shuts down.
8e5d894529f30fe6c9671495f678daf0ae7f6478Timo Sirainen chrisd: Has this been fixed by the changes for PR 38737?
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * We do not properly substitute the prefix-variables in the configuration
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch scripts or generated-configs. (i.e. if sysconfdir is etc,
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch httpd-std.conf points to conf.)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * If any request gets through ap_process_request_internal() and is
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch scheduled to be served by the core handler, without a flag that this
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch r->filename was tested by dir/file_walk, we need to 500 at the very
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch end of the ap_process_request_internal() processing so sub_req-esters
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch know this request cannot be run. This provides authors of older
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch modules better compatibility, while still improving the security and
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch robustness of 2.0.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Status: still need to decide where this goes, OtherBill comments...
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Message-ID: <065701c14526$495203b0$96c0b0d0@roweclan.net>
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch [Deleted comments regarding the ap_run_handler phase, as irrelevant
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch as BillS points out that "common case will be caught in
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch default_handler already (with the r->finfo.filetype == 0 check)"
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch and the issue is detecting this -before- we try to run the req.]
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch gregames says: can this happen somehow without a broken module
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch being involved? If not, why waste cycles trying to defend against
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch potential broken modules? It seems futile.
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch wrowe counters: no, it shouldn't happen unless the module is broken.
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch But the right answer is to fail the request up-front in dir/file
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch walk if the path was entirely invalid; and we can't do that either
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch UNTIL 2.1 or we break modules that haven't hooked map_to_storage.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Can a static httpd be built reliably?
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Message-ID: <20020207142751.T31582@clove.org>
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Usage of APR_BRIGADE_NORMALIZE in core_input_filter should be
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch removed if possible.
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch <Pine.LNX.4.33.0201202232430.318-100000@deepthought.cs.virginia.edu>
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch Jeff wonders if we still care about this. It is no longer an
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch API issue but simply an extra trip through the brigade.
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch * Try to get libtool inter-library dependency code working on AIX.
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch Message-ID: <cm3n10lx555.fsf@rdu163-40-092.nc.rr.com>
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Justin says: If we get it working on AIX, we can enable this
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch on all platforms and clean up our build system somewhat.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Jeff says: I thought I tested a patch for you sometime in
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch January that you were going to commit within a few days.
efe78d3ba24fc866af1c79b9223dc0809ba26cadStephan Bosch * Handling of %2f in URIs. Currently both 1.3 and 2.0
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch completely disallow %2f in the request URI path (see
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch ap_unescape_url() in util.c). It's permitted and passed
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch through in the query string, however. Roy says the
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch original reason for disallowing it, from five years ago,
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch was to protect CGI scripts that applied PATH_INFO to
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch a filesystem location and which might be tricked by
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch ..%2f..%2f(...). We *should* allow path-info of the
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch form 'http://foo.com/index.cgi/path/to/path%2finfo'.
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch Since we've revamped a lot of our processing of path
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch segments, it would be nice to allow this, or at least
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch allow it conditionally with a directive.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch OtherBill adds that %2f as the SECOND character of a multibyte
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch sequence causes the request to fail! This happens notably in
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch the ja-jis encoding.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * There is increasing demand from module writers for an API
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch that will allow them to control the server à la apachectl.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Reasons include sole-function servers that need to die if
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch an external dependency (e.g., a database) fails, et cetera.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Perhaps something in the (ever more abused) scoreboard?
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch On the other hand, we already have a pipe that goes between parent
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch and child for graceful shutdown events, along with an API that
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch can be used to send a message down that pipe. In threaded MPMs,
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch it is easy enough to make that one pipe be used for graceful
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch and graceless events, and it is also easy to open that pipe
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch to both parent and child for writing. Then we just need to
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch figure out how to do graceless on non-threaded MPMs.
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch * Allow the DocumentRoot directive within <Location > scopes? This
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch allows the beloved (crusty) Alias /foo/ /somepath/foo/ followed
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch by a <Directory /somepath/foo> to become simply
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch <Location /foo/> DocumentRoot /somefile/foo (IMHO a bit more legible
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch and in-your-face.) DocumentRoot unset would be accepted [and would
ecc3dad2120f048a5c0344f9a0af546487f4d43cStephan Bosch not permit content to be served, only virtual resources such as
efe78d3ba24fc866af1c79b9223dc0809ba26cadStephan Bosch server-info or server-status.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch This proposed change would _not_ depricate Alias.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch striker: See the thread starting with Message-ID:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch JLEGKKNELMHCJPNMOKHOGEEJFBAA.striker@apache.org.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Win32: Rotatelogs sometimes is not terminated when Apache
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch goes down hard. FirstBill was looking at possibly tracking the
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch child's-child processes in the parent process.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch stoddard: Shared scoreboard might offer a good way for the parent
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch to keep track of 'other child' processes and whack them if the child
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch Other thoughts on walking the process chain using the NT kernel
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch have also been proposed on APR.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Eliminate unnecessary creation of pipes in mod_cgid
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch * Combine log_child and piped_log_spawn. Clean up http_log.c.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Common logging API.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Platforms that do not support fork (primarily Win32 and AS/400)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Architect start-up code that avoids initializing all the modules
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch in the parent process on platforms that do not support fork.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * There are still a number of places in the code where we are
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch losing error status (i.e. throwing away the error returned by a
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch system call and replacing it with a generic error code)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Mass vhosting version of suEXEC.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * All DBMs suffer from confusion in support/dbmmanage (perl script) since
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch the dbmmanage employs the first-matched dbm format. This is not
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch necessarily the library that Apache was built with. Aught to
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch rewrite dbmmanage upon installation to bin/ with the proper library
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch for predictable mod_auth_dbm administration.
df97665f0f471d21566b38cc856253a0adadc3b7Stephan Bosch Questions; htdbm exists, time to kill dbmmanage, or does it remain
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch useful as a perl dbm management example? If we keep it,
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch do we address the issue above?
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Integrate mod_dav.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Some additional items remaining:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch - case_preserved_filename stuff
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch (use the new canonical name stuff?)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch - find a new home for ap_text(_header)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch - is it possible to remove the DAV: namespace stuff from util_xml?
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * ap_core_translate() and its use by mod_mmap_static and mod_file_cache
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch are a bit wonky. The function should probably be exposed as a utility
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch function (such as ap_translate_url2fs() or ap_validate_fs_url() or
69c5fffbba024bc531106f00a9469c1f810a084dStephan Bosch something). Another approach would be a new hook phase after
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch "translate" which would allow the module to munge what the
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch translation has decided to do.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Status: Greg +1 (volunteers)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Explore use of a post-config hook for the code in http_main.c which
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch calls ap_fixup_virutal_hosts(), ap_fini_vhost_config(), and
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch ap_sort_hooks() [to reduce the logic in main()]
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * read the config tree just once, and process N times (as necessary)
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * (possibly) use UUIDs in mod_unique_id and/or mod_usertrack
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * (possibly) port the bug fix for PR 6942 (segv when LoadModule is put
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch into a VirtualHost container) to 2.0.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * shift stuff to mod_core.h
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * callers of ap_run_create_request() should check the return value
df97665f0f471d21566b38cc856253a0adadc3b7Stephan Bosch for failure (Doug volunteers)
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch * Fix the worker MPM to use POD to kill child processes instead
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch of ap_os_killpg, regardless of how they should die.
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch chrisd: Is this done, by any chance? See r92598 and r93358.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Scoreboard structures could be changed in the future such that
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch proper alignment is not maintained, leading to segfaults on
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch some systems. Cliff posted a patch to deal with this issue but
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch later recanted. See this message to dev@apr.apache.org:
96fbbe9c6fd04bca25770020bf94eec50888bc9bStephan Bosch <Pine.LNX.4.44.0203011354090.16457-200000@deepthought.cs.virginia.edu>
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * APXS either needs to be fixed completely for use when apr is out of tree,
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch or it should drop query mode altogether, and we just grow an
96fbbe9c6fd04bca25770020bf94eec50888bc9bStephan Bosch httpd-config or similar arrangement.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch To quote a discussion in STATUS earlier:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch thommay: this doesn't fix all the problems with apxs and out of
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch tree apr/apr-util, but it's a good start. There's still the
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch query cases; but I'm beginning to think that in these cases
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch the app should be querying ap{r,u}-config directly
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch deprecate -q: add htpd-config: gstein, pquerna, minfrin, pgollucci
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan BoschTODO ISSUES REMAINING IN MOD_SSL:
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch * Do we need SSL_set_read_ahead()?
96fbbe9c6fd04bca25770020bf94eec50888bc9bStephan Bosch * the ssl_expr api is NOT THREAD SAFE. race conditions exist:
96fbbe9c6fd04bca25770020bf94eec50888bc9bStephan Bosch -in ssl_expr_comp() if SSLRequire is used in .htaccess
b22a040754a7011cfbb4f5cca02fb525a496e73fStephan Bosch (ssl_expr_info is global)
b22a040754a7011cfbb4f5cca02fb525a496e73fStephan Bosch -is ssl_expr_eval() if there is an error
b22a040754a7011cfbb4f5cca02fb525a496e73fStephan Bosch (ssl_expr_error is global)
b22a040754a7011cfbb4f5cca02fb525a496e73fStephan Bosch * SSLRequire directive (parsing of) leaks memory
b22a040754a7011cfbb4f5cca02fb525a496e73fStephan Bosch * Diffie-Hellman-Parameters for temporary keys are hardcoded in
b22a040754a7011cfbb4f5cca02fb525a496e73fStephan Bosch ssl_engine_dh.c, while the comment in ssl_engine_kernel.c says:
b22a040754a7011cfbb4f5cca02fb525a496e73fStephan Bosch "it is suggested that keys be changed daily or every 500
b22a040754a7011cfbb4f5cca02fb525a496e73fStephan Bosch transactions, and more often if possible."
b22a040754a7011cfbb4f5cca02fb525a496e73fStephan Bosch * ssl_var_lookup could be rewritten to be MUCH faster
96fbbe9c6fd04bca25770020bf94eec50888bc9bStephan Bosch * CRL callback should be pluggable
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch * session cache store should be pluggable
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * init functions should return status code rather than ssl_die()
96fbbe9c6fd04bca25770020bf94eec50888bc9bStephan Bosch * ssl_engine_pphrase.c needs to be reworked so it is generic enough
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch to also decrypt proxy keys
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * output warning when allowing SSL v2.0 ? its so old
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * mod_proxy: Ability to run SSL over proxy gateway connections,
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch encrypting (or reencrypting) at the proxy.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * mod_cache: Handle ESI tags.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * mod_cache: Resolve issue of how to cache page fragments (or perhaps
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch -if- we want to cache page fragments). Today, mod_cache/mod_mem_cache
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch will cache #include 'virtual' requests (but not #include 'file'
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch requests). This was accomplished by making CACHE_IN a
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch CONTENT_SET-1 filter to force it to run before the SUBREQ_CORE
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch filter. But now responses cannot be cached that include the
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch effects of having been run through CONTENT_SET filters
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch (mod_deflate, mod_expires, etc). We could rerun all the
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch CONTENT_SET filters on the cached response, but this will not
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch work in all cases. For example, mod_expires relies on installing
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch the EXPIRATION filter during fixups. Contents served out of
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch mod_cache (out of the quick_handler) bypass -all- the request
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch line server hooks (Ryan really hated this. It is great for
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch performance, but bad because of the complications listed above).
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch * mod_mem_cache: Consider adding a RevalidateTimeout directive to
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch specify time at which local cached content is to be revalidated
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch (ie, underlying file stat'ed to see if it has changed).
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * mod_cache: CacheEnable/CacheDisable should accept regular expressions.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch jerenkrantz says: Too slow. Get regexs away from speedy caches by
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch default. Introduce a new CacheEnableRegex if you want.
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch * mod_mem_cache/mod_disk_cache: Need to be able to query cache
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch status (num of entries, cache object properties, etc.).
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch mod_status could be extended to query optional hooks defined
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch by modules for the purpose of reporting module status.
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch mod_cache (et. al.) could define optional hooks that are called
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch to collect status. Status should be queryable by
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch HTTP or SNMP?
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch jerenkrantz says: Yawn. Who cares.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * MaxRequestsPerChild measures connections, not requests.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Until someone has a better way, we'll probably just rename it
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch "MaxConnectionsPerChild".
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch * Regex containers don't work in an intutive way
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Status: No one has come up with an efficient way to fix this
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch behavior. Dean has suggested getting rid of regex containers
df97665f0f471d21566b38cc856253a0adadc3b7Stephan Bosch OtherBill suggests: We at least seem to agree on eliminating
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch the <Container ~ foo> forms, and using only
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch <ContainerMatch foo> semantics.
636e3a4d35aeed92e50ceb72c4e0e815bb77ed0aStephan Bosch * orig_ct in the byterange/multipart handling may not be
df97665f0f471d21566b38cc856253a0adadc3b7Stephan Bosch needed. Apache 1.3 just never stashed "multipart" into
df97665f0f471d21566b38cc856253a0adadc3b7Stephan Bosch r->content_type. We should probably follow suit since the
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch byterange stuff doesn't want the rest of the code to see the
df97665f0f471d21566b38cc856253a0adadc3b7Stephan Bosch multipart content-type; the other code should still think it is
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch dealing with the <orig_ct> stuff.
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Status: Greg volunteers to investigate (esp. since he was most
659f431cb56f26ec07e308db4d6c563c0eaa5300Stephan Bosch likely the one to break it :-)
df97665f0f471d21566b38cc856253a0adadc3b7Stephan BoschEXPERIMENTAL MODULES:
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch Experimental modules should eventually be be promoted to fully supported
96fbbe9c6fd04bca25770020bf94eec50888bc9bStephan Bosch status or removed from the repository entirely (ie, the
df97665f0f471d21566b38cc856253a0adadc3b7Stephan Bosch 'experiment' failed). This section tracks what needs to happen to
f9511e684858bf5f6ac77ab12254b85b737beae8Stephan Bosch get the modules promoted to fully supported status.