STATUS revision 6491ae8cad7bb8403ffb3a6ce17fafbe826ac817
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncAPACHE 2.3 STATUS: -*-text-*-
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncLast modified at [$Date$]
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncThe current version of this file can be found at:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * http://svn.apache.org/repos/asf/httpd/httpd/trunk/STATUS
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncDocumentation status is maintained seperately and can be found at:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * docs/STATUS in this source tree, or
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * http://svn.apache.org/repos/asf/httpd/httpd/trunk/docs/STATUS
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncConsult the following STATUS files for information on related projects:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * http://svn.apache.org/repos/asf/apr/apr/trunk/STATUS
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * http://svn.apache.org/repos/asf/apr/apr-util/trunk/STATUS
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncPatches considered for backport are noted in their branches' STATUS:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * http://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x/STATUS
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * http://svn.apache.org/repos/asf/httpd/httpd/branches/2.0.x/STATUS
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * http://svn.apache.org/repos/asf/httpd/httpd/branches/2.2.x/STATUS
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncRelease history:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync [NOTE that x.{odd}.z versions are strictly Alpha/Beta releases,
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync while x.{even}.z versions are Stable/GA releases.]
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync 2.3.7 : In development.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync 2.3.6 : Tagged June 11, 2010.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync 2.3.5 : Released on January 26, 2010.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync 2.3.4 : Released on December 8, 2009.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync 2.3.3 : Tagged on November 11, 2009, not released.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync 2.3.2 : Tagged on March 23, 2009, not released.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync 2.3.1 : Tagged on January 2, 2009, not released.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync 2.3.0 : Tagged on December 6, 2008, not released.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncContributors looking for a mission:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Just do an egrep on "TODO" or "XXX" in the source.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Review the bug database at: http://issues.apache.org/bugzilla/
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Review the "PatchAvailable" bugs in the bug database:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync https://issues.apache.org/bugzilla/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&product=Apache+httpd-2&keywords=PatchAvailable
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync After testing, you can append a comment saying "Reviewed and tested".
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Open bugs in the bug database.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncCURRENT RELEASE NOTES:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncRELEASE SHOWSTOPPERS:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Modules without documentation need to be moved to experimental or be
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * There is no working equivalent to 'Satisfy any' to authorize by
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync user _or_ IP address:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync http://mail-archives.apache.org/mod_mbox/httpd-dev/200912.mbox/<4B28E73C.4050209%40kippdata.de>
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Not all MPMs are updated to set conn_rec::current_thread correctly.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync (Prefork, Worker, Event, Simple are updated).
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync jim sez: Then we just ship with those... mark any others as
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync experimental
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Running the log_transaction hook from pool cleanup is fubar:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync http://marc.info/?l=apache-httpd-dev&m=123910381908293&w=1
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync sf: Has been changed to a pre-cleanup and should be ok now.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * MPM event (maybe others, too) closes open files only after the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync connection has been closed. This could be fixed in apr-util or httpd:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync http://mail-archives.apache.org/mod_mbox/httpd-dev/201005.mbox/<201005172311.39558.sf@sfritsch.de>
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync FOR NEXT ALPHA:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync OLD ISSUES THAT WERE THOUGHT TO BE SHOWSTOPPERS FOR 2.2 BUT OBVIOUSLY WEREN'T:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Handling of non-trailing / config by non-default handler is broken
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync http://marc.theaimsgroup.com/?l=apache-httpd-dev&m=105451701628081&w=2
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync jerenkrantz asks: Why should this block a release?
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync wsanchez agrees: this may be a change in behavior, but isn't
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync clearly wrong, and even if so, it doesn't seem like a
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync showstopper.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * the edge connection filter cannot be removed
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync http://marc.theaimsgroup.com/?l=apache-httpd-dev&m=105366252619530&w=2
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync jerenkrantz asks: Why should this block a release?
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync stas replies: because it requires a rewrite of the filters stack
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync implementation (you have suggested that) and once 2.2 is
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync released you can't do that anymore.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync pgollucci: this affects mod_perl I'm pretty sure.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncCURRENT VOTES:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * If the parent process dies, should the remaining child processes
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync "gracefully" self-terminate. Or maybe we should make it a runtime
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync option, or have a concept of 2 parent processes (one being a
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync "hot spare").
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync See: Message-ID: <3C58232C.FE91F19F@Golux.Com>
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Self-destruct: Ken, Martin, Lars, sctemme (parent shouldn't die, ever)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Not self-destruct: BrianP, Ian, Cliff, BillS
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Make it runtime configurable: Aaron, jim, Justin, wrowe, rederpj, nd, pgollucci
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync /* The below was a concept on *how* to handle the problem */
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Have 2 parents: +1: jim
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync -1: Justin, wrowe, rederpj, nd, pgollucci
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync +0: Lars, Martin (while standing by, could it do
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync something useful?)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Make the worker MPM the default MPM for threaded Unix boxes.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync +1: Justin, Ian, Cliff, BillS, striker, wrowe, nd, pgollucci, sctemme
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync +0: BrianP, Aaron (mutex contention is looking better with the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync latest code, let's continue tuning and testing), rederpj, jim
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync pquerna: Do we want to change this for *2.4*?
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync wrowe: Replies "yes"
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Name the Server (version 2.4 or 3.0, depending on the final call)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Recent discussion indicates we should designate a (short name).
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync This is not yet a [Vote] - Your nominations please:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Apache HTTP Server (httpd)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync +1: sctemme (why mess with it?)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncRELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Patches submitted to the bug database:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync http://issues.apache.org/bugzilla/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&product=Apache+httpd-2&keywords=PatchAvailable
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Filter stacks and subrequests, redirects and fast redirects.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync There's at least one PR that suffers from the current unclean behaviour
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync (which lets the server send garbage): PR 17629
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync nd says: Every subrequest should get its own filter stack with the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync subreq_core filter as bottom-most. That filter does two things:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync - swallow EOS buckets
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync - redirect the data stream to the upper request's (rr->main)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync filter chain directly after the subrequest's starting
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Once we have a clean solution, we can try to optimize
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync it, so that the server won't be slow down too much.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * RFC 2616 violations.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Closed PRs: 15852, 15857, 15859, 15861, 15864, 15869, 15870, 16120,
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync 16125, 16135, 16136, 16137, 16138, 16139, 16140, 16518,
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Open PRs: 15865, 15866, 15868, 16126, 16133, 16142, 16521
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync jerenkrantz says: need to decide how many we need to backport and/or
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync if these rise to showstopper status.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync wrowe suggests: it would be nice to see "MUST" v.s. "SHOULD" v.s. "MAY"
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync out of this list, without reviewing them individually.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * There is a bug in how we sort some hooks, at least the pre-config
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync hook. The first time we call the hooks, they are in the correct
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync order, but the second time, we don't sort them correctly. Currently,
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync the modules/http/config.m4 file has been renamed to
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync modules/http/config2.m4 to work around this problem, it should moved
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync back when this is fixed.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync OtherBill offers that this is a SERIOUS problem. We do not sort
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync correctly by the ordering arguments passed to the register hook
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync functions. This was proven when I reordered the open_logs hook
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync to attempt to open the error logs prior to the access logs. Possibly
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync the entire sorting code needs to be refactored.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * pipes deadlock on all platforms with limited pipe buffers (e.g. both
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Linux and Win32, as opposed to only Win32 on 1.3). The right solution
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync is either GStein's proposal for a "CGI Brigade", or OtherBill's proposal
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync for "Poll Buckets" for "Polling Filter Chains". Or maybe both :-)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * All handlers should always send content down even if r->header_only
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync is set. If not, it means that the HEAD requests don't generate the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync same headers as a GET which is wrong.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * exec cmd and suexec arg-passing enhancements
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Status: Patches proposed
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Message-ID: <20020526041748.A29148@prodigy.Redbrick.DCU.IE>
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync (see the "proc.patch" and "suexec-shell.patch" links in this message)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * The 2.0.36 worker MPM graceless shutdown changes work but are
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync a bit clunky on some platforms; eg, on Linux, the loop to
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync join each worker thread seems to hang, and the parent ends up
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync killing off the child with SIGKILL. But at least it shuts down.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync chrisd: Has this been fixed by the changes for PR 38737?
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * We do not properly substitute the prefix-variables in the configuration
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync scripts or generated-configs. (i.e. if sysconfdir is etc,
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync httpd-std.conf points to conf.)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * If any request gets through ap_process_request_internal() and is
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync scheduled to be served by the core handler, without a flag that this
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync r->filename was tested by dir/file_walk, we need to 500 at the very
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync end of the ap_process_request_internal() processing so sub_req-esters
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync know this request cannot be run. This provides authors of older
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync modules better compatibility, while still improving the security and
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync robustness of 2.0.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Status: still need to decide where this goes, OtherBill comments...
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Message-ID: <065701c14526$495203b0$96c0b0d0@roweclan.net>
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync [Deleted comments regarding the ap_run_handler phase, as irrelevant
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync as BillS points out that "common case will be caught in
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync default_handler already (with the r->finfo.filetype == 0 check)"
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync and the issue is detecting this -before- we try to run the req.]
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync gregames says: can this happen somehow without a broken module
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync being involved? If not, why waste cycles trying to defend against
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync potential broken modules? It seems futile.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync wrowe counters: no, it shouldn't happen unless the module is broken.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync But the right answer is to fail the request up-front in dir/file
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync walk if the path was entirely invalid; and we can't do that either
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync UNTIL 2.1 or we break modules that haven't hooked map_to_storage.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Can a static httpd be built reliably?
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Message-ID: <20020207142751.T31582@clove.org>
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Usage of APR_BRIGADE_NORMALIZE in core_input_filter should be
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync removed if possible.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Message-ID:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync <Pine.LNX.4.33.0201202232430.318-100000@deepthought.cs.virginia.edu>
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Jeff wonders if we still care about this. It is no longer an
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync API issue but simply an extra trip through the brigade.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Try to get libtool inter-library dependency code working on AIX.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Message-ID: <cm3n10lx555.fsf@rdu163-40-092.nc.rr.com>
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Justin says: If we get it working on AIX, we can enable this
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync on all platforms and clean up our build system somewhat.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Jeff says: I thought I tested a patch for you sometime in
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync January that you were going to commit within a few days.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Handling of %2f in URIs. Currently both 1.3 and 2.0
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync completely disallow %2f in the request URI path (see
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync ap_unescape_url() in util.c). It's permitted and passed
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync through in the query string, however. Roy says the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync original reason for disallowing it, from five years ago,
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync was to protect CGI scripts that applied PATH_INFO to
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync a filesystem location and which might be tricked by
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync ..%2f..%2f(...). We *should* allow path-info of the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync form 'http://foo.com/index.cgi/path/to/path%2finfo'.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Since we've revamped a lot of our processing of path
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync segments, it would be nice to allow this, or at least
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync allow it conditionally with a directive.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync OtherBill adds that %2f as the SECOND character of a multibyte
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync sequence causes the request to fail! This happens notably in
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync the ja-jis encoding.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * There is increasing demand from module writers for an API
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync that will allow them to control the server à la apachectl.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Reasons include sole-function servers that need to die if
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync an external dependency (e.g., a database) fails, et cetera.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Perhaps something in the (ever more abused) scoreboard?
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync On the other hand, we already have a pipe that goes between parent
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync and child for graceful shutdown events, along with an API that
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync can be used to send a message down that pipe. In threaded MPMs,
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync it is easy enough to make that one pipe be used for graceful
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync and graceless events, and it is also easy to open that pipe
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync to both parent and child for writing. Then we just need to
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync figure out how to do graceless on non-threaded MPMs.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Allow the DocumentRoot directive within <Location > scopes? This
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync allows the beloved (crusty) Alias /foo/ /somepath/foo/ followed
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync by a <Directory /somepath/foo> to become simply
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync <Location /foo/> DocumentRoot /somefile/foo (IMHO a bit more legible
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync and in-your-face.) DocumentRoot unset would be accepted [and would
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync not permit content to be served, only virtual resources such as
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync server-info or server-status.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync This proposed change would _not_ depricate Alias.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync striker: See the thread starting with Message-ID:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync JLEGKKNELMHCJPNMOKHOGEEJFBAA.striker@apache.org.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Win32: Rotatelogs sometimes is not terminated when Apache
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync goes down hard. FirstBill was looking at possibly tracking the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync child's-child processes in the parent process.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync stoddard: Shared scoreboard might offer a good way for the parent
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync to keep track of 'other child' processes and whack them if the child
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Other thoughts on walking the process chain using the NT kernel
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync have also been proposed on APR.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Eliminate unnecessary creation of pipes in mod_cgid
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Combine log_child and piped_log_spawn. Clean up http_log.c.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Common logging API.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Platforms that do not support fork (primarily Win32 and AS/400)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Architect start-up code that avoids initializing all the modules
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync in the parent process on platforms that do not support fork.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * There are still a number of places in the code where we are
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync losing error status (i.e. throwing away the error returned by a
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync system call and replacing it with a generic error code)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Mass vhosting version of suEXEC.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * All DBMs suffer from confusion in support/dbmmanage (perl script) since
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync the dbmmanage employs the first-matched dbm format. This is not
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync necessarily the library that Apache was built with. Aught to
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync rewrite dbmmanage upon installation to bin/ with the proper library
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync for predictable mod_auth_dbm administration.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Questions; htdbm exists, time to kill dbmmanage, or does it remain
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync useful as a perl dbm management example? If we keep it,
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync do we address the issue above?
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Integrate mod_dav.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Some additional items remaining:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync - case_preserved_filename stuff
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync (use the new canonical name stuff?)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync - find a new home for ap_text(_header)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync - is it possible to remove the DAV: namespace stuff from util_xml?
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * ap_core_translate() and its use by mod_mmap_static and mod_file_cache
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync are a bit wonky. The function should probably be exposed as a utility
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync function (such as ap_translate_url2fs() or ap_validate_fs_url() or
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync something). Another approach would be a new hook phase after
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync "translate" which would allow the module to munge what the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync translation has decided to do.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Status: Greg +1 (volunteers)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Explore use of a post-config hook for the code in http_main.c which
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync calls ap_fixup_virutal_hosts(), ap_fini_vhost_config(), and
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync ap_sort_hooks() [to reduce the logic in main()]
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * read the config tree just once, and process N times (as necessary)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * (possibly) use UUIDs in mod_unique_id and/or mod_usertrack
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * (possibly) port the bug fix for PR 6942 (segv when LoadModule is put
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync into a VirtualHost container) to 2.0.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * shift stuff to mod_core.h
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * callers of ap_run_create_request() should check the return value
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync for failure (Doug volunteers)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Fix the worker MPM to use POD to kill child processes instead
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync of ap_os_killpg, regardless of how they should die.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync chrisd: Is this done, by any chance? See r92598 and r93358.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Scoreboard structures could be changed in the future such that
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync proper alignment is not maintained, leading to segfaults on
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync some systems. Cliff posted a patch to deal with this issue but
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync later recanted. See this message to dev@apr.apache.org:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Message-ID:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync <Pine.LNX.4.44.0203011354090.16457-200000@deepthought.cs.virginia.edu>
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * APXS either needs to be fixed completely for use when apr is out of tree,
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync or it should drop query mode altogether, and we just grow an
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync httpd-config or similar arrangement.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync To quote a discussion in STATUS earlier:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync thommay: this doesn't fix all the problems with apxs and out of
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync tree apr/apr-util, but it's a good start. There's still the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync query cases; but I'm beginning to think that in these cases
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync the app should be querying ap{r,u}-config directly
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync deprecate -q: add htpd-config: gstein, pquerna, minfrin, pgollucci
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncTODO ISSUES REMAINING IN MOD_SSL:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Do we need SSL_set_read_ahead()?
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * the ssl_expr api is NOT THREAD SAFE. race conditions exist:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync -in ssl_expr_comp() if SSLRequire is used in .htaccess
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync (ssl_expr_info is global)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync -is ssl_expr_eval() if there is an error
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync (ssl_expr_error is global)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * SSLRequire directive (parsing of) leaks memory
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Diffie-Hellman-Parameters for temporary keys are hardcoded in
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync ssl_engine_dh.c, while the comment in ssl_engine_kernel.c says:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync "it is suggested that keys be changed daily or every 500
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync transactions, and more often if possible."
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * ssl_var_lookup could be rewritten to be MUCH faster
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * CRL callback should be pluggable
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * session cache store should be pluggable
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * init functions should return status code rather than ssl_die()
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * ssl_engine_pphrase.c needs to be reworked so it is generic enough
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync to also decrypt proxy keys
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * output warning when allowing SSL v2.0 ? its so old
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * mod_proxy: Ability to run SSL over proxy gateway connections,
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync encrypting (or reencrypting) at the proxy.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * mod_cache: Handle ESI tags.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * mod_cache: Resolve issue of how to cache page fragments (or perhaps
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync -if- we want to cache page fragments). Today, mod_cache/mod_mem_cache
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync will cache #include 'virtual' requests (but not #include 'file'
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync requests). This was accomplished by making CACHE_IN a
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync CONTENT_SET-1 filter to force it to run before the SUBREQ_CORE
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync filter. But now responses cannot be cached that include the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync effects of having been run through CONTENT_SET filters
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync (mod_deflate, mod_expires, etc). We could rerun all the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync CONTENT_SET filters on the cached response, but this will not
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync work in all cases. For example, mod_expires relies on installing
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync the EXPIRATION filter during fixups. Contents served out of
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync mod_cache (out of the quick_handler) bypass -all- the request
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync line server hooks (Ryan really hated this. It is great for
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync performance, but bad because of the complications listed above).
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * mod_mem_cache: Consider adding a RevalidateTimeout directive to
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync specify time at which local cached content is to be revalidated
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync (ie, underlying file stat'ed to see if it has changed).
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * mod_cache: CacheEnable/CacheDisable should accept regular expressions.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync jerenkrantz says: Too slow. Get regexs away from speedy caches by
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync default. Introduce a new CacheEnableRegex if you want.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * mod_mem_cache/mod_disk_cache: Need to be able to query cache
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync status (num of entries, cache object properties, etc.).
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync mod_status could be extended to query optional hooks defined
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync by modules for the purpose of reporting module status.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync mod_cache (et. al.) could define optional hooks that are called
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync to collect status. Status should be queryable by
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync HTTP or SNMP?
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync jerenkrantz says: Yawn. Who cares.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * MaxRequestsPerChild measures connections, not requests.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Until someone has a better way, we'll probably just rename it
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync "MaxConnectionsPerChild".
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * Regex containers don't work in an intutive way
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Status: No one has come up with an efficient way to fix this
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync behavior. Dean has suggested getting rid of regex containers
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync completely.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync OtherBill suggests: We at least seem to agree on eliminating
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync the <Container ~ foo> forms, and using only
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync <ContainerMatch foo> semantics.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync * orig_ct in the byterange/multipart handling may not be
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync needed. Apache 1.3 just never stashed "multipart" into
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync r->content_type. We should probably follow suit since the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync byterange stuff doesn't want the rest of the code to see the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync multipart content-type; the other code should still think it is
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync dealing with the <orig_ct> stuff.
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Status: Greg volunteers to investigate (esp. since he was most
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync likely the one to break it :-)
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsyncEXPERIMENTAL MODULES:
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync Experimental modules should eventually be be promoted to fully supported
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync status or removed from the repository entirely (ie, the
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync 'experiment' failed). This section tracks what needs to happen to
03532efdc331b598d3dedc8cc7477641c3b0dd12vboxsync get the modules promoted to fully supported status.