STATUS revision 05e4bc4973958917a458e0f27b260c117414df71
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsyncAPACHE 2.0 STATUS: -*-text-*-
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsyncLast modified at [$Date: 2002/03/08 21:34:20 $]
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.34 : in development
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.33 : tagged March 6, 2002.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.32 : released Feburary 16, 2002.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.31 : rolled Feburary 1, 2002. not released.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.30 : tagged January 8, 2002. not rolled.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.29 : tagged November 27, 2001. not rolled.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.28 : released November 13, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.27 : rolled November 6, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.26 : tagged October 16, 2001. not rolled.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.25 : rolled August 29, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.24 : rolled August 18, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.23 : rolled August 9, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.22 : rolled July 29, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.21 : rolled July 20, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.20 : rolled July 8, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.19 : rolled June 27, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.18 : rolled May 18, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.17 : rolled April 17, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.16 : rolled April 4, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.15 : rolled March 21, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0.14 : rolled March 7, 2001
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0a9 : released December 12, 2000
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0a8 : released November 20, 2000
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0a7 : released October 8, 2000
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0a6 : released August 18, 2000
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0a5 : released August 4, 2000
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0a4 : released June 7, 2000
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0a3 : released April 28, 2000
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0a2 : released March 31, 2000
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync 2.0a1 : released March 10, 2000
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsyncPlease consult the following STATUS files for information
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsyncon related projects:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsyncCURRENT RELEASE NOTES:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * 34 status: Let's get all API changes and showstoppers in this one.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsyncFINAL RELEASE SHOWSTOPPERS:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * If any request gets to the core handler, without a flag that this
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync r->filename was tested by dir/file_walk, we need to 500 at the very
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync end of the ap_process_request_internal() processing. This provides
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync authors of older modules better compatibility, while still improving
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync the security and robustness of 2.0.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Status: still need to decide where this goes, OtherBill comments...
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Message-ID: <065701c14526$495203b0$96c0b0d0@roweclan.net>
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync we need to look at halting this in the 'default handler' case,
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync and that implies pushing the 'handler election' into the request
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync internal processing phase from the run request phase.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Jim asks: would a stopgap be something bogus like adding another
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync flag to request_rec ala eos_sent and before we OK, if not set
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Jeff says: reviewing the original message and the one
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync follow-up (also from OtherBill) it looks like OtherBill had a
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync good handle on the problem, though I wonder why not just put a
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync simple check in default_handler to see if dir/file_walk has
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync been done (a footprint left by dir/file_walk doesn't have to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync be in request_rec; a better place is core_request_config)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * We need to find out the right place to add the AddOutputFilterByType
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync directive. mod_mime? No. core with fixups? That's what it is now.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync ap_pass_brigade() hook? As a filter that runs at HTTP_HEADER stage?
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync This needs to be resolved before the next public release.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Message-ID: <018701c1b99c$f0f24310$0a01230a@KOJ>
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * API changes planned for 2.0 that should happen before the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync GA release:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Free lists for bucket allocation
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Pool allocator change
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * We do not properly substitute the prefix-variables in the configuration
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync scripts or generated-configs. (i.e. if sysconfdir is etc,
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync httpd-std.conf points to conf.)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Aaron says: This is not a showstopper, these problems have existed
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync for as long as I can remember. It would be nice to fix
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync them but they are not new.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Address popular PRs
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Win32 doesn't install as service correctly [9863, 9914, 9961]
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Don't be stupid and cd to a blank directory when doing installs
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsyncCURRENT VOTES:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Should we always build binaries statically unless otherwise
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Message-ID: <20020129210006.B23512@Lithium.MeepZor.Com>
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync -1: Justin, Ian
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * If the parent process dies, should the remaining child processes
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync "gracefully" self-terminate. Or maybe we should make it a runtime
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync option, or have a concept of 2 parent processes (one being a
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync "hot spare").
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync See: Message-ID: <3C58232C.FE91F19F@Golux.Com>
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Self-destruct: Ken, Martin
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Not self-destruct: BrianP, Ian, Cliff, BillS
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Make it runtime configurable: Aaron, Jim, Justin
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Have 2 parents: +1: Jim
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync +0: Martin (while standing by, could it do
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync something useful?)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Make the worker MPM the default MPM for threaded Unix boxes.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync +1: Justin, Ian, Cliff
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync -0: Aaron (premature decision, needs more discussion), Lars
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsyncRELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Convert all instances of the old apr_lock_t type to the new
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Status: Aaron has converted all but the perchild MPM to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync use the new lock API. Since perchild has been put
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync on the back burner, this is no longer a showstopper.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Aaron will patch perchild as soon as it becomes
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync buildable again.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * With AP_MODE_EXHAUSTIVE in the core, it is finally clear to me
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync how the Perchild MPM should be re-written. It hasn't worked
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync correctly since filters were added because it wasn't possible to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync get the content that had already been written and the socket at
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync the same time. This mode lets us do that, so the MPM can be
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * htpasswd blindly processes the file you give it, and does no
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync sanity checking before totally corrupting whatever file it was
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync you thought you had. It should check the input file and bail
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync if it finds non-comment lines that do not contain exactly 1
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync ':' character.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Message-ID: <20020217150457.A31632@clove.org>
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Can a static httpd be built reliably?
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Message-ID: <20020207142751.T31582@clove.org>
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * [Ken] Test suite failures:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync o worker is also failing some of the 'cgi' subtests
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync (see <URL:http://Source-Zone.Org/Apache/regression/>):
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Justin says: "Worker should be fine and passes httpd-test here.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync If you can provide evidence that it can be reproduced
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync outside of httpd-test, then it's a showstopper. I
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync think it's a perl or a httpd-test problem."
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Not a showstopper: Justin
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Usage of APR_BRIGADE_NORMALIZE in core_input_filter should be
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync removed if possible.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Message-ID: <Pine.LNX.4.33.0201202232430.318-100000@deepthought.cs.virginia.edu>
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * There is a bug in how we sort some hooks, at least the pre-config
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync hook. The first time we call the hooks, they are in the correct
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync order, but the second time, we don't sort them correctly. Currently,
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync the modules/http/config.m4 file has been renamed to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync modules/http/config2.m4 to work around this problem, it should moved
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync back when this is fixed. rbb
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Justin says: "Is this really a showstopper? This has been here
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync forever. What's wrong? Does this have to do with
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync autoconf or m4?"
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Not a showstopper: Justin, BrianP, trawick, gregames
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * The Add...Filter and Set...Filter directives do not allow the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync administrator to order filters, beyond the order of filename (mime)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync extensions. It isn't clear if Set...Filter(s) should be inserted
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync before or after the Add...Filter(s) which are ordered by sequence of
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync filename extensions. At minimum, some sort of +-[0-10] syntax seems
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync like the quickest fix for a 2.0 gold release.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Justin says: "Could we delay this for a point release or 2.1?"
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Not a showstopper: justin, wrowe, trawick, stoddard, Jim, Ian, Aaron,
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Get perchild to work on platforms other than Linux. This
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync will require a portable mechanism to pass data and file/socket
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync descriptors between vhost child groups. An API was proposed
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync on dev@apr:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Message-ID: <20020111115006.K1529@clove.org>
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * CGI single-byte reads
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync BrianP suggests that this is caused by the ap_scan_script_header_err()
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync routine, which will do single-byte reads until it finds the end
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync of the header, at which point it constructs a pipe-bucket (buffered)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync to read from.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Proposed solution in:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Message-ID: <3C36ADAF.60601@cnet.com>
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Try to get libtool inter-library dependency code working on AIX.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Message-ID: <cm3n10lx555.fsf@rdu163-40-092.nc.rr.com>
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Justin says: If we get it working on AIX, we can enable this
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync on all platforms and clean up our build system
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Jeff says: I thought I tested a patch for you sometime in
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync January that you were going to commit within a few
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Handling of %2f in URIs. Currently both 1.3 and 2.0
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync completely disallow %2f in the request URI path (see
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync ap_unescape_url() in util.c). It's permitted and passed
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync through in the query string, however. Roy says the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync original reason for disallowing it, from five years ago,
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync was to protect CGI scripts that applied PATH_INFO to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync a filesystem location and which might be tricked by
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync ..%2f..%2f(...). We *should* allow path-info of the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync form 'http://foo.com/index.cgi/path/to/path%2finfo'.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Since we've revamped a lot of our processing of path
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync segments, it would be nice to allow this, or at least
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync allow it conditionally with a directive.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * FreeBSD, threads, and worker MPM. All seems to work fine
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync if you only have one worker process with many threads. Add
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync a second worker process and the accept lock seems to be
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync lost. This might be an APR issue with how it deals with
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync the child_init hook (i.e. the fcntl lock needs to be resynced).
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync More examination and analysis is required.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Status: This has also been reported on Cygwin.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Message-ID: <3C2CC514.8EF3BED1@wapme-systems.de> (cygnus)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Justin says: So, FreeBSD-CURRENT and Cywin have the same
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync problem. Yum. If another platform has this
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync with worker, this becomes a showstopper.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Aaron says: I spent some time disecting this and have come to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync the conclusion that it is not a problem in the worker MPM
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync (or at least, it is not isolated to a problem in worker).
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync I'll list some of the problems I'm seeing in case someone
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync else wants to pick up where I've left off:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync - Delivery of just about any signal to one of the child
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync processes will send it into an infinite loop as well.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync - Even though the parent is spinning out of control,
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync at first the child or children will appear to work
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync properly. At times it is possible to get it into a state,
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync however, where a request will hang until another concurrent
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync request "kicks" the first, at which point the second will
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync hang. My theory is that this has to do with the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync pthread_cond_*() implementation in FreeBSD, but it's still
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync possible that it is in APR.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Justin adds: Oh, FreeBSD threads are implemented entirely with
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync select()/poll()/longjmp(). Welcome to the nightmare.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync So, that means a ktrace output also has the thread
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync scheduling internals in it (since it is all the same to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync the kernel). Which makes it hard to distinguish between
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync our select() calls and their select() calls.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync *bangs head on wall repeatedly* But, some of the libc_r
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync files have a DBG_MSG #define. This is moderately helpful
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync when used with -DNO_DETACH. The kernel scheduler isn't
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync waking up the threads on a select(). Yum. And, I bet
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync those decrementing select calls have to do with the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync scheduler. Time to brush up on our OS fundamentals.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * There is increasing demand from module writers for an API
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync that will allow them to control the server � la apachectl.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Reasons include sole-function servers that need to die if
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync an external dependency (e.g., a database) fails, et cetera.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Perhaps something in the (ever more abused) scoreboard?
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync rbb: I don't believe the scoreboard is the correct mechanism
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync for this. We already have a pipe that goes between parent
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync and child for graceful shutdown events, along with an API that
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync can be used to send a message down that pipe. In threaded MPMs,
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync it is easy enough to make that one pipe be used for graceful
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync and graceless events, and it is also easy to open that pipe
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync to both parent and child for writing. Then we just need to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync figure out how to do graceless on non-threaded MPMs.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Allow the DocumentRoot directive within <Location > scopes? This
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync allows the beloved (crusty) Alias /foo/ /somepath/foo/ followed
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync by a <Directory /somepath/foo> to become simply
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync <Location /foo/> DocumentRoot /somefile/foo (IMHO a bit more legible
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync and in-your-face.) DocumentRoot unset would be accepted [and would
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync not permit content to be served, only virtual resources such as
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync server-info or server-status.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync This proposed change would _not_ depricate Alias.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Win32: Rotatelogs sometimes is not terminated when Apache
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync goes down hard. FirstBill was looking at possibly tracking the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync child's-child processes in the parent process.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync OtherBill asks, wasn't this fixed?
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync stoddard: Not fixed. Shared scoreboard might offer a good
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync way for the parent to keep track of 'other child' processes
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync and whack them if the child goes down.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Win32: Add a simple hold console open patch (wait for close or
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync the ESC key, with a nice message) if the server died a bad
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync death (non-zero exit code) in console mode.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Resolution: bring forward same ugly hacks from 1.3.13-.20
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Port of mod_ssl to Apache 2.0:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync The current porting state is summarized in modules/ssl/README. The
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync remaining work includes:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync (1) stablizing/optimizing the SSL filter logic
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync (2) Enabling the various SSL caching mechanisms (shmcb, shmht)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync (3) Enabling SSL extentions
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync (4) Trying to seperate the https filter logic from mod_ssl -
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync This is to facilitate other modules that wish to use the https
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync filter or the mod_ssl logic or both as required.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Justin: mod_ssl filter logic is redone, so that should be fine.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Madhu has submitted a patch for SSL caching - however, I
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync am -0 on that patch as I *think* we could implement the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync shared memory another way that is much cleaner (i.e.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync treat shmem directly as a dbm via APR routines). Justin
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync also thinks that the https filter logic may be sufficiently
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync decoupled now, but isn't really sure.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Performance & Debug: Eliminate most (and perhaps all) of the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync malloc/free calls in the bucket brigade code. Need some
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync light weight memory management functions that allow freeing
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync memory (putting it back into a memory pool) when it is no
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync longer needed. Enabling simple debugging features like guard
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync bands, double free detection, etc. would be cool but certainly
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync not a hard requirement.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Status: Cliff started to implement this using SMS as has
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync been discussed at length for months, but since
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync SMS is not being used anywhere else in the server,
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync several people expressed the opinion that we should
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync get rid of it entirely, meaning that the buckets
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync need their own memory management (free list) functions.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Cliff will implement that this weekend so we at least
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync have something to look at/compare with.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Eliminate unnecessary creation of pipes in mod_cgid
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Status: Ken asks, didn't Jeff fix this when he fixed the queue
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Combine log_child and piped_log_spawn. Clean up http_log.c.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Common logging API.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Document mod_file_cache.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Platforms that do not support fork (primarily Win32 and AS/400)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Architect start-up code that avoids initializing all the modules
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync in the parent process on platforms that do not support fork.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Win32: Migrate the MPM over to use APR thread/process calls. This
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync would eliminate some code in the Win32 branch that essentially
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync duplicates what is in APR.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * There are still a number of places in the code where we are
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync losing error status (i.e. throwing away the error returned by a
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync system call and replacing it with a generic error code)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Mass vhosting version of suEXEC.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * All DBMs suffer from confusion in support/dbmmanage (perl script) since
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync the dbmmanage employs the first-matched dbm format. This is not
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync necessarily the library that Apache was built with. Aught to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync rewrite dbmmanage upon installation to bin/ with the proper library
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync for predictable mod_auth_dbm administration.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Questions; htdbm exists, time to kill dbmmanage, or does it remain
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync useful as a perl dbm management example? If we keep it,
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync do we address the issue above?
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Integrate mod_dav.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Some additional items remaining:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync - case_preserved_filename stuff
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync (use the new canonical name stuff?)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync - find a new home for ap_text(_header)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync - is it possible to remove the DAV: namespace stuff from util_xml?
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * ap_core_translate() and its use by mod_mmap_static and mod_file_cache
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync are a bit wonky. The function should probably be exposed as a utility
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync function (such as ap_translate_url2fs() or ap_validate_fs_url() or
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync something). Another approach would be a new hook phase after
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync "translate" which would allow the module to munge what the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync translation has decided to do.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Status: Greg +1 (volunteers), Ryan +1
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Explore use of a post-config hook for the code in http_main.c which
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync calls ap_fixup_virutal_hosts(), ap_fini_vhost_config(), and
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync ap_sort_hooks() [to reduce the logic in main()]
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * read the config tree just once, and process N times (as necessary)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * (possibly) use UUIDs in mod_unique_id and/or mod_usertrack
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * (possibly) port the bug fix for PR 6942 (segv when LoadModule is put
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync into a VirtualHost container) to 2.0.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * shift stuff to mod_core.h
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * callers of ap_run_create_request() should check the return value
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync for failure (Doug volunteers)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Win32: Get Apache working on Windows 95/98. The following work
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync (at least) needs to be done:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync - Document warning that OSR2 is required (for Crypt functions, in
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync rand.c, at least.) This could be resolved with an SSL library, or
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync randomization in APR itself.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync - Bring the Win9xConHook.dll from 1.3 into 2.0 (no sense till it
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync actually works) and add in a splash of Win9x service code.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * In order to use a DSO version of mod_ssl we have to link with
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync -lssl and -lcrypto. A workaround is in place right now where the
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync entire EXTRA_LIBS macro is being appended to the objects list, but
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync this is a hack. We should either revamp the APACHE_CHECK_SSL_TOOLKIT
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync autoconf function or come up with some other autoconf checks to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync search for libssl and libcrypto and properly add them to mod_ssl's
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync link flags.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * Fix the worker MPM to use POD to kill child processes instead
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync of ap_os_killpg, regardless of how they should die. (Ryan Bloom)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * mod_cache: handle cache_control: no_cache "field_name" to enable
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync cacheing the response w/o header "field_name"
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync See RFC2616 section 14.9.1
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsyncPRs that have been suspended forever waiting for someone to
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsyncput them into 'the next release':
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#76: general
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync missing call to "setlocale();"
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#78: mod_include
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Additional status for XBitHack directive
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#362: mod_proxy
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Mod_proxy doesn't allow change of error pages
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#370: mod_env
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Modified PATH environemnt variable is not passed, instead
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync system's is used
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#440: mod_proxy
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Proxy doesn't deliver documents if not connected
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#534: mod_proxy
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync proxy converts ~name to %7Ename when name starts with a dot (.)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#537: mod_access
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync mod_access syntax allows hosts that should be restricted
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#557: mod_auth-any
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync ~UserHome directories are not honored in absolute pathname
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync requests (.htaccess)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#612: mod_proxy
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Proxy FTP Authentication Fails
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#623: mod_include
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync A smarter "Last Modified" value for SSI documents (see PR number 600)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#628: config
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Request of "Options SymLinksIfGroupMatch"
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#700: mod_proxy
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Proxy doesn't do links right for OpenVMS files through ftp:
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#759: mod_imap
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync imap should read <MAP><AREA>*</MAP> too!
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#793: general
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync RLimitCPU and RLimitMEM don't apply to all children like they should
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#921: suexec
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Uses cwd before filling it in, doesn't use syslog
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#922: config
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync it is useful to allow specifiction that root-owned symlinks
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync should always be followed
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#980: mod_proxy
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Controlling Access to Remote Proxies would be nice...
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#994: mod_proxy
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Adding authentication "on the fly" through the proxy module
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1004: apache-api
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync request_config field in request_rec is moderately bogus
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1028: other
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync DoS attacks involving memory consumption
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1050: mod_log-any
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Logging of virtual server to error_log as well
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1085: mod_proxy
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync ProxyRemote make a dead cycle.
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1117: mod_auth-any
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Using NIS passwd.byname dbm files with AuthDBMUserFile
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1120: suexec
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync suexec does not parse arguments to #exec cmd
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1145: mod_include
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Allow for Last-Modified: without resorting to XBitHack
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1158: apache-api
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync improvements to child spawning API
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1166: mod_proxy
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync ``nph-'' not honored (no buffering) for ProxyRemote mapping
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1176: mod_cgi
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Apache cannot handle continuation line in headers
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1191: general
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync setlogin() is not called, causing problems with e.g. identd
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1204: general
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync regerror() exists, use it
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1233: apache-api
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync there is no way to keep per-connection per-module state
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1263: mod_autoexec
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Add frame-safe anchor attribute to mod_autoindex links
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1268: suexec
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync CGI scripts running as Apache user: security (suexec etc.)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1285: suexec
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Error messages could be easier to spot in cgi.log file for suexec.c
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1287: mod_access
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync add allow,deny/deny,allow warning to mod_access
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1290: mod_proxy
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Need to know "hit-rate" on proxy cache
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1358: mod_log-any
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync Selective url-encode of log fields (or maybe a pseudo
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync log_rewrite module?)
a734c64bff58bda2fa48c2795453e092167b0ff7vboxsync * PR#1383: mod_headers
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