STATUS revision db3ccce11afac4fc1d4f51a65424412f7480c46c
998a0c99bc22357406f359ace2f602c5d6e376c6wroweApache 2.0 STATUS:
a601d863bd772fefc4dc82a883589d8be6a44811wroweLast modified at [$Date: 2000/06/06 21:46:40 $]
a601d863bd772fefc4dc82a883589d8be6a44811wrowe 2.0a4 : vague plans to release mid-May
a601d863bd772fefc4dc82a883589d8be6a44811wrowe 2.0a3 : released April 28, 2000
a601d863bd772fefc4dc82a883589d8be6a44811wrowe 2.0a2 : released March 31, 2000
a601d863bd772fefc4dc82a883589d8be6a44811wrowe 2.0a1 : released March 10, 2000
6f51bbc3054846e0c0a897d5f16ceba1726bebc6jerenkrantzRELEASE SHOWSTOPPERS:
6f51bbc3054846e0c0a897d5f16ceba1726bebc6jerenkrantz * Win32: Get mod_auth_digest working under win32
6f51bbc3054846e0c0a897d5f16ceba1726bebc6jerenkrantz - APR_HAS_RANDOM should be defined on windows and there is a
6f51bbc3054846e0c0a897d5f16ceba1726bebc6jerenkrantz lib/apr/misc/win32/rand.c which is basically a copy of what
6f51bbc3054846e0c0a897d5f16ceba1726bebc6jerenkrantz mod_auth_digest used to use.
ff920f8ae37ba372801b67ea7c3d5bf1dfb55585dreid * Re-work configuration for top level Apache. Work should start with
ff920f8ae37ba372801b67ea7c3d5bf1dfb55585dreid trying to clean the autoconf stuff. If and only if this proves
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe impossible to do (very unlikely), autoconf should be removed and we
ff920f8ae37ba372801b67ea7c3d5bf1dfb55585dreid will roll our own config implementation.
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe NOTE: When this is implemented, it would be really nice if the
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe default cgi module on Unix with a threaded MPM is the cgid
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe module. This should be override-able of course.
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe Status: Jim Jagielski is looking into this.
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe * suEXEC doesn't work
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe Status: Manoj has posted an patch to fix this.
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe <19991103003605.A20612@samosa.mindspring.com>
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe * Win32: Enable the Windows MPM to honor max_requests_per_child
ff920f8ae37ba372801b67ea7c3d5bf1dfb55585dreid Status: Bill will fix this.
a601d863bd772fefc4dc82a883589d8be6a44811wrowe * Win32: Get Apache working on Windows 95/98. The following work
a601d863bd772fefc4dc82a883589d8be6a44811wrowe (at least) needs to be done:
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe - winnt MPM: Fix 95/98 code paths in the winnt MPM. There is some NT
a601d863bd772fefc4dc82a883589d8be6a44811wrowe specific code that is still not in NT only code paths
a601d863bd772fefc4dc82a883589d8be6a44811wrowe - IOL binds to APR sendfile, implemented with TransmitFile, which
a601d863bd772fefc4dc82a883589d8be6a44811wrowe is not available on 95/98.
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe * Win32: Test access logging with multiple threads. Will the
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe native file I/O calls serialize automagically like the
a601d863bd772fefc4dc82a883589d8be6a44811wrowe CRT calls or do we need to add region locking each time
a601d863bd772fefc4dc82a883589d8be6a44811wrowe we access the logs?
a601d863bd772fefc4dc82a883589d8be6a44811wrowe * Win32: Complete the revamp the service environment and relocation
a601d863bd772fefc4dc82a883589d8be6a44811wrowe into the WinNT MPM. Changes ServerRoot service registry
a601d863bd772fefc4dc82a883589d8be6a44811wrowe parameter into ConfigArgs for multiple service startup parameters.
a601d863bd772fefc4dc82a883589d8be6a44811wrowe Problems to fix in the revamp: -k shutdown/restart are broken,
a601d863bd772fefc4dc82a883589d8be6a44811wrowe signals are not being acknowledged. Close window and shutdown
a601d863bd772fefc4dc82a883589d8be6a44811wrowe also seem out of sorts.
a601d863bd772fefc4dc82a883589d8be6a44811wrowe OtherBill is working on this
fc9e01023a2fb7f7af9b25621ab080bbe7a95611jerenkrantz * Win32: fix build/run time environment to remove ApacheCore.dll
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe * We need a thread-safe resolver, at least on Unix.
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe Status: The best known candidate would be something from
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe Status: Greg asks, "why? doesn't gethostname_r() handle this?"
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe * Fix up ./configure to properly detect the various dynamic load
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe libraries/flags on the different platforms (current, we default to
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe -ldl which is wrong).
998a0c99bc22357406f359ace2f602c5d6e376c6wroweRELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe * OS/2: Get loadable modules working again. Requires shared core support
bfae8a5d3c37df19dea49b411b109198c7e84a4fbrianp which doesn't appear to be catered for in the current build system.
bfae8a5d3c37df19dea49b411b109198c7e84a4fbrianp * OS/2: Make mod_status work for spmt_os2 MPM.
7dfb10a15cdfd48a23a9aa5713a2deb129821761wrowe * Build scripts do not recognise AIX 4.2.1 pthreads, so the
998a0c99bc22357406f359ace2f602c5d6e376c6wrowe pthread MPMs will not build.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Win32: Reuse accept socket after transmitfile/close
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein This is not a bug, but would be nice to get this feature in
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein before ship.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Win32: Enable the winnt MPM to use the new scoreboard API
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Win32: Implement ap_shm_ functions in APR.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Win32: Win9x console window still won't play nice with the
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein close window, logoff and shutdown scenarios.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Win32: Add a simple hold console open patch (wait for close or
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein the ESC key, with a nice message) if the server died a bad
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein death (non-zero exit code) in console mode.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Platforms that do not support fork (primarily Win32 and AS/400)
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein Consider introducing HAVE_FORK feature macro. Architect start-up code
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe that avoids initializing all the modules in the parent process on
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe platforms that do not support fork.
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe * Clean the code. There are a lot of places we used APR but didn't
29b927351fa38aca768d31d70116acd0a71ed415jerenkrantz remove the hacks that were required for the cross-platform code in
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein 1.3. We need to make the code look like APR was supposed to be there.
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein * Go throught the 1.3 Bug DB and research the bugs marked "suspended".
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein People were told these would be considered for inclusion in Apache 2.0,
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein it would be nice to actually do so.
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein * Win32: Migrate the MPM over to use APR thread/process calls. This
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein would eliminate some code in the Win32 branch that essentially
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein duplicates what is in APR.
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein Bill says we need a new procattr, APR_CREATE_SUSPENDED (or
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein something similar) to direct ap_create_process to create the
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein process suspended. We also need a call to wake up the suspended
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein process This may not be able to be implemented everywhere though.
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein * Move I/O layering into APR.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * There are still a number of places in the code where we are
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein loosing error status (i.e. throwing away the error returned by a
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe system call and replacing it with a generic error code)
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe * Win32: Implement reliable piped logs on Windows
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Switch to autoconf, et al. for configuration.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein Status: Manoj has placed an initial version into the 2.0
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein repository. Todos include:
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe * Use APR to get rid of more platform dependancies.
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe Status: Ryan Bloom <rbb@covalenet.net> is working on this.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * The connection status table is not very efficient. Also, very few stats
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein are exported to the connection status table (easy to fix), and mod_status
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe * Optimise the name-based <VirtualHost> implementation so that it uses
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe a hash table instead of a linear linked list scan.
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe * Mass vhosting version of suEXEC.
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe * Provide a sane API for handling the request's environment variables.
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe * configuration option to use *DBM
cdc56e100a8fa11e989d1633914502db1c8e0818wrowe Status: Greg +1 (volunteers)
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * add SDBM into src/lib/sdbm/ as a default/fallback DBM implementation.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein SDBM is used by Perl, mod_dav, mod_sssl, others for basic DBM support.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein Status: Greg +1 (volunteers)
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Get mod_mmap_static working on 2.0. Greg Ames posted a patch on
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein 05/26/2000 to get it running and to improve performance.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Integrate mod_dav.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein Message-id: forthcoming (plan to be posted before integration begins)
10a6f4803b893e9e77f5ad60ccb387ca1a15f84djerenkrantz Status: Greg +1 (volunteers :-)
10a6f4803b893e9e77f5ad60ccb387ca1a15f84djerenkrantzOther bugs that need fixing:
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * MaxRequestsPerChild measures connections, not requests.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein Until someone has a better way, we'll probably just rename it
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein "MaxConnectionsPerChild".
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Regex containers don't work in an intutive way
10a6f4803b893e9e77f5ad60ccb387ca1a15f84djerenkrantz Status: No one has come up with an efficient way to fix this
10a6f4803b893e9e77f5ad60ccb387ca1a15f84djerenkrantz behavior. Dean has suggested getting rid of regex containers
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * SIGSEGV on Linux (glibc 2.1.2) isn't caught properly by a
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein sigwaiting thread. We need to work around this, perhaps unless
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein there is hope soon for a fixed glibc.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gsteinOther features that need writing:
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Finish infrastructure in core for async MPMs
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein Status: post 2.0
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * TODO in source -- just do an egrep on "TODO" and see what's there
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gsteinDocumentation that needs writing:
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Mod_status docs are needed.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * The concept of MPMs, especially if we ship more than one MPM for a
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein given platform
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * New directives in the various MPMs and appropriate links from
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein obsolete directives in core.html to the MPM documentation.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Revise manual/stopping.html and the last part of
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein manual/misc/perf-tuning.html to take account of the MPMs.
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * API documentation
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein Status: Ben Laurie has written some hooks documentation
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein * Changes since 1.3.9 can be more easily seen in the commitlog file
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein dev.apache.org:/home/cvs/CVSROOT/commitlogs/apache-2.0
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein which includes some of Roy's comments when the changes were
27ddebab333df2a3d82b0f4ea63878d1d9f38ae8gstein committed in rough change-sets by purpose. Note that the commitlog
29b927351fa38aca768d31d70116acd0a71ed415jerenkrantz does not show the contents of new files until later.
29b927351fa38aca768d31d70116acd0a71ed415jerenkrantzAvailable Patches:
29b927351fa38aca768d31d70116acd0a71ed415jerenkrantz * Mike Abbott's <mja@trudge.engr.sgi.com> patches to improve
29b927351fa38aca768d31d70116acd0a71ed415jerenkrantz Status: These were written for 1.3, and are awaiting a port to
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein * What do we do about mod_proxy?
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein * Which MPMs will be included with Apache 2.0?
39d8628e74736d0bd8c4114d8e615dffa8fbbcf7gstein * Is conf/highperformance.conf-dist obsolete? It looks obsolete.