STATUS revision fbfdb4a45ff0e32dbfef0d414c8d75fe8aebcbb9
55e5c51e661e23e24573db84114a3837817745c9Evan HuntApache 2.0 STATUS:
431a83fb29482c5170b3e4026e59bb14849a6707Tinderbox UserLast modified at [$Date: 1999/08/31 21:46:35 $]
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt 2.0: In pre-alpha development
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Vague goal of an alpha or beta release in 1999. Commit-then-review
55e5c51e661e23e24573db84114a3837817745c9Evan HuntRELEASE SHOWSTOPPERS:
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * CGI doesn't work
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Status: Ben Laurie is working on this.
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * mod_status doesn't work
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Status: Manoj Kasichainula <manojk@io.com> is working on this. A
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt patch has been posted that starts to fix this:
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt (<19990825025653.A19065@samosa.mindspring.com> and
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt <19990825022146.A10096@io.com>)
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * suEXEC doesn't work
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Ben Laurie's work to fix CGIs will also fix this.
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Windows NT port isn't done
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Status: Bill is working on this through his MPM work, and Ryan
dcf426e9b546d4bcc0681904551752af43c1bcd6Evan Hunt is with APR.
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Current 2.0 code is not tested on many Unix platforms. Make 2.0
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt work on most, if not all the systems 1.3 did
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Status: Autoconf and APR will get us most of the way there.
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * The module API is a weird combination of the old table and the new
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt hook system. Switch completely to the new hook system
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Status: Ben Laurie is working on this.
55e5c51e661e23e24573db84114a3837817745c9Evan HuntRELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP:
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Switch to autoconf, et al. for configuration.
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Use APR to get rid of more platform dependancies.
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Status: Ryan Bloom <rbb@raleigh.ibm.com> is working on this.
55e5c51e661e23e24573db84114a3837817745c9Evan HuntOther bugs that need fixing:
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * MaxRequestsPerChild measures connections, not requests.
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Regex containers don't work in an intutive way
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Status: No one has come up with an efficient way to fix this
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt behavior. Dean has suggested getting rid of regex containers
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Areas where APRFile is being used need to be cleaned-up
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * SIGSEGV on Linux seems to only kill a thread not a whole process;
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt we need to work around this, probably by bouncing the signal to
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt the sigwait thread. But this will hurt debugability.
55e5c51e661e23e24573db84114a3837817745c9Evan HuntOther features that need writing:
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Finish infrastructure in core for async MPMs
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * TODO in source -- just do an egrep on "TODO" and see what's there
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Odd comments in source (egrep for "ZZZ") need to be cleaned-up
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Work on other MPMs. Possible MPMs discussed or in progress include:
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt - David Reid's <abb37@dial.pipex.com> Beos MPM
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Status: in progress
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt - Dean Gaudet's async MPM
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt - Zach Brown's <zab@zabbo.net> Linux siginfo MPM
55e5c51e661e23e24573db84114a3837817745c9Evan HuntDocumentation that needs writing:
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * The concept of MPMs, especially if we ship more than one MPM for a
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt given platform
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * New directives in the various MPMs
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * API documentation
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Status: Ben Laurie has written some hooks documentation
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Changes since 1.3.9 can be more easily seen in the commitlog file
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt dev.apache.org:/home/cvs/CVSROOT/commitlogs/apache-2.0
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt which includes some of Roy's comments when the changes were
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt committed in rough change-sets by purpose. Note that the commitlog
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt does not show the contents of new files until later.
55e5c51e661e23e24573db84114a3837817745c9Evan HuntAvailable Patches:
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Mike Abbott's <mja@trudge.engr.sgi.com> patches to improve
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt Status: These were written for 1.3, and are awaiting a port to
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * What do we do about mod_proxy?
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Which MPMs will be included with Apache 2.0?
55e5c51e661e23e24573db84114a3837817745c9Evan Hunt * Is conf/highperformance.conf-dist obsolete? It looks obsolete.