Return-Path: Delivered-To: apmail-apache-cvs-archive@apache.org Received: (qmail 84930 invoked by uid 500); 2 Nov 2000 18:03:31 -0000 Mailing-List: contact apache-cvs-help@apache.org; run by ezmlm Precedence: bulk Reply-To: new-httpd@apache.org list-help: list-unsubscribe: list-post: Delivered-To: mailing list apache-cvs@apache.org Received: (qmail 84911 invoked by uid 500); 2 Nov 2000 18:03:28 -0000 Delivered-To: apmail-apache-1.3-cvs@apache.org Date: 2 Nov 2000 18:03:23 -0000 Message-ID: <20001102180323.84904.qmail@locus.apache.org> From: wrowe@locus.apache.org To: apache-1.3-cvs@apache.org Subject: cvs commit: apache-1.3 STATUS wrowe 00/11/02 10:03:19 Modified: . STATUS Log: Would every committer please take just a minute to assure you haven't applied/solved one of the lingering 1.3.x issues in the recent or even the distant past? Revision Changes Path 1.854 +35 -7 apache-1.3/STATUS Index: STATUS =================================================================== RCS file: /home/cvs/apache-1.3/STATUS,v retrieving revision 1.853 retrieving revision 1.854 diff -u -r1.853 -r1.854 --- STATUS 2000/10/14 09:35:17 1.853 +++ STATUS 2000/11/02 18:03:14 1.854 @@ -1,5 +1,5 @@ 1.3 STATUS: - Last modified at [$Date: 2000/10/14 09:35:17 $] + Last modified at [$Date: 2000/11/02 18:03:14 $] Release: @@ -30,12 +30,39 @@ RELEASE SHOWSTOPPERS: + * Parse the ServerName arg for invalid characters at startup. + Bug report 6787 and others are very confused that these bugs + don't appear until requests to the invalid hostname are + rejected in http_protocol.c - wasting bug reviewers time. + + * This shouldn't be, but it is since non-core modules install + with backslashes in the Win32 httpd.conf file. Need to either + (#1) accept backslashes back into the core, or (#2) always + canonicalize the elements before testing if an absolute paths + when concatinating them to the ServerRoot. See the thread + titled: [Patch] 1.3.15 ap_os_is_path_absolute + Status #1: OtherBill -1, Brian Harvard +1 + Status #2: OtherBill +1, Brian Harvard -1 + + * Likewise, mod_alias paths are never canonicalized today. Needs + a fix to canonicalize all but perhaps the regex Aliases, and + while we are at it, additional feature to accept Alias relative + to ServerRoot could be good. See the above showstopper and the + thread titled: Resolution - 1.3.15 ap_os_is_path_absolute + Status: Patch proposed; won't work for OS2; still in review. + RELEASE NON-SHOWSTOPPERS BUT WOULD BE REAL NICE TO WRAP THESE UP: - * Root all file systems with for WIN32/OS2/NW permissions - Status: NETWARE still does not handle all these variants properly: - - Brad Nicholes is researching + * Patch to piped logs to prevent procs from being spawned over + and over again once they return a 'fatal' error (e.g. volume + full, bad args, etc.) + Status: Victor Orlikowski is working up a patch + + * mod_autoindex results return no timestamp, so they are cached + past their legitimate lifetime. Patch to add ETag and + Last Modified based on directory mtime posted to new-httpd + Message ID <001f01c0441e$b719c180$93c0b0d0@covalentwrowe1> + Status: * long pathnames with many components and no AllowOverride None Workaround is to define with AllowOverride None, @@ -116,7 +143,8 @@ Status: * Stipe Tolj's Cygwin32 port - PR#: 2936 + New Message-ID: <39F6BD04.7C8922B0@wapme-systems.de> for 1.3.14 + Original PR#: 2936 Status: Lars +1 (on concept), Martin +1 (on concept), Ken +1 (concept), Jim +1 (concept) @@ -392,7 +420,7 @@ from OtherBill <002001bff8b6$cb53b420$345985d0@corecomm.net> Status: Is this done, now that we use scan_script_headers_err_strs? - * Apache console on Windows 95 (98?) won't close, close button/menu + * Apache console on Windows 95 alone won't close, close button/menu option can't (?) be intercepted, any workarounds out there? There is also an unverified report that Apache in a Win98 (95?) console window is ignoring shutdown/logoff, but this is not a