summaryrefslogtreecommitdiff
path: root/www/apache
AgeCommit message (Collapse)AuthorFilesLines
2004-04-07Add three patches to resolve security issue:reed4-1/+121
SECURITY [CAN-2003-0020]: escape arbitrary data before writing into the errorlog The three patches are from Apache cvs. http://cvs.apache.org/viewcvs.cgi/apache-1.3/src/ Also bump PKGREVISION too.
2004-03-29Add a BUILDLINK_PKGBASE.<pkg> definition where it's not equal to <pkg>,jlam2-2/+4
e.g. "BUILDLINK_PKGBASE.gtk?= gtk+". This is mandated by the example buildlink[23].mk files in bsd.buildlink[23].mk.
2004-03-05Reorder location and setting of BUILDLINK_PACKAGES to match templatejlam1-7/+11
buildlink3.mk file in revision 1.101 of bsd.buildlink3.mk.
2004-02-28Forgot to commit this part. MAKE_ENV needs CHGRP="${CHGRP}".snj1-2/+2
2004-02-28Use ${CHGRP} instead of chgrp. From Georg Schwarz in PR pkg/24594.snj2-7/+10
2004-02-18Include "../../mk/compiler.mk" to get the value of CC_VERSION beforejlam1-1/+3
testing it.
2004-02-11Fix pasto(?) in BUILDLINK_PKGSRCDIR.apache's value.seb1-2/+2
2004-02-09USE_GCC[23], USE_SUNPRO, and USE_MIPSPRO are all deprecated.jlam1-2/+2
2004-01-24Append to BUILDLINK_DEPENDS.<pkg> instead of setting a default value. Injlam1-2/+2
the normal case when BUILDLINK_DEPENDS.<pkg> isn't specified, it receives a value only once due to the multiple inclusion protection in the bulldlink3.mk files. In the case where a package includes several buildlink3.mk files that each want a slightly different version of another dependency, having BUILDLINK_DEPENDS.<pkg> be a list allows for the strictest <pkg> dependency to be matched.
2004-01-05Use S/+$// instead of C/\+$// to save a backslash. Very highlyjlam1-2/+2
recommended by seb :)
2004-01-04Re-arrange to match example buildlink3.mk file in bsd.buildlink3.mk.jlam1-10/+9
2004-01-03Initial sprinkling of work-in-progress buildlink3.mk files for using thejlam1-0/+29
buildlink3 framework.
2003-11-02Updated apache to 1.3.29.grant7-128/+15
Major changes since 1.3.28: Security vulnerabilities * CAN-2003-0542 (cve.mitre.org) Fix buffer overflows in mod_alias and mod_rewrite which occurred if one configured a regular expression with more than 9 captures. Bugs fixed The following noteworthy bugs were found in Apache 1.3.28 (or earlier) and have been fixed in Apache 1.3.29: * Within ap_bclose(), ap_pclosesocket() is now called * consistently for sockets and ap_pclosef() for files. Also, closesocket() is used consistenly to close socket fd's. The previous confusion between socket and file fd's would cause problems with some applications now that we proactively close fd's to prevent leakage. * Fixed mod_usertrack to not get false positive matches on the user-tracking cookie's name. * Prevent creation of subprocess Zombies when using CGI wrappers such as suEXEC and cgiwrap.
2003-10-29Import security fix from 1.3.29 distribution for a buffer overflow incube6-3/+119
mod_rewrite and mod_alias, referenced CAN-2003-0542. Bump PKGREVISION.
2003-10-11bump PKGREVISIONs after bump of expat BUILDLINK_DEPENDS.grant1-1/+2
2003-09-17get rid of USE_GCC2/3 in pkg Makefiles, and set GCC_REQD orgrant1-4/+5
USE_PKGSRC_GCC as appropriate, as this is handled by compiler.mk now.
2003-07-19Update "apache" package to version 1.3.28. Changes since version 1.3.27:tron3-13/+18
- SECURITY: CAN-2003-0460 (cve.mitre.org) Fix the rotatelogs support program on Win32 and OS/2 to ignore special control characters received over the pipe. Previously such characters could cause it to quit logging and exit. [André Malo] - Prevent the server from crashing when entering infinite loops. The new LimitInternalRecursion directive configures limits of subsequent internal redirects and nested subrequests, after which the request will be aborted. PR 19753 (and probably others). [William Rowe, Jeff Trawick, Jim Jagielski, André Malo] - Fix NULL-pointer issue in ab when parsing an incomplete or non-HTTP response. PR 21085. [Glenn Nielsen <glenn@apache.org>, André Malo] - Removed BIND_NOSTART from HP/UX shl_load() logic for loadable Apache modules, so that statics are initialized when the module is loaded (especially critical for c++ modules on HPUX.) [William Rowe, Noah Arliss <narliss@netegrity.com>] - Win32 build system changes; always recompile buildmark.c (used for Apache -v 'server built' messages) even when Apache is built from within the IDE; build test_char.h and uri_delims.h from within the ApacheCore.dsp project. PR 12706. [William Rowe] - Introduce Win32 .pdb diagnostic symbols into the Apache 1.3 build (as created in Apache 2.0.45 and later.) Makes debugging and analysis of crash dumps and Dr. Watson logs trivial. Requires the Win32 binary builder to set aside the exact .pdb files that match the released binaries (.exe/.so files) for reference by users and developers. [William Rowe] - Make sure the accept mutex is released before calling child exit hooks and cleanups. Otherwise, modules can segfault in such code and, with pthread mutexes, leave the server deadlocked. Even if the module doesn't segfault, if it performs extensive processing it can temporarily prevent the server from accepting new connections. [Jeff Trawick] - Fix mod_rewrite's handling of absolute URIs. The escaping routines now work scheme dependent and the query string will only be appended if supported by the particular scheme. [André Malo] - Use appropriate language codes for Czech (cs) and Traditional Chinese (zh-tw) in default config files. PR 9427. [André Malo] - Don't block synchronous signals (e.g., SIGSEGV) while waiting for and holding a pthread accept mutex. [Jeff Trawick] - AIX: Change the default accept mechanism from pthread back to fcntl. Idle child cleanup doesn't work when the child selected for termination by the parent is waiting on a pthread mutex, and because the AIX kernel's notion of hot process is apparently the same as Apache's, it is common for the Apache parent to continually select a child for termination that the kernel will leave waiting on the mutex for extended periods of time. There are other concerns with pthread mutexes as well, such as the ability to deadlock the server if a child process segfaults while holding the mutex. [Jeff Trawick] - Fix a pair of potential buffer overflows in htdigest [Martin Schulze <joey@infodrom.org>, Thom May] - A newly created child now has a start_time of 0, to prevent mod_status from displaying a bogus value for the "time to process most recent request" column for freshly-started children in a previously-used scoreboard slot. [Martin Kraemer] - When using Redirect in directory context, append requested query string if there's no one supplied by configuration. PR 10961. [André Malo] - Fix path handling of mod_rewrite, especially on non-unix systems. There was some confusion between local paths and URL paths. PR 12902. [André Malo] - backport from 2.x series: Prevent endless loops of internal redirects in mod_rewrite by aborting after exceeding a limit of internal redirects. The limit defaults to 10 and can be changed using the RewriteOptions directive. PR 17462. [André Malo] - Use the correct locations of srm.conf and access.conf when tailoring the httpd.conf during the install process. PR 9446. [Stanislav Brabec <utx@penguin.cz>] - suexec: Be more pedantic when cleaning environment. Clean it immediately after startup. PR 2790, 10449. [Jeff Stewart <jws@purdue.edu>, André Malo] - Fix apxs to insert LoadModule/AddModule directives only outside of sections. PR 8712, 9012. [André Malo] - Fix suexec compile error under SUNOS4, where strerror() doesn't exist. PR 5913, 9977. [Jonathan W Miner <Jonathan.W.Miner@lmco.com>] - Unix build: Add support for environment variable EXTRA_LDFLAGS_SHLIB, which allows the user to add to the hard-coded ld flags specified for DSOs. Compare with the existing LDFLAGS_SHLIB environment variable, which allows the user to completely replace the hard-coded ld flags specified for DSOs. [Jeff Trawick] - mod_auth_digest no longer tries to guess AuthDigestDomain, if it's not specified. Now it assumes "/" as already documented. PR 16937. [André Malo] - In configure always assume suexec-umask to be an octal value by prepending a "0". PR 16984. [André Malo] - Fix typo in suexec -V output. PR 9034. [Youichirou Koga <y-koga@apache.or.jp>] - Fix bug where 'Satisfy Any' without an AuthType resulted in an "Internal Server Error" response. PR 9076. [André Malo] - mod_rewrite: Allow "RewriteEngine Off" even if no "Options FollowSymlinks" (or SymlinksIfOwnermatch) is set. PR 12395. [André Malo] - Change the log messages for setsockopt(TCP_NODELAY) and getsockname() failures to log the client IP address and to change the log level to debug. [Jeff Trawick] - Correction to mod_negotation for Win32, OS2, Netware etc, where case insensitive requests such as the HEADER or README search from autoindex would fail to match HEADER.html (because the system internally looked for the case-sensitive header.* pattern.) PR 7300 [William Rowe] - Correction to mod_autoindex so that only text/* files (prefering /html, then /plain, then some other flavor) can be recovered from a multiview-based HEADER or README subrequest. [William Rowe] - Improvements to mod_usertrack that allows for a regular (verbose) as well as "compact" version of the tracking cookie (the new 'CookieFormat' directive), and the ability to prepend a string to the cookie via the 'CookiePrefix' directive. [Pål Løberg <pallo@initio.no>, with cleanup by Jim Jagielski] - Certain 3rd party modules would bypass the Apache API and not invoke ap_cleanup_for_exec() before creating sub-processes. To such a child process, Apache's file descriptors (lock fd's, log files, sockets) were accessible, allowing them direct access to Apache log file etc. Where the OS allows, we now add proactive close functions to prevent these file descriptors from leaking to the child processes. [Jim Jagielski, Martin Kraemer] - Prevent obscenely large values of precision in ap_vformatter from clobbering a buffer. [Sander Striker, Jim Jagielski] - NetWare: implemented ap_os_default_port() to resolve the correct default port based on the request method. This fixes a URL reconstruction problem on a redirect. [Pavel Novy (novy@feld.cvut.cz)] - Added new ap_register_cleanup_ex() API function which allows for a "magic" cleanup function to be run at register time rather than at cleanup time. Also added the ap_note_cleanups_for_(socket|fd|file)_ex() API functions which allows for control over whether that magic cleanup should be called or not. This does not change the default behavior of the non-"ex" function (eg: ap_register_cleanup). [Jim Jagielski, concept by Ben Laurie] - PORT: Take advantage of OpenBSD's arc4random() function for the initial secret [Henning Brauer <hb-apache-dev at bsws.de>] - If Listen directive is not a port, but just an IP, emit an error condition as this case is ambiguous. [Rich Bowen, Justin Erenkrantz, Cliff Woolley] - Update timeout algorithm in free_proc_chain. If a subprocess did not exit immediately, the thread would sleep for 3 seconds before checking the subprocess exit status again. In a very common case when the subprocess was an HTTP server CGI script, the CGI script actually exited a fraction of a second into the 3 second sleep, which effectively limited the server to serving one CGI request every 3 seconds across a persistent connection. PRs 6961, 8664 [Bill Stoddard] - mod_setenvif: Add SERVER_ADDR special keyword to allow envariable setting according to the server IP address which received the request. [Ken Coar] - PORT: Enable SINGLE_LISTEN_UNSERIALIZED_ACCEPT for AIX 4.3.2 and above. Update AIX configure logic to allow higher AIX release numbers without having to change Apache. [Jeff Trawick]
2003-07-18gcc.buildlink2.mk is now handled automatically.grant1-4/+1
2003-07-17s/netbsd.org/NetBSD.org/grant1-3/+3
2003-06-06replace ${WHOLE_ARCHIVE_FLAG} -lgcc ${NO_WHOLE_ARCHIVE_FLAG} by justdrochner1-2/+2
${LINK_ALL_LIBGCC_HACK}
2003-05-31tidy up logic for adding libgcc and libgcc_s. no functional change.grant1-7/+7
2003-05-31make sure modules are built with the correct version of gcc.grant1-4/+9
2003-05-31for consistency, use WHOLE_ARCHIVE_FLAG instead ofgrant1-2/+2
-Wl,--whole-archive.
2003-05-31drop some unnecessary parensgrant1-2/+2
2003-05-31link against shared libgcc_s when built with gcc3 so Apache modulesgrant1-2/+10
can resolve the required symbols.
2003-04-15be quiet in all targets except post-install.grant1-11/+11
2003-04-10Fix fetch stage. Noted broken by grant.wiz1-2/+2
2003-04-10Update to 1.3.27nb4: compile with mod-ssl-2.8.14.wiz2-16/+10
Remove quite some lint while here.
2003-03-14(1) Publicly export the value of _OPSYS_RPATH_NAME as RPATH_FLAG;jlam1-2/+2
Makefiles simply need to use this value often, for better or for worse. (2) Create a new variable FIX_RPATH that lists variables that should be cleansed of -R or -rpath values if ${_USE_RPATH} is "no". By default, FIX_RPATH contains LIBS, X11_LDFLAGS, and LDFLAGS, and additional variables may be appended from package Makefiles.
2003-02-23Use MASTER_SITE_APACHE.jmmv1-3/+3
2003-02-17also pass CPPFLAGS to apxs.grant1-3/+3
2003-02-17pass LDFLAGS to apxs.grant1-2/+2
2003-02-17set the default source files to *.c, as this is common in our apachegrant1-2/+4
modules.
2003-02-17USE_PERL5 is already defined in apache/buildlink2.mk.grant1-3/+1
2003-02-17add makefile fragment for Apache modules.grant1-0/+47
This provides a build and install target to simplify the writing of Apache module Makefiles. ok'd by wiz.
2003-01-28Instead of including bsd.pkg.install.mk directly in a package Makefile,jlam1-2/+2
have it be automatically included by bsd.pkg.mk if USE_PKGINSTALL is set to "YES". This enforces the requirement that bsd.pkg.install.mk be included at the end of a package Makefile. Idea suggested by Julio M. Merino Vidal <jmmv at menta.net>.
2003-01-27Use XARGS="${XARGS}" rather than XARGS=${XARGS} when adding to MAKE_ENV.abs1-3/+3
Avoids breakage on Linux which has XARGS as "usr/bin/xargs -r"
2003-01-20Make this use ${CHOWN}, ${CHMOD}, ${XARGS} and ${FIND}, rather than thejschauma3-20/+61
hard-coded commands, which may not be in the PATH depending on the operating system. This partly addresses PR pkg/19804.
2002-12-22Revert to previous behavior and use db1 functions in libc if they aregrant1-4/+12
present, rather than installing databases/db (affects NetBSD, Linux, Darwin and possibly others). As suggested by Kimmo Suominen.
2002-12-19Tidy up conditional use of databases/db/buildlink2.mk.grant1-5/+7
Fixes build problem on Solaris.
2002-11-19As per discussion in pkg/18854, remove support for apache_start.conf asjlam1-19/+8
the precedence of the contents of this file changes depending on whether it's started at boot time or started manually, and it's not really necessary to add the extra complexity since it's valid (and easier) to just set apache_start in /etc/rc.conf.
2002-11-18Fix glitch in last commit.tron1-2/+2
2002-11-18Use "${NONBINMODE}" instead of mode "0" so that "pkg_admin check" stilltron1-3/+3
works for a normal user. Also invoke "chmod" only once because fork() and exec() is expensive on certain platforms.
2002-11-18Fix chmod (PR#19093 by Adrian Portelli). Bump PKGREVISION.martti1-3/+4
2002-11-15Updated apache to 1.3.27nb2martti1-2/+2
chmod 0 libexec/cgi-bin/{printenv,test-cgi} (to make nessus happy).
2002-11-15chmod 0 libexec/cgi-bin/{printenv,test-cgi} (to make nessus happy).martti1-1/+2
2002-10-25Unused.wiz1-60/+0
2002-10-25Update "apache" package to version 1.3.27nb1 which is now based versiontron2-5/+6
2.8.12-1.3.27 of "mod_ssl" module so that one doesn't need two version of the "mod_ssl" sources to build "apache".
2002-10-18Use newer way to convert a build dependency to a full dependency.jlam1-2/+2
2002-10-14Only set apache_start to "start" if it's not already set.lukem1-3/+3
Previously, if apache_start was set in /etc/rc.conf and /etc/rc.d/apache was loaded as part of the /etc/rc start sequence, apache_start's value would be overridden by "apache_start=start" in this script, because /etc/rc.conf would have already been loaded and load_rc_config() would not reload it again. This problem would not have been seen if /etc/rc.d/apache was started manually, or /etc/rc.conf.d/apache or @PKG_SYSCONFDIR@/apache_start.conf was used to set apache_start. (I am using /etc/rc.conf, and was wondering why apache wasn't starting with ssl support at boot, but worked after a manual restart...)