in an unkillable process if shutdown_pending doesn't manage to trigger
process exit.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96349 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96348 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96347 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96346 13f79535-47bb-0310-9956-ffa450edef68

first and then performs a check of the version, only warning the
user if they chose, or we selected, an old version of OpenSSL.
This change also allows the code to work for non-openssl libraries
selected via the --with-ssl=dir option, which can override the
automated library check in any case.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96345 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96334 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96333 13f79535-47bb-0310-9956-ffa450edef68

Submitted by: Mathihalli Madhusudan <madhusudan_mathihalli@hp.com>


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96328 13f79535-47bb-0310-9956-ffa450edef68

  Security rollup for 2.0.40 release.  Tag and roll baby.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96327 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96326 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96325 13f79535-47bb-0310-9956-ffa450edef68

  Notations for win32/iconv, and fix a typo


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96324 13f79535-47bb-0310-9956-ffa450edef68

  Reenable apr-iconv for ongoing development effort.  Sorry that we had
  to defer this, but building with the patch [which will be placed in
  patches] will allow anyone to build an xlate-enabled aprutil for ongoing
  svn and other development.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96323 13f79535-47bb-0310-9956-ffa450edef68

  Since library detection is _not_ complete, and this code is extremely
  alpha quality, backout the apr-iconv for Win32 changes from the .40
  release.  Many have objected [not -1 exactly] to forcing iconv for any
  user to build xlate, and the detection/fixup patches are a blocker for
  releasing this overdue version.

  Recommiting immediately after tagging this sans-iconv point for .40, and
  create a patch anyone can apply to enable this for the release tarball.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96322 13f79535-47bb-0310-9956-ffa450edef68

extra eyeballs would be appreciated.

If it's not really dead, then we need to re-arrange this function so
that earlier changes to the r aren't lost.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96321 13f79535-47bb-0310-9956-ffa450edef68

chain.
PR:
Obtained from:
Submitted by:	Peter Van Biesen <peter.vanbiesen@vlafo.be>
Reviewed by:	Graham Leggett <minfrin@apache.org>


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96320 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96319 13f79535-47bb-0310-9956-ffa450edef68

This will allow things like...
SetEnv gzip-only-text/html 1
BrowserMatch "GECKO" gzip-only-text/html=0

Obtained from: Andr� Schild <A.Schild@aarboard.ch>
Reviewed by:   Ian Holsman


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96318 13f79535-47bb-0310-9956-ffa450edef68

return anything.

Submitted by: Thom May <thom@planetarytramp.net>


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96317 13f79535-47bb-0310-9956-ffa450edef68

  Pardon je, it seems the most obvious solutions are always the easiest
  to incorporate new bugs.  Solve the 0index.html bad uri bug.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96316 13f79535-47bb-0310-9956-ffa450edef68

  Wasn't the only one with the essential idea, co-credit where credit is
  due.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96315 13f79535-47bb-0310-9956-ffa450edef68

  This was never a 'vulnerability'... the APR_FILEPATH_SECUREROOT flag
  passed to apr_filepath_merge refused to merge any rooted 'addpath'.
  However, that isn't the traditional 1.3 behavior, so fly past any
  leading '/'s on the way to merging the uri to the DocumentRoot.

PR: 10946


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96314 13f79535-47bb-0310-9956-ffa450edef68

  worth noting


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96313 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96312 13f79535-47bb-0310-9956-ffa450edef68

Submitted by: Zeno <zeno@cgisecurity.net>


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96311 13f79535-47bb-0310-9956-ffa450edef68

  This was an identity... sort of silly really.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96309 13f79535-47bb-0310-9956-ffa450edef68

  INVALID_SOCKET is -1 and appropriate to test throughout.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96308 13f79535-47bb-0310-9956-ffa450edef68

  Lots o bugs and emits here.  accept_socket is a handle, although vc and
  the psdk aren't good about remembering that.  never use GetLastError()
  to ap_log_error.  remember that we could have a socket closed by our
  neighbor while we were blocking.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96307 13f79535-47bb-0310-9956-ffa450edef68

  We generate an .rc now, not a .h


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96306 13f79535-47bb-0310-9956-ffa450edef68

  Outch!  Always check completion against actual socket/pipe/file handles,
  never look at the completion event itself.  Could be part of the zone
  alarm mischief.  Note also that waiting on two events is better than
  waking every few minutes.  Factoring out max_requests_per_child_hit
  or whatever that event name is will help here.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96305 13f79535-47bb-0310-9956-ffa450edef68

  Just an observation, we potentially read/retain far too much of the body,
  and this can be optimized.  Not a problem for sendfile based byterange
  requests, but potentially lethal to serve byterange requests of any
  parsed or cgi generated responses.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96304 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96303 13f79535-47bb-0310-9956-ffa450edef68

  We describe body lengths throughout apr in apr_off_t lengths.  While we
  all agree we won't ever cache an entity that huge in memory, the disk
  cache is another matter [e.g. if we cache an existing file.]


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96302 13f79535-47bb-0310-9956-ffa450edef68

  Revert to CRYPTO_malloc_init for non-rsa [gotta read the diffs more
  closely the first time.]


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96301 13f79535-47bb-0310-9956-ffa450edef68

  Support rsaref compilation


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96300 13f79535-47bb-0310-9956-ffa450edef68

  catch up with the apr_atoll -> apr_atoi64 rename


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96299 13f79535-47bb-0310-9956-ffa450edef68

  Ignore goodness for VC6 .mak file exporters


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96298 13f79535-47bb-0310-9956-ffa450edef68

  More ignore goodness


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96297 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96296 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96294 13f79535-47bb-0310-9956-ffa450edef68