Jim Jagielski [Mon, 3 Jun 2002 12:28:27 +0000 (12:28 +0000)]
Attached is a patch for allowing user changes on the cygwin platform
and a #define wrapper for the timeout singal we use to kill of pending
open childs that do not react on "usual" signals.
The signalling issue seems to be a problem on the Cygwin platform, but
it's abstracted, so other platforms may benefit from it. Again,
nothing else is changed in behaviour.
Changes are:
* src/include/ap_config.h: added the system uid for Cygwin that is
the "root" user on Cygwin
* src/main/http_main.c: some cygwin specific #defines around
setpgrp() and getuid() calls. Adding the #define SIG_TIMEOUT_KILL to
define which singal should be used to kill of timed out childs.
Defaulting to the know value for all other plaforms.
* src/modules/proxy/proxy_cache.c: cygwin specific #define around
setpgrp()
Jim Jagielski [Thu, 30 May 2002 13:05:40 +0000 (13:05 +0000)]
Push out the T&R 2 days to account for the latest additions to
the proxy code (for testing). Also, to allow time for one more
vote (or not) on the Bugz 9181 patch.
PR:
Obtained from:
Submitted by:
Reviewed by:
Graham Leggett [Thu, 30 May 2002 10:19:49 +0000 (10:19 +0000)]
Add X-Forwarded-Host and X-Forwarded-Server to X-Forwarded-For
to the proxy.
PR:
Obtained from:
Submitted by: Thomas Eibner <thomas@stderr.net>
Reviewed by: Graham Leggett
Martin Kraemer [Wed, 29 May 2002 20:39:16 +0000 (20:39 +0000)]
Fix a problem in mod_proxy: it would not set the number of bytes
transferred, so other modules could not access the value from
the request_rec->bytes_sent field.
(And indeed, I also observed some time ago that the bytes_sent field was not set.)
PR: 6841
Submitted by: Anthony Howe <achowe (at) snert.com>
Reviewed by: Martin Kraemer
Doug MacEachern [Thu, 23 May 2002 03:25:20 +0000 (03:25 +0000)]
PR:
Obtained from:
Submitted by:
Reviewed by:
ap_escape_logitem referenced c2x() before it was declared, fatal error
with hpux cc. move c2x definition before ap_escape_logitem.
Martin Kraemer [Tue, 21 May 2002 13:03:56 +0000 (13:03 +0000)]
Apply a stricter check to the request line syntax, in order to prevent
arbitrary user input to end up (unescaped) in the access_log and error_log
files. Until now, garbage could be injected to spoof accesses to nonexistent
(or inaccessible) resources -- of course without the client actually
getting access to them.
Now anything but whitespace following the "<method> <url> HTTP/x.y" request
line is disallowed, and special characters in the request are escaped
in the log.
Martin Kraemer [Tue, 21 May 2002 12:43:32 +0000 (12:43 +0000)]
John Von Essen <john at essenz.com> contributes the diskimg.gif icon:
> Hopefully this is the right place to send this comment. I have alot of
> disk images (.img) on my server. By default, there is no icon for a disk
> image.
>
> So I made one...
>
> AddIcon /icons/diskimg.gif .img
>
> The diskimg.gif can be found at:
>
> http://www.essenz.com/icons/diskimg.gif
>
> Could this be added into future releases? Disk images are fairly popular.
> Especially when making boot floppies or driver disks for linux.
Martin Kraemer [Tue, 21 May 2002 12:24:59 +0000 (12:24 +0000)]
Get rid of DEFAULT_XFERLOG as it is not used anywhere. It was
preserved by the build system, printed with "httpd -V", but
apart from that completely ignored: the default transfer log
is to not produce any transfer log.
At least on Win32 ... close the book on nice-to-haves with no patch,
or no point (.exe's available from 2.0 builds, duplicating ApacheMonitor
in both trees is a waste and a maintenance headache.)
Besides, we are -discouraging- future use. If 1.3.2x is 'initial release'
quality, 2.0 is much closer to fine.
I suspect this might be fixed by now, if anyone wants to test the
remaining half of this issue (IndexIgnore .foo .bar should hide both
.foo and .bar). Also interesting to determine if it's good in 2.0.
Jim Jagielski [Wed, 15 May 2002 14:13:29 +0000 (14:13 +0000)]
No doubt, people are already either handling
this themselves, by clearing out env or else it's not a concern
for many people, or whatever... No need to track this anymore
Adds a '-F' flag to httpd's main() - which causes the mother
or supervisor process to not fork&detach - as to allow integration
with deamontools (http://cr.yp.to/daemontools.html).
PR: http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7628
Obtained from: http://marc.theaimsgroup.com/?l=apache-httpd-dev&m=101467598720760&w=2
Michael Handler <handler-grendel.net>
Submitted by: http://www.catnook.com/patches/apache-1.3.24-daemontools.patch
Jos Backus <jos-catnook.com>
Reviewed by: Dirk-Willem van Gulik
Martin Kraemer [Mon, 13 May 2002 08:13:06 +0000 (08:13 +0000)]
Fix propagation of ${OPTIM} to subdirectories (which used to work in
early 1.3's at least), so that it is possible to do a
make OPTIM=-g
from the src/ directory to temporarily override the default
(e.g., for debugging).
Martin Kraemer [Mon, 6 May 2002 08:34:13 +0000 (08:34 +0000)]
src/support/apxs.pl: re-arranged things to handle DSO extensions
in a less hardcoded way. (.so) because the shared building chain would
fail for Cygwin which uses native Win32 extensions (.dll).
Jim Jagielski [Thu, 2 May 2002 13:28:46 +0000 (13:28 +0000)]
OK. This is admittedly anal. But the whole idea behind cpp macros
is to avoid things like "we know NSLINKMODULE_OPTION_NONE is 0"
and making such shortcuts as this. This makes it clear what exactly
we are setting, and though admittedly verbose, the tradeoff of a
few bytes of source is worth it :)
Jim Jagielski [Mon, 22 Apr 2002 12:36:18 +0000 (12:36 +0000)]
For supported versions of Darwin, place dynamically loaded
Apache extensions' public symbols into the global symbol
table. This allows dynamically loaded PHP extensions.
NOTE: this works as long as multiple modules don't
include overlapping symbols. Jim added the wrapper to
check for the existance of the option (which doesn't
exist in Rhapsody).
PR:
Obtained from:
Submitted by: Marko Karppinen <markonen@php.net>
Reviewed by: Jim Jagielski, Wilfredo Sanchez
Graham Leggett [Sun, 21 Apr 2002 21:16:39 +0000 (21:16 +0000)]
Add a FIXME note around some broken code. The write is writing a buffer
of bogus data with a length that was never initialised.
PR:
Obtained from:
Submitted by:
Reviewed by:
Graham Leggett [Fri, 19 Apr 2002 11:15:19 +0000 (11:15 +0000)]
Correct proxy to be able to handle the unexpected 100-continue
reponses sent during PUT or POST requests. (ie Hotmail works now)
PR:
Obtained from:
Submitted by:
Reviewed by:
Martin Kraemer [Mon, 15 Apr 2002 09:41:22 +0000 (09:41 +0000)]
Sometimes (when a read happened to end on a chunk boundary) the proxy
would just hang trying to read a CRLF that did not come (remaining==0
because end-of-chunk, not because chunksize==0). Also, when it
received its data from an upstream proxy which did not really support
HTTP/1.1 but claimed it did (ICS, Squid, both using nonstandard
Proxy-Connection: headers), Apache would not clear out the
"Transfer-Enconding: chunked" header and failed to correctly serve the
request.
First, use addition to test against an unsigned for chunk_start after
we are certain chunk_start is a positive value. Second, ap_bread
is prototyped as an -int- ... not a size_t, not an off_t.
These changes eliminate all mismatch type/sign errors on Win32.
Martin Kraemer [Tue, 9 Apr 2002 12:02:49 +0000 (12:02 +0000)]
Do a "proper" indent by moving some curly braces outside of #if's.
Fix an error message to print the correct filename (not tempfile).
Remove the tempfile for Win32 on a failed rename(2).
PR:
Obtained from:
Submitted by:
Reviewed by:
Graham Leggett [Sun, 7 Apr 2002 19:12:27 +0000 (19:12 +0000)]
Correct a timeout problem within proxy which would force long
or slow POST requests to close after 300 seconds.
PR: 7572
Obtained from:
Submitted by: Martin Lichtin <martin@lichtin.net>, Brian Bothwell <brian.bothwell@wisdomtools.com>
Reviewed by:
Graham Leggett [Sat, 6 Apr 2002 13:20:48 +0000 (13:20 +0000)]
Remove the last little bit of the multiple-headers bug in proxy. Before,
only the last instance of a header would have been cached.
PR:
Obtained from:
Submitted by:
Reviewed by:
Graham Leggett [Sat, 6 Apr 2002 13:12:36 +0000 (13:12 +0000)]
Fix a log line message to be clearer. There have been queries in the past
as to whether this message was an Apache problem, when it warns of a
downstream server problem.
PR:
Obtained from:
Submitted by:
Reviewed by:
Fixed incompliance of AB when used with a proxy as reported by
Taisuke Yamada <tai@iij.ad.jp>. Added error trapping to the
write function (important on systems with defered errors
after a connect). Documented the weakness of the write()
function - it assumes the whole RQ can always be fully
written to the socket. Fixed a compiler warning which
pointed out that 'bad' was both a label and a variable.
Joshua Slive [Thu, 28 Mar 2002 18:56:08 +0000 (18:56 +0000)]
Update the Redhat layout to match v7. I decided to just replace the old layout
on the principal that tracking every version of every os in config.layout would
be a nightmare.
Graham Leggett [Mon, 25 Mar 2002 22:45:05 +0000 (22:45 +0000)]
Tighten up the overridden-Server-header bugfix in the proxy, by
only overriding if the request is a proxy request. It has been
pointed out that the previous fix allows CGIs and modules to
override the Server header, which is change to previous behavior.
PR:
Obtained from:
Submitted by: Graham Leggett, Joshua Slive
Reviewed by:
projects / thirdparty / apache / httpd.git / log
