Jeff Trawick [Wed, 22 Aug 2001 18:58:36 +0000 (18:58 +0000)]
simplify some code in output_results() just a bit to avoid a
bug in gcc 2.7.2.3 on FreeBSD 3.4; didn't hit the problem with
-O2; did hit the problem with no -O
Change the new-httpd@apache.org references to dev@httpd.apache.org.
There are some references just to "new-httpd" throughout - I left
these intact as I'm not sure what the nickname for the new list is
(dev@httpd?).
Doug MacEachern [Wed, 22 Aug 2001 16:59:26 +0000 (16:59 +0000)]
rather than creating small 1024 byte buckets of output data,
create a transient bucket pointing directly to the BIO mem buff.
this makes for a dramatic increase in performance. previously,
downloading large files (2Mb-5Mb-ish) made my laptop start to
smoke from the fan spinning so fast to cool the cpu.
also, apache stylize churn_output()
PR:
Obtained from:
Submitted by:
Reviewed by:
Ensure that libtool does the installation for us. This way we will actually
copy over the binaries rather than libtool wrapper scripts which would
copy the binaries to .libs/lt-* at first execution.
Reviewed by: Greg Stein
Doug MacEachern [Wed, 22 Aug 2001 15:30:37 +0000 (15:30 +0000)]
destroy the brigade when we are done with it, rather than remove
one bucket at a time. prevents a problem when downloading large files.
also change ssl_io_filter_Output to apache style
and change some variable names that should make the code easier to
read/understand, e.g. pbbIn -> bb, pbktIn -> bucket
PR:
Obtained from:
Submitted by:
Reviewed by:
We don't have rules.mk.in, so autoconf/m4/sed will overwrite the good
rules.mk with a blank file.
This gets us building again. From what I see of rbb's libtool 1.4 commit,
I didn't see any change to rules.mk that would require variable
substitution. Ryan?
Ryan Bloom [Tue, 21 Aug 2001 23:41:35 +0000 (23:41 +0000)]
Allow Apache to use libtool 1.4. In order to do this, we check the
libtool version of the machine, and we add the correct flags to the build.
We also had to remove the --disable-shared in order to make this work
correctly.
Doug MacEachern [Tue, 21 Aug 2001 05:57:13 +0000 (05:57 +0000)]
authentication/authorization hooks were backwards
make authentication hook run APR_HOOK_FIRST for FakeBasicAuth
PR:
Obtained from:
Submitted by:
Reviewed by:
Joshua Slive [Tue, 21 Aug 2001 01:31:23 +0000 (01:31 +0000)]
I hope I don't get anybody upset here, but I'm going to go super-simple
on this for now. I removed all the fancy stuff. If someone wants
to put it back in after we've thought a little about things like
the location of the graphic file, I will have no objections.
First, let's start with something simple that works.
Fix build breakage on Linux glibc 2.1+ systems that prevented db1 from being
correctly found for mod_auth_dbm. This is similar to the mod_auth_db check.
Ryan Bloom [Mon, 20 Aug 2001 22:30:17 +0000 (22:30 +0000)]
Add the openssl/include/openssl directory to the INCLUDES variable.
This allows us to remove the openssl from the #include lines in the
mod_ssl files. This makes it easier to use a different SSL library,
with fewer changes to the mod_ssl files.
Create the abs.dsp win32 project to support building ab with /USE_SSL,
and modify Makefile.win to install the ssl-enabled abs.exe binary
as 'bin/ab.exe' if srclib/openssl exists.
We might want to disable this, by default, when we get ready for GA.
ITMT, this provides a fantastic stress-test for interoperability of
the ap_die handling, the includes filter, the negotiated typemap
behavior and the new typemap-body functions.
Keep adding to the list of errors as they are converted from the
httpd-2.0/docs/docroot/error individual files into docs/error.
Paul J. Reder [Mon, 20 Aug 2001 19:47:52 +0000 (19:47 +0000)]
The prefork and OS/2 MPMs are overwriting the pid file when a second copy
of httpd is started and shuts down due to socket conflict. Moving the
call to ap_log_pid solves the problem.
*) Changed the late-1.3 log_config substitution %c to %X (logs the
status of the closed connection) as it conflicts with the far more
common, historical ssl logging directive %...{var}c.
My bad. We need to use the constants defined in httpd.h. The essential
reason is that these errors will grow to include 'other' things, like
custom error responses, DAV error responses, and so on. They may also
start including 'sub errors' - local langauge explanations of specific
failures. Let's use the httpd.h constants rather than numbers, they
are easier to read.
This is the Body:tag\n<Content>*tag\n patch for type map files. If the
content can be served, it sends a file bucket of the type map file
itself, forward spaced to the <Content>, set to the length of <Content>.
<Content> may contain any binary data. The end tag must exist, anything
between the tag and \n is ignored. This version (v.s. the patch sent to
the list) includes error reporting if the end tag cannot be located.
We need to change the etag code to accept an additional argument, so we
can set the cache controls properly.
Brian Havard [Mon, 20 Aug 2001 16:10:57 +0000 (16:10 +0000)]
Catch exceptions in worker threads so a segfault doesn't kill all active
connections in the process. When this occurs, a graceful shutdown of the
child is initiated so any leaked resources & hosed memory is disposed of ASAP.
- Switch to AC_SEARCH_LIBS to check multiple libraries for mod_auth_db
(the AC_* way of doing what Martin committed)
- Fix the configuration check for mod_auth_digest so that:
- Everything is on separate lines so that the preprocessor doesn't scream
- It adds the path to APR. (APR_SOURCE_DIR looks right, but I'm not sure.)
By the time the modules are configured, the CPPFLAGS and such are not
setup to point at APR. This should probably be rectified and then this
can be taken out - so you could assume that apr.h is in your CPPFLAGS
somewhere when configuring the modules.
Martin Kraemer [Mon, 20 Aug 2001 07:00:23 +0000 (07:00 +0000)]
On modern systems (e.g., FreeBSD), the db* functions reside in libc
rather than in libdb. Also, be more precise in telling what is actually
checked for (we are NOT testing for main(), but we are testing for dbopen(),
right?).
Without the first fix, configuration would fail completely for FreeBSD
Doug MacEachern [Mon, 20 Aug 2001 02:43:02 +0000 (02:43 +0000)]
change filter name to CaseFilterIn (mod_case_filter.c already uses CaseFilter)
allow filter to be configured without 'CaseFilterIn On'
PR:
Obtained from:
Submitted by:
Reviewed by:
This Variant type-map file is the authoritative reference to our
"It Worked" page variants. Of course it isn't yet (although it is
complete), please review httpd-std.conf as well as this list
for inconsistencies (I documented those I discovered).
This Variant type-map file is the authoritative reference to our
"It Worked" page variants. Of course it isn't yet (although it is
complete), please review httpd-std.conf as well as this list
for inconsistencies (I documented those I discovered).
Per discussion on new-httpd, change the logic to do the following:
--enable-modules=most will disable a module if it can not be built
--enable-modules=all will error if a module can not be built
--enable-modules=foo will error if the specified module can not be built
(even if --enable-modules=most is set)
This is Apache 2.0, which has been essentially rewritten.
This citation was inconsistent with the rest of our credits.
We still thank Ambarish Malpani of ValiCert, Inc. for his
contribution, but we have many, many others to thank, and this
one accolade was out of place.
projects / apache / log