Commits · faa904ab81ae41a154964f44c962379e0ceeedec · CYBER - Cyber Security / TS 103 523 MSP / TLMSP / TLMSP Apache Httpd

Jan 06, 2002

Bypass a strdup and an 8KB local variable in the common case where · faa904ab

Brian Pane authored Jan 06, 2002

the logger is using the default time format


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92752 13f79535-47bb-0310-9956-ffa450edef68

faa904ab

Cleaned up a couple of ap_ssi_parse_string calls that I missed · dc3c7404

Brian Pane authored Jan 06, 2002

in the last commit


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92748 13f79535-47bb-0310-9956-ffa450edef68

dc3c7404

Eliminated various large (8KB) string buffers from the stack in · 83b46c64

Brian Pane authored Jan 06, 2002

mod_include by adding support for a power-of-two pool-based allocator
in ap_ssi_parse_string(). (The default operation of this function
is backward-compatible, to support the other modules that call it.)
This change should help reduce memory usage for servers delivering
shtml pages.

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92747 13f79535-47bb-0310-9956-ffa450edef68

83b46c64

This issue has turned into a FAQ. · 2a5227d5

Justin Erenkrantz authored Jan 06, 2002

Start tracking it in STATUS so we can point people to this if they ask
about this again.

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92744 13f79535-47bb-0310-9956-ffa450edef68

2a5227d5

Fix a typo in mod_deflate's m4 config section. · 16057a62

Justin Erenkrantz authored Jan 06, 2002

Submitted by: albert chin (china@thewrittenword.com)
Reviewed by: Justin Erenkrantz

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92743 13f79535-47bb-0310-9956-ffa450edef68

16057a62

Note this for the record. I have no idea if/when I'll have the time to · 4be34aab
Justin Erenkrantz authored Jan 06, 2002
```
implement this.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92742 13f79535-47bb-0310-9956-ffa450edef68
```
4be34aab

Jan 04, 2002

Reworked the accept() loop to avoid an extra call to select() just to · 8681fa20

Bradley Nicholes authored Jan 04, 2002

determine if there is work to do.  Made the calls to accept() non-blocking.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92737 13f79535-47bb-0310-9956-ffa450edef68

8681fa20

Changed to 4 byte packing · f248afad

Bradley Nicholes authored Jan 04, 2002


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92736 13f79535-47bb-0310-9956-ffa450edef68

f248afad

for proxy CONNECT handling: · cc9618fe

Jeff Trawick authored Jan 04, 2002

  don't hard-code the address family of the target; respect
  what the resolver told us

  get a new socket each time we try to connect to one of the
  target addresses as reported by the resolver; you can't
  portably attempt to connect more than once on the same
  socket

this is the same fix committed to proxy http support yesterday


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92735 13f79535-47bb-0310-9956-ffa450edef68

cc9618fe

cache now can cache things for minutes. · 62ffd967

Ian Holsman authored Jan 04, 2002

add remove_entity function to type on mem_cache (so we can remove items from the cache)
PR:
Obtained from:
Submitted by:
Reviewed by:


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92734 13f79535-47bb-0310-9956-ffa450edef68

62ffd967

2.0.29 isn't going to be rolled. · 46fe66f3

Justin Erenkrantz authored Jan 04, 2002


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92732 13f79535-47bb-0310-9956-ffa450edef68

46fe66f3

split mem_cache out of mod_cache (ala mod_proxy & proxy_http) · 4daa4fe7

Ian Holsman authored Jan 04, 2002

PR:
Obtained from:
Submitted by:
Reviewed by:


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92731 13f79535-47bb-0310-9956-ffa450edef68

4daa4fe7

Fix a couple of mod_proxy problems forwarding HTTP connections: · 98b414fd

Jeff Trawick authored Jan 04, 2002

(1) PR #9190  It failed to connect to IPv6 hosts.
(2) It failed to connect when the first IP address returned by
    the resolver was unreachable but a secondary IP address was.

PR:    9190 (partly)


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92727 13f79535-47bb-0310-9956-ffa450edef68

98b414fd

Jan 03, 2002

Keep my comments to myself. · 491b5c63

Bill Stoddard authored Jan 03, 2002


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92725 13f79535-47bb-0310-9956-ffa450edef68

491b5c63

get the declarations for strncasecmp() on AIX · 8713a0c5

Jeff Trawick authored Jan 03, 2002


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92724 13f79535-47bb-0310-9956-ffa450edef68

8713a0c5

Use ap_getline() in favor of ap_proxy_string_read() in ap_proxy_http_process_response(). · d588ed65
Bill Stoddard authored Jan 03, 2002
```
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92723 13f79535-47bb-0310-9956-ffa450edef68
```
d588ed65

Fix seg fault in HTTP_IN when trying to handle the LimitRequestBody · e3a9f55a

Bill Stoddard authored Jan 03, 2002

directive when reading a proxy response. We should bypass LimitRequestBody
activities when handling a response from the proxied server.

I don't think this will circumvent requestbody checking on a large request
coming into the server that is destined to be proxied. (not tested though).

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92722 13f79535-47bb-0310-9956-ffa450edef68

e3a9f55a

Fix the module identifier as shown in the docs for some · 80fec2ab

Jeff Trawick authored Jan 03, 2002

modules.

PR:           #9338
Submitted by: James Watson <ap2bug@sowega.org>


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92716 13f79535-47bb-0310-9956-ffa450edef68

80fec2ab

You know, I got an A in my "Principles of Operating Systems" class · 252749ab

Justin Erenkrantz authored Jan 03, 2002

last quarter.  Ain't worth a cent now.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92715 13f79535-47bb-0310-9956-ffa450edef68

252749ab

OS/2: Allow but ignore unix config directives so the default config files · 2352a367

Brian Havard authored Jan 03, 2002

don't have to be munged further.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92713 13f79535-47bb-0310-9956-ffa450edef68

2352a367

Jan 02, 2002

MMN = Things that go bump in the night. · 11e4e7e6

Justin Erenkrantz authored Jan 02, 2002

(limit_req_body is now an apr_off_t)


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92712 13f79535-47bb-0310-9956-ffa450edef68

11e4e7e6

Making sure that the global variables have been initialized to avoid linker · d9a9dcd1

Bradley Nicholes authored Jan 02, 2002

problems at least on NetWare
Submitted by: Pavel Novy


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92711 13f79535-47bb-0310-9956-ffa450edef68

d9a9dcd1

· 1641c07a

William A. Rowe Jr authored Jan 02, 2002

Merge a singular path for dir_walk to parse, if we have an r->filename
which is an APR_DIR, and path_info contents. Also, al la Mr. Pane,
optimize our canonical_filename by simply noting the length of the
identity match, and refresh canonical_filename when we are finished.

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92707 13f79535-47bb-0310-9956-ffa450edef68

1641c07a

· b03dcb01

William A. Rowe Jr authored Jan 02, 2002

Eliminate a duplicate absolute path test, and NEVER serve a request
for an APR_DIR file from the cache when we have path_info, it is a
contradition (APR_DIR always forces dir_walk to gather the next segment
from path_info, even if it is APR_NOFILE, until we have no path_info.)

So we can't use a predetermined filename/path_info combo, ever, if the
filename resolves to an APR_DIR.

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92706 13f79535-47bb-0310-9956-ffa450edef68

b03dcb01

Threads on FreeBSD are not my friend. · d81878d4

Aaron Bannert authored Jan 02, 2002


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92705 13f79535-47bb-0310-9956-ffa450edef68

d81878d4

Change the return code from APR_TIMEUP to APR_ECONNABORTED, which seems · 7759b835

Bill Stoddard authored Jan 02, 2002

to be a bit more descriptive. Move the check to inside the inner while()
loop and add an additional check for eos. If we get an EOS bucket, there
is no point in going further. Hopefully this will fix the last seg fault
in the function.

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92704 13f79535-47bb-0310-9956-ffa450edef68

7759b835

showstoppers--; · aa77b776

Justin Erenkrantz authored Jan 02, 2002

This appears to be due to an API change in LWP 5.61. A fix has been
committed to httpd-test. (Works on Linux with prefork and worker now.)

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92703 13f79535-47bb-0310-9956-ffa450edef68

aa77b776

Add note about worker MPM weirdness observed on FreeBSD and Cygwin. · fb5a2373

Justin Erenkrantz authored Jan 02, 2002

Does anyone else see this?


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92702 13f79535-47bb-0310-9956-ffa450edef68

fb5a2373

showstoppers--; · cb90037b

Justin Erenkrantz authored Jan 02, 2002


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92701 13f79535-47bb-0310-9956-ffa450edef68

cb90037b

Fix LimitRequestBody directive by moving the relevant code from · 53c5df33

Justin Erenkrantz authored Jan 02, 2002

ap_*_client_block to ap_http_filter (aka HTTP_IN). This is the
only appropriate place for limit checking to occur (otherwise,
chunked input is not correctly limited).

Also changed the type of limit_req_body to apr_off_t to match the
other types inside of HTTP_IN. Also made the strtol call for
limit_req_body a bit more robust.

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92700 13f79535-47bb-0310-9956-ffa450edef68

53c5df33

Remove a needless immortal bucket creation. We can just leave the · 91aac8f0

Justin Erenkrantz authored Jan 02, 2002

brigade as-is.

Suggested by: Cliff


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92698 13f79535-47bb-0310-9956-ffa450edef68

91aac8f0

Jan 01, 2002

Generalized the recent prep_walk_cache optimizations to allow other · dc013bfa

Brian Pane authored Jan 01, 2002

modules to register "notes" within the array of working data in
the core_request_config

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92697 13f79535-47bb-0310-9956-ffa450edef68

dc013bfa

· 47f57ee6

William A. Rowe Jr authored Jan 01, 2002

Need a null language_priority array for an either/or choice in the
conf merge. This does so. The merges should be significantly quicker.

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92696 13f79535-47bb-0310-9956-ffa450edef68

47f57ee6

· 2160c665

William A. Rowe Jr authored Jan 01, 2002

Thanks Mr. Trawick, backtraces can be very enlightening.

This fixes two bugs, the segfault when researching index numbers,
and another. We don't hint anywhere that LanguagePriority is a
cumulative setting, in fact the syntax doesn't suggest so either.
Merges are always slow, there is no reason to make this even slower.

So the patch has the LanguagePriority directive override any existing
LanguagePriority list from an earlier container.

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92695 13f79535-47bb-0310-9956-ffa450edef68

2160c665

Dec 31, 2001

· 91c025db

William A. Rowe Jr authored Dec 31, 2001

  Time to commit... before I get lost in the changes


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92694 13f79535-47bb-0310-9956-ffa450edef68

91c025db

Fix mod_proxy seg fault when the proxied server returns · 2dd7322e

Bill Stoddard authored Dec 31, 2001

an HTTP/0.9 response or a bogus status line.

Submitted by: Adam Sussman
Reviewed by:	Bill Stoddard


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92693 13f79535-47bb-0310-9956-ffa450edef68

2dd7322e

Prevent mod_proxy from truncating one character off the · 9efcfae4

Bill Stoddard authored Dec 31, 2001

end of the status line returned from the proxied server.
Note that buffer is NULL terminated by the call to
ap_proxy_string_read() so there is no need to terminate it
again.

Submitted by:	Adam Sussman
Reviewed by:	Bill Stoddard


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92692 13f79535-47bb-0310-9956-ffa450edef68

9efcfae4

Change the call to be blocking to eliminate a loop caused when the upstream · 8645ba98

Bill Stoddard authored Dec 31, 2001

host does not send any bytes back our way.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92691 13f79535-47bb-0310-9956-ffa450edef68

8645ba98

Eliminate loop in ap_proxy_string_read(). Need to be able to handle · fcc54fbe

Bill Stoddard authored Dec 31, 2001

getting an empty brigade from ap_get_brigade(). Also made sure that
we always call ap_get_brigade() in readline mode.

Submitted/Reviewed by: Adam Sussman & Bill Stoddard

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92690 13f79535-47bb-0310-9956-ffa450edef68

fcc54fbe

Performance fix for prep_walk_cache(): · 74ac1a96

Brian Pane authored Dec 31, 2001

Moved the directory/location/file-walk caches from the
request's pool userdata hash table to the core_request_config
struct.

This change removes about 60% of the processing time from
prep_walk_cache().


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@92684 13f79535-47bb-0310-9956-ffa450edef68

74ac1a96