for 80 character folding and legibility.
Folds several identical errors into the same APLOGNO() groups.
Catches several errors which were missing APLOGNO() entries, in the
existing groups and in one new message.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1611165 13f79535-47bb-0310-9956-ffa450edef68
service in the WinNT MPM used in all Windows installations.
Workaround: AcceptFilter <protocol> {none|connect}
Submitted by: trawick
Reviewed by: jorton, covener, jim
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1610652 13f79535-47bb-0310-9956-ffa450edef68
Use SYNCHRONIZE instead of PROCESS_ALL_ACCESS because
a. it is sufficient
b. it avoids an issue where PROCESS_ALL_ACCESS is larger on
newer SDKs, resulting in a run-time error when running on
older Windows
Close the handle.
Submitted by: Ivan Zhakov <ivan visualsvn.com>
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1527220 13f79535-47bb-0310-9956-ffa450edef68
Submitted by: Oracle, via trawick
The original modification was made some years ago for Oracle HTTP Server
by an Oracle employee. trawick made additional changes for style and
for trunk/2.4.x changes.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1526666 13f79535-47bb-0310-9956-ffa450edef68
context->accept_socket. This FD will be closed a second time, often
shortly after a worker picks it up in this same FD being reused. The
first recv fails with WSAENOTSOCK since the same FD was closed in the
listener thread while the worker was pulling it off the queue
(The second close is of the underlying FD/socket, not a shared
apr_socket_t, so it's not short-circuited)
This patch makes it a bit more 2.2.x-ish and solves my problem -- the
context->accept_socket gets zapped at the bottom of the loop if
!disconnected.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1523387 13f79535-47bb-0310-9956-ffa450edef68
directive on Windows => crash-o-matic
Or,
WinNT MPM: Don't crash during child process initialization if the
Listen protocol is unrecognized.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1520368 13f79535-47bb-0310-9956-ffa450edef68
r1385216 broke building of mpm event with serf support. To fix,
* introduce APACHE_MPMPATH_INIT/FINISH similar to
APACHE_MODPATH_INIT/FINISH to create modules.mk and add
MOD_* vars to it.
* change APACHE_MPM_MODULE to only append to modules.mk, not
overwrite
* add APACHE_MPMPATH_INIT/FINISH to all MPMs
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1395252 13f79535-47bb-0310-9956-ffa450edef68
workers were allocating from pchild without holding '
child_lock. Allocate from ptrans instead.
Related to: PR 52196
The reporter suggests that the apr_thread_t for r->connection->current_thread
should have a unique pool. This revision handles that for WinNT MPM,
but not for other MPMs.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1374874 13f79535-47bb-0310-9956-ffa450edef68
the function pointers dynamically.
There's no indication that this will resolve any operational
glitches, but it might make the accept path slightly faster.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1363312 13f79535-47bb-0310-9956-ffa450edef68
allocates a new one. This resolves a crash seen on WinXP, caused
by r1103595.
Bad revision found by: Gregg L Smith <gls gknw.net>
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1142264 13f79535-47bb-0310-9956-ffa450edef68
everywhere.
We know that the core module has module_index 0. Therefore we can save
some pointer operations in ap_get_module_config(cv, &core_module) and
ap_set_module_config(cv, &core_module, val). As these are called rather often,
this may actually have some (small) measurable effect.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1132781 13f79535-47bb-0310-9956-ffa450edef68
We override exit() throughout the httpd build in order to intercept and
report our status through the service control manager.
We must include process.h prior to overriding exit().
I seem to remember that this is the reason apr.hw once included
process.h unilaterally, to avoid this conflict over exit.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1129878 13f79535-47bb-0310-9956-ffa450edef68
trunk mod_status relies on this value for proper formatting
of the workers; now we see '-' for the idle WinNT MPM workers
instead of ' '
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1096814 13f79535-47bb-0310-9956-ffa450edef68
processes. Add end_generation hook for notification when the last
MPM child of a generation exits.
end_generation is implemented completely by core using the
child_status hook run by the MPM.
simple and mpmt_os2 MPMs don't currently run the child_status
hook, so neither hook is invoked with those MPMs.
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1096609 13f79535-47bb-0310-9956-ffa450edef68
mpm_winnt.c: Fail to compile if enabled, and add a hint about the
issue to address before it will work.
configure.in: Move the --enable-v4-mapped check to below the
MPM logic, and default to --disable-v4-mapped if
using the WinNT MPM
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1091467 13f79535-47bb-0310-9956-ffa450edef68
between parent and child
the extra info is helpful when using a generic listener on an IPv6-
enabled box when either IPV6_V6ONLY isn't defined for the APR build
or httpd is built with IPv4-mapped addresses enabled
in that case you'll see the parent writing one socket but the child
expecting too, before it bails out with
(OS 109)The pipe has been ended. : Child: Unable to read socket data
from parent
git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1091421 13f79535-47bb-0310-9956-ffa450edef68
