@- Fixed signal() bug that could cause the Apache master process to

@  die. (Thies)

the problem was that we *did* register the signal in the apache
master process as well.
php_module_startup() -> REGISTER_INI_ENTRIES() -> OnChangeMaxExecutionTime() -> php_set_timeout().

from now on the .ini mechanism only sets PG(max_execution_time) and that gets
"converted" to a timeout-signal() in php_request_startup()!

we still need to think about the ZTS timeout mechanism.
This commit is contained in:
Thies C. Arntzen 2000-02-05 19:32:27 +00:00
parent 82dbc9c74f
commit 1616b5193c

View File

@ -1,5 +1,5 @@
/*
+----------------------------------------------------------------------+
+----------------------------------------------------------------------
| PHP version 4.0 |
+----------------------------------------------------------------------+
| Copyright (c) 1997, 1998, 1999, 2000 The PHP Group |
@ -120,22 +120,6 @@ static PHP_INI_MH(OnSetPrecision)
}
static PHP_INI_MH(OnChangeMaxExecutionTime)
{
int new_timeout;
PLS_FETCH();
if (new_value) {
new_timeout = atoi(new_value);
} else {
new_timeout = 0;
}
PG(max_execution_time) = new_timeout;
php_set_timeout(new_timeout);
return SUCCESS;
}
#if MEMORY_LIMIT
static PHP_INI_MH(OnChangeMemoryLimit)
{
@ -240,7 +224,7 @@ PHP_INI_BEGIN()
PHP_INI_ENTRY_EX("define_syslog_variables", "0", PHP_INI_ALL, NULL, php_ini_boolean_displayer_cb)
PHP_INI_ENTRY("max_execution_time", "30", PHP_INI_ALL, OnChangeMaxExecutionTime)
STD_PHP_INI_ENTRY("max_execution_time", "30", PHP_INI_ALL, OnUpdateInt, max_execution_time, php_core_globals, core_globals)
#if MEMORY_LIMIT
PHP_INI_ENTRY("memory_limit", "8388608", PHP_INI_ALL, OnChangeMemoryLimit)
#endif