lighttpd
/
lighttpd1.4
2
0
Fork 0
Code Issues Releases Wiki Activity

remove zombies again

Browse Source 
git-svn-id: svn://svn.lighttpd.net/lighttpd/branches/lighttpd-merge-1.4.x@902 152afb58-edef-0310-8abb-c4023f1b3aa9
svn/tags/lighttpd-1.4.11
Jan Kneschke 18 years ago
parent 8f871d4f04
commit b789ce87a1
1 changed files with 34 additions and 35 deletions
Show Stats Download Patch File Download Diff File

69
src/mod_fastcgi.c
Unescape Escape View File

@ -1579,6 +1579,13 @@ static connection_result_t fcgi_establish_connection(server *srv, handler_ctx *h
return CONNECTION_DELAYED;
} else if (errno == EAGAIN) {
if (hctx->conf.debug) {
log_error_write(srv, __FILE__, __LINE__, "sd",
"This means that the you have more incoming requests than your fastcgi-backend can handle in parallel. "
"Perhaps it helps to spawn more fastcgi backend or php-children, if not decrease server.max-connections."
"The load for this fastcgi backend is:", proc->load);
}
return CONNECTION_OVERLOADED;
} else {
log_error_write(srv, __FILE__, __LINE__, "ssdb",
@ -1586,15 +1593,6 @@ static connection_result_t fcgi_establish_connection(server *srv, handler_ctx *h
strerror(errno),
proc->port, proc->socket);
#if 0
if (errno == EAGAIN) {
log_error_write(srv, __FILE__, __LINE__, "sd",
"This means that the you have more incoming requests than your fastcgi-backend can handle in parallel. "
"Perhaps it helps to spawn more fastcgi backend or php-children, if not decrease server.max-connections."
"The load for this fastcgi backend is:", proc->load);
}
#endif
return CONNECTION_DEAD;
}
}
@ -2659,6 +2657,7 @@ static handler_t fcgi_write_request(server *srv, handler_ctx *hctx) {
return HANDLER_ERROR;
}
/* we can't handle this in the switch as we have to fall through in it */
if (hctx->state == FCGI_STATE_CONNECT_DELAYED) {
int socket_error;
socklen_t socket_error_len = sizeof(socket_error);
@ -2679,12 +2678,9 @@ static handler_t fcgi_write_request(server *srv, handler_ctx *hctx) {
"port:", hctx->proc->port);
}
hctx->proc->state = PROC_STATE_DISABLED;
hctx->proc->disabled_until = srv->cur_ts + 10;
host->active_procs--;
status_counter_inc(srv, CONST_STR_LEN("fastcgi.backend.<hostid>.died"));
status_counter_set(srv, CONST_STR_LEN("fastcgi.backend.<hostid>.disable"), 1);
return HANDLER_ERROR;
}
@ -2762,11 +2758,9 @@ static handler_t fcgi_write_request(server *srv, handler_ctx *hctx) {
host->unixsocket);
hctx->proc->state = PROC_STATE_DISABLED;
hctx->proc->disabled_until = srv->cur_ts + 2;
host->active_procs--;
status_counter_inc(srv, CONST_STR_LEN("fastcgi.backend.<hostid>.overloaded"));
status_counter_set(srv, CONST_STR_LEN("fastcgi.backend.<hostid>.disable"), 1);
return HANDLER_ERROR;
case CONNECTION_DEAD:
@ -2777,12 +2771,9 @@ static handler_t fcgi_write_request(server *srv, handler_ctx *hctx) {
* for check if the host is back in 10 seconds
* */
hctx->proc->state = PROC_STATE_DISABLED;
hctx->proc->disabled_until = srv->cur_ts + 10;
host->active_procs--;
status_counter_inc(srv, CONST_STR_LEN("fastcgi.backend.<hostid>.died"));
status_counter_set(srv, CONST_STR_LEN("fastcgi.backend.<hostid>.disable"), 1);
return HANDLER_ERROR;
case CONNECTION_OK:
@ -2979,30 +2970,38 @@ SUBREQUEST_FUNC(mod_fastcgi_handle_subrequest) {
/* connect() or getsockopt() failed,
* restart the request-handling
*/
if (proc && proc->is_local) {
if (proc) {
if (proc->is_local) {
if (p->conf.debug) {
log_error_write(srv, __FILE__, __LINE__, "sbdb", "connect() to fastcgi failed, restarting the request-handling:",
if (p->conf.debug) {
log_error_write(srv, __FILE__, __LINE__, "sbdb",
"connect() to fastcgi failed, restarting the request-handling:",
host->host,
proc->port,
proc->socket);
}
}
/*
* several hctx might reference the same proc
*
* Only one of them should mark the proc as dead all the other
* ones should just take a new one.
*
* If a new proc was started with the old struct this might lead
* the mark a perfect proc as dead otherwise
*
*/
if (proc->state == PROC_STATE_RUNNING &&
hctx->pid == proc->pid) {
proc->state = PROC_STATE_DIED_WAIT_FOR_PID;
/*
* several hctx might reference the same proc
*
* Only one of them should mark the proc as dead all the other
* ones should just take a new one.
*
* If a new proc was started with the old struct this might lead
* the mark a perfect proc as dead otherwise
*
*/
if (proc->state == PROC_STATE_RUNNING &&
hctx->pid == proc->pid) {
proc->state = PROC_STATE_DIED_WAIT_FOR_PID;
}
} else {
proc->state = PROC_STATE_DISABLED;
}
host->active_procs--;
status_counter_set(srv, CONST_STR_LEN("fastcgi.backend.<hostid>.disable"), 1);
}
fcgi_restart_dead_procs(srv, p, host);
/* cleanup this request and let the request handler start this request again */

Write Preview
Loading…
Cancel
Save