[MAJOR] add a connection error state to the stream_interface
Tracking connection status changes was hard, and some code was
redundant. A new SI_ST_CER state was added to the stream interface
to indicate a past connection error, and an SI_FL_ERR flag was
added to report past I/O error. The stream_sock code does not set
the connection to SI_ST_CLO anymore in case of I/O error, it's
the upper layer which does it. This makes it possible to know
exactly when the file descriptors are allocated.
The new SI_ST_CER state permitted to split tcp_connection_status()
in two parts, one processing SI_ST_CON and the other one SI_ST_CER.
Synchronous connection errors now make use of this last state, hence
eliminating duplicate code.
Some ib<->ob copy paste errors were found and fixed, and all entities
setting SI_ST_CLO also shut the buffers down.
Some of these stream_interface specific functions and structures
have migrated to a new stream_interface.c file.
Some types of errors are still not detected by the buffers. For
instance, let's assume the following scenario in one single pass
of process_session: a connection sits in SI_ST_TAR state during
a retry. At TAR expiration, a new connection attempt is made, the
connection is obtained and srv->cur_sess is increased. Then the
buffer timeout is fires and everything is cleared, the new state
becomes SI_ST_CLO. The cleaning code checks that previous state
was either SI_ST_CON or SI_ST_EST to release the connection. But
that's wrong because last state is still SI_ST_TAR. So the
server's connection count does not get decreased.
This means that prev_state must not be used, and must be replaced
by some transition detection instead of level detection.
The following debugging line was useful to track state changes :
fprintf(stderr, "%s:%d: cs=%d ss=%d(%d) rqf=0x%08x rpf=0x%08x\n", __FUNCTION__, __LINE__,
s->si[0].state, s->si[1].state, s->si[1].err_type, s->req->flags, s-> rep->flags);
2008-11-03 05:26:53 +00:00
|
|
|
/*
|
2009-10-18 21:56:35 +00:00
|
|
|
* include/proto/stream_interface.h
|
|
|
|
* This file contains stream_interface function prototypes
|
|
|
|
*
|
2011-02-13 12:16:36 +00:00
|
|
|
* Copyright (C) 2000-2011 Willy Tarreau - w@1wt.eu
|
2009-10-18 21:56:35 +00:00
|
|
|
*
|
|
|
|
* This library is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU Lesser General Public
|
|
|
|
* License as published by the Free Software Foundation, version 2.1
|
|
|
|
* exclusively.
|
|
|
|
*
|
|
|
|
* This library is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
|
|
* Lesser General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU Lesser General Public
|
|
|
|
* License along with this library; if not, write to the Free Software
|
|
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
|
|
|
|
*/
|
[MAJOR] add a connection error state to the stream_interface
Tracking connection status changes was hard, and some code was
redundant. A new SI_ST_CER state was added to the stream interface
to indicate a past connection error, and an SI_FL_ERR flag was
added to report past I/O error. The stream_sock code does not set
the connection to SI_ST_CLO anymore in case of I/O error, it's
the upper layer which does it. This makes it possible to know
exactly when the file descriptors are allocated.
The new SI_ST_CER state permitted to split tcp_connection_status()
in two parts, one processing SI_ST_CON and the other one SI_ST_CER.
Synchronous connection errors now make use of this last state, hence
eliminating duplicate code.
Some ib<->ob copy paste errors were found and fixed, and all entities
setting SI_ST_CLO also shut the buffers down.
Some of these stream_interface specific functions and structures
have migrated to a new stream_interface.c file.
Some types of errors are still not detected by the buffers. For
instance, let's assume the following scenario in one single pass
of process_session: a connection sits in SI_ST_TAR state during
a retry. At TAR expiration, a new connection attempt is made, the
connection is obtained and srv->cur_sess is increased. Then the
buffer timeout is fires and everything is cleared, the new state
becomes SI_ST_CLO. The cleaning code checks that previous state
was either SI_ST_CON or SI_ST_EST to release the connection. But
that's wrong because last state is still SI_ST_TAR. So the
server's connection count does not get decreased.
This means that prev_state must not be used, and must be replaced
by some transition detection instead of level detection.
The following debugging line was useful to track state changes :
fprintf(stderr, "%s:%d: cs=%d ss=%d(%d) rqf=0x%08x rpf=0x%08x\n", __FUNCTION__, __LINE__,
s->si[0].state, s->si[1].state, s->si[1].err_type, s->req->flags, s-> rep->flags);
2008-11-03 05:26:53 +00:00
|
|
|
|
|
|
|
#ifndef _PROTO_STREAM_INTERFACE_H
|
|
|
|
#define _PROTO_STREAM_INTERFACE_H
|
|
|
|
|
|
|
|
#include <stdlib.h>
|
|
|
|
|
|
|
|
#include <common/config.h>
|
|
|
|
#include <types/stream_interface.h>
|
|
|
|
|
|
|
|
|
|
|
|
/* main event functions used to move data between sockets and buffers */
|
2009-09-20 18:14:49 +00:00
|
|
|
int stream_int_check_timeouts(struct stream_interface *si);
|
[MAJOR] add a connection error state to the stream_interface
Tracking connection status changes was hard, and some code was
redundant. A new SI_ST_CER state was added to the stream interface
to indicate a past connection error, and an SI_FL_ERR flag was
added to report past I/O error. The stream_sock code does not set
the connection to SI_ST_CLO anymore in case of I/O error, it's
the upper layer which does it. This makes it possible to know
exactly when the file descriptors are allocated.
The new SI_ST_CER state permitted to split tcp_connection_status()
in two parts, one processing SI_ST_CON and the other one SI_ST_CER.
Synchronous connection errors now make use of this last state, hence
eliminating duplicate code.
Some ib<->ob copy paste errors were found and fixed, and all entities
setting SI_ST_CLO also shut the buffers down.
Some of these stream_interface specific functions and structures
have migrated to a new stream_interface.c file.
Some types of errors are still not detected by the buffers. For
instance, let's assume the following scenario in one single pass
of process_session: a connection sits in SI_ST_TAR state during
a retry. At TAR expiration, a new connection attempt is made, the
connection is obtained and srv->cur_sess is increased. Then the
buffer timeout is fires and everything is cleared, the new state
becomes SI_ST_CLO. The cleaning code checks that previous state
was either SI_ST_CON or SI_ST_EST to release the connection. But
that's wrong because last state is still SI_ST_TAR. So the
server's connection count does not get decreased.
This means that prev_state must not be used, and must be replaced
by some transition detection instead of level detection.
The following debugging line was useful to track state changes :
fprintf(stderr, "%s:%d: cs=%d ss=%d(%d) rqf=0x%08x rpf=0x%08x\n", __FUNCTION__, __LINE__,
s->si[0].state, s->si[1].state, s->si[1].err_type, s->req->flags, s-> rep->flags);
2008-11-03 05:26:53 +00:00
|
|
|
void stream_int_report_error(struct stream_interface *si);
|
2008-11-30 18:48:07 +00:00
|
|
|
void stream_int_retnclose(struct stream_interface *si, const struct chunk *msg);
|
[MAJOR] add a connection error state to the stream_interface
Tracking connection status changes was hard, and some code was
redundant. A new SI_ST_CER state was added to the stream interface
to indicate a past connection error, and an SI_FL_ERR flag was
added to report past I/O error. The stream_sock code does not set
the connection to SI_ST_CLO anymore in case of I/O error, it's
the upper layer which does it. This makes it possible to know
exactly when the file descriptors are allocated.
The new SI_ST_CER state permitted to split tcp_connection_status()
in two parts, one processing SI_ST_CON and the other one SI_ST_CER.
Synchronous connection errors now make use of this last state, hence
eliminating duplicate code.
Some ib<->ob copy paste errors were found and fixed, and all entities
setting SI_ST_CLO also shut the buffers down.
Some of these stream_interface specific functions and structures
have migrated to a new stream_interface.c file.
Some types of errors are still not detected by the buffers. For
instance, let's assume the following scenario in one single pass
of process_session: a connection sits in SI_ST_TAR state during
a retry. At TAR expiration, a new connection attempt is made, the
connection is obtained and srv->cur_sess is increased. Then the
buffer timeout is fires and everything is cleared, the new state
becomes SI_ST_CLO. The cleaning code checks that previous state
was either SI_ST_CON or SI_ST_EST to release the connection. But
that's wrong because last state is still SI_ST_TAR. So the
server's connection count does not get decreased.
This means that prev_state must not be used, and must be replaced
by some transition detection instead of level detection.
The following debugging line was useful to track state changes :
fprintf(stderr, "%s:%d: cs=%d ss=%d(%d) rqf=0x%08x rpf=0x%08x\n", __FUNCTION__, __LINE__,
s->si[0].state, s->si[1].state, s->si[1].err_type, s->req->flags, s-> rep->flags);
2008-11-03 05:26:53 +00:00
|
|
|
|
2009-09-05 18:57:35 +00:00
|
|
|
/* functions used when running a stream interface as a task */
|
|
|
|
void stream_int_update(struct stream_interface *si);
|
|
|
|
void stream_int_update_embedded(struct stream_interface *si);
|
|
|
|
void stream_int_shutr(struct stream_interface *si);
|
|
|
|
void stream_int_shutw(struct stream_interface *si);
|
|
|
|
void stream_int_chk_rcv(struct stream_interface *si);
|
|
|
|
void stream_int_chk_snd(struct stream_interface *si);
|
|
|
|
|
|
|
|
struct task *stream_int_register_handler(struct stream_interface *si,
|
2011-02-13 12:16:36 +00:00
|
|
|
struct si_applet *app);
|
2009-09-05 18:57:35 +00:00
|
|
|
struct task *stream_int_register_handler_task(struct stream_interface *si,
|
|
|
|
struct task *(*fct)(struct task *));
|
|
|
|
void stream_int_unregister_handler(struct stream_interface *si);
|
|
|
|
|
2011-03-10 13:03:36 +00:00
|
|
|
static inline void clear_target(struct target *dest)
|
|
|
|
{
|
|
|
|
dest->type = TARG_TYPE_NONE;
|
|
|
|
dest->ptr.v = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void set_target_server(struct target *dest, struct server *s)
|
|
|
|
{
|
|
|
|
dest->type = TARG_TYPE_SERVER;
|
|
|
|
dest->ptr.s = s;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void set_target_proxy(struct target *dest, struct proxy *p)
|
|
|
|
{
|
|
|
|
dest->type = TARG_TYPE_PROXY;
|
|
|
|
dest->ptr.p = p;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void set_target_applet(struct target *dest, struct si_applet *a)
|
|
|
|
{
|
|
|
|
dest->type = TARG_TYPE_APPLET;
|
|
|
|
dest->ptr.a = a;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void set_target_task(struct target *dest, struct task *t)
|
|
|
|
{
|
|
|
|
dest->type = TARG_TYPE_TASK;
|
|
|
|
dest->ptr.t = t;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline struct target *copy_target(struct target *dest, struct target *src)
|
|
|
|
{
|
|
|
|
*dest = *src;
|
|
|
|
return dest;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int target_match(struct target *a, struct target *b)
|
|
|
|
{
|
|
|
|
return a->type == b->type && a->ptr.v == b->ptr.v;
|
|
|
|
}
|
|
|
|
|
2011-03-10 15:55:02 +00:00
|
|
|
static inline struct server *target_srv(struct target *t)
|
|
|
|
{
|
|
|
|
if (!t || t->type != TARG_TYPE_SERVER)
|
|
|
|
return NULL;
|
|
|
|
return t->ptr.s;
|
|
|
|
}
|
|
|
|
|
[MAJOR] add a connection error state to the stream_interface
Tracking connection status changes was hard, and some code was
redundant. A new SI_ST_CER state was added to the stream interface
to indicate a past connection error, and an SI_FL_ERR flag was
added to report past I/O error. The stream_sock code does not set
the connection to SI_ST_CLO anymore in case of I/O error, it's
the upper layer which does it. This makes it possible to know
exactly when the file descriptors are allocated.
The new SI_ST_CER state permitted to split tcp_connection_status()
in two parts, one processing SI_ST_CON and the other one SI_ST_CER.
Synchronous connection errors now make use of this last state, hence
eliminating duplicate code.
Some ib<->ob copy paste errors were found and fixed, and all entities
setting SI_ST_CLO also shut the buffers down.
Some of these stream_interface specific functions and structures
have migrated to a new stream_interface.c file.
Some types of errors are still not detected by the buffers. For
instance, let's assume the following scenario in one single pass
of process_session: a connection sits in SI_ST_TAR state during
a retry. At TAR expiration, a new connection attempt is made, the
connection is obtained and srv->cur_sess is increased. Then the
buffer timeout is fires and everything is cleared, the new state
becomes SI_ST_CLO. The cleaning code checks that previous state
was either SI_ST_CON or SI_ST_EST to release the connection. But
that's wrong because last state is still SI_ST_TAR. So the
server's connection count does not get decreased.
This means that prev_state must not be used, and must be replaced
by some transition detection instead of level detection.
The following debugging line was useful to track state changes :
fprintf(stderr, "%s:%d: cs=%d ss=%d(%d) rqf=0x%08x rpf=0x%08x\n", __FUNCTION__, __LINE__,
s->si[0].state, s->si[1].state, s->si[1].err_type, s->req->flags, s-> rep->flags);
2008-11-03 05:26:53 +00:00
|
|
|
#endif /* _PROTO_STREAM_INTERFACE_H */
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Local variables:
|
|
|
|
* c-indent-level: 8
|
|
|
|
* c-basic-offset: 8
|
|
|
|
* End:
|
|
|
|
*/
|