1
0
mirror of https://github.com/samba-team/samba.git synced 2024-12-25 23:21:54 +03:00
samba-mirror/source4/smbd/process_standard.c
Volker Lendecke e73ccc06ef Rely on /dev/urandom
This removes quite a bit of code. All reasonable systems have /dev/urandom
these days. Linux, Solaris and the BSDs do.  In case we find a system
without /dev/urandom, we will have to go hunting in other libraries.

The main reason for this is speed: On Ubuntu 14.04 doing direct reads from
/dev/urandom is 2-3 times faster than our md4 based code. On virtualized
FreeBSD 10 the difference is even larger.

My first approach was to use fopen/fread. It was even faster, but less
than twice as fast. So I thought we could save the additional complexity
when having to deal with throwing away buffers when forking and the
additional memory footprint per process.

With this simple generate_random_buffer it will be easier to adapt new
syscalls to get randomness.

Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>

Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Oct 13 04:25:39 CEST 2015 on sn-devel-104
2015-10-13 04:25:38 +02:00

413 lines
11 KiB
C

/*
Unix SMB/CIFS implementation.
process model: standard (1 process per client connection)
Copyright (C) Andrew Tridgell 1992-2005
Copyright (C) James J Myers 2003 <myersjj@samba.org>
Copyright (C) Stefan (metze) Metzmacher 2004
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3 of the License, or
(at your option) any later version.
This program 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 General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#include "includes.h"
#include "lib/events/events.h"
#include "smbd/process_model.h"
#include "system/filesys.h"
#include "cluster/cluster.h"
#include "param/param.h"
#include "ldb_wrap.h"
struct standard_child_state {
const char *name;
pid_t pid;
int to_parent_fd;
int from_child_fd;
struct tevent_fd *from_child_fde;
};
NTSTATUS process_model_standard_init(void);
/* we hold a pipe open in the parent, and the any child
processes wait for EOF on that pipe. This ensures that
children die when the parent dies */
static int child_pipe[2] = { -1, -1 };
/*
called when the process model is selected
*/
static void standard_model_init(void)
{
int rc;
rc = pipe(child_pipe);
if (rc < 0) {
smb_panic("Failed to initialze pipe!");
}
}
/*
handle EOF on the parent-to-all-children pipe in the child
*/
static void standard_pipe_handler(struct tevent_context *event_ctx, struct tevent_fd *fde,
uint16_t flags, void *private_data)
{
DEBUG(10,("Child %d exiting\n", (int)getpid()));
exit(0);
}
/*
handle EOF on the child pipe in the parent, so we know when a
process terminates without using SIGCHLD or waiting on all possible pids.
We need to ensure we do not ignore SIGCHLD because we need it to
work to get a valid error code from samba_runcmd_*().
*/
static void standard_child_pipe_handler(struct tevent_context *ev,
struct tevent_fd *fde,
uint16_t flags,
void *private_data)
{
struct standard_child_state *state
= talloc_get_type_abort(private_data, struct standard_child_state);
int status = 0;
pid_t pid;
/* the child has closed the pipe, assume its dead */
errno = 0;
pid = waitpid(state->pid, &status, 0);
if (pid != state->pid) {
if (errno == ECHILD) {
/*
* this happens when the
* parent has set SIGCHLD to
* SIG_IGN. In that case we
* can only get error
* information for the child
* via its logging. We should
* stop using SIG_IGN on
* SIGCHLD in the standard
* process model.
*/
DEBUG(0, ("Error in waitpid() unexpectedly got ECHILD "
"for child %d (%s) - %s, someone has set SIGCHLD "
"to SIG_IGN!\n",
(int)state->pid, state->name,
strerror(errno)));
TALLOC_FREE(state);
return;
}
DEBUG(0, ("Error in waitpid() for child %d (%s) - %s \n",
(int)state->pid, state->name, strerror(errno)));
if (errno == 0) {
errno = ECHILD;
}
TALLOC_FREE(state);
return;
}
if (WIFEXITED(status)) {
status = WEXITSTATUS(status);
DEBUG(2, ("Child %d (%s) exited with status %d\n",
(int)state->pid, state->name, status));
} else if (WIFSIGNALED(status)) {
status = WTERMSIG(status);
DEBUG(0, ("Child %d (%s) terminated with signal %d\n",
(int)state->pid, state->name, status));
}
TALLOC_FREE(state);
return;
}
static struct standard_child_state *setup_standard_child_pipe(struct tevent_context *ev,
const char *name)
{
struct standard_child_state *state;
int parent_child_pipe[2];
int ret;
/*
* Prepare a pipe to allow us to know when the child exits,
* because it will trigger a read event on this private
* pipe.
*
* We do all this before the accept and fork(), so we can
* clean up if it fails.
*/
state = talloc_zero(ev, struct standard_child_state);
if (state == NULL) {
return NULL;
}
if (name == NULL) {
name = "";
}
state->name = talloc_strdup(state, name);
if (state->name == NULL) {
TALLOC_FREE(state);
return NULL;
}
ret = pipe(parent_child_pipe);
if (ret == -1) {
DEBUG(0, ("Failed to create parent-child pipe to handle "
"SIGCHLD to track new process for socket\n"));
TALLOC_FREE(state);
return NULL;
}
smb_set_close_on_exec(parent_child_pipe[0]);
smb_set_close_on_exec(parent_child_pipe[1]);
state->from_child_fd = parent_child_pipe[0];
state->to_parent_fd = parent_child_pipe[1];
/*
* The basic purpose of calling this handler is to ensure we
* call waitpid() and so avoid zombies (now that we no longer
* user SIGIGN on for SIGCHLD), but it also allows us to clean
* up other resources in the future.
*/
state->from_child_fde = tevent_add_fd(ev, state,
state->from_child_fd,
TEVENT_FD_READ,
standard_child_pipe_handler,
state);
if (state->from_child_fde == NULL) {
TALLOC_FREE(state);
return NULL;
}
tevent_fd_set_auto_close(state->from_child_fde);
return state;
}
/*
called when a listening socket becomes readable.
*/
static void standard_accept_connection(struct tevent_context *ev,
struct loadparm_context *lp_ctx,
struct socket_context *sock,
void (*new_conn)(struct tevent_context *,
struct loadparm_context *, struct socket_context *,
struct server_id , void *),
void *private_data)
{
NTSTATUS status;
struct socket_context *sock2;
pid_t pid;
struct socket_address *c, *s;
struct standard_child_state *state;
state = setup_standard_child_pipe(ev, NULL);
if (state == NULL) {
return;
}
/* accept an incoming connection. */
status = socket_accept(sock, &sock2);
if (!NT_STATUS_IS_OK(status)) {
DEBUG(0,("standard_accept_connection: accept: %s\n",
nt_errstr(status)));
/* this looks strange, but is correct. We need to throttle things until
the system clears enough resources to handle this new socket */
sleep(1);
close(state->to_parent_fd);
state->to_parent_fd = -1;
TALLOC_FREE(state);
return;
}
pid = fork();
if (pid != 0) {
close(state->to_parent_fd);
state->to_parent_fd = -1;
if (pid > 0) {
state->pid = pid;
} else {
TALLOC_FREE(state);
}
/* parent or error code ... */
talloc_free(sock2);
/* go back to the event loop */
return;
}
/* this leaves state->to_parent_fd open */
TALLOC_FREE(state);
pid = getpid();
/* This is now the child code. We need a completely new event_context to work with */
if (tevent_re_initialise(ev) != 0) {
smb_panic("Failed to re-initialise tevent after fork");
}
/* this will free all the listening sockets and all state that
is not associated with this new connection */
talloc_free(sock);
/* we don't care if the dup fails, as its only a select()
speed optimisation */
socket_dup(sock2);
/* tdb needs special fork handling */
ldb_wrap_fork_hook();
tevent_add_fd(ev, ev, child_pipe[0], TEVENT_FD_READ,
standard_pipe_handler, NULL);
if (child_pipe[1] != -1) {
close(child_pipe[1]);
child_pipe[1] = -1;
}
/* setup the process title */
c = socket_get_peer_addr(sock2, ev);
s = socket_get_my_addr(sock2, ev);
if (s && c) {
setproctitle("conn c[%s:%u] s[%s:%u] server_id[%d]",
c->addr, c->port, s->addr, s->port, (int)pid);
}
talloc_free(c);
talloc_free(s);
/* setup this new connection. Cluster ID is PID based for this process model */
new_conn(ev, lp_ctx, sock2, cluster_id(pid, 0), private_data);
/* we can't return to the top level here, as that event context is gone,
so we now process events in the new event context until there are no
more to process */
tevent_loop_wait(ev);
talloc_free(ev);
exit(0);
}
/*
called to create a new server task
*/
static void standard_new_task(struct tevent_context *ev,
struct loadparm_context *lp_ctx,
const char *service_name,
void (*new_task)(struct tevent_context *, struct loadparm_context *lp_ctx, struct server_id , void *),
void *private_data)
{
pid_t pid;
struct standard_child_state *state;
state = setup_standard_child_pipe(ev, service_name);
if (state == NULL) {
return;
}
pid = fork();
if (pid != 0) {
close(state->to_parent_fd);
state->to_parent_fd = -1;
if (pid > 0) {
state->pid = pid;
} else {
TALLOC_FREE(state);
}
/* parent or error code ... go back to the event loop */
return;
}
/* this leaves state->to_parent_fd open */
TALLOC_FREE(state);
pid = getpid();
/* this will free all the listening sockets and all state that
is not associated with this new connection */
if (tevent_re_initialise(ev) != 0) {
smb_panic("Failed to re-initialise tevent after fork");
}
/* ldb/tdb need special fork handling */
ldb_wrap_fork_hook();
tevent_add_fd(ev, ev, child_pipe[0], TEVENT_FD_READ,
standard_pipe_handler, NULL);
if (child_pipe[1] != -1) {
close(child_pipe[1]);
child_pipe[1] = -1;
}
setproctitle("task %s server_id[%d]", service_name, (int)pid);
/* setup this new task. Cluster ID is PID based for this process model */
new_task(ev, lp_ctx, cluster_id(pid, 0), private_data);
/* we can't return to the top level here, as that event context is gone,
so we now process events in the new event context until there are no
more to process */
tevent_loop_wait(ev);
talloc_free(ev);
exit(0);
}
/* called when a task goes down */
_NORETURN_ static void standard_terminate(struct tevent_context *ev, struct loadparm_context *lp_ctx,
const char *reason)
{
DEBUG(2,("standard_terminate: reason[%s]\n",reason));
talloc_free(ev);
/* this reload_charcnv() has the effect of freeing the iconv context memory,
which makes leak checking easier */
reload_charcnv(lp_ctx);
/* terminate this process */
exit(0);
}
/* called to set a title of a task or connection */
static void standard_set_title(struct tevent_context *ev, const char *title)
{
if (title) {
setproctitle("%s", title);
} else {
setproctitle(NULL);
}
}
static const struct model_ops standard_ops = {
.name = "standard",
.model_init = standard_model_init,
.accept_connection = standard_accept_connection,
.new_task = standard_new_task,
.terminate = standard_terminate,
.set_title = standard_set_title,
};
/*
initialise the standard process model, registering ourselves with the process model subsystem
*/
NTSTATUS process_model_standard_init(void)
{
return register_process_model(&standard_ops);
}