Aren
/
NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
NetBSD/dist/bind/bin/named/ns_main.c

2923 lines
78 KiB
C
Raw Blame History

/* $NetBSD: ns_main.c,v 1.11 2001/08/24 13:25:57 mrg Exp $ */
#if !defined(lint) && !defined(SABER)
static const char sccsid[] = "@(#)ns_main.c 4.55 (Berkeley) 7/1/91";
static const char rcsid[] = "Id: ns_main.c,v 8.145 2001/03/16 12:07:57 marka Exp";
#endif /* not lint */
/*
* Copyright (c) 1986, 1989, 1990
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the University of
* California, Berkeley and its contributors.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
/*
* Portions Copyright (c) 1993 by Digital Equipment Corporation.
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies, and that
* the name of Digital Equipment Corporation not be used in advertising or
* publicity pertaining to distribution of the document or software without
* specific, written prior permission.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND DIGITAL EQUIPMENT CORP. DISCLAIMS ALL
* WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL DIGITAL EQUIPMENT
* CORPORATION BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL
* DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR
* PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
* ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
* SOFTWARE.
*/
/*
* Portions Copyright (c) 1996-2000 by Internet Software Consortium.
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND INTERNET SOFTWARE CONSORTIUM DISCLAIMS
* ALL WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL INTERNET SOFTWARE
* CONSORTIUM BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL
* DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR
* PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
* ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
* SOFTWARE.
*/
#if !defined(lint) && !defined(SABER)
char copyright[] =
"@(#) Copyright (c) 1986, 1989, 1990 The Regents of the University of California.\n"
"portions Copyright (c) 1993 Digital Equipment Corporation\n"
"portions Copyright (c) 1995-1999 Internet Software Consortium\n"
"portions Copyright (c) 1999 Check Point Software Technologies\n"
"All rights reserved.\n";
#endif /* not lint */
/*
* Internet Name server (see RCF1035 & others).
*/
#include "port_before.h"
#include <sys/types.h>
#include <sys/param.h>
#include <sys/file.h>
#include <sys/stat.h>
#include <sys/wait.h>
#include <sys/ioctl.h>
#include <sys/socket.h>
#include <sys/un.h>
#ifdef SVR4 /* XXX */
# include <sys/sockio.h>
#else
# include <sys/mbuf.h>
#endif
#include <netinet/in.h>
#include <net/route.h>
#include <net/if.h>
#include <arpa/nameser.h>
#include <arpa/inet.h>
#include <ctype.h>
#include <errno.h>
#include <fcntl.h>
#include <grp.h>
#include <irs.h>
#include <stdio.h>
#include <stdlib.h>
#include <signal.h>
#include <netdb.h>
#include <pwd.h>
#include <resolv.h>
#include <string.h>
#include <syslog.h>
#include <time.h>
#include <unistd.h>
#include <isc/eventlib.h>
#include <isc/logging.h>
#include <isc/memcluster.h>
#include <isc/list.h>
#include "port_after.h"
#ifdef HAVE_GETRUSAGE /* XXX */
#include <sys/resource.h>
#endif
#define MAIN_PROGRAM
#include "named.h"
#undef MAIN_PROGRAM
typedef void (*handler)(void);
typedef struct _savedg {
struct sockaddr_in from;
int dfd;
interface * ifp;
time_t gen;
u_char * buf;
u_int16_t buflen;
} savedg;
/* list of interfaces */
static LIST(struct _interface) iflist;
static int iflist_initialized = 0;
static int iflist_dont_rescan = 0;
static const int drbufsize = 32 * 1024, /* UDP rcv buf size */
dsbufsize = 48 * 1024, /* UDP snd buf size */
sbufsize = 16 * 1024, /* TCP snd buf size */
#ifdef BROKEN_RECVFROM
nudptrans = 1,
#else
nudptrans = 20, /* #/udps per select */
#endif
listenmax = 50;
static u_int16_t nsid_state;
static u_int16_t *nsid_pool; /* optional query id pool */
static u_int16_t *nsid_vtable; /* optional shuffle table */
static u_int32_t nsid_hash_state;
static u_int16_t nsid_a1, nsid_a2, nsid_a3;
static u_int16_t nsid_c1, nsid_c2, nsid_c3;
static u_int16_t nsid_state2;
static int nsid_algorithm;
static int needs = 0, needs_exit = 0;
static handler handlers[main_need_num];
static void savedg_waitfunc(evContext, void*, const void*);
static void need_waitfunc(evContext, void *, const void *);
static int drain_rcvbuf(evContext, interface *, int,
int *, int *);
static int drain_all_rcvbuf(evContext);
static struct qstream *sq_add(void);
static int opensocket_d(interface *),
opensocket_s(interface *);
static void sq_query(struct qstream *),
dq_remove(interface *);
static int sq_dowrite(struct qstream *);
static void use_desired_debug(void);
static void stream_write(evContext, void *, int, int);
static interface * if_find(struct in_addr, u_int16_t port);
static void deallocate_everything(void),
stream_accept(evContext, void *, int,
const void *, int,
const void *, int),
stream_getlen(evContext, void *, int, int),
stream_getmsg(evContext, void *, int, int),
datagram_read(evContext, void *, int, int),
dispatch_message(u_char *, int, int,
struct qstream *,
struct sockaddr_in, int,
interface *);
static void stream_send(evContext, void *, int,
const void *, int,
const void *, int);
static int only_digits(const char *);
static void init_needs(void),
handle_needs(void),
exit_handler(void);
#ifndef HAVE_CUSTOM
static void custom_init(void),
custom_shutdown(void);
#endif
static void
usage() {
fprintf(stderr,
"Usage: named [-d #] [-q] [-r] [-v] [-f] [-p port] [[-b|-c] configfile]\n");
#ifdef CAN_CHANGE_ID
fprintf(stderr,
" [-u (username|uid)] [-g (groupname|gid)]\n");
#endif
#ifdef HAVE_CHROOT
fprintf(stderr,
" [-t directory]\n");
#endif
exit(1);
}
static const char bad_p_option[] =
"-p remote/local obsolete; use 'listen-on' in config file to specify local";
static const char bad_directory[] = "chdir failed for directory '%s': %s";
/*ARGSUSED*/
int
main(int argc, char *argv[], char *envp[]) {
int n;
char *p;
int ch;
struct passwd *pw;
struct group *gr;
#ifdef _AUX_SOURCE
set42sig();
#endif
debugfile = savestr(_PATH_DEBUG, 1);
user_id = getuid();
group_id = getgid();
ns_port = htons(NAMESERVER_PORT);
desired_debug = debug;
/* BSD has a better random number generator but it's not clear
* that we need it here.
*/
gettime(&tt);
srand(((unsigned)getpid()) + (unsigned)tt.tv_usec);
(void) umask(022);
/* Save argv[] before getopt() destroys it -- needed for execvp(). */
saved_argv = malloc(sizeof(char *) * (argc + 1));
INSIST(saved_argv != NULL);
for (n = 0; n < argc; n++) {
saved_argv[n] = strdup(argv[n]);
INSIST(saved_argv[n] != NULL);
}
saved_argv[argc] = NULL;
/* XXX we need to free() this for clean shutdowns. */
while ((ch = getopt(argc, argv, "b:c:d:g:p:t:u:vw:qrf")) != -1) {
switch (ch) {
case 'b':
case 'c':
if (conffile != NULL)
freestr(conffile);
conffile = savestr(optarg, 1);
break;
case 'd':
desired_debug = atoi(optarg);
if (desired_debug <= 0)
desired_debug = 1;
break;
case 'p':
/* use nonstandard port number.
* usage: -p remote/local
* remote is the port number to which
* we send queries. local is the port
* on which we listen for queries.
* local defaults to same as remote.
*/
ns_port = htons((u_int16_t) atoi(optarg));
p = strchr(optarg, '/');
if (p) {
syslog(LOG_WARNING, bad_p_option);
fprintf(stderr, bad_p_option);
fputc('\n', stderr);
}
break;
case 'w':
if (chdir(optarg) < 0) {
syslog(LOG_CRIT, bad_directory, optarg,
strerror(errno));
fprintf(stderr, bad_directory, optarg,
strerror(errno));
fputc('\n', stderr);
exit(1);
}
break;
#ifdef QRYLOG
case 'q':
qrylog = 1;
break;
#endif
case 'r':
ns_setoption(OPTION_NORECURSE);
break;
case 'f':
foreground = 1;
break;
case 't':
chroot_dir = savestr(optarg, 1);
break;
case 'v':
fprintf(stdout, "%s\n", Version);
exit(0);
#ifdef CAN_CHANGE_ID
case 'u':
user_name = savestr(optarg, 1);
if (only_digits(user_name))
user_id = atoi(user_name);
else {
pw = getpwnam(user_name);
if (pw == NULL) {
fprintf(stderr,
"user \"%s\" unknown\n",
user_name);
exit(1);
}
user_id = pw->pw_uid;
if (group_name == NULL) {
char name[256];
sprintf(name, "%lu",
(u_long)pw->pw_gid);
group_name = savestr(name, 1);
group_id = pw->pw_gid;
}
}
break;
case 'g':
if (group_name != NULL)
freestr(group_name);
group_name = savestr(optarg, 1);
if (only_digits(group_name))
group_id = atoi(group_name);
else {
gr = getgrnam(group_name);
if (gr == NULL) {
fprintf(stderr,
"group \"%s\" unknown\n",
group_name);
exit(1);
}
group_id = gr->gr_gid;
}
break;
#endif /* CAN_CHANGE_ID */
case '?':
default:
usage();
}
}
argc -= optind;
argv += optind;
if (argc) {
if (conffile != NULL)
freestr(conffile);
conffile = savestr(*argv, 1);
argc--, argv++;
}
if (argc)
usage();
if (conffile == NULL)
conffile = savestr(_PATH_CONF, 1);
/*
* Make sure we don't inherit any open descriptors
* other than those that daemon() can deal with.
*/
for (n = sysconf(_SC_OPEN_MAX) - 1; n >= 0; n--)
if (n != STDIN_FILENO &&
n != STDOUT_FILENO &&
n != STDERR_FILENO)
(void) close(n);
/*
* Chroot if desired.
*/
if (chroot_dir != NULL) {
#ifdef HAVE_CHROOT
if (chroot(chroot_dir) < 0) {
fprintf(stderr, "chroot %s failed: %s\n", chroot_dir,
strerror(errno));
exit(1);
}
if (chdir("/") < 0) {
fprintf(stderr, "chdir(\"/\") failed: %s\n",
strerror(errno));
exit(1);
}
#else
fprintf(stderr, "warning: chroot() not available\n");
freestr(chroot_dir);
chroot_dir = NULL;
#endif
}
/* Establish global event context. */
evCreate(&ev);
/* Establish global resolver context. */
res_ninit(&res);
res.options &= ~(RES_DEFNAMES | RES_DNSRCH | RES_RECURSE);
/*
* Set up logging.
*/
n = LOG_PID;
#ifdef LOG_NOWAIT
n |= LOG_NOWAIT;
#endif
#ifdef LOG_NDELAY
n |= LOG_NDELAY;
#endif
#if defined(LOG_CONS) && defined(USE_LOG_CONS)
n |= LOG_CONS;
#endif
#ifdef SYSLOG_42BSD
openlog("named", n);
#else
openlog("named", n, ISC_FACILITY);
#endif
init_logging();
set_assertion_failure_callback(ns_assertion_failed);
#ifdef DEBUG
use_desired_debug();
#endif
/* Perform system-dependent initialization */
custom_init();
init_needs();
init_signals();
ns_notice(ns_log_default, "starting (%s). %s", conffile, Version);
/*
* Initialize and load database.
*/
gettime(&tt);
buildservicelist();
buildprotolist();
confmtime = ns_init(conffile);
time(&boottime);
resettime = boottime;
nsid_init();
/*
* Fork and go into background now that
* we've done any slow initialization
* and are ready to answer queries.
*/
if (foreground == 0) {
if (daemon(1, 0))
ns_panic(ns_log_default, 1, "daemon: %s",
strerror(errno));
update_pid_file();
}
/* Check that udp checksums are on. */
ns_udp();
/*
* We waited until now to log this because we wanted logging to
* be set up the way the user prefers.
*/
if (chroot_dir != NULL)
ns_info(ns_log_security, "chrooted to %s", chroot_dir);
#ifdef CAN_CHANGE_ID
/*
* Set user and group if desired.
*/
if (group_name != NULL) {
if (setgid(group_id) < 0)
ns_panic(ns_log_security, 1, "setgid(%s): %s",
group_name, strerror(errno));
ns_info(ns_log_security, "group = %s", group_name);
}
if (user_name != NULL) {
if (getuid() == 0 && initgroups(user_name, group_id) < 0)
ns_panic(ns_log_security, 1, "initgroups(%s, %d): %s",
user_name, (int)group_id, strerror(errno));
endgrent();
endpwent();
if (setuid(user_id) < 0)
ns_panic(ns_log_security, 1, "setuid(%s): %s",
user_name, strerror(errno));
ns_info(ns_log_security, "user = %s", user_name);
if (user_id != 0)
iflist_dont_rescan++;
}
#endif /* CAN_CHANGE_ID */
ns_notice(ns_log_default, "Ready to answer queries.");
gettime(&tt);
prime_cache();
while (!needs_exit) {
evEvent event;
ns_debug(ns_log_default, 15, "main loop");
if (needs != 0)
handle_needs();
else if (evGetNext(ev, &event, EV_WAIT) != -1)
INSIST_ERR(evDispatch(ev, event) != -1);
else
INSIST_ERR(errno == EINTR);
}
ns_info(ns_log_default, "named shutting down");
#ifdef BIND_UPDATE
dynamic_about_to_exit();
#endif
if (server_options && server_options->pid_filename)
(void)unlink(server_options->pid_filename);
ns_logstats(ev, NULL, evNowTime(), evConsTime(0, 0));
if (NS_OPTION_P(OPTION_DEALLOC_ON_EXIT))
deallocate_everything();
else
shutdown_configuration();
/* Cleanup for system-dependent stuff */
custom_shutdown();
return (0);
}
static int
ns_socket(int domain, int type, int protocol) {
int fd;
fd = socket(domain, type, protocol);
if (fd == -1)
return (-1);
#ifdef F_DUPFD /* XXX */
/*
* Leave a space for stdio to work in.
*/
if (fd >= 0 && fd <= 20) {
int new, tmp;
if ((new = fcntl(fd, F_DUPFD, 20)) == -1)
ns_notice(ns_log_default, "fcntl(fd, F_DUPFD, 20): %s",
strerror(errno));
tmp = errno;
close(fd);
errno = tmp;
fd = new;
}
#endif
return (fd);
}
#ifndef IP_OPT_BUF_SIZE
/* arbitrary size */
#define IP_OPT_BUF_SIZE 50
#endif
static void
stream_accept(evContext lev, void *uap, int rfd,
const void *lav, int lalen,
const void *rav, int ralen)
{
interface *ifp = uap;
struct qstream *sp;
struct iovec iov;
int len, n;
const int on = 1;
#ifdef IP_OPTIONS /* XXX */
u_char ip_opts[IP_OPT_BUF_SIZE];
#endif
const struct sockaddr_in *la, *ra;
la = (const struct sockaddr_in *)lav;
ra = (const struct sockaddr_in *)rav;
INSIST(ifp != NULL);
#ifdef F_DUPFD
/*
* Leave a space for stdio to work in.
*/
if (rfd >= 0 && rfd <= 20) {
int new, tmp;
new = fcntl(rfd, F_DUPFD, 20);
tmp = errno;
if (new == -1)
ns_notice(ns_log_default,
"fcntl(rfd, F_DUPFD, 20): %s",
strerror(errno));
close(rfd);
errno = tmp;
rfd = new;
}
#endif
if (rfd < 0) {
switch (errno) {
case EINTR:
case EAGAIN:
#if (EWOULDBLOCK != EAGAIN)
case EWOULDBLOCK:
#endif
case ECONNABORTED:
#ifdef EPROTO
case EPROTO:
#endif
case EHOSTUNREACH:
case EHOSTDOWN:
case ENETUNREACH:
case ENETDOWN:
case ECONNREFUSED:
#ifdef ENONET
case ENONET:
#endif
/*
* These errors are expected and harmless, so
* we ignore them.
*/
return;
case EBADF:
case ENOTSOCK:
case EFAULT:
/*
* If one these happens, we're broken.
*/
ns_panic(ns_log_default, 1, "accept: %s",
strerror(errno));
case EMFILE:
/*
* If we're out of file descriptors, find the least
* busy fd and close it. Then we'll return to the
* eventlib which will call us right back.
*/
if (streamq) {
struct qstream *nextsp;
struct qstream *candidate = NULL;
time_t lasttime, maxctime = 0;
for (sp = streamq; sp; sp = nextsp) {
nextsp = sp->s_next;
if (sp->s_refcnt)
continue;
gettime(&tt);
lasttime = tt.tv_sec - sp->s_time;
if (lasttime >= VQEXPIRY)
sq_remove(sp);
else if (lasttime > maxctime) {
candidate = sp;
maxctime = lasttime;
}
}
if (candidate)
sq_remove(candidate);
return;
}
/* fall through */
default:
/*
* Either we got an error we didn't expect, or we
* got EMFILE and didn't have anything left to close.
* Log it and press on.
*/
ns_info(ns_log_default, "accept: %s", strerror(errno));
return;
}
}
/* Condition the socket. */
#ifndef CANNOT_SET_SNDBUF
if (setsockopt(rfd, SOL_SOCKET, SO_SNDBUF,
(char*)&sbufsize, sizeof sbufsize) < 0) {
ns_info(ns_log_default, "setsockopt(rfd, SO_SNDBUF, %d): %s",
sbufsize, strerror(errno));
(void) close(rfd);
return;
}
#endif
if (setsockopt(rfd, SOL_SOCKET, SO_KEEPALIVE,
(char *)&on, sizeof on) < 0) {
ns_info(ns_log_default, "setsockopt(rfd, KEEPALIVE): %s",
strerror(errno));
(void) close(rfd);
return;
}
if ((n = fcntl(rfd, F_GETFL, 0)) == -1) {
ns_info(ns_log_default, "fcntl(rfd, F_GETFL): %s",
strerror(errno));
(void) close(rfd);
return;
}
if (fcntl(rfd, F_SETFL, n|PORT_NONBLOCK) == -1) {
ns_info(ns_log_default, "fcntl(rfd, NONBLOCK): %s",
strerror(errno));
(void) close(rfd);
return;
}
/*
* We don't like IP options. Turn them off if the connection came in
* with any. log this event since it usually indicates a security
* problem.
*/
#if defined(IP_OPTIONS) /* XXX */
len = sizeof ip_opts;
if (getsockopt(rfd, IPPROTO_IP, IP_OPTIONS,
(char *)ip_opts, &len) < 0) {
ns_info(ns_log_default, "getsockopt(rfd, IP_OPTIONS): %s",
strerror(errno));
(void) close(rfd);
return;
}
if (len != 0) {
nameserIncr(ra->sin_addr, nssRcvdOpts);
if (!haveComplained(ina_ulong(ra->sin_addr),
(u_long)"rcvd ip options")) {
ns_info(ns_log_default,
"rcvd IP_OPTIONS from %s (ignored)",
sin_ntoa(*ra));
}
if (setsockopt(rfd, IPPROTO_IP, IP_OPTIONS, NULL, 0) < 0) {
ns_info(ns_log_default, "setsockopt(!IP_OPTIONS): %s",
strerror(errno));
(void) close(rfd);
}
}
#endif
/* Create and populate a qsp for this socket. */
if ((sp = sq_add()) == NULL) {
(void) close(rfd);
return;
}
sp->s_rfd = rfd; /* stream file descriptor */
gettime(&tt);
sp->s_time = tt.tv_sec; /* last transaction time */
sp->s_from = *ra; /* address to respond to */
sp->s_ifp = ifp;
INSIST(sizeof sp->s_temp >= INT16SZ);
iov = evConsIovec(sp->s_temp, INT16SZ);
INSIST_ERR(evRead(lev, rfd, &iov, 1, stream_getlen, sp, &sp->evID_r)
!= -1);
sp->flags |= STREAM_READ_EV;
ns_debug(ns_log_default, 1, "IP/TCP connection from %s (fd %d)",
sin_ntoa(sp->s_from), rfd);
}
int
tcp_send(struct qinfo *qp) {
struct qstream *sp;
struct sockaddr_in src;
int on = 1, n;
ns_debug(ns_log_default, 1, "tcp_send");
if ((sp = sq_add()) == NULL) {
return (SERVFAIL);
}
if ((sp->s_rfd = ns_socket(AF_INET, SOCK_STREAM, PF_UNSPEC)) == -1) {
sq_remove(sp);
return (SERVFAIL);
}
if (sp->s_rfd > evHighestFD(ev)) {
sq_remove(sp);
return (SERVFAIL);
}
if (setsockopt(sp->s_rfd, SOL_SOCKET, SO_REUSEADDR,
(char*)&on, sizeof(on)) < 0)
ns_info(ns_log_default,
"tcp_send: setsockopt(SO_REUSEADDR): %s",
strerror(errno));
#ifdef SO_REUSEPORT
if (setsockopt(sp->s_rfd, SOL_SOCKET, SO_REUSEPORT,
(char*)&on, sizeof(on)) < 0)
ns_info(ns_log_default,
"tcp_send: setsockopt(SO_REUSEPORT): %s",
strerror(errno));
#endif
src = server_options->query_source;
src.sin_port = htons(0);
if (bind(sp->s_rfd, (struct sockaddr *)&src, sizeof(src)) < 0)
ns_info(ns_log_default, "tcp_send: bind(query_source): %s",
strerror(errno));
if (fcntl(sp->s_rfd, F_SETFD, 1) < 0) {
sq_remove(sp);
return (SERVFAIL);
}
if ((n = fcntl(sp->s_rfd, F_GETFL, 0)) == -1) {
sq_remove(sp);
return (SERVFAIL);
}
if (fcntl(sp->s_rfd, F_SETFL, n|PORT_NONBLOCK) == -1) {
sq_remove(sp);
return (SERVFAIL);
}
if (sq_openw(sp, qp->q_msglen + INT16SZ) == -1) {
sq_remove(sp);
return (SERVFAIL);
}
if (sq_write(sp, qp->q_msg, qp->q_msglen) == -1) {
sq_remove(sp);
return (SERVFAIL);
}
if (setsockopt(sp->s_rfd, SOL_SOCKET, SO_KEEPALIVE,
(char*)&on, sizeof(on)) < 0)
ns_info(ns_log_default,
"tcp_send: setsockopt(SO_KEEPALIVE): %s",
strerror(errno));
gettime(&tt);
sp->s_size = -1;
sp->s_time = tt.tv_sec; /* last transaction time */
sp->s_refcnt = 1;
sp->flags |= STREAM_DONE_CLOSE;
sp->s_from = qp->q_addr[qp->q_curaddr].ns_addr;
if (evConnect(ev, sp->s_rfd, &sp->s_from, sizeof(sp->s_from),
stream_send, sp, &sp->evID_c) == -1) {
sq_remove(sp);
return (SERVFAIL);
}
sp->flags |= STREAM_CONNECT_EV;
return (NOERROR);
}
static void
stream_send(evContext lev, void *uap, int fd, const void *la, int lalen,
const void *ra, int ralen) {
struct qstream *sp = uap;
ns_debug(ns_log_default, 1, "stream_send");
sp->flags &= ~STREAM_CONNECT_EV;
if (fd == -1) {
/* connect failed */
sq_remove(sp);
return;
}
if (evSelectFD(ev, sp->s_rfd, EV_WRITE,
stream_write, sp, &sp->evID_w) < 0) {
sq_remove(sp);
return;
}
sp->flags |= STREAM_WRITE_EV;
}
static void
stream_write(evContext ctx, void *uap, int fd, int evmask) {
struct qstream *sp = uap;
struct iovec iov;
ns_debug(ns_log_default, 1, "stream_write");
INSIST(evmask & EV_WRITE);
INSIST(fd == sp->s_rfd);
if (sq_dowrite(sp) < 0) {
sq_remove(sp);
return;
}
if (sp->s_wbuf_free != sp->s_wbuf_send)
return;
if (sp->s_wbuf) {
memput(sp->s_wbuf, sp->s_wbuf_end - sp->s_wbuf);
sp->s_wbuf_send = sp->s_wbuf_free = NULL;
sp->s_wbuf_end = sp->s_wbuf = NULL;
}
(void) evDeselectFD(ev, sp->evID_w);
sp->flags &= ~STREAM_WRITE_EV;
sp->s_refcnt = 0;
iov = evConsIovec(sp->s_temp, INT16SZ);
INSIST_ERR(evRead(ctx, fd, &iov, 1, stream_getlen, sp, &sp->evID_r) !=
-1);
sp->flags |= STREAM_READ_EV;
}
static void
stream_getlen(evContext lev, void *uap, int fd, int bytes) {
struct qstream *sp = uap;
struct iovec iov;
sp->flags &= ~STREAM_READ_EV;
if (bytes != INT16SZ) {
/*
* bytes == 0 is normal EOF; see if something unusual
* happened.
*/
if (bytes < 0) {
/*
* ECONNRESET happens frequently and is not worth
* logging.
*/
if (errno != ECONNRESET)
ns_info(ns_log_default,
"stream_getlen(%s): %s",
sin_ntoa(sp->s_from), strerror(errno));
} else if (bytes != 0)
ns_error(ns_log_default,
"stream_getlen(%s): unexpected byte count %d",
sin_ntoa(sp->s_from), bytes);
sq_remove(sp);
return;
}
/*
* Unpack the size, allocate memory for the query. This is
* tricky since in a low memory situation with possibly very
* large (64KB) queries, we want to make sure we can read at
* least the header since we need it to send back a SERVFAIL
* (owing to the out-of-memory condition).
*/
sp->s_size = ns_get16(sp->s_temp);
ns_debug(ns_log_default, 5, "stream message: %d bytes", sp->s_size);
if (sp->s_size < HFIXEDSZ) {
ns_error(ns_log_default,
"stream_getlen(%s): request too small",
sin_ntoa(sp->s_from));
sq_remove(sp);
return;
}
if (!(sp->flags & STREAM_MALLOC)) {
sp->s_bufsize = 64*1024-1; /* maximum tcp message size */
sp->s_buf = (u_char *)memget(sp->s_bufsize);
if (sp->s_buf != NULL)
sp->flags |= STREAM_MALLOC;
else {
sp->s_buf = sp->s_temp;
sp->s_bufsize = HFIXEDSZ;
}
}
iov = evConsIovec(sp->s_buf, (sp->s_size <= sp->s_bufsize) ?
sp->s_size : sp->s_bufsize);
if (evRead(lev, sp->s_rfd, &iov, 1, stream_getmsg, sp, &sp->evID_r)
== -1)
ns_panic(ns_log_default, 1, "evRead(fd %d): %s",
sp->s_rfd, strerror(errno));
sp->flags |= STREAM_READ_EV;
}
static void
stream_getmsg(evContext lev, void *uap, int fd, int bytes) {
struct qstream *sp = uap;
sp->flags &= ~STREAM_READ_EV;
if (bytes == -1) {
ns_info(ns_log_default, "stream_getmsg(%s): %s",
sin_ntoa(sp->s_from), strerror(errno));
sq_remove(sp);
return;
}
gettime(&tt);
sp->s_time = tt.tv_sec;
if (ns_wouldlog(ns_log_default,5)) {
ns_debug(ns_log_default, 5, "sp %p rfd %d size %d time %ld next %p",
sp, sp->s_rfd, sp->s_size, (long)sp->s_time, sp->s_next);
ns_debug(ns_log_default, 5, "\tbufsize %d bytes %d", sp->s_bufsize,
bytes);
}
/*
* Do we have enough memory for the query? If not, and if we have a
* query id, then we will send a SERVFAIL error back to the client.
*/
if (bytes != sp->s_size) {
HEADER *hp = (HEADER *)sp->s_buf;
hp->qr = 1;
hp->ra = (NS_OPTION_P(OPTION_NORECURSE) == 0);
hp->ancount = htons(0);
hp->qdcount = htons(0);
hp->nscount = htons(0);
hp->arcount = htons(0);
hp->rcode = SERVFAIL;
writestream(sp, sp->s_buf, HFIXEDSZ);
sp->flags |= STREAM_DONE_CLOSE;
return;
}
nameserIncr(sp->s_from.sin_addr, nssRcvdTCP);
sq_query(sp);
dispatch_message(sp->s_buf, bytes, sp->s_bufsize, sp, sp->s_from, -1,
sp->s_ifp);
}
static void
datagram_read(evContext lev, void *uap, int fd, int evmask) {
interface *ifp = uap;
struct sockaddr_in from;
int from_len = sizeof from;
int n, nudp;
union {
HEADER h; /* Force alignment of 'buf'. */
u_char buf[PACKETSZ+1];
} u;
tt = evTimeVal(evNowTime());
nudp = 0;
more:
n = recvfrom(fd, (char *)u.buf, sizeof u.buf, 0,
(struct sockaddr *)&from, &from_len);
if (n < 0) {
switch (errno) {
case EINTR:
case EAGAIN:
#if (EWOULDBLOCK != EAGAIN)
case EWOULDBLOCK:
#endif
case EHOSTUNREACH:
case EHOSTDOWN:
case ENETUNREACH:
case ENETDOWN:
case ECONNREFUSED:
#ifdef ENONET
case ENONET:
#endif
/*
* These errors are expected and harmless, so we
* ignore them.
*/
return;
default:
/*
* An error we don't expect. Log it and press
* on.
*/
ns_info(ns_log_default, "recvfrom: %s",
strerror(errno));
return;
}
}
/* Handle bogosity on systems that need it. */
if (n == 0)
return;
if (ns_wouldlog(ns_log_default, 1)) {
ns_debug(ns_log_default, 1, "datagram from %s, fd %d, len %d",
sin_ntoa(from), fd, n);
}
if (n > PACKETSZ) {
/*
* The message is too big. It's probably a response to
* one of our questions, so we truncate it and press on.
*/
n = trunc_adjust(u.buf, PACKETSZ, PACKETSZ);
ns_debug(ns_log_default, 1, "truncated oversize UDP packet");
}
dispatch_message(u.buf, n, PACKETSZ, NULL, from, fd, ifp);
if (++nudp < nudptrans)
goto more;
}
static void
savedg_waitfunc(evContext ctx, void *uap, const void *tag) {
savedg *dg = (savedg *)uap;
if (!EMPTY(iflist) && HEAD(iflist)->gen == dg->gen) {
u_char buf[PACKETSZ];
memcpy(buf, dg->buf, dg->buflen);
dispatch_message(buf, dg->buflen, sizeof buf, NULL,
dg->from, dg->dfd, dg->ifp);
}
memput(dg->buf, dg->buflen);
memput(dg, sizeof *dg);
}
static void
dispatch_message(u_char *msg, int msglen, int buflen, struct qstream *qsp,
struct sockaddr_in from, int dfd, interface *ifp)
{
HEADER *hp = (HEADER *)msg;
if (msglen < HFIXEDSZ) {
ns_debug(ns_log_default, 1, "dropping undersize message");
if (qsp) {
qsp->flags |= STREAM_DONE_CLOSE;
sq_done(qsp);
}
return;
}
if (server_options->blackhole_acl != NULL &&
ip_match_address(server_options->blackhole_acl,
from.sin_addr) == 1) {
ns_debug(ns_log_default, 1,
"dropping blackholed %s from %s",
hp->qr ? "response" : "query",
sin_ntoa(from));
if (qsp) {
qsp->flags |= STREAM_DONE_CLOSE;
sq_done(qsp);
}
return;
}
/* Drop UDP packets from port zero. They are invariable forged. */
if (qsp == NULL && ntohs(from.sin_port) == 0) {
ns_notice(ns_log_security,
"dropping source port zero packet from %s",
sin_ntoa(from));
return;
}
if (hp->qr) {
ns_resp(msg, msglen, from, qsp);
if (qsp)
sq_done(qsp);
/* Now is a safe time for housekeeping. */
if (needs_prime_cache)
prime_cache();
} else if (ifp != NULL)
ns_req(msg, msglen, buflen, qsp, from, dfd);
else {
ns_notice(ns_log_security,
"refused query on non-query socket from %s",
sin_ntoa(from));
if (qsp) {
qsp->flags |= STREAM_DONE_CLOSE;
sq_done(qsp);
}
/* XXX Send refusal here. */
}
}
void
getnetconf(int periodic_scan) {
struct ifconf ifc;
struct ifreq ifreq;
struct in_addr ina;
interface *ifp;
char *buf, *cp, *cplim;
static int bufsiz = 4095;
time_t my_generation = time(NULL);
int s, cpsize, n;
int found;
listen_info li;
ip_match_element ime;
u_char *mask_ptr;
struct in_addr mask;
if (iflist_initialized) {
if (iflist_dont_rescan)
return;
} else {
INIT_LIST(iflist);
iflist_initialized = 1;
}
ns_debug(ns_log_default, 1, "getnetconf(generation %lu)",
(u_long)my_generation);
/* Get interface list from system. */
if ((s = socket(AF_INET, SOCK_DGRAM, 0)) < 0) {
if (!periodic_scan)
ns_panic(ns_log_default, 1, "socket(SOCK_RAW): %s",
strerror(errno));
ns_error(ns_log_default, "socket(SOCK_RAW): %s",
strerror(errno));
return;
}
if (local_addresses != NULL)
free_ip_match_list(local_addresses);
local_addresses = new_ip_match_list();
if (local_networks != NULL)
free_ip_match_list(local_networks);
local_networks = new_ip_match_list();
for (;;) {
buf = memget(bufsiz);
if (!buf)
ns_panic(ns_log_default, 1,
"memget(interface)");
ifc.ifc_len = bufsiz;
ifc.ifc_buf = buf;
#ifdef IRIX_EMUL_IOCTL_SIOCGIFCONF
/*
* This is a fix for IRIX OS in which the call to ioctl with
* the flag SIOCGIFCONF may not return an entry for all the
* interfaces like most flavors of Unix.
*/
if (emul_ioctl(&ifc) >= 0)
break;
#else
if ((n = ioctl(s, SIOCGIFCONF, (char *)&ifc)) != -1) {
/*
* Some OS's just return what will fit rather
* than set EINVAL if the buffer is too small
* to fit all the interfaces in. If
* ifc.ifc_len is too near to the end of the
* buffer we will grow it just in case and
* retry.
*/
if (ifc.ifc_len + 2 * sizeof(ifreq) < bufsiz)
break;
}
#endif
if ((n == -1) && errno != EINVAL)
ns_panic(ns_log_default, 1,
"get interface configuration: %s",
strerror(errno));
if (bufsiz > 1000000)
ns_panic(ns_log_default, 1,
"get interface configuration: maximum buffer size exceeded");
memput(buf, bufsiz);
bufsiz += 4096;
}
ns_debug(ns_log_default, 2, "getnetconf: SIOCGIFCONF: ifc_len = %d",
ifc.ifc_len);
/* Parse system's interface list and open some sockets. */
cplim = buf + ifc.ifc_len; /* skip over if's with big ifr_addr's */
for (cp = buf; cp < cplim; cp += cpsize) {
memcpy(&ifreq, cp, sizeof ifreq);
#ifdef HAVE_SA_LEN
#ifdef FIX_ZERO_SA_LEN
if (ifreq.ifr_addr.sa_len == 0)
ifreq.ifr_addr.sa_len = 16;
#endif
#ifdef HAVE_MINIMUM_IFREQ
ns_debug(ns_log_default, 2, "%s sa_len = %d",
ifreq.ifr_name, (int)ifreq.ifr_addr.sa_len);
cpsize = sizeof ifreq;
if (ifreq.ifr_addr.sa_len > sizeof (struct sockaddr))
cpsize += (int)ifreq.ifr_addr.sa_len -
(int)(sizeof (struct sockaddr));
#else
cpsize = sizeof ifreq.ifr_name + ifreq.ifr_addr.sa_len;
#endif /* HAVE_MINIMUM_IFREQ */
#elif defined SIOCGIFCONF_ADDR
cpsize = sizeof ifreq;
#else
cpsize = sizeof ifreq.ifr_name;
if (ioctl(s, SIOCGIFADDR, (char *)&ifreq) < 0) {
ns_notice(ns_log_default,
"get interface addr (%s): %s",
ifreq.ifr_name, strerror(errno));
continue;
}
#endif
if (ifreq.ifr_addr.sa_family != AF_INET) {
ns_debug(ns_log_default, 2,
"getnetconf: %s AF %d != INET",
ifreq.ifr_name, ifreq.ifr_addr.sa_family);
continue;
}
ina = ina_get((u_char *)&((struct sockaddr_in *)
&ifreq.ifr_addr)->sin_addr);
ns_debug(ns_log_default, 1,
"getnetconf: considering %s [%s]",
ifreq.ifr_name, inet_ntoa(ina));
/*
* Don't test IFF_UP, packets may still be received at this
* address if any other interface is up.
*/
if (ina_hlong(ina) == INADDR_ANY) {
ns_debug(ns_log_default, 2,
"getnetconf: INADDR_ANY, ignoring.");
continue;
}
INSIST(server_options != NULL);
INSIST(server_options->listen_list != NULL);
found=0;
for (li = server_options->listen_list->first;
li != NULL;
li = li->next) {
if (ip_match_address(li->list, ina) > 0) {
found++;
/*
* Look for an already existing source
* interface address/port pair.
* This happens mostly when reinitializing.
* Also, if the machine has multiple point to
* point interfaces, then the local address
* may appear more than once.
*/
ifp = if_find(ina, li->port);
if (ifp != NULL) {
ns_debug(ns_log_default, 1,
"dup interface addr [%s].%u (%s)",
inet_ntoa(ina),
ntohs(li->port),
ifreq.ifr_name);
ifp->gen = my_generation;
continue;
}
ifp = (interface *)memget(sizeof *ifp);
if (!ifp)
ns_panic(ns_log_default, 1,
"memget(interface)");
memset(ifp, 0, sizeof *ifp);
INIT_LINK(ifp, link);
APPEND(iflist, ifp, link);
ifp->addr = ina;
ifp->port = li->port;
ifp->gen = my_generation;
ifp->flags = 0;
ifp->dfd = -1;
ifp->sfd = -1;
if (opensocket_d(ifp) < 0 ||
opensocket_s(ifp) < 0) {
dq_remove(ifp);
found = 0;
break;
}
ns_info(ns_log_default,
"listening on [%s].%u (%s)",
inet_ntoa(ina), ntohs(li->port),
ifreq.ifr_name);
}
}
if (!found)
ns_debug(ns_log_default, 1,
"not listening on addr [%s] (%s)",
inet_ntoa(ina), ifreq.ifr_name);
/*
* Add this interface's address to the list of local
* addresses if we haven't added it already.
*/
if (ip_match_address(local_addresses, ina) < 0) {
ime = new_ip_match_pattern(ina, 32);
add_to_ip_match_list(local_addresses, ime);
}
/*
* Get interface flags.
*/
if (ioctl(s, SIOCGIFFLAGS, (char *)&ifreq) < 0) {
ns_notice(ns_log_default, "get interface flags: %s",
strerror(errno));
continue;
}
if ((ifreq.ifr_flags & IFF_POINTOPOINT)) {
/*
* The local network for a PPP link is just the
* two ends of the link, so for each endpoint we
* add a pattern that will only match the endpoint.
*/
if (ioctl(s, SIOCGIFDSTADDR, (char *)&ifreq) < 0) {
ns_notice(ns_log_default, "get dst addr: %s",
strerror(errno));
continue;
}
mask.s_addr = htonl(INADDR_BROADCAST);
/*
* Our end.
*
* Only add it if we haven't seen it before.
*/
if (ip_match_network(local_networks, ina, mask) < 0) {
ime = new_ip_match_pattern(ina, 32);
add_to_ip_match_list(local_networks, ime);
}
/*
* The other end.
*/
ina = ((struct sockaddr_in *)
&ifreq.ifr_addr)->sin_addr;
/*
* Only add it if we haven't seen it before.
*/
if (ip_match_network(local_networks, ina, mask) < 0) {
ime = new_ip_match_pattern(ina, 32);
add_to_ip_match_list(local_networks, ime);
}
} else {
/*
* Add this interface's network and netmask to the
* list of local networks.
*/
#ifdef SIOCGIFNETMASK /* XXX */
if (ioctl(s, SIOCGIFNETMASK, (char *)&ifreq) < 0) {
ns_notice(ns_log_default, "get netmask: %s",
strerror(errno));
continue;
}
/*
* Use ina_get because the ifreq structure might not
* be aligned.
*/
mask_ptr = (u_char *)
&((struct sockaddr_in *)&ifreq.ifr_addr)->sin_addr;
mask = ina_get(mask_ptr);
#else
mask = net_mask(ina);
#endif
ina.s_addr &= mask.s_addr; /* make network address */
/*
* Only add it if we haven't seen it before.
*/
if (ip_match_network(local_networks, ina, mask) < 0) {
ime = new_ip_match_mask(ina, mask);
add_to_ip_match_list(local_networks, ime);
}
}
}
close(s);
memput(buf, bufsiz);
ns_debug(ns_log_default, 7, "local addresses:");
dprint_ip_match_list(ns_log_default, local_addresses, 2, "", "");
ns_debug(ns_log_default, 7, "local networks:");
dprint_ip_match_list(ns_log_default, local_networks, 2, "", "");
/*
* now go through the iflist and delete anything that
* does not have the current generation number. this is
* how we catch interfaces that go away or change their
* addresses. note that 0.0.0.0 is the wildcard element
* and should never be deleted by this code.
*/
dq_remove_gen(my_generation);
if (EMPTY(iflist))
ns_warning(ns_log_default, "not listening on any interfaces");
}
/* opensocket_d(ifp)
* Open datagram socket bound to interface address.
* Returns:
* 0 on success.
* -1 on failure.
*/
static int
opensocket_d(interface *ifp) {
struct sockaddr_in nsa;
const int on = 1;
int m, n;
int fd;
memset(&nsa, 0, sizeof nsa);
nsa.sin_family = AF_INET;
nsa.sin_addr = ifp->addr;
nsa.sin_port = ifp->port;
if ((ifp->dfd = ns_socket(AF_INET, SOCK_DGRAM, 0)) < 0) {
ns_error(ns_log_default, "socket(SOCK_DGRAM): %s",
strerror(errno));
return (-1);
}
if (ifp->dfd > evHighestFD(ev)) {
ns_error(ns_log_default, "socket too high: %d", ifp->dfd);
close(ifp->dfd);
return (-1);
}
if ((n = fcntl(ifp->dfd, F_GETFL, 0)) == -1) {
ns_info(ns_log_default, "fcntl(ifp->dfd, F_GETFL): %s",
strerror(errno));
(void) close(ifp->dfd);
return (-1);
}
if (fcntl(ifp->dfd, F_SETFL, n|PORT_NONBLOCK) == -1) {
ns_info(ns_log_default, "fcntl(ifp->dfd, NONBLOCK): %s",
strerror(errno));
(void) close(ifp->dfd);
return (-1);
}
if (fcntl(ifp->dfd, F_SETFD, 1) < 0) {
ns_error(ns_log_default, "F_SETFD: %s", strerror(errno));
close(ifp->dfd);
return (-1);
}
ns_debug(ns_log_default, 1, "ifp->addr %s d_dfd %d",
sin_ntoa(nsa), ifp->dfd);
if (setsockopt(ifp->dfd, SOL_SOCKET, SO_REUSEADDR,
(char *)&on, sizeof(on)) != 0) {
ns_notice(ns_log_default, "setsockopt(REUSEADDR): %s",
strerror(errno));
/* XXX press on regardless, this is not too serious. */
}
#ifdef SO_RCVBUF /* XXX */
m = sizeof n;
if ((getsockopt(ifp->dfd, SOL_SOCKET, SO_RCVBUF, (char*)&n, &m) >= 0)
&& (m == sizeof n)
&& (n < drbufsize)) {
(void) setsockopt(ifp->dfd, SOL_SOCKET, SO_RCVBUF,
(char *)&drbufsize, sizeof drbufsize);
}
#endif /* SO_RCVBUF */
#ifndef CANNOT_SET_SNDBUF
if (setsockopt(ifp->dfd, SOL_SOCKET, SO_SNDBUF,
(char*)&dsbufsize, sizeof dsbufsize) < 0) {
ns_info(ns_log_default,
"setsockopt(dfd=%d, SO_SNDBUF, %d): %s",
ifp->dfd, dsbufsize, strerror(errno));
/* XXX press on regardless, this is not too serious. */
}
#endif
#ifdef SO_BSDCOMPAT
if (setsockopt(ifp->dfd, SOL_SOCKET, SO_BSDCOMPAT,
(char*)&on, sizeof on) < 0) {
ns_info(ns_log_default,
"setsockopt(dfd=%d, SO_BSDCOMPAT): %s",
ifp->dfd, strerror(errno));
}
#endif
if (bind(ifp->dfd, (struct sockaddr *)&nsa, sizeof nsa)) {
ns_error(ns_log_default, "bind(dfd=%d, %s): %s",
ifp->dfd, sin_ntoa(nsa), strerror(errno));
return (-1);
}
if (evSelectFD(ev, ifp->dfd, EV_READ, datagram_read, ifp,
&ifp->evID_d) == -1) {
ns_error(ns_log_default, "evSelectFD(dfd=%d): %s",
ifp->dfd, strerror(errno));
return (-1);
}
ifp->flags |= INTERFACE_FILE_VALID;
return (0);
}
static int
drain_rcvbuf(evContext ctx, interface *ifp, int fd, int *mread, int *mstore) {
int drop = 0;
drop = 0;
for (; *mread > 0; (*mread)--) {
union {
HEADER h;
u_char buf[PACKETSZ+1];
} u;
struct sockaddr_in from;
int from_len = sizeof from;
savedg *dg;
int n;
n = recvfrom(fd, (char *)u.buf, sizeof u.buf, 0,
(struct sockaddr *)&from, &from_len);
if (n <= 0)
break; /* Socket buffer assumed empty. */
drop++; /* Pessimistic assumption. */
if (n > PACKETSZ)
continue; /* Oversize message - EDNS0 needed. */
if (from.sin_family != AF_INET)
continue; /* Not IPv4 - IPv6 needed. */
if (u.h.opcode == ns_o_query && u.h.qr == 0)
continue; /* Query - what we're here to axe. */
if (*mstore <= 0)
continue; /* Reached storage quota, ignore. */
if ((dg = memget(sizeof *dg)) == NULL)
continue; /* No memory - probably fatal. */
if ((dg->buf = memget(n)) == NULL) {
memput(dg, sizeof *dg);
continue; /* No memory - probably fatal. */
}
dg->from = from;
dg->dfd = fd;
dg->ifp = ifp;
dg->gen = ifp->gen;
dg->buflen = n;
memcpy(dg->buf, u.buf, n);
if (evWaitFor(ctx, (void *)drain_all_rcvbuf, savedg_waitfunc,
dg, NULL) < 0)
{
memput(dg->buf, dg->buflen);
memput(dg, sizeof *dg);
continue; /* No memory - probably fatal. */
}
drop--; /* Pessimism was inappropriate. */
(*mstore)--;
}
return (drop);
}
static int
drain_all_rcvbuf(evContext ctx) {
interface *ifp;
int mread = MAX_SYNCDRAIN;
int mstore = MAX_SYNCSTORE;
int drop = 0;
for (ifp = HEAD(iflist); ifp != NULL; ifp = NEXT(ifp, link))
if (ifp->dfd != -1)
drop += drain_rcvbuf(ctx, ifp, ifp->dfd,
&mread, &mstore);
if (mstore < MAX_SYNCSTORE)
INSIST_ERR(evDo(ctx, (void *)drain_all_rcvbuf) != -1);
return (drop);
}
/* opensocket_s(ifp)
* Open stream (listener) socket bound to interface address.
* Returns:
* 0 on success.
* -1 on failure.
*/
static int
opensocket_s(interface *ifp) {
struct sockaddr_in nsa;
const int on = 1;
int n;
int fd;
memset(&nsa, 0, sizeof nsa);
nsa.sin_family = AF_INET;
nsa.sin_addr = ifp->addr;
nsa.sin_port = ifp->port;
/*
* Open stream (listener) port.
*/
n = 0;
again:
if ((ifp->sfd = ns_socket(AF_INET, SOCK_STREAM, 0)) < 0) {
ns_error(ns_log_default, "socket(SOCK_STREAM): %s",
strerror(errno));
return (-1);
}
if (ifp->sfd > evHighestFD(ev)) {
ns_error(ns_log_default, "socket too high: %d", ifp->sfd);
close(ifp->sfd);
return (-1);
}
if (fcntl(ifp->sfd, F_SETFD, 1) < 0) {
ns_error(ns_log_default, "F_SETFD: %s", strerror(errno));
close(ifp->sfd);
return (-1);
}
if (setsockopt(ifp->sfd, SOL_SOCKET, SO_REUSEADDR,
(char *)&on, sizeof on) != 0) {
ns_notice(ns_log_default, "setsockopt(REUSEADDR): %s",
strerror(errno));
/* Consider that your first warning of trouble to come. */
}
if (bind(ifp->sfd, (struct sockaddr *)&nsa, sizeof nsa) < 0) {
if (errno != EADDRINUSE || ++n > 4) {
if (errno == EADDRINUSE)
ns_error(ns_log_default,
"There may be a name server already running on %s",
sin_ntoa(nsa));
else
ns_error(ns_log_default,
"bind(sfd=%d, %s): %s", ifp->sfd,
sin_ntoa(nsa), strerror(errno));
return (-1);
}
/* Retry opening the socket a few times */
close(ifp->sfd);
ifp->sfd = -1;
sleep(30);
goto again;
}
if (evListen(ev, ifp->sfd, listenmax, stream_accept, ifp, &ifp->evID_s)
== -1) {
ns_error(ns_log_default, "evListen(sfd=%d): %s",
ifp->sfd, strerror(errno));
return (-1);
}
ifp->flags |= INTERFACE_CONN_VALID;
return (0);
}
/* opensocket_f()
* Open datagram socket bound to no particular interface; use for ns_forw
* and sysquery.
*/
void
opensocket_f() {
static struct sockaddr_in prev_qsrc;
static int been_here;
static interface *prev_ifp;
struct sockaddr_in nsa;
const int on = 1;
int n, need_close;
interface *ifp;
need_close = 0;
if (been_here) {
if (prev_ifp != NULL)
prev_ifp->flags &= ~INTERFACE_FORWARDING;
else if (server_options->query_source.sin_port == htons(0) ||
prev_qsrc.sin_addr.s_addr !=
server_options->query_source.sin_addr.s_addr ||
prev_qsrc.sin_port !=
server_options->query_source.sin_port)
need_close = 1;
} else
ds = -1;
been_here = 1;
INSIST(server_options != NULL);
if (need_close) {
evDeselectFD(ev, ds_evID);
close(ds);
ds = -1;
}
/*
* If we're already listening on the query_source address and port,
* we don't need to open another socket. We mark the interface, so
* we'll notice we're in trouble if it goes away.
*/
ifp = if_find(server_options->query_source.sin_addr,
server_options->query_source.sin_port);
if (ifp != NULL) {
ifp->flags |= INTERFACE_FORWARDING;
prev_ifp = ifp;
ds = ifp->dfd;
ns_info(ns_log_default, "forwarding source address is %s",
sin_ntoa(server_options->query_source));
return;
}
/*
* If we're already using the correct query source, we're done.
*/
if (ds >= 0)
return;
prev_qsrc = server_options->query_source;
prev_ifp = NULL;
if ((ds = socket(AF_INET, SOCK_DGRAM, 0)) < 0)
ns_panic(ns_log_default, 1, "socket(SOCK_DGRAM): %s",
strerror(errno));
if (ds > evHighestFD(ev))
ns_panic(ns_log_default, 1, "socket too high: %d", ds);
if (fcntl(ds, F_SETFD, 1) < 0)
ns_panic(ns_log_default, 1, "F_SETFD: %s", strerror(errno));
if (setsockopt(ds, SOL_SOCKET, SO_REUSEADDR,
(char *)&on, sizeof on) != 0) {
ns_notice(ns_log_default, "setsockopt(REUSEADDR): %s",
strerror(errno));
/* XXX press on regardless, this is not too serious. */
}
#ifdef SO_BSDCOMPAT
if (setsockopt(ds, SOL_SOCKET, SO_BSDCOMPAT,
(char *)&on, sizeof on) != 0) {
ns_notice(ns_log_default, "setsockopt(BSDCOMPAT): %s",
strerror(errno));
/* XXX press on regardless, this is not too serious. */
}
#endif
if (bind(ds, (struct sockaddr *)&server_options->query_source,
sizeof server_options->query_source) < 0)
ns_panic(ns_log_default, 0, "opensocket_f: bind(%s): %s",
sin_ntoa(server_options->query_source),
strerror(errno));
n = sizeof nsa;
if (getsockname(ds, (struct sockaddr *)&nsa, &n) < 0)
ns_panic(ns_log_default, 1, "opensocket_f: getsockaddr: %s",
strerror(errno));
ns_debug(ns_log_default, 1, "fwd ds %d addr %s", ds, sin_ntoa(nsa));
ns_info(ns_log_default, "Forwarding source address is %s",
sin_ntoa(nsa));
if (evSelectFD(ev, ds, EV_READ, datagram_read, NULL, &ds_evID) == -1)
ns_panic(ns_log_default, 1, "evSelectFD(fd %d): %s",
ds, strerror(errno));
/* XXX: should probably use a different FileFunc that only accepts
* responses, since requests on this socket make no sense.
*/
}
static void
setdebug(int new_debug) {
#ifdef DEBUG
int old_debug;
if (!new_debug)
ns_debug(ns_log_default, 1, "Debug off");
old_debug = debug;
debug = new_debug;
log_option(log_ctx, LOG_OPTION_DEBUG, debug);
log_option(log_ctx, LOG_OPTION_LEVEL, debug);
evSetDebug(ev, debug, log_get_stream(eventlib_channel));
if (debug) {
if (!old_debug)
open_special_channels();
ns_debug(ns_log_default, 1, "Debug level %d", debug);
if (!old_debug) {
ns_debug(ns_log_default, 1, "Version = %s", Version);
ns_debug(ns_log_default, 1, "conffile = %s", conffile);
}
}
#endif
}
/*
** Routines for managing stream queue
*/
static struct qstream *
sq_add() {
struct qstream *sqp;
if (!(sqp = (struct qstream *)memget(sizeof *sqp))) {
ns_error(ns_log_default, "sq_add: memget: %s",
strerror(errno));
return (NULL);
}
memset(sqp, 0, sizeof *sqp);
ns_debug(ns_log_default, 3, "sq_add(%#lx)", (u_long)sqp);
sqp->flags = 0;
/* XXX should init other fields too? */
sqp->s_next = streamq;
streamq = sqp;
return (sqp);
}
/* sq_remove(qp)
* remove stream queue structure `qp'.
* no current queries may refer to this stream when it is removed.
* side effects:
* memory is deallocated. sockets are closed. lists are relinked.
*/
void
sq_remove(struct qstream *qp) {
struct qstream *qsp;
ns_debug(ns_log_default, 2, "sq_remove(%#lx, %d) rfcnt=%d",
(u_long)qp, qp->s_rfd, qp->s_refcnt);
if (qp->s_wbuf != NULL) {
memput(qp->s_wbuf, qp->s_wbuf_end - qp->s_wbuf);
qp->s_wbuf_send = qp->s_wbuf_free = NULL;
qp->s_wbuf_end = qp->s_wbuf = NULL;
}
if (qp->flags & STREAM_MALLOC)
memput(qp->s_buf, qp->s_bufsize);
if (qp->flags & STREAM_READ_EV)
INSIST_ERR(evCancelRW(ev, qp->evID_r) != -1);
if (qp->flags & STREAM_WRITE_EV)
INSIST_ERR(evDeselectFD(ev, qp->evID_w) != -1);
if (qp->flags & STREAM_CONNECT_EV)
INSIST_ERR(evCancelConn(ev, qp->evID_c) != -1);
if (qp->flags & STREAM_AXFR || qp->flags & STREAM_AXFRIXFR)
ns_freexfr(qp);
(void) close(qp->s_rfd);
if (qp == streamq)
streamq = qp->s_next;
else {
for (qsp = streamq;
qsp && (qsp->s_next != qp);
qsp = qsp->s_next)
(void)NULL;
if (qsp)
qsp->s_next = qp->s_next;
}
memput(qp, sizeof *qp);
}
/* void
* sq_flush(allbut)
* call sq_remove() on all open streams except `allbut'
* side effects:
* global list `streamq' modified
* idiocy:
* is N^2 due to the scan inside of sq_remove()
*/
void
sq_flush(struct qstream *allbut) {
struct qstream *sp, *spnext;
for (sp = streamq; sp != NULL; sp = spnext) {
spnext = sp->s_next;
if (sp != allbut)
sq_remove(sp);
}
}
/* int
* sq_openw(qs, buflen)
* add a write buffer to a stream
* return:
* 0 = success
* -1 = failure (check errno)
*/
int
sq_openw(struct qstream *qs, int buflen) {
#ifdef DO_SO_LINGER /* XXX */
static const struct linger ll = { 1, 120 };
#endif
INSIST(qs->s_wbuf == NULL);
qs->s_wbuf = (u_char *)memget(buflen);
if (qs->s_wbuf == NULL)
return (-1);
qs->s_wbuf_send = qs->s_wbuf;
qs->s_wbuf_free = qs->s_wbuf;
qs->s_wbuf_end = qs->s_wbuf + buflen;
#ifdef DO_SO_LINGER /* XXX */
/* kernels that map pages for IO end up failing if the pipe is full
* at exit and we take away the final buffer. this is really a kernel
* bug but it's harmless on systems that are not broken, so...
*/
setsockopt(qs->s_rfd, SOL_SOCKET, SO_LINGER, (char *)&ll, sizeof ll);
#endif
return (0);
}
/* static void
* sq_dowrite(qs)
* try to submit data to the system, remove it from our queue.
*/
static int
sq_dowrite(struct qstream *qs) {
if (qs->s_wbuf_free > qs->s_wbuf_send) {
int n = write(qs->s_rfd, qs->s_wbuf_send,
qs->s_wbuf_free - qs->s_wbuf_send);
INSIST(qs->s_wbuf != NULL);
if (n < 0) {
if (errno != EINTR && errno != EAGAIN
#if (EWOULDBLOCK != EAGAIN)
&& errno != EWOULDBLOCK
#endif
)
return (-1);
return (0);
}
qs->s_wbuf_send += n;
if (qs->s_wbuf_free > qs->s_wbuf_send) {
/* XXX: need some kind of delay here during which the
* socket will be deselected so we don't spin.
*/
n = qs->s_wbuf_free - qs->s_wbuf_send;
memmove(qs->s_wbuf, qs->s_wbuf_send, n);
qs->s_wbuf_send = qs->s_wbuf;
qs->s_wbuf_free = qs->s_wbuf + n;
}
}
if (qs->s_wbuf_free == qs->s_wbuf_send)
qs->s_wbuf_free = qs->s_wbuf_send = qs->s_wbuf;
return (0);
}
/* void
* sq_flushw(qs)
* called when the socket becomes writable and we want to flush our
* buffers and the system's socket buffers. use as a closure with
* sq_writeh().
*/
void
sq_flushw(struct qstream *qs) {
if (qs->s_wbuf_free == qs->s_wbuf_send) {
sq_writeh(qs, NULL);
sq_done(qs);
}
}
/* static void
* sq_writable(ctx, uap, fd, evmask)
* glue between eventlib closures and qstream closures
*/
static void
sq_writable(evContext ctx, void *uap, int fd, int evmask) {
struct qstream *qs = uap;
INSIST(evmask & EV_WRITE);
INSIST(fd == qs->s_rfd);
if (sq_dowrite(qs) < 0) {
sq_remove(qs);
return;
}
if (qs->s_wbuf_closure
&& qs->s_wbuf_end - qs->s_wbuf_free >= HFIXEDSZ+2) /* XXX guess */
(*qs->s_wbuf_closure)(qs);
if (sq_dowrite(qs) < 0) {
sq_remove(qs);
return;
}
}
/* int
* sq_writeh(qs, closure)
* register a closure to be called when a stream becomes writable
* return:
* 0 = success
* -1 = failure (check errno)
*/
int
sq_writeh(struct qstream *qs, sq_closure c) {
if (c) {
if (!qs->s_wbuf_closure) {
if (evSelectFD(ev, qs->s_rfd, EV_WRITE,
sq_writable, qs, &qs->evID_w) < 0) {
return (-1);
}
qs->flags |= STREAM_WRITE_EV;
}
} else {
(void) evDeselectFD(ev, qs->evID_w);
qs->flags &= ~STREAM_WRITE_EV;
}
qs->s_wbuf_closure = c;
return (0);
}
/* int
* sq_write(qs, buf, len)
* queue a message onto the stream, prepended by a two byte length field
* return:
* 0 = success
* -1 = failure (check errno; E2BIG means we can't handle this right now)
*/
int
sq_write(struct qstream *qs, const u_char *buf, int len) {
INSIST(qs->s_wbuf != NULL);
if (NS_INT16SZ + len > qs->s_wbuf_end - qs->s_wbuf_free) {
if (sq_dowrite(qs) < 0)
return (-1);
if (NS_INT16SZ + len > qs->s_wbuf_end - qs->s_wbuf_free) {
errno = E2BIG;
return (-1);
}
}
__putshort(len, qs->s_wbuf_free);
qs->s_wbuf_free += NS_INT16SZ;
memcpy(qs->s_wbuf_free, buf, len);
qs->s_wbuf_free += len;
return (0);
}
/*
* Initiate query on stream;
* mark as referenced and stop selecting for input.
*/
static void
sq_query(struct qstream *sp) {
sp->s_refcnt++;
}
/*
* Note that the current request on a stream has completed,
* and that we should continue looking for requests on the stream.
*/
void
sq_done(struct qstream *sp) {
struct iovec iov;
if (sp->s_wbuf != NULL) {
INSIST(sp->s_wbuf_send == sp->s_wbuf_free);
memput(sp->s_wbuf, sp->s_wbuf_end - sp->s_wbuf);
sp->s_wbuf_send = sp->s_wbuf_free = NULL;
sp->s_wbuf_end = sp->s_wbuf = NULL;
}
if (sp->flags & STREAM_AXFR || sp->flags & STREAM_AXFRIXFR)
ns_freexfr(sp);
sp->s_refcnt = 0;
sp->s_time = tt.tv_sec;
if (sp->flags & STREAM_DONE_CLOSE) {
/* XXX */
sq_remove(sp);
return;
}
iov = evConsIovec(sp->s_temp, INT16SZ);
if (evRead(ev, sp->s_rfd, &iov, 1, stream_getlen, sp, &sp->evID_r) ==
-1)
ns_panic(ns_log_default, 1, "evRead(fd %d): %s",
sp->s_rfd, strerror(errno));
sp->flags |= STREAM_READ_EV;
}
/* void
* dq_remove_gen(gen)
* close/deallocate all the udp sockets (except 0.0.0.0) which are
* not from the current generation.
* side effects:
* global list `iflist' is modified.
*/
void
dq_remove_gen(time_t gen) {
interface *this, *next;
for (this = HEAD(iflist); this != NULL; this = next) {
next = NEXT(this, link);
if (this->gen != gen && ina_hlong(this->addr) != INADDR_ANY)
dq_remove(this);
}
}
/* void
* dq_remove_all()
* close/deallocate all interfaces.
* side effects:
* global list `iflist' is modified.
*/
void
dq_remove_all() {
interface *this, *next;
for (this = HEAD(iflist); this != NULL; this = next) {
next = NEXT(this, link);
/*
* Clear the forwarding flag so we don't panic the server.
*/
this->flags &= ~INTERFACE_FORWARDING;
dq_remove(this);
}
}
/* void
* dq_remove(interface *this)
* close/deallocate an interface's sockets. called on errors
* or if the interface disappears.
* side effects:
* global list `iflist' is modified.
*/
static void
dq_remove(interface *this) {
ns_notice(ns_log_default, "deleting interface [%s].%u",
inet_ntoa(this->addr), ntohs(this->port));
if ((this->flags & INTERFACE_FORWARDING) != 0)
ns_panic(ns_log_default, 0,
"forwarding interface [%s].%u gone",
inet_ntoa(this->addr),
ntohs(this->port));
/* Deallocate fields. */
if ((this->flags & INTERFACE_FILE_VALID) != 0)
(void) evDeselectFD(ev, this->evID_d);
if (this->dfd >= 0)
(void) close(this->dfd);
if ((this->flags & INTERFACE_CONN_VALID) != 0)
(void) evCancelConn(ev, this->evID_s);
if (this->sfd >= 0)
(void) close(this->sfd);
UNLINK(iflist, this, link);
memput(this, sizeof *this);
}
/* struct in_addr
* net_mask(ina)
* makes a classful assumption in a classless world, and returns it.
*/
struct in_addr
net_mask(struct in_addr ina) {
u_long hl = ina_hlong(ina);
struct in_addr ret;
if (IN_CLASSA(hl))
hl = IN_CLASSA_NET;
else if (IN_CLASSB(hl))
hl = IN_CLASSB_NET;
else if (IN_CLASSC(hl))
hl = IN_CLASSC_NET;
else
hl = INADDR_BROADCAST;
ina_ulong(ret) = htonl(hl);
return (ret);
}
/* aIsUs(addr)
* scan our list of interface addresses for "addr".
* returns:
* 0: address isn't one of our interfaces
* >0: address is one of our interfaces, or INADDR_ANY
*/
int
aIsUs(struct in_addr addr) {
if (ina_hlong(addr) == INADDR_ANY || if_find(addr, 0) != NULL)
return (1);
return (0);
}
/* interface *
* if_find(addr, port)
* scan our list of interface addresses for "addr" and port.
* port == 0 means match any port
* returns:
* pointer to interface with this address/port, or NULL if there isn't
* one.
*/
static interface *
if_find(struct in_addr addr, u_int16_t port) {
interface *ifp;
for (ifp = HEAD(iflist); ifp != NULL; ifp = NEXT(ifp, link))
if (ina_equal(addr, ifp->addr))
if (port == 0 || ifp->port == port)
break;
return (ifp);
}
/*
* These are here in case we ever want to get more clever, like perhaps
* using a bitmap to keep track of outstanding queries and a random
* allocation scheme to make it a little harder to predict them. Note
* that the resolver will need the same protection so the cleverness
* should be put there rather than here; this is just an interface layer.
*
* This is true but ... most clients only send out a few queries, they
* use varying port numbers, and the queries aren't sent to the outside
* world which we know is full of spoofers. Doing a good job of randomizing
* ids may also be to expensive for each client. Queries forwarded by the
* server always come from the same port (unless you let 8.x pick a port
* and restart it periodically - maybe it should open several and use
* them randomly). The server sends out lots more queries, and if it's
* cache is corrupted, it has the potential to affect more clients.
* NOTE: - randomizing the ID or source port doesn't help a bit if the
* queries can be sniffed.
* -- DL
*/
/*
* Allow the user to pick one of two ID randomization algorithms.
*
* The first algorithm is an adaptation of the sequence shuffling
* algorithm discovered by Carter Bays and S. D. Durham [ACM Trans. Math.
* Software 2 (1976), 59-64], as documented as Algorithm B in Chapter
* 3.2.2 in Volume 2 of Knuth's "The Art of Computer Programming". We use
* a randomly selected linear congruential random number generator with a
* modulus of 2^16, whose increment is a randomly picked odd number, and
* whose multiplier is picked from a set which meets the following
* criteria:
* Is of the form 8*n+5, which ensures "high potency" according to
* principle iii in the summary chapter 3.6. This form also has a
* gcd(a-1,m) of 4 which is good according to principle iv.
*
* Is between 0.01 and 0.99 times the modulus as specified by
* principle iv.
*
* Passes the spectral test "with flying colors" (ut >= 1) in
* dimensions 2 through 6 as calculated by Algorithm S in Chapter
* 3.3.4 and the ratings calculated by formula 35 in section E.
*
* Of the multipliers that pass this test, pick the set that is
* best according to the theoretical bounds of the serial
* correlation test. This was calculated using a simplified
* version of Knuth's Theorem K in Chapter 3.3.3.
*
* These criteria may not be important for this use, but we might as well
* pick from the best generators since there are so many possible ones and
* we don't have that many random bits to do the picking.
*
* We use a modulus of 2^16 instead of something bigger so that we will
* tend to cycle through all the possible IDs before repeating any,
* however the shuffling will perturb this somewhat. Theoretically there
* is no minimimum interval between two uses of the same ID, but in
* practice it seems to be >64000.
*
* Our adaptatation of Algorithm B mixes the hash state which has
* captured various random events into the shuffler to perturb the
* sequence.
*
* One disadvantage of this algorithm is that if the generator parameters
* were to be guessed, it would be possible to mount a limited brute force
* attack on the ID space since the IDs are only shuffled within a limited
* range.
*
* The second algorithm uses the same random number generator to populate
* a pool of 65536 IDs. The hash state is used to pick an ID from a window
* of 4096 IDs in this pool, then the chosen ID is swapped with the ID
* at the beginning of the window and the window position is advanced.
* This means that the interval between uses of the ID will be no less
* than 65536-4096. The ID sequence in the pool will become more random
* over time.
*
* For both algorithms, two more linear congruential random number generators
* are selected. The ID from the first part of algorithm is used to seed
* the first of these generators, and its output is used to seed the second.
* The strategy is use these generators as 1 to 1 hashes to obfuscate the
* properties of the generator used in the first part of either algorithm.
*
* The first algorithm may be suitable for use in a client resolver since
* its memory requirements are fairly low and it's pretty random out of
* the box. It is somewhat succeptible to a limited brute force attack,
* so the second algorithm is probably preferable for a longer running
* program that issues a large number of queries and has time to randomize
* the pool.
*/
#define NSID_SHUFFLE_TABLE_SIZE 100 /* Suggested by Knuth */
/*
* Pick one of the next 4096 IDs in the pool.
* There is a tradeoff here between randomness and how often and ID is reused.
*/
#define NSID_LOOKAHEAD 4096 /* Must be a power of 2 */
#define NSID_SHUFFLE_ONLY 1 /* algorithm 1 */
#define NSID_USE_POOL 2 /* algorithm 2 */
/*
* Keep a running hash of various bits of data that we'll use to
* stir the ID pool or perturb the ID generator
*/
void
nsid_hash(u_char *data, size_t len) {
/*
* Hash function similar to the one we use for hashing names.
* We don't fold case or toss the upper bit here, though.
* This hash doesn't do much interesting when fed binary zeros,
* so there may be a better hash function.
* This function doesn't need to be very strong since we're
* only using it to stir the pool, but it should be reasonably
* fast.
*/
while (len-- > 0) {
HASHROTATE(nsid_hash_state);
nsid_hash_state += *data++;
}
}
/*
* Table of good linear congruential multipliers for modulus 2^16
* in order of increasing serial correlation bounds (so trim from
* the end).
*/
static const u_int16_t nsid_multiplier_table[] = {
17565, 25013, 11733, 19877, 23989, 23997, 24997, 25421,
26781, 27413, 35901, 35917, 35973, 36229, 38317, 38437,
39941, 40493, 41853, 46317, 50581, 51429, 53453, 53805,
11317, 11789, 12045, 12413, 14277, 14821, 14917, 18989,
19821, 23005, 23533, 23573, 23693, 27549, 27709, 28461,
29365, 35605, 37693, 37757, 38309, 41285, 45261, 47061,
47269, 48133, 48597, 50277, 50717, 50757, 50805, 51341,
51413, 51581, 51597, 53445, 11493, 14229, 20365, 20653,
23485, 25541, 27429, 29421, 30173, 35445, 35653, 36789,
36797, 37109, 37157, 37669, 38661, 39773, 40397, 41837,
41877, 45293, 47277, 47845, 49853, 51085, 51349, 54085,
56933, 8877, 8973, 9885, 11365, 11813, 13581, 13589,
13613, 14109, 14317, 15765, 15789, 16925, 17069, 17205,
17621, 17941, 19077, 19381, 20245, 22845, 23733, 24869,
25453, 27213, 28381, 28965, 29245, 29997, 30733, 30901,
34877, 35485, 35613, 36133, 36661, 36917, 38597, 40285,
40693, 41413, 41541, 41637, 42053, 42349, 45245, 45469,
46493, 48205, 48613, 50861, 51861, 52877, 53933, 54397,
55669, 56453, 56965, 58021, 7757, 7781, 8333, 9661,
12229, 14373, 14453, 17549, 18141, 19085, 20773, 23701,
24205, 24333, 25261, 25317, 27181, 30117, 30477, 34757,
34885, 35565, 35885, 36541, 37957, 39733, 39813, 41157,
41893, 42317, 46621, 48117, 48181, 49525, 55261, 55389,
56845, 7045, 7749, 7965, 8469, 9133, 9549, 9789,
10173, 11181, 11285, 12253, 13453, 13533, 13757, 14477,
15053, 16901, 17213, 17269, 17525, 17629, 18605, 19013,
19829, 19933, 20069, 20093, 23261, 23333, 24949, 25309,
27613, 28453, 28709, 29301, 29541, 34165, 34413, 37301,
37773, 38045, 38405, 41077, 41781, 41925, 42717, 44437,
44525, 44613, 45933, 45941, 47077, 50077, 50893, 52117,
5293, 55069, 55989, 58125, 59205, 6869, 14685, 15453,
16821, 17045, 17613, 18437, 21029, 22773, 22909, 25445,
25757, 26541, 30709, 30909, 31093, 31149, 37069, 37725,
37925, 38949, 39637, 39701, 40765, 40861, 42965, 44813,
45077, 45733, 47045, 50093, 52861, 52957, 54181, 56325,
56365, 56381, 56877, 57013, 5741, 58101, 58669, 8613,
10045, 10261, 10653, 10733, 11461, 12261, 14069, 15877,
17757, 21165, 23885, 24701, 26429, 26645, 27925, 28765,
29197, 30189, 31293, 39781, 39909, 40365, 41229, 41453,
41653, 42165, 42365, 47421, 48029, 48085, 52773, 5573,
57037, 57637, 58341, 58357, 58901, 6357, 7789, 9093,
10125, 10709, 10765, 11957, 12469, 13437, 13509, 14773,
15437, 15773, 17813, 18829, 19565, 20237, 23461, 23685,
23725, 23941, 24877, 25461, 26405, 29509, 30285, 35181,
37229, 37893, 38565, 40293, 44189, 44581, 45701, 47381,
47589, 48557, 4941, 51069, 5165, 52797, 53149, 5341,
56301, 56765, 58581, 59493, 59677, 6085, 6349, 8293,
8501, 8517, 11597, 11709, 12589, 12693, 13517, 14909,
17397, 18085, 21101, 21269, 22717, 25237, 25661, 29189,
30101, 31397, 33933, 34213, 34661, 35533, 36493, 37309,
40037, 4189, 42909, 44309, 44357, 44389, 4541, 45461,
46445, 48237, 54149, 55301, 55853, 56621, 56717, 56901,
5813, 58437, 12493, 15365, 15989, 17829, 18229, 19341,
21013, 21357, 22925, 24885, 26053, 27581, 28221, 28485,
30605, 30613, 30789, 35437, 36285, 37189, 3941, 41797,
4269, 42901, 43293, 44645, 45221, 46893, 4893, 50301,
50325, 5189, 52109, 53517, 54053, 54485, 5525, 55949,
56973, 59069, 59421, 60733, 61253, 6421, 6701, 6709,
7101, 8669, 15797, 19221, 19837, 20133, 20957, 21293,
21461, 22461, 29085, 29861, 30869, 34973, 36469, 37565,
38125, 38829, 39469, 40061, 40117, 44093, 47429, 48341,
50597, 51757, 5541, 57629, 58405, 59621, 59693, 59701,
61837, 7061, 10421, 11949, 15405, 20861, 25397, 25509,
25893, 26037, 28629, 28869, 29605, 30213, 34205, 35637,
36365, 37285, 3773, 39117, 4021, 41061, 42653, 44509,
4461, 44829, 4725, 5125, 52269, 56469, 59085, 5917,
60973, 8349, 17725, 18637, 19773, 20293, 21453, 22533,
24285, 26333, 26997, 31501, 34541, 34805, 37509, 38477,
41333, 44125, 46285, 46997, 47637, 48173, 4925, 50253,
50381, 50917, 51205, 51325, 52165, 52229, 5253, 5269,
53509, 56253, 56341, 5821, 58373, 60301, 61653, 61973,
62373, 8397, 11981, 14341, 14509, 15077, 22261, 22429,
24261, 28165, 28685, 30661, 34021, 34445, 39149, 3917,
43013, 43317, 44053, 44101, 4533, 49541, 49981, 5277,
54477, 56357, 57261, 57765, 58573, 59061, 60197, 61197,
62189, 7725, 8477, 9565, 10229, 11437, 14613, 14709,
16813, 20029, 20677, 31445, 3165, 31957, 3229, 33541,
36645, 3805, 38973, 3965, 4029, 44293, 44557, 46245,
48917, 4909, 51749, 53709, 55733, 56445, 5925, 6093,
61053, 62637, 8661, 9109, 10821, 11389, 13813, 14325,
15501, 16149, 18845, 22669, 26437, 29869, 31837, 33709,
33973, 34173, 3677, 3877, 3981, 39885, 42117, 4421,
44221, 44245, 44693, 46157, 47309, 5005, 51461, 52037,
55333, 55693, 56277, 58949, 6205, 62141, 62469, 6293,
10101, 12509, 14029, 17997, 20469, 21149, 25221, 27109,
2773, 2877, 29405, 31493, 31645, 4077, 42005, 42077,
42469, 42501, 44013, 48653, 49349, 4997, 50101, 55405,
56957, 58037, 59429, 60749, 61797, 62381, 62837, 6605,
10541, 23981, 24533, 2701, 27333, 27341, 31197, 33805,
3621, 37381, 3749, 3829, 38533, 42613, 44381, 45901,
48517, 51269, 57725, 59461, 60045, 62029, 13805, 14013,
15461, 16069, 16157, 18573, 2309, 23501, 28645, 3077,
31541, 36357, 36877, 3789, 39429, 39805, 47685, 47949,
49413, 5485, 56757, 57549, 57805, 58317, 59549, 62213,
62613, 62853, 62933, 8909, 12941, 16677, 20333, 21541,
24429, 26077, 26421, 2885, 31269, 33381, 3661, 40925,
42925, 45173, 4525, 4709, 53133, 55941, 57413, 57797,
62125, 62237, 62733, 6773, 12317, 13197, 16533, 16933,
18245, 2213, 2477, 29757, 33293, 35517, 40133, 40749,
4661, 49941, 62757, 7853, 8149, 8573, 11029, 13421,
21549, 22709, 22725, 24629, 2469, 26125, 2669, 34253,
36709, 41013, 45597, 46637, 52285, 52333, 54685, 59013,
60997, 61189, 61981, 62605, 62821, 7077, 7525, 8781,
10861, 15277, 2205, 22077, 28517, 28949, 32109, 33493,
3685, 39197, 39869, 42621, 44997, 48565, 5221, 57381,
61749, 62317, 63245, 63381, 23149, 2549, 28661, 31653,
33885, 36341, 37053, 39517, 42805, 45853, 48997, 59349,
60053, 62509, 63069, 6525, 1893, 20181, 2365, 24893,
27397, 31357, 32277, 33357, 34437, 36677, 37661, 43469,
43917, 50997, 53869, 5653, 13221, 16741, 17893, 2157,
28653, 31789, 35301, 35821, 61613, 62245, 12405, 14517,
17453, 18421, 3149, 3205, 40341, 4109, 43941, 46869,
48837, 50621, 57405, 60509, 62877, 8157, 12933, 12957,
16501, 19533, 3461, 36829, 52357, 58189, 58293, 63053,
17109, 1933, 32157, 37701, 59005, 61621, 13029, 15085,
16493, 32317, 35093, 5061, 51557, 62221, 20765, 24613,
2629, 30861, 33197, 33749, 35365, 37933, 40317, 48045,
56229, 61157, 63797, 7917, 17965, 1917, 1973, 20301,
2253, 33157, 58629, 59861, 61085, 63909, 8141, 9221,
14757, 1581, 21637, 26557, 33869, 34285, 35733, 40933,
42517, 43501, 53653, 61885, 63805, 7141, 21653, 54973,
31189, 60061, 60341, 63357, 16045, 2053, 26069, 33997,
43901, 54565, 63837, 8949, 17909, 18693, 32349, 33125,
37293, 48821, 49053, 51309, 64037, 7117, 1445, 20405,
23085, 26269, 26293, 27349, 32381, 33141, 34525, 36461,
37581, 43525, 4357, 43877, 5069, 55197, 63965, 9845,
12093, 2197, 2229, 32165, 33469, 40981, 42397, 8749,
10853, 1453, 18069, 21693, 30573, 36261, 37421, 42533
};
#define NSID_MULT_TABLE_SIZE \
((sizeof nsid_multiplier_table)/(sizeof nsid_multiplier_table[0]))
void
nsid_init(void) {
struct timeval now;
pid_t mypid;
u_int16_t a1ndx, a2ndx, a3ndx, c1ndx, c2ndx, c3ndx;
int i;
if (nsid_algorithm != 0)
return;
gettimeofday(&now, NULL);
mypid = getpid();
/* Initialize the state */
nsid_hash_state = 0;
nsid_hash((u_char *)&now, sizeof now);
nsid_hash((u_char *)&mypid, sizeof mypid);
/*
* Select our random number generators and initial seed.
* We could really use more random bits at this point,
* but we'll try to make a silk purse out of a sows ear ...
*/
/* generator 1 */
a1ndx = ((u_long) NSID_MULT_TABLE_SIZE *
(nsid_hash_state & 0xFFFF)) >> 16;
nsid_a1 = nsid_multiplier_table[a1ndx];
c1ndx = (nsid_hash_state >> 9) & 0x7FFF;
nsid_c1 = 2*c1ndx + 1;
/* generator 2, distinct from 1 */
a2ndx = ((u_long) (NSID_MULT_TABLE_SIZE - 1) *
((nsid_hash_state >> 10) & 0xFFFF)) >> 16;
if (a2ndx >= a1ndx)
a2ndx++;
nsid_a2 = nsid_multiplier_table[a2ndx];
c2ndx = nsid_hash_state % 32767;
if (c2ndx >= c1ndx)
c2ndx++;
nsid_c2 = 2*c2ndx + 1;
/* generator 3, distinct from 1 and 2 */
a3ndx = ((u_long) (NSID_MULT_TABLE_SIZE - 2) *
((nsid_hash_state >> 20) & 0xFFFF)) >> 16;
if (a3ndx >= a1ndx || a3ndx >= a2ndx)
a3ndx++;
if (a3ndx >= a1ndx && a3ndx >= a2ndx)
a3ndx++;
nsid_a3 = nsid_multiplier_table[a3ndx];
c3ndx = nsid_hash_state % 32766;
if (c3ndx >= c1ndx || c3ndx >= c2ndx)
c3ndx++;
if (c3ndx >= c1ndx && c3ndx >= c2ndx)
c3ndx++;
nsid_c3 = 2*c3ndx + 1;
nsid_state = ((nsid_hash_state >> 16) ^ (nsid_hash_state)) & 0xFFFF;
/* Do the algorithm specific initialization */
INSIST(server_options != NULL);
if (NS_OPTION_P(OPTION_USE_ID_POOL) == 0) {
/* Algorithm 1 */
nsid_algorithm = NSID_SHUFFLE_ONLY;
nsid_vtable = memget(NSID_SHUFFLE_TABLE_SIZE *
(sizeof(u_int16_t)) );
if (!nsid_vtable)
ns_panic(ns_log_default, 1, "memget(nsid_vtable)");
for (i = 0; i < NSID_SHUFFLE_TABLE_SIZE; i++) {
nsid_vtable[i] = nsid_state;
nsid_state = (((u_long) nsid_a1 * nsid_state) + nsid_c1)
& 0xFFFF;
}
nsid_state2 = nsid_state;
} else {
/* Algorithm 2 */
nsid_algorithm = NSID_USE_POOL;
nsid_pool = memget(0x10000 * (sizeof(u_int16_t)));
if (!nsid_pool)
ns_panic(ns_log_default, 1, "memget(nsid_pool)");
for (i = 0; ; i++) {
nsid_pool[i] = nsid_state;
nsid_state = (((u_long) nsid_a1 * nsid_state) + nsid_c1) & 0xFFFF;
if (i == 0xFFFF)
break;
}
}
}
#define NSID_RANGE_MASK (NSID_LOOKAHEAD - 1)
#define NSID_POOL_MASK 0xFFFF /* used to wrap the pool index */
u_int16_t
nsid_next() {
u_int16_t id, compressed_hash;
compressed_hash = ((nsid_hash_state >> 16) ^ (nsid_hash_state)) &
0xFFFF;
if (nsid_algorithm == NSID_SHUFFLE_ONLY) {
u_int16_t j;
/*
* This is the original Algorithm B
* j = ((u_long) NSID_SHUFFLE_TABLE_SIZE * nsid_state2)
* >> 16;
*
* We'll perturb it with some random stuff ...
*/
j = ((u_long) NSID_SHUFFLE_TABLE_SIZE *
(nsid_state2 ^ compressed_hash)) >> 16;
nsid_state2 = id = nsid_vtable[j];
nsid_state = (((u_long) nsid_a1 * nsid_state) + nsid_c1) &
0xFFFF;
nsid_vtable[j] = nsid_state;
} else if (nsid_algorithm == NSID_USE_POOL) {
u_int16_t pick;
pick = compressed_hash & NSID_RANGE_MASK;
id = nsid_pool[(nsid_state + pick) & NSID_POOL_MASK];
if (pick != 0) {
/* Swap two IDs to stir the pool */
nsid_pool[(nsid_state + pick) & NSID_POOL_MASK] =
nsid_pool[nsid_state];
nsid_pool[nsid_state] = id;
}
/* increment the base pointer into the pool */
if (nsid_state == 65535)
nsid_state = 0;
else
nsid_state++;
} else
ns_panic(ns_log_default, 1, "Unknown ID algorithm");
/* Now lets obfuscate ... */
id = (((u_long) nsid_a2 * id) + nsid_c2) & 0xFFFF;
id = (((u_long) nsid_a3 * id) + nsid_c3) & 0xFFFF;
return (id);
}
/* Note: this function CAN'T deallocate the saved_argv[]. */
static void
deallocate_everything(void) {
FILE *f;
f = write_open(server_options->memstats_filename);
ns_freestats();
qflush();
sq_flush(NULL);
free_addinfo();
ns_shutdown();
dq_remove_all();
db_lame_destroy();
if (local_addresses != NULL)
free_ip_match_list(local_addresses);
if (local_networks != NULL)
free_ip_match_list(local_networks);
destroyservicelist();
destroyprotolist();
shutdown_logging();
evDestroy(ev);
if (conffile != NULL)
freestr(conffile);
conffile = NULL;
if (debugfile != NULL)
freestr(debugfile);
debugfile = NULL;
if (user_name != NULL)
freestr(user_name);
user_name = NULL;
if (group_name != NULL)
freestr(group_name);
group_name = NULL;
if (chroot_dir != NULL)
freestr(chroot_dir);
chroot_dir = NULL;
if (nsid_pool != NULL)
memput(nsid_pool, 0x10000 * (sizeof(u_int16_t)));
nsid_pool = NULL;
irs_destroy();
if (f != NULL) {
memstats(f);
(void)fclose(f);
}
}
static void
ns_restart(void) {
ns_info(ns_log_default, "named restarting");
#ifdef BIND_UPDATE
dynamic_about_to_exit();
#endif
if (server_options && server_options->pid_filename)
(void)unlink(server_options->pid_filename);
ns_logstats(ev, NULL, evNowTime(), evConsTime(0, 0));
if (NS_OPTION_P(OPTION_DEALLOC_ON_EXIT))
deallocate_everything();
else
shutdown_configuration();
execvp(saved_argv[0], saved_argv);
abort();
}
static void
use_desired_debug(void) {
#ifdef DEBUG
sigset_t set;
/* Protect against race conditions by blocking debugging signals. */
if (sigemptyset(&set) < 0) {
ns_error(ns_log_os,
"sigemptyset failed in use_desired_debug: %s",
strerror(errno));
return;
}
if (sigaddset(&set, SIGUSR1) < 0) {
ns_error(ns_log_os,
"sigaddset SIGUSR1 failed in use_desired_debug: %s",
strerror(errno));
return;
}
if (sigaddset(&set, SIGUSR2) < 0) {
ns_error(ns_log_os,
"sigaddset SIGUSR2 failed in use_desired_debug: %s",
strerror(errno));
return;
}
if (sigprocmask(SIG_BLOCK, &set, NULL) < 0) {
ns_error(ns_log_os,
"sigprocmask to block USR1 and USR2 failed: %s",
strerror(errno));
return;
}
setdebug(desired_debug);
if (sigprocmask(SIG_UNBLOCK, &set, NULL) < 0)
ns_error(ns_log_os,
"sigprocmask to unblock USR1 and USR2 failed: %s",
strerror(errno));
#endif
}
void
toggle_qrylog(void) {
qrylog = !qrylog;
ns_notice(ns_log_default, "query log %s\n", qrylog ?"on" :"off");
}
static void
wild(void) {
ns_panic(ns_log_default, 1, "wild need");
}
/*
* This is a functional interface to the global needs and options.
*/
static void
init_needs(void) {
int need;
for (need = 0; need < main_need_num; need++)
handlers[need] = wild;
handlers[main_need_zreload] = ns_zreload;
handlers[main_need_reload] = ns_reload;
handlers[main_need_reconfig] = ns_reconfig;
handlers[main_need_endxfer] = endxfer;
handlers[main_need_zoneload] = loadxfer;
handlers[main_need_dump] = doadump;
handlers[main_need_statsdump] = ns_stats;
handlers[main_need_statsdumpandclear] = ns_stats_dumpandclear;
handlers[main_need_exit] = exit_handler;
handlers[main_need_qrylog] = toggle_qrylog;
handlers[main_need_debug] = use_desired_debug;
handlers[main_need_restart] = ns_restart;
handlers[main_need_reap] = reapchild;
handlers[main_need_noexpired] = ns_noexpired;
}
static void
handle_needs(void) {
int need, queued = 0;
ns_debug(ns_log_default, 15, "handle_needs()");
block_signals();
for (need = 0; need < main_need_num; need++)
if ((needs & (1 << need)) != 0) {
INSIST_ERR(evWaitFor(ev, (void *)handle_needs,
need_waitfunc,
(void *)handlers[need],
NULL) != -1);
queued++;
}
needs = 0;
unblock_signals();
ns_debug(ns_log_default, 15, "handle_needs(): queued %d", queued);
if (queued != 0) {
INSIST_ERR(evDo(ev, (void *)handle_needs) != -1);
return;
}
ns_panic(ns_log_default, 1, "ns_handle_needs: queued == 0", NULL);
}
static void
need_waitfunc(evContext ctx, void *uap, const void *tag) {
handler hand = (handler) uap;
time_t begin;
long syncdelay;
begin = time(NULL);
(*hand)();
syncdelay = time(NULL) - begin;
if (syncdelay > MAX_SYNCDELAY)
ns_notice(ns_log_default, "drained %d queries (delay %ld sec)",
drain_all_rcvbuf(ctx), syncdelay);
}
void
ns_need(enum need need) {
block_signals();
ns_need_unsafe(need);
unblock_signals();
}
/* Note: this function should only be called with signals blocked. */
void
ns_need_unsafe(enum need need) {
needs |= (1 << need);
}
static void
exit_handler(void) {
needs_exit = 1;
}
void
ns_setoption(int option) {
ns_warning(ns_log_default, "used obsolete ns_setoption(%d)", option);
}
void
writestream(struct qstream *sp, const u_char *msg, int msglen) {
if (sq_openw(sp, msglen + INT16SZ) == -1) {
sq_remove(sp);
return;
}
if (sq_write(sp, msg, msglen) == -1) {
sq_remove(sp);
return;
}
sq_writeh(sp, sq_flushw);
}
static int
only_digits(const char *s) {
if (*s == '\0')
return (0);
while (*s != '\0') {
if (!isdigit(*s))
return (0);
s++;
}
return (1);
}
#if defined(__GNUC__) && defined(__BOUNDS_CHECKING_ON)
/* Use bounds checking malloc, etc. */
void *
memget(size_t len) {
return (malloc(len));
}
void
memput(void *addr, size_t len) {
free(addr);
}
int
meminit(size_t init_max_size, size_t target_size) {
return (0);
}
void *
memget_debug(size_t size, const char *file, int line) {
void *ptr;
ptr = __memget(size);
fprintf(stderr, "%s:%d: memget(%lu) -> %p\n", file, line,
(u_long)size, ptr);
return (ptr);
}
void
memput_debug(void *ptr, size_t size, const char *file, int line) {
fprintf(stderr, "%s:%d: memput(%p, %lu)\n", file, line, ptr,
(u_long)size);
__memput(ptr, size);
}
void
memstats(FILE *out) {
fputs("No memstats\n", out);
}
#endif
#ifndef HAVE_CUSTOM
/* Standard implementation has nothing here */
static void
custom_init(void) {
/* Noop. */
}
static void
custom_shutdown(void) {
/* Noop. */
}
#endif
Reference in New Issue View Git Blame Copy Permalink