xref: /netbsd-src/external/bsd/nsd/dist/xfrd-tcp.c (revision bdc22b2e01993381dcefeff2bc9b56ca75a4235c)
1 /*
2  * xfrd-tcp.c - XFR (transfer) Daemon TCP system source file. Manages tcp conn.
3  *
4  * Copyright (c) 2001-2006, NLnet Labs. All rights reserved.
5  *
6  * See LICENSE for the license.
7  *
8  */
9 
10 #include "config.h"
11 #include <assert.h>
12 #include <errno.h>
13 #include <fcntl.h>
14 #include <unistd.h>
15 #include <stdlib.h>
16 #include <sys/uio.h>
17 #include "nsd.h"
18 #include "xfrd-tcp.h"
19 #include "buffer.h"
20 #include "packet.h"
21 #include "dname.h"
22 #include "options.h"
23 #include "namedb.h"
24 #include "xfrd.h"
25 #include "xfrd-disk.h"
26 #include "util.h"
27 
28 /* sort tcppipe, first on IP address, for an IPaddresss, sort on num_unused */
29 static int
30 xfrd_pipe_cmp(const void* a, const void* b)
31 {
32 	const struct xfrd_tcp_pipeline* x = (struct xfrd_tcp_pipeline*)a;
33 	const struct xfrd_tcp_pipeline* y = (struct xfrd_tcp_pipeline*)b;
34 	int r;
35 	if(x == y)
36 		return 0;
37 	if(y->ip_len != x->ip_len)
38 		/* subtraction works because nonnegative and small numbers */
39 		return (int)y->ip_len - (int)x->ip_len;
40 	r = memcmp(&x->ip, &y->ip, x->ip_len);
41 	if(r != 0)
42 		return r;
43 	/* sort that num_unused is sorted ascending, */
44 	if(x->num_unused != y->num_unused) {
45 		return (x->num_unused < y->num_unused) ? -1 : 1;
46 	}
47 	/* different pipelines are different still, even with same numunused*/
48 	return (uintptr_t)x < (uintptr_t)y ? -1 : 1;
49 }
50 
51 struct xfrd_tcp_set* xfrd_tcp_set_create(struct region* region)
52 {
53 	int i;
54 	struct xfrd_tcp_set* tcp_set = region_alloc(region,
55 		sizeof(struct xfrd_tcp_set));
56 	memset(tcp_set, 0, sizeof(struct xfrd_tcp_set));
57 	tcp_set->tcp_count = 0;
58 	tcp_set->tcp_waiting_first = 0;
59 	tcp_set->tcp_waiting_last = 0;
60 	for(i=0; i<XFRD_MAX_TCP; i++)
61 		tcp_set->tcp_state[i] = xfrd_tcp_pipeline_create(region);
62 	tcp_set->pipetree = rbtree_create(region, &xfrd_pipe_cmp);
63 	return tcp_set;
64 }
65 
66 struct xfrd_tcp_pipeline*
67 xfrd_tcp_pipeline_create(region_type* region)
68 {
69 	int i;
70 	struct xfrd_tcp_pipeline* tp = (struct xfrd_tcp_pipeline*)
71 		region_alloc_zero(region, sizeof(*tp));
72 	tp->num_unused = ID_PIPE_NUM;
73 	assert(sizeof(tp->unused)/sizeof(tp->unused[0]) == ID_PIPE_NUM);
74 	for(i=0; i<ID_PIPE_NUM; i++)
75 		tp->unused[i] = (uint16_t)i;
76 	tp->tcp_r = xfrd_tcp_create(region, QIOBUFSZ);
77 	tp->tcp_w = xfrd_tcp_create(region, 512);
78 	return tp;
79 }
80 
81 void
82 xfrd_setup_packet(buffer_type* packet,
83 	uint16_t type, uint16_t klass, const dname_type* dname, uint16_t qid)
84 {
85 	/* Set up the header */
86 	buffer_clear(packet);
87 	ID_SET(packet, qid);
88 	FLAGS_SET(packet, 0);
89 	OPCODE_SET(packet, OPCODE_QUERY);
90 	QDCOUNT_SET(packet, 1);
91 	ANCOUNT_SET(packet, 0);
92 	NSCOUNT_SET(packet, 0);
93 	ARCOUNT_SET(packet, 0);
94 	buffer_skip(packet, QHEADERSZ);
95 
96 	/* The question record. */
97 	buffer_write(packet, dname_name(dname), dname->name_size);
98 	buffer_write_u16(packet, type);
99 	buffer_write_u16(packet, klass);
100 }
101 
102 static socklen_t
103 #ifdef INET6
104 xfrd_acl_sockaddr(acl_options_type* acl, unsigned int port,
105 	struct sockaddr_storage *sck)
106 #else
107 xfrd_acl_sockaddr(acl_options_type* acl, unsigned int port,
108 	struct sockaddr_in *sck, const char* fromto)
109 #endif /* INET6 */
110 {
111 	/* setup address structure */
112 #ifdef INET6
113 	memset(sck, 0, sizeof(struct sockaddr_storage));
114 #else
115 	memset(sck, 0, sizeof(struct sockaddr_in));
116 #endif
117 	if(acl->is_ipv6) {
118 #ifdef INET6
119 		struct sockaddr_in6* sa = (struct sockaddr_in6*)sck;
120 		sa->sin6_family = AF_INET6;
121 		sa->sin6_port = htons(port);
122 		sa->sin6_addr = acl->addr.addr6;
123 		return sizeof(struct sockaddr_in6);
124 #else
125 		log_msg(LOG_ERR, "xfrd: IPv6 connection %s %s attempted but no \
126 INET6.", fromto, acl->ip_address_spec);
127 		return 0;
128 #endif
129 	} else {
130 		struct sockaddr_in* sa = (struct sockaddr_in*)sck;
131 		sa->sin_family = AF_INET;
132 		sa->sin_port = htons(port);
133 		sa->sin_addr = acl->addr.addr;
134 		return sizeof(struct sockaddr_in);
135 	}
136 }
137 
138 socklen_t
139 #ifdef INET6
140 xfrd_acl_sockaddr_to(acl_options_type* acl, struct sockaddr_storage *to)
141 #else
142 xfrd_acl_sockaddr_to(acl_options_type* acl, struct sockaddr_in *to)
143 #endif /* INET6 */
144 {
145 	unsigned int port = acl->port?acl->port:(unsigned)atoi(TCP_PORT);
146 #ifdef INET6
147 	return xfrd_acl_sockaddr(acl, port, to);
148 #else
149 	return xfrd_acl_sockaddr(acl, port, to, "to");
150 #endif /* INET6 */
151 }
152 
153 socklen_t
154 #ifdef INET6
155 xfrd_acl_sockaddr_frm(acl_options_type* acl, struct sockaddr_storage *frm)
156 #else
157 xfrd_acl_sockaddr_frm(acl_options_type* acl, struct sockaddr_in *frm)
158 #endif /* INET6 */
159 {
160 	unsigned int port = acl->port?acl->port:0;
161 #ifdef INET6
162 	return xfrd_acl_sockaddr(acl, port, frm);
163 #else
164 	return xfrd_acl_sockaddr(acl, port, frm, "from");
165 #endif /* INET6 */
166 }
167 
168 void
169 xfrd_write_soa_buffer(struct buffer* packet,
170 	const dname_type* apex, struct xfrd_soa* soa)
171 {
172 	size_t rdlength_pos;
173 	uint16_t rdlength;
174 	buffer_write(packet, dname_name(apex), apex->name_size);
175 
176 	/* already in network order */
177 	buffer_write(packet, &soa->type, sizeof(soa->type));
178 	buffer_write(packet, &soa->klass, sizeof(soa->klass));
179 	buffer_write(packet, &soa->ttl, sizeof(soa->ttl));
180 	rdlength_pos = buffer_position(packet);
181 	buffer_skip(packet, sizeof(rdlength));
182 
183 	/* uncompressed dnames */
184 	buffer_write(packet, soa->prim_ns+1, soa->prim_ns[0]);
185 	buffer_write(packet, soa->email+1, soa->email[0]);
186 
187 	buffer_write(packet, &soa->serial, sizeof(uint32_t));
188 	buffer_write(packet, &soa->refresh, sizeof(uint32_t));
189 	buffer_write(packet, &soa->retry, sizeof(uint32_t));
190 	buffer_write(packet, &soa->expire, sizeof(uint32_t));
191 	buffer_write(packet, &soa->minimum, sizeof(uint32_t));
192 
193 	/* write length of RR */
194 	rdlength = buffer_position(packet) - rdlength_pos - sizeof(rdlength);
195 	buffer_write_u16_at(packet, rdlength_pos, rdlength);
196 }
197 
198 struct xfrd_tcp*
199 xfrd_tcp_create(region_type* region, size_t bufsize)
200 {
201 	struct xfrd_tcp* tcp_state = (struct xfrd_tcp*)region_alloc(
202 		region, sizeof(struct xfrd_tcp));
203 	memset(tcp_state, 0, sizeof(struct xfrd_tcp));
204 	tcp_state->packet = buffer_create(region, bufsize);
205 	tcp_state->fd = -1;
206 
207 	return tcp_state;
208 }
209 
210 static struct xfrd_tcp_pipeline*
211 pipeline_find(struct xfrd_tcp_set* set, xfrd_zone_type* zone)
212 {
213 	rbnode_type* sme = NULL;
214 	struct xfrd_tcp_pipeline* r;
215 	/* smaller buf than a full pipeline with 64kb ID array, only need
216 	 * the front part with the key info, this front part contains the
217 	 * members that the compare function uses. */
218 	const size_t keysize = sizeof(struct xfrd_tcp_pipeline) -
219 		ID_PIPE_NUM*(sizeof(struct xfrd_zone*) + sizeof(uint16_t));
220 	/* void* type for alignment of the struct,
221 	 * divide the keysize by ptr-size and then add one to round up */
222 	void* buf[ (keysize / sizeof(void*)) + 1 ];
223 	struct xfrd_tcp_pipeline* key = (struct xfrd_tcp_pipeline*)buf;
224 	key->node.key = key;
225 	key->ip_len = xfrd_acl_sockaddr_to(zone->master, &key->ip);
226 	key->num_unused = ID_PIPE_NUM;
227 	/* lookup existing tcp transfer to the master with highest unused */
228 	if(rbtree_find_less_equal(set->pipetree, key, &sme)) {
229 		/* exact match, strange, fully unused tcp cannot be open */
230 		assert(0);
231 	}
232 	if(!sme)
233 		return NULL;
234 	r = (struct xfrd_tcp_pipeline*)sme->key;
235 	/* <= key pointed at, is the master correct ? */
236 	if(r->ip_len != key->ip_len)
237 		return NULL;
238 	if(memcmp(&r->ip, &key->ip, key->ip_len) != 0)
239 		return NULL;
240 	/* correct master, is there a slot free for this transfer? */
241 	if(r->num_unused == 0)
242 		return NULL;
243 	return r;
244 }
245 
246 /* remove zone from tcp waiting list */
247 static void
248 tcp_zone_waiting_list_popfirst(struct xfrd_tcp_set* set, xfrd_zone_type* zone)
249 {
250 	assert(zone->tcp_waiting);
251 	set->tcp_waiting_first = zone->tcp_waiting_next;
252 	if(zone->tcp_waiting_next)
253 		zone->tcp_waiting_next->tcp_waiting_prev = NULL;
254 	else	set->tcp_waiting_last = 0;
255 	zone->tcp_waiting_next = 0;
256 	zone->tcp_waiting = 0;
257 }
258 
259 /* remove zone from tcp pipe write-wait list */
260 static void
261 tcp_pipe_sendlist_remove(struct xfrd_tcp_pipeline* tp, xfrd_zone_type* zone)
262 {
263 	if(zone->in_tcp_send) {
264 		if(zone->tcp_send_prev)
265 			zone->tcp_send_prev->tcp_send_next=zone->tcp_send_next;
266 		else	tp->tcp_send_first=zone->tcp_send_next;
267 		if(zone->tcp_send_next)
268 			zone->tcp_send_next->tcp_send_prev=zone->tcp_send_prev;
269 		else	tp->tcp_send_last=zone->tcp_send_prev;
270 		zone->in_tcp_send = 0;
271 	}
272 }
273 
274 /* remove first from write-wait list */
275 static void
276 tcp_pipe_sendlist_popfirst(struct xfrd_tcp_pipeline* tp, xfrd_zone_type* zone)
277 {
278 	tp->tcp_send_first = zone->tcp_send_next;
279 	if(tp->tcp_send_first)
280 		tp->tcp_send_first->tcp_send_prev = NULL;
281 	else	tp->tcp_send_last = NULL;
282 	zone->in_tcp_send = 0;
283 }
284 
285 /* remove zone from tcp pipe ID map */
286 static void
287 tcp_pipe_id_remove(struct xfrd_tcp_pipeline* tp, xfrd_zone_type* zone)
288 {
289 	assert(tp->num_unused < ID_PIPE_NUM && tp->num_unused >= 0);
290 	assert(tp->id[zone->query_id] == zone);
291 	tp->id[zone->query_id] = NULL;
292 	tp->unused[tp->num_unused] = zone->query_id;
293 	/* must remove and re-add for sort order in tree */
294 	(void)rbtree_delete(xfrd->tcp_set->pipetree, &tp->node);
295 	tp->num_unused++;
296 	(void)rbtree_insert(xfrd->tcp_set->pipetree, &tp->node);
297 }
298 
299 /* stop the tcp pipe (and all its zones need to retry) */
300 static void
301 xfrd_tcp_pipe_stop(struct xfrd_tcp_pipeline* tp)
302 {
303 	int i, conn = -1;
304 	assert(tp->num_unused < ID_PIPE_NUM); /* at least one 'in-use' */
305 	assert(ID_PIPE_NUM - tp->num_unused > tp->num_skip); /* at least one 'nonskip' */
306 	/* need to retry for all the zones connected to it */
307 	/* these could use different lists and go to a different nextmaster*/
308 	for(i=0; i<ID_PIPE_NUM; i++) {
309 		if(tp->id[i] && tp->id[i] != TCP_NULL_SKIP) {
310 			xfrd_zone_type* zone = tp->id[i];
311 			conn = zone->tcp_conn;
312 			zone->tcp_conn = -1;
313 			zone->tcp_waiting = 0;
314 			tcp_pipe_sendlist_remove(tp, zone);
315 			tcp_pipe_id_remove(tp, zone);
316 			xfrd_set_refresh_now(zone);
317 		}
318 	}
319 	assert(conn != -1);
320 	/* now release the entire tcp pipe */
321 	xfrd_tcp_pipe_release(xfrd->tcp_set, tp, conn);
322 }
323 
324 static void
325 tcp_pipe_reset_timeout(struct xfrd_tcp_pipeline* tp)
326 {
327 	int fd = tp->handler.ev_fd;
328 	struct timeval tv;
329 	tv.tv_sec = xfrd->tcp_set->tcp_timeout;
330 	tv.tv_usec = 0;
331 	if(tp->handler_added)
332 		event_del(&tp->handler);
333 	event_set(&tp->handler, fd, EV_PERSIST|EV_TIMEOUT|EV_READ|
334 		(tp->tcp_send_first?EV_WRITE:0), xfrd_handle_tcp_pipe, tp);
335 	if(event_base_set(xfrd->event_base, &tp->handler) != 0)
336 		log_msg(LOG_ERR, "xfrd tcp: event_base_set failed");
337 	if(event_add(&tp->handler, &tv) != 0)
338 		log_msg(LOG_ERR, "xfrd tcp: event_add failed");
339 	tp->handler_added = 1;
340 }
341 
342 /* handle event from fd of tcp pipe */
343 void
344 xfrd_handle_tcp_pipe(int ATTR_UNUSED(fd), short event, void* arg)
345 {
346 	struct xfrd_tcp_pipeline* tp = (struct xfrd_tcp_pipeline*)arg;
347 	if((event & EV_WRITE)) {
348 		tcp_pipe_reset_timeout(tp);
349 		if(tp->tcp_send_first) {
350 			DEBUG(DEBUG_XFRD,1, (LOG_INFO, "xfrd: event tcp write, zone %s",
351 				tp->tcp_send_first->apex_str));
352 			xfrd_tcp_write(tp, tp->tcp_send_first);
353 		}
354 	}
355 	if((event & EV_READ) && tp->handler_added) {
356 		DEBUG(DEBUG_XFRD,1, (LOG_INFO, "xfrd: event tcp read"));
357 		tcp_pipe_reset_timeout(tp);
358 		xfrd_tcp_read(tp);
359 	}
360 	if((event & EV_TIMEOUT) && tp->handler_added) {
361 		/* tcp connection timed out */
362 		DEBUG(DEBUG_XFRD,1, (LOG_INFO, "xfrd: event tcp timeout"));
363 		xfrd_tcp_pipe_stop(tp);
364 	}
365 }
366 
367 /* add a zone to the pipeline, it starts to want to write its query */
368 static void
369 pipeline_setup_new_zone(struct xfrd_tcp_set* set, struct xfrd_tcp_pipeline* tp,
370 	xfrd_zone_type* zone)
371 {
372 	/* assign the ID */
373 	int idx;
374 	assert(tp->num_unused > 0);
375 	/* we pick a random ID, even though it is TCP anyway */
376 	idx = random_generate(tp->num_unused);
377 	zone->query_id = tp->unused[idx];
378 	tp->unused[idx] = tp->unused[tp->num_unused-1];
379 	tp->id[zone->query_id] = zone;
380 	/* decrement unused counter, and fixup tree */
381 	(void)rbtree_delete(set->pipetree, &tp->node);
382 	tp->num_unused--;
383 	(void)rbtree_insert(set->pipetree, &tp->node);
384 
385 	/* add to sendlist, at end */
386 	zone->tcp_send_next = NULL;
387 	zone->tcp_send_prev = tp->tcp_send_last;
388 	zone->in_tcp_send = 1;
389 	if(tp->tcp_send_last)
390 		tp->tcp_send_last->tcp_send_next = zone;
391 	else	tp->tcp_send_first = zone;
392 	tp->tcp_send_last = zone;
393 
394 	/* is it first in line? */
395 	if(tp->tcp_send_first == zone) {
396 		xfrd_tcp_setup_write_packet(tp, zone);
397 		/* add write to event handler */
398 		tcp_pipe_reset_timeout(tp);
399 	}
400 }
401 
402 void
403 xfrd_tcp_obtain(struct xfrd_tcp_set* set, xfrd_zone_type* zone)
404 {
405 	struct xfrd_tcp_pipeline* tp;
406 	assert(zone->tcp_conn == -1);
407 	assert(zone->tcp_waiting == 0);
408 
409 	if(set->tcp_count < XFRD_MAX_TCP) {
410 		int i;
411 		assert(!set->tcp_waiting_first);
412 		set->tcp_count ++;
413 		/* find a free tcp_buffer */
414 		for(i=0; i<XFRD_MAX_TCP; i++) {
415 			if(set->tcp_state[i]->tcp_r->fd == -1) {
416 				zone->tcp_conn = i;
417 				break;
418 			}
419 		}
420 		/** What if there is no free tcp_buffer? return; */
421 		if (zone->tcp_conn < 0) {
422 			return;
423 		}
424 
425 		tp = set->tcp_state[zone->tcp_conn];
426 		zone->tcp_waiting = 0;
427 
428 		/* stop udp use (if any) */
429 		if(zone->zone_handler.ev_fd != -1)
430 			xfrd_udp_release(zone);
431 
432 		if(!xfrd_tcp_open(set, tp, zone)) {
433 			zone->tcp_conn = -1;
434 			set->tcp_count --;
435 			xfrd_set_refresh_now(zone);
436 			return;
437 		}
438 		/* ip and ip_len set by tcp_open */
439 		tp->node.key = tp;
440 		tp->num_unused = ID_PIPE_NUM;
441 		tp->num_skip = 0;
442 		tp->tcp_send_first = NULL;
443 		tp->tcp_send_last = NULL;
444 		memset(tp->id, 0, sizeof(tp->id));
445 		for(i=0; i<ID_PIPE_NUM; i++) {
446 			tp->unused[i] = i;
447 		}
448 
449 		/* insert into tree */
450 		(void)rbtree_insert(set->pipetree, &tp->node);
451 		xfrd_deactivate_zone(zone);
452 		xfrd_unset_timer(zone);
453 		pipeline_setup_new_zone(set, tp, zone);
454 		return;
455 	}
456 	/* check for a pipeline to the same master with unused ID */
457 	if((tp = pipeline_find(set, zone))!= NULL) {
458 		int i;
459 		if(zone->zone_handler.ev_fd != -1)
460 			xfrd_udp_release(zone);
461 		for(i=0; i<XFRD_MAX_TCP; i++) {
462 			if(set->tcp_state[i] == tp)
463 				zone->tcp_conn = i;
464 		}
465 		xfrd_deactivate_zone(zone);
466 		xfrd_unset_timer(zone);
467 		pipeline_setup_new_zone(set, tp, zone);
468 		return;
469 	}
470 
471 	/* wait, at end of line */
472 	DEBUG(DEBUG_XFRD,2, (LOG_INFO, "xfrd: max number of tcp "
473 		"connections (%d) reached.", XFRD_MAX_TCP));
474 	zone->tcp_waiting_next = 0;
475 	zone->tcp_waiting_prev = set->tcp_waiting_last;
476 	zone->tcp_waiting = 1;
477 	if(!set->tcp_waiting_last) {
478 		set->tcp_waiting_first = zone;
479 		set->tcp_waiting_last = zone;
480 	} else {
481 		set->tcp_waiting_last->tcp_waiting_next = zone;
482 		set->tcp_waiting_last = zone;
483 	}
484 	xfrd_deactivate_zone(zone);
485 	xfrd_unset_timer(zone);
486 }
487 
488 int
489 xfrd_tcp_open(struct xfrd_tcp_set* set, struct xfrd_tcp_pipeline* tp,
490 	xfrd_zone_type* zone)
491 {
492 	int fd, family, conn;
493 	struct timeval tv;
494 	assert(zone->tcp_conn != -1);
495 
496 	/* if there is no next master, fallback to use the first one */
497 	/* but there really should be a master set */
498 	if(!zone->master) {
499 		zone->master = zone->zone_options->pattern->request_xfr;
500 		zone->master_num = 0;
501 	}
502 
503 	DEBUG(DEBUG_XFRD,1, (LOG_INFO, "xfrd: zone %s open tcp conn to %s",
504 		zone->apex_str, zone->master->ip_address_spec));
505 	tp->tcp_r->is_reading = 1;
506 	tp->tcp_r->total_bytes = 0;
507 	tp->tcp_r->msglen = 0;
508 	buffer_clear(tp->tcp_r->packet);
509 	tp->tcp_w->is_reading = 0;
510 	tp->tcp_w->total_bytes = 0;
511 	tp->tcp_w->msglen = 0;
512 	tp->connection_established = 0;
513 
514 	if(zone->master->is_ipv6) {
515 #ifdef INET6
516 		family = PF_INET6;
517 #else
518 		xfrd_set_refresh_now(zone);
519 		return 0;
520 #endif
521 	} else {
522 		family = PF_INET;
523 	}
524 	fd = socket(family, SOCK_STREAM, IPPROTO_TCP);
525 	if(fd == -1) {
526 		/* squelch 'Address family not supported by protocol' at low
527 		 * verbosity levels */
528 		if(errno != EAFNOSUPPORT || verbosity > 2)
529 		    log_msg(LOG_ERR, "xfrd: %s cannot create tcp socket: %s",
530 			zone->master->ip_address_spec, strerror(errno));
531 		xfrd_set_refresh_now(zone);
532 		return 0;
533 	}
534 	if(fcntl(fd, F_SETFL, O_NONBLOCK) == -1) {
535 		log_msg(LOG_ERR, "xfrd: fcntl failed: %s", strerror(errno));
536 		close(fd);
537 		xfrd_set_refresh_now(zone);
538 		return 0;
539 	}
540 
541 	if(xfrd->nsd->outgoing_tcp_mss > 0) {
542 #if defined(IPPROTO_TCP) && defined(TCP_MAXSEG)
543 		if(setsockopt(fd, IPPROTO_TCP, TCP_MAXSEG,
544 			(void*)&xfrd->nsd->outgoing_tcp_mss,
545 			sizeof(xfrd->nsd->outgoing_tcp_mss)) < 0) {
546 			log_msg(LOG_ERR, "xfrd: setsockopt(TCP_MAXSEG)"
547 					"failed: %s", strerror(errno));
548 		}
549 #else
550 		log_msg(LOG_ERR, "setsockopt(TCP_MAXSEG) unsupported");
551 #endif
552 	}
553 
554 	tp->ip_len = xfrd_acl_sockaddr_to(zone->master, &tp->ip);
555 
556 	/* bind it */
557 	if (!xfrd_bind_local_interface(fd, zone->zone_options->pattern->
558 		outgoing_interface, zone->master, 1)) {
559 		close(fd);
560 		xfrd_set_refresh_now(zone);
561 		return 0;
562         }
563 
564 	conn = connect(fd, (struct sockaddr*)&tp->ip, tp->ip_len);
565 	if (conn == -1 && errno != EINPROGRESS) {
566 		log_msg(LOG_ERR, "xfrd: connect %s failed: %s",
567 			zone->master->ip_address_spec, strerror(errno));
568 		close(fd);
569 		xfrd_set_refresh_now(zone);
570 		return 0;
571 	}
572 	tp->tcp_r->fd = fd;
573 	tp->tcp_w->fd = fd;
574 
575 	/* set the tcp pipe event */
576 	if(tp->handler_added)
577 		event_del(&tp->handler);
578 	event_set(&tp->handler, fd, EV_PERSIST|EV_TIMEOUT|EV_READ|EV_WRITE,
579 		xfrd_handle_tcp_pipe, tp);
580 	if(event_base_set(xfrd->event_base, &tp->handler) != 0)
581 		log_msg(LOG_ERR, "xfrd tcp: event_base_set failed");
582 	tv.tv_sec = set->tcp_timeout;
583 	tv.tv_usec = 0;
584 	if(event_add(&tp->handler, &tv) != 0)
585 		log_msg(LOG_ERR, "xfrd tcp: event_add failed");
586 	tp->handler_added = 1;
587 	return 1;
588 }
589 
590 void
591 xfrd_tcp_setup_write_packet(struct xfrd_tcp_pipeline* tp, xfrd_zone_type* zone)
592 {
593 	struct xfrd_tcp* tcp = tp->tcp_w;
594 	assert(zone->tcp_conn != -1);
595 	assert(zone->tcp_waiting == 0);
596 	/* start AXFR or IXFR for the zone */
597 	if(zone->soa_disk_acquired == 0 || zone->master->use_axfr_only ||
598 		zone->master->ixfr_disabled ||
599 		/* if zone expired, after the first round, do not ask for
600 		 * IXFR any more, but full AXFR (of any serial number) */
601 		(zone->state == xfrd_zone_expired && zone->round_num != 0)) {
602 		DEBUG(DEBUG_XFRD,1, (LOG_INFO, "request full zone transfer "
603 						"(AXFR) for %s to %s",
604 			zone->apex_str, zone->master->ip_address_spec));
605 
606 		xfrd_setup_packet(tcp->packet, TYPE_AXFR, CLASS_IN, zone->apex,
607 			zone->query_id);
608 	} else {
609 		DEBUG(DEBUG_XFRD,1, (LOG_INFO, "request incremental zone "
610 						"transfer (IXFR) for %s to %s",
611 			zone->apex_str, zone->master->ip_address_spec));
612 
613 		xfrd_setup_packet(tcp->packet, TYPE_IXFR, CLASS_IN, zone->apex,
614 			zone->query_id);
615         	NSCOUNT_SET(tcp->packet, 1);
616 		xfrd_write_soa_buffer(tcp->packet, zone->apex, &zone->soa_disk);
617 	}
618 	/* old transfer needs to be removed still? */
619 	if(zone->msg_seq_nr)
620 		xfrd_unlink_xfrfile(xfrd->nsd, zone->xfrfilenumber);
621 	zone->msg_seq_nr = 0;
622 	zone->msg_rr_count = 0;
623 	if(zone->master->key_options && zone->master->key_options->tsig_key) {
624 		xfrd_tsig_sign_request(tcp->packet, &zone->tsig, zone->master);
625 	}
626 	buffer_flip(tcp->packet);
627 	DEBUG(DEBUG_XFRD,1, (LOG_INFO, "sent tcp query with ID %d", zone->query_id));
628 	tcp->msglen = buffer_limit(tcp->packet);
629 	tcp->total_bytes = 0;
630 }
631 
632 static void
633 tcp_conn_ready_for_reading(struct xfrd_tcp* tcp)
634 {
635 	tcp->total_bytes = 0;
636 	tcp->msglen = 0;
637 	buffer_clear(tcp->packet);
638 }
639 
640 int conn_write(struct xfrd_tcp* tcp)
641 {
642 	ssize_t sent;
643 
644 	if(tcp->total_bytes < sizeof(tcp->msglen)) {
645 		uint16_t sendlen = htons(tcp->msglen);
646 #ifdef HAVE_WRITEV
647 		struct iovec iov[2];
648 		iov[0].iov_base = (uint8_t*)&sendlen + tcp->total_bytes;
649 		iov[0].iov_len = sizeof(sendlen) - tcp->total_bytes;
650 		iov[1].iov_base = buffer_begin(tcp->packet);
651 		iov[1].iov_len = buffer_limit(tcp->packet);
652 		sent = writev(tcp->fd, iov, 2);
653 #else /* HAVE_WRITEV */
654 		sent = write(tcp->fd,
655 			(const char*)&sendlen + tcp->total_bytes,
656 			sizeof(tcp->msglen) - tcp->total_bytes);
657 #endif /* HAVE_WRITEV */
658 
659 		if(sent == -1) {
660 			if(errno == EAGAIN || errno == EINTR) {
661 				/* write would block, try later */
662 				return 0;
663 			} else {
664 				return -1;
665 			}
666 		}
667 
668 		tcp->total_bytes += sent;
669 		if(sent > (ssize_t)sizeof(tcp->msglen))
670 			buffer_skip(tcp->packet, sent-sizeof(tcp->msglen));
671 		if(tcp->total_bytes < sizeof(tcp->msglen)) {
672 			/* incomplete write, resume later */
673 			return 0;
674 		}
675 #ifdef HAVE_WRITEV
676 		if(tcp->total_bytes == tcp->msglen + sizeof(tcp->msglen)) {
677 			/* packet done */
678 			return 1;
679 		}
680 #endif
681 		assert(tcp->total_bytes >= sizeof(tcp->msglen));
682 	}
683 
684 	assert(tcp->total_bytes < tcp->msglen + sizeof(tcp->msglen));
685 
686 	sent = write(tcp->fd,
687 		buffer_current(tcp->packet),
688 		buffer_remaining(tcp->packet));
689 	if(sent == -1) {
690 		if(errno == EAGAIN || errno == EINTR) {
691 			/* write would block, try later */
692 			return 0;
693 		} else {
694 			return -1;
695 		}
696 	}
697 
698 	buffer_skip(tcp->packet, sent);
699 	tcp->total_bytes += sent;
700 
701 	if(tcp->total_bytes < tcp->msglen + sizeof(tcp->msglen)) {
702 		/* more to write when socket becomes writable again */
703 		return 0;
704 	}
705 
706 	assert(tcp->total_bytes == tcp->msglen + sizeof(tcp->msglen));
707 	return 1;
708 }
709 
710 void
711 xfrd_tcp_write(struct xfrd_tcp_pipeline* tp, xfrd_zone_type* zone)
712 {
713 	int ret;
714 	struct xfrd_tcp* tcp = tp->tcp_w;
715 	assert(zone->tcp_conn != -1);
716 	assert(zone == tp->tcp_send_first);
717 	/* see if for non-established connection, there is a connect error */
718 	if(!tp->connection_established) {
719 		/* check for pending error from nonblocking connect */
720 		/* from Stevens, unix network programming, vol1, 3rd ed, p450 */
721 		int error = 0;
722 		socklen_t len = sizeof(error);
723 		if(getsockopt(tcp->fd, SOL_SOCKET, SO_ERROR, &error, &len) < 0){
724 			error = errno; /* on solaris errno is error */
725 		}
726 		if(error == EINPROGRESS || error == EWOULDBLOCK)
727 			return; /* try again later */
728 		if(error != 0) {
729 			log_msg(LOG_ERR, "%s: Could not tcp connect to %s: %s",
730 				zone->apex_str, zone->master->ip_address_spec,
731 				strerror(error));
732 			xfrd_tcp_pipe_stop(tp);
733 			return;
734 		}
735 	}
736 	ret = conn_write(tcp);
737 	if(ret == -1) {
738 		log_msg(LOG_ERR, "xfrd: failed writing tcp %s", strerror(errno));
739 		xfrd_tcp_pipe_stop(tp);
740 		return;
741 	}
742 	if(tcp->total_bytes != 0 && !tp->connection_established)
743 		tp->connection_established = 1;
744 	if(ret == 0) {
745 		return; /* write again later */
746 	}
747 	/* done writing this message */
748 
749 	/* remove first zone from sendlist */
750 	tcp_pipe_sendlist_popfirst(tp, zone);
751 
752 	/* see if other zone wants to write; init; let it write (now) */
753 	/* and use a loop, because 64k stack calls is a too much */
754 	while(tp->tcp_send_first) {
755 		/* setup to write for this zone */
756 		xfrd_tcp_setup_write_packet(tp, tp->tcp_send_first);
757 		/* attempt to write for this zone (if success, continue loop)*/
758 		ret = conn_write(tcp);
759 		if(ret == -1) {
760 			log_msg(LOG_ERR, "xfrd: failed writing tcp %s", strerror(errno));
761 			xfrd_tcp_pipe_stop(tp);
762 			return;
763 		}
764 		if(ret == 0)
765 			return; /* write again later */
766 		tcp_pipe_sendlist_popfirst(tp, tp->tcp_send_first);
767 	}
768 
769 	/* if sendlist empty, remove WRITE from event */
770 
771 	/* listen to READ, and not WRITE events */
772 	assert(tp->tcp_send_first == NULL);
773 	tcp_pipe_reset_timeout(tp);
774 }
775 
776 int
777 conn_read(struct xfrd_tcp* tcp)
778 {
779 	ssize_t received;
780 	/* receive leading packet length bytes */
781 	if(tcp->total_bytes < sizeof(tcp->msglen)) {
782 		received = read(tcp->fd,
783 			(char*) &tcp->msglen + tcp->total_bytes,
784 			sizeof(tcp->msglen) - tcp->total_bytes);
785 		if(received == -1) {
786 			if(errno == EAGAIN || errno == EINTR) {
787 				/* read would block, try later */
788 				return 0;
789 			} else {
790 #ifdef ECONNRESET
791 				if (verbosity >= 2 || errno != ECONNRESET)
792 #endif /* ECONNRESET */
793 				log_msg(LOG_ERR, "tcp read sz: %s", strerror(errno));
794 				return -1;
795 			}
796 		} else if(received == 0) {
797 			/* EOF */
798 			return -1;
799 		}
800 		tcp->total_bytes += received;
801 		if(tcp->total_bytes < sizeof(tcp->msglen)) {
802 			/* not complete yet, try later */
803 			return 0;
804 		}
805 
806 		assert(tcp->total_bytes == sizeof(tcp->msglen));
807 		tcp->msglen = ntohs(tcp->msglen);
808 
809 		if(tcp->msglen == 0) {
810 			buffer_set_limit(tcp->packet, tcp->msglen);
811 			return 1;
812 		}
813 		if(tcp->msglen > buffer_capacity(tcp->packet)) {
814 			log_msg(LOG_ERR, "buffer too small, dropping connection");
815 			return 0;
816 		}
817 		buffer_set_limit(tcp->packet, tcp->msglen);
818 	}
819 
820 	assert(buffer_remaining(tcp->packet) > 0);
821 
822 	received = read(tcp->fd, buffer_current(tcp->packet),
823 		buffer_remaining(tcp->packet));
824 	if(received == -1) {
825 		if(errno == EAGAIN || errno == EINTR) {
826 			/* read would block, try later */
827 			return 0;
828 		} else {
829 #ifdef ECONNRESET
830 			if (verbosity >= 2 || errno != ECONNRESET)
831 #endif /* ECONNRESET */
832 			log_msg(LOG_ERR, "tcp read %s", strerror(errno));
833 			return -1;
834 		}
835 	} else if(received == 0) {
836 		/* EOF */
837 		return -1;
838 	}
839 
840 	tcp->total_bytes += received;
841 	buffer_skip(tcp->packet, received);
842 
843 	if(buffer_remaining(tcp->packet) > 0) {
844 		/* not complete yet, wait for more */
845 		return 0;
846 	}
847 
848 	/* completed */
849 	assert(buffer_position(tcp->packet) == tcp->msglen);
850 	return 1;
851 }
852 
853 void
854 xfrd_tcp_read(struct xfrd_tcp_pipeline* tp)
855 {
856 	xfrd_zone_type* zone;
857 	struct xfrd_tcp* tcp = tp->tcp_r;
858 	int ret;
859 	enum xfrd_packet_result pkt_result;
860 
861 	ret = conn_read(tcp);
862 	if(ret == -1) {
863 		xfrd_tcp_pipe_stop(tp);
864 		return;
865 	}
866 	if(ret == 0)
867 		return;
868 	/* completed msg */
869 	buffer_flip(tcp->packet);
870 	/* see which ID number it is, if skip, handle skip, NULL: warn */
871 	if(tcp->msglen < QHEADERSZ) {
872 		/* too short for DNS header, skip it */
873 		DEBUG(DEBUG_XFRD,1, (LOG_INFO,
874 			"xfrd: tcp skip response that is too short"));
875 		tcp_conn_ready_for_reading(tcp);
876 		return;
877 	}
878 	zone = tp->id[ID(tcp->packet)];
879 	if(!zone || zone == TCP_NULL_SKIP) {
880 		/* no zone for this id? skip it */
881 		DEBUG(DEBUG_XFRD,1, (LOG_INFO,
882 			"xfrd: tcp skip response with %s ID",
883 			zone?"set-to-skip":"unknown"));
884 		tcp_conn_ready_for_reading(tcp);
885 		return;
886 	}
887 	assert(zone->tcp_conn != -1);
888 
889 	/* handle message for zone */
890 	pkt_result = xfrd_handle_received_xfr_packet(zone, tcp->packet);
891 	/* setup for reading the next packet on this connection */
892 	tcp_conn_ready_for_reading(tcp);
893 	switch(pkt_result) {
894 		case xfrd_packet_more:
895 			/* wait for next packet */
896 			break;
897 		case xfrd_packet_newlease:
898 			/* set to skip if more packets with this ID */
899 			tp->id[zone->query_id] = TCP_NULL_SKIP;
900 			tp->num_skip++;
901 			/* fall through to remove zone from tp */
902 			/* fallthrough */
903 		case xfrd_packet_transfer:
904 			if(zone->zone_options->pattern->multi_master_check) {
905 				xfrd_tcp_release(xfrd->tcp_set, zone);
906 				xfrd_make_request(zone);
907 				break;
908 			}
909 			xfrd_tcp_release(xfrd->tcp_set, zone);
910 			assert(zone->round_num == -1);
911 			break;
912 		case xfrd_packet_notimpl:
913 			xfrd_disable_ixfr(zone);
914 			xfrd_tcp_release(xfrd->tcp_set, zone);
915 			/* query next server */
916 			xfrd_make_request(zone);
917 			break;
918 		case xfrd_packet_bad:
919 		case xfrd_packet_tcp:
920 		default:
921 			/* set to skip if more packets with this ID */
922 			tp->id[zone->query_id] = TCP_NULL_SKIP;
923 			tp->num_skip++;
924 			xfrd_tcp_release(xfrd->tcp_set, zone);
925 			/* query next server */
926 			xfrd_make_request(zone);
927 			break;
928 	}
929 }
930 
931 void
932 xfrd_tcp_release(struct xfrd_tcp_set* set, xfrd_zone_type* zone)
933 {
934 	int conn = zone->tcp_conn;
935 	struct xfrd_tcp_pipeline* tp = set->tcp_state[conn];
936 	DEBUG(DEBUG_XFRD,1, (LOG_INFO, "xfrd: zone %s released tcp conn to %s",
937 		zone->apex_str, zone->master->ip_address_spec));
938 	assert(zone->tcp_conn != -1);
939 	assert(zone->tcp_waiting == 0);
940 	zone->tcp_conn = -1;
941 	zone->tcp_waiting = 0;
942 
943 	/* remove from tcp_send list */
944 	tcp_pipe_sendlist_remove(tp, zone);
945 	/* remove it from the ID list */
946 	if(tp->id[zone->query_id] != TCP_NULL_SKIP)
947 		tcp_pipe_id_remove(tp, zone);
948 	DEBUG(DEBUG_XFRD,1, (LOG_INFO, "xfrd: released tcp pipe now %d unused",
949 		tp->num_unused));
950 	/* if pipe was full, but no more, then see if waiting element is
951 	 * for the same master, and can fill the unused ID */
952 	if(tp->num_unused == 1 && set->tcp_waiting_first) {
953 #ifdef INET6
954 		struct sockaddr_storage to;
955 #else
956 		struct sockaddr_in to;
957 #endif
958 		socklen_t to_len = xfrd_acl_sockaddr_to(
959 			set->tcp_waiting_first->master, &to);
960 		if(to_len == tp->ip_len && memcmp(&to, &tp->ip, to_len) == 0) {
961 			/* use this connection for the waiting zone */
962 			zone = set->tcp_waiting_first;
963 			assert(zone->tcp_conn == -1);
964 			zone->tcp_conn = conn;
965 			tcp_zone_waiting_list_popfirst(set, zone);
966 			if(zone->zone_handler.ev_fd != -1)
967 				xfrd_udp_release(zone);
968 			xfrd_unset_timer(zone);
969 			pipeline_setup_new_zone(set, tp, zone);
970 			return;
971 		}
972 		/* waiting zone did not go to same server */
973 	}
974 
975 	/* if all unused, or only skipped leftover, close the pipeline */
976 	if(tp->num_unused >= ID_PIPE_NUM || tp->num_skip >= ID_PIPE_NUM - tp->num_unused)
977 		xfrd_tcp_pipe_release(set, tp, conn);
978 }
979 
980 void
981 xfrd_tcp_pipe_release(struct xfrd_tcp_set* set, struct xfrd_tcp_pipeline* tp,
982 	int conn)
983 {
984 	DEBUG(DEBUG_XFRD,1, (LOG_INFO, "xfrd: tcp pipe released"));
985 	/* one handler per tcp pipe */
986 	if(tp->handler_added)
987 		event_del(&tp->handler);
988 	tp->handler_added = 0;
989 
990 	/* fd in tcp_r and tcp_w is the same, close once */
991 	if(tp->tcp_r->fd != -1)
992 		close(tp->tcp_r->fd);
993 	tp->tcp_r->fd = -1;
994 	tp->tcp_w->fd = -1;
995 
996 	/* remove from pipetree */
997 	(void)rbtree_delete(xfrd->tcp_set->pipetree, &tp->node);
998 
999 	/* a waiting zone can use the free tcp slot (to another server) */
1000 	/* if that zone fails to set-up or connect, we try to start the next
1001 	 * waiting zone in the list */
1002 	while(set->tcp_count == XFRD_MAX_TCP && set->tcp_waiting_first) {
1003 		int i;
1004 
1005 		/* pop first waiting process */
1006 		xfrd_zone_type* zone = set->tcp_waiting_first;
1007 		/* start it */
1008 		assert(zone->tcp_conn == -1);
1009 		zone->tcp_conn = conn;
1010 		tcp_zone_waiting_list_popfirst(set, zone);
1011 
1012 		/* stop udp (if any) */
1013 		if(zone->zone_handler.ev_fd != -1)
1014 			xfrd_udp_release(zone);
1015 		if(!xfrd_tcp_open(set, tp, zone)) {
1016 			zone->tcp_conn = -1;
1017 			xfrd_set_refresh_now(zone);
1018 			/* try to start the next zone (if any) */
1019 			continue;
1020 		}
1021 		/* re-init this tcppipe */
1022 		/* ip and ip_len set by tcp_open */
1023 		tp->node.key = tp;
1024 		tp->num_unused = ID_PIPE_NUM;
1025 		tp->num_skip = 0;
1026 		tp->tcp_send_first = NULL;
1027 		tp->tcp_send_last = NULL;
1028 		memset(tp->id, 0, sizeof(tp->id));
1029 		for(i=0; i<ID_PIPE_NUM; i++) {
1030 			tp->unused[i] = i;
1031 		}
1032 
1033 		/* insert into tree */
1034 		(void)rbtree_insert(set->pipetree, &tp->node);
1035 		/* setup write */
1036 		xfrd_unset_timer(zone);
1037 		pipeline_setup_new_zone(set, tp, zone);
1038 		/* started a task, no need for cleanups, so return */
1039 		return;
1040 	}
1041 	/* no task to start, cleanup */
1042 	assert(!set->tcp_waiting_first);
1043 	set->tcp_count --;
1044 	assert(set->tcp_count >= 0);
1045 }
1046 
1047