worker.c 52.9 KB
Newer Older
1
/*  Copyright (C) 2014-2017 CZ.NIC, z.s.p.o. <knot-dns@labs.nic.cz>
Marek Vavruša's avatar
Marek Vavruša committed
2 3 4 5 6 7 8 9 10 11 12 13

    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation, either version 3 of the License, or
    (at your option) any later version.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public License
14
    along with this program.  If not, see <https://www.gnu.org/licenses/>.
Marek Vavruša's avatar
Marek Vavruša committed
15 16
 */

17
#include <uv.h>
18
#include <lua.h>
19
#include <libknot/packet/pkt.h>
20
#include <libknot/descriptor.h>
21 22
#include <contrib/ucw/lib.h>
#include <contrib/ucw/mempool.h>
23
#include <contrib/wire.h>
Marek Vavruša's avatar
Marek Vavruša committed
24 25 26
#if defined(__GLIBC__) && defined(_GNU_SOURCE)
#include <malloc.h>
#endif
27
#include <assert.h>
28 29
#include <sys/types.h>
#include <unistd.h>
30
#include <gnutls/gnutls.h>
31
#include "lib/utils.h"
32
#include "lib/layer.h"
33
#include "daemon/worker.h"
34
#include "daemon/bindings.h"
35
#include "daemon/engine.h"
36
#include "daemon/io.h"
37
#include "daemon/tls.h"
38
#include "daemon/zimport.h"
39
#include "daemon/session.h"
40

41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56

/* Magic defaults for the worker. */
#ifndef MP_FREELIST_SIZE
# ifdef __clang_analyzer__
#  define MP_FREELIST_SIZE 0
# else
#  define MP_FREELIST_SIZE 64 /**< Maximum length of the worker mempool freelist */
# endif
#endif
#ifndef QUERY_RATE_THRESHOLD
#define QUERY_RATE_THRESHOLD (2 * MP_FREELIST_SIZE) /**< Nr of parallel queries considered as high rate */
#endif
#ifndef MAX_PIPELINED
#define MAX_PIPELINED 100
#endif

57
#define VERBOSE_MSG(qry, ...) QRVERBOSE(qry, "wrkr", __VA_ARGS__)
58

59 60 61 62 63 64 65 66 67 68 69 70 71
/** Client request state. */
struct request_ctx
{
	struct kr_request req;
	struct {
		union inaddr addr;
		union inaddr dst_addr;
		/* uv_handle_t *handle; */

		/** NULL if the request didn't come over network. */
		struct session *session;
	} source;
	struct worker_ctx *worker;
72
	struct qr_task *task;
73 74 75 76 77 78 79 80
};

/** Query resolution task. */
struct qr_task
{
	struct request_ctx *ctx;
	knot_pkt_t *pktbuf;
	qr_tasklist_t waiting;
81
	struct session *pending[MAX_PENDING];
82 83 84 85 86 87 88 89 90
	uint16_t pending_count;
	uint16_t addrlist_count;
	uint16_t addrlist_turn;
	uint16_t timeouts;
	uint16_t iter_count;
	struct sockaddr *addrlist;
	uint32_t refs;
	bool finished : 1;
	bool leading  : 1;
91
	uint64_t creation_time;
92 93
};

94

95 96 97 98
/* Convenience macros */
#define qr_task_ref(task) \
	do { ++(task)->refs; } while(0)
#define qr_task_unref(task) \
99
	do { if (task && --(task)->refs == 0) { qr_task_free(task); } } while (0)
100 101 102 103 104

/** @internal get key for tcp session
 *  @note kr_straddr() return pointer to static string
 */
#define tcpsess_key(addr) kr_straddr(addr)
105 106

/* Forward decls */
107
static void qr_task_free(struct qr_task *task);
108 109 110
static int qr_task_step(struct qr_task *task,
			const struct sockaddr *packet_source,
			knot_pkt_t *packet);
111
static int qr_task_send(struct qr_task *task, struct session *session,
112 113 114 115
			struct sockaddr *addr, knot_pkt_t *pkt);
static int qr_task_finalize(struct qr_task *task, int state);
static void qr_task_complete(struct qr_task *task);
static struct session* worker_find_tcp_connected(struct worker_ctx *worker,
116
						 const struct sockaddr *addr);
117 118 119 120 121 122
static int worker_add_tcp_waiting(struct worker_ctx *worker,
				  const struct sockaddr *addr,
				  struct session *session);
static int worker_del_tcp_waiting(struct worker_ctx *worker,
				  const struct sockaddr *addr);
static struct session* worker_find_tcp_waiting(struct worker_ctx *worker,
123
					       const struct sockaddr *addr);
124
static void on_tcp_connect_timeout(uv_timer_t *timer);
125 126 127 128 129 130 131

/** @internal Get singleton worker. */
static inline struct worker_ctx *get_worker(void)
{
	return uv_default_loop()->data;
}

132 133
/*! @internal Create a UDP/TCP handle for an outgoing AF_INET* connection.
 *  socktype is SOCK_* */
134
static uv_handle_t *ioreq_spawn(struct worker_ctx *worker, int socktype, sa_family_t family)
135
{
136 137 138
	bool precond = (socktype == SOCK_DGRAM || socktype == SOCK_STREAM)
			&& (family == AF_INET  || family == AF_INET6);
	if (!precond) {
139 140
		/* assert(false); see #245 */
		kr_log_verbose("[work] ioreq_spawn: pre-condition failed\n");
141 142 143
		return NULL;
	}

144
	/* Create connection for iterative query */
145 146
	uv_handle_t *handle = malloc(socktype == SOCK_DGRAM
					? sizeof(uv_udp_t) : sizeof(uv_tcp_t));
147 148 149
	if (!handle) {
		return NULL;
	}
150 151 152 153 154 155
	int ret = io_create(worker->loop, handle, socktype, family);
	if (ret) {
		if (ret == UV_EMFILE) {
			worker->too_many_open = true;
			worker->rconcurrent_highwatermark = worker->stats.rconcurrent;
		}
156
		free(handle);
157 158
		return NULL;
	}
159 160 161 162

	/* Bind to outgoing address, according to IP v4/v6. */
	union inaddr *addr;
	if (family == AF_INET) {
163
		addr = (union inaddr *)&worker->out_addr4;
164
	} else {
165
		addr = (union inaddr *)&worker->out_addr6;
166 167 168 169
	}
	if (addr->ip.sa_family != AF_UNSPEC) {
		assert(addr->ip.sa_family == family);
		if (socktype == SOCK_DGRAM) {
170 171 172 173 174
			uv_udp_t *udp = (uv_udp_t *)handle;
			ret = uv_udp_bind(udp, &addr->ip, 0);
		} else if (socktype == SOCK_STREAM){
			uv_tcp_t *tcp = (uv_tcp_t *)handle;
			ret = uv_tcp_bind(tcp, &addr->ip, 0);
175 176 177
		}
	}

178
	if (ret != 0) {
179
		io_deinit(handle);
180
		free(handle);
181 182
		return NULL;
	}
183 184 185 186

	/* Set current handle as a subrequest type. */
	struct session *session = handle->data;
	session_flags(session)->outgoing = true;
187
	/* Connect or issue query datagram */
188
	return handle;
189 190
}

191
static void ioreq_kill_pending(struct qr_task *task)
192
{
193
	for (uint16_t i = 0; i < task->pending_count; ++i) {
194
		session_kill_ioreq(task->pending[i], task);
195 196 197 198
	}
	task->pending_count = 0;
}

199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222
/** @cond This memory layout is internal to mempool.c, use only for debugging. */
#if defined(__SANITIZE_ADDRESS__)
struct mempool_chunk {
  struct mempool_chunk *next;
  size_t size;
};
static void mp_poison(struct mempool *mp, bool poison)
{
	if (!poison) { /* @note mempool is part of the first chunk, unpoison it first */
		kr_asan_unpoison(mp, sizeof(*mp));
	}
	struct mempool_chunk *chunk = mp->state.last[0];
	void *chunk_off = (void *)chunk - chunk->size;
	if (poison) {
		kr_asan_poison(chunk_off, chunk->size);
	} else {
		kr_asan_unpoison(chunk_off, chunk->size);
	}
}
#else
#define mp_poison(mp, enable)
#endif
/** @endcond */

223
/** Get a mempool.  (Recycle if possible.)  */
224
static inline struct mempool *pool_borrow(struct worker_ctx *worker)
225 226
{
	struct mempool *mp = NULL;
227 228 229 230
	if (worker->pool_mp.len > 0) {
		mp = array_tail(worker->pool_mp);
		array_pop(worker->pool_mp);
		mp_poison(mp, 0);
231 232 233 234 235 236
	} else { /* No mempool on the freelist, create new one */
		mp = mp_new (4 * CPU_PAGE_SIZE);
	}
	return mp;
}

237
/** Return a mempool.  (Cache them up to some count.) */
238 239
static inline void pool_release(struct worker_ctx *worker, struct mempool *mp)
{
240
	if (worker->pool_mp.len < MP_FREELIST_SIZE) {
241
		mp_flush(mp);
242
		array_push(worker->pool_mp, mp);
243
		mp_poison(mp, 1);
244 245 246 247 248
	} else {
		mp_delete(mp);
	}
}

249 250 251 252 253 254 255 256 257 258 259 260
/** Create a key for an outgoing subrequest: qname, qclass, qtype.
 * @param key Destination buffer for key size, MUST be SUBREQ_KEY_LEN or larger.
 * @return key length if successful or an error
 */
static const size_t SUBREQ_KEY_LEN = KR_RRKEY_LEN;
static int subreq_key(char *dst, knot_pkt_t *pkt)
{
	assert(pkt);
	return kr_rrkey(dst, knot_pkt_qclass(pkt), knot_pkt_qname(pkt),
			knot_pkt_qtype(pkt), knot_pkt_qtype(pkt));
}

261 262 263 264 265 266 267
/** Create and initialize a request_ctx (on a fresh mempool).
 *
 * handle and addr point to the source of the request, and they are NULL
 * in case the request didn't come from network.
 */
static struct request_ctx *request_create(struct worker_ctx *worker,
					  uv_handle_t *handle,
268 269
					  const struct sockaddr *addr,
					  uint32_t uid)
270
{
271
	knot_mm_t pool = {
272
		.ctx = pool_borrow(worker),
273
		.alloc = (knot_mm_alloc_t) mp_alloc
274
	};
275

276 277 278 279
	/* Create request context */
	struct request_ctx *ctx = mm_alloc(&pool, sizeof(*ctx));
	if (!ctx) {
		pool_release(worker, pool.ctx);
280 281
		return NULL;
	}
282

283 284 285 286
	memset(ctx, 0, sizeof(*ctx));

	/* TODO Relocate pool to struct request */
	ctx->worker = worker;
287 288
	struct session *s = handle ? handle->data : NULL;
	if (s) {
289
		assert(session_flags(s)->outgoing == false);
290
	}
291
	ctx->source.session = s;
292 293 294

	struct kr_request *req = &ctx->req;
	req->pool = pool;
295
	req->vars_ref = LUA_NOREF;
296
	req->uid = uid;
297
	req->daemon_context = worker;
298

299
	/* Remember query source addr */
300 301 302
	if (!addr || (addr->sa_family != AF_INET && addr->sa_family != AF_INET6)) {
		ctx->source.addr.ip.sa_family = AF_UNSPEC;
	} else {
303
		memcpy(&ctx->source.addr, addr, kr_sockaddr_len(addr));
304
		ctx->req.qsource.addr = &ctx->source.addr.ip;
305
	}
306 307 308 309 310

	worker->stats.rconcurrent += 1;

	if (!handle) {
		return ctx;
311
	}
312

313
	/* Remember the destination address. */
314 315 316 317 318 319
	int addr_len = sizeof(ctx->source.dst_addr);
	struct sockaddr *dst_addr = &ctx->source.dst_addr.ip;
	ctx->source.dst_addr.ip.sa_family = AF_UNSPEC;
	if (handle->type == UV_UDP) {
		if (uv_udp_getsockname((uv_udp_t *)handle, dst_addr, &addr_len) == 0) {
			req->qsource.dst_addr = dst_addr;
320
		}
321
		req->qsource.flags.tcp = false;
322
		req->qsource.flags.tls = false;
323 324 325
	} else if (handle->type == UV_TCP) {
		if (uv_tcp_getsockname((uv_tcp_t *)handle, dst_addr, &addr_len) == 0) {
			req->qsource.dst_addr = dst_addr;
326
		}
327
		req->qsource.flags.tcp = true;
328
		req->qsource.flags.tls = s && session_flags(s)->has_tls;
329
	}
330 331

	return ctx;
332 333
}

334 335
/** More initialization, related to the particular incoming query/packet. */
static int request_start(struct request_ctx *ctx, knot_pkt_t *query)
336
{
337 338 339 340 341
	assert(query && ctx);
	size_t answer_max = KNOT_WIRE_MIN_PKTSIZE;
	struct kr_request *req = &ctx->req;

	/* source.session can be empty if request was generated by kresd itself */
342 343
	struct session *s = ctx->source.session;
	if (!s || session_get_handle(s)->type == UV_TCP) {
344 345 346 347
		answer_max = KNOT_WIRE_MAX_PKTSIZE;
	} else if (knot_pkt_has_edns(query)) { /* EDNS */
		answer_max = MAX(knot_edns_get_payload(query->opt_rr),
				 KNOT_WIRE_MIN_PKTSIZE);
348
	}
349
	req->qsource.size = query->size;
350 351 352
	if (knot_pkt_has_tsig(query)) {
		req->qsource.size += query->tsig_wire.len;
	}
353

354 355
	knot_pkt_t *answer = knot_pkt_new(NULL, answer_max, &req->pool);
	if (!answer) { /* Failed to allocate answer */
356 357 358
		return kr_error(ENOMEM);
	}

359 360 361
	knot_pkt_t *pkt = knot_pkt_new(NULL, req->qsource.size, &req->pool);
	if (!pkt) {
		return kr_error(ENOMEM);
362
	}
363 364 365

	int ret = knot_pkt_copy(pkt, query);
	if (ret != KNOT_EOK && ret != KNOT_ETRAIL) {
366
		return kr_error(ENOMEM);
367
	}
368 369
	req->qsource.packet = pkt;

370 371 372
	/* Start resolution */
	struct worker_ctx *worker = ctx->worker;
	struct engine *engine = worker->engine;
373
	kr_resolve_begin(req, &engine->resolver, answer);
374 375 376 377 378 379 380 381 382 383 384
	worker->stats.queries += 1;
	/* Throttle outbound queries only when high pressure */
	if (worker->stats.concurrent < QUERY_RATE_THRESHOLD) {
		req->options.NO_THROTTLE = true;
	}
	return kr_ok();
}

static void request_free(struct request_ctx *ctx)
{
	struct worker_ctx *worker = ctx->worker;
385 386 387 388 389 390 391 392 393 394 395 396 397 398
	/* Dereference any Lua vars table if exists */
	if (ctx->req.vars_ref != LUA_NOREF) {
		lua_State *L = worker->engine->L;
		/* Get worker variables table */
		lua_rawgeti(L, LUA_REGISTRYINDEX, worker->vars_table_ref);
		/* Get next free element (position 0) and store it under current reference (forming a list) */
		lua_rawgeti(L, -1, 0);
		lua_rawseti(L, -2, ctx->req.vars_ref);
		/* Set current reference as the next free element */
		lua_pushinteger(L, ctx->req.vars_ref);
		lua_rawseti(L, -2, 0);
		lua_pop(L, 1);
		ctx->req.vars_ref = LUA_NOREF;
	}
399
	/* Return mempool to ring or free it if it's full */
400
	pool_release(worker, ctx->req.pool.ctx);
401
	/* @note The 'task' is invalidated from now on. */
Marek Vavruša's avatar
Marek Vavruša committed
402
	/* Decommit memory every once in a while */
403
	static int mp_delete_count = 0;
404 405 406
	if (++mp_delete_count == 100000) {
		lua_gc(worker->engine->L, LUA_GCCOLLECT, 0);
#if defined(__GLIBC__) && defined(_GNU_SOURCE)
Marek Vavruša's avatar
Marek Vavruša committed
407
		malloc_trim(0);
408
#endif
Marek Vavruša's avatar
Marek Vavruša committed
409
		mp_delete_count = 0;
410
	}
411
	worker->stats.rconcurrent -= 1;
412
}
413

414 415 416 417 418 419 420 421
static struct qr_task *qr_task_create(struct request_ctx *ctx)
{
	/* How much can client handle? */
	struct engine *engine = ctx->worker->engine;
	size_t pktbuf_max = KR_EDNS_PAYLOAD;
	if (engine->resolver.opt_rr) {
		pktbuf_max = MAX(knot_edns_get_payload(engine->resolver.opt_rr),
				 pktbuf_max);
422 423
	}

424 425 426 427
	/* Create resolution task */
	struct qr_task *task = mm_alloc(&ctx->req.pool, sizeof(*task));
	if (!task) {
		return NULL;
428
	}
429
	memset(task, 0, sizeof(*task)); /* avoid accidentally unintialized fields */
430

431 432 433 434 435
	/* Create packet buffers for answer and subrequests */
	knot_pkt_t *pktbuf = knot_pkt_new(NULL, pktbuf_max, &ctx->req.pool);
	if (!pktbuf) {
		mm_free(&ctx->req.pool, task);
		return NULL;
436
	}
437
	pktbuf->size = 0;
438

439 440 441 442
	task->ctx = ctx;
	task->pktbuf = pktbuf;
	array_init(task->waiting);
	task->refs = 0;
443 444 445 446
	assert(ctx->task == NULL);
	ctx->task = task;
	/* Make the primary reference to task. */
	qr_task_ref(task);
447
	task->creation_time = kr_now();
448 449
	ctx->worker->stats.concurrent += 1;
	return task;
450 451
}

452 453 454 455 456 457 458 459 460
/* This is called when the task refcount is zero, free memory. */
static void qr_task_free(struct qr_task *task)
{
	struct request_ctx *ctx = task->ctx;

	assert(ctx);

	struct worker_ctx *worker = ctx->worker;

461
	if (ctx->task == NULL) {
462 463 464 465 466 467 468
		request_free(ctx);
	}

	/* Update stats */
	worker->stats.concurrent -= 1;
}

469 470 471
/*@ Register new qr_task within session. */
static int qr_task_register(struct qr_task *task, struct session *session)
{
472
	assert(!session_flags(session)->outgoing && session_get_handle(session)->type == UV_TCP);
473 474 475 476 477 478 479 480 481 482 483

	session_tasklist_add(session, task);

	struct request_ctx *ctx = task->ctx;
	assert(ctx && (ctx->source.session == NULL || ctx->source.session == session));
	ctx->source.session = session;
	/* Soft-limit on parallel queries, there is no "slow down" RCODE
	 * that we could use to signalize to client, but we can stop reading,
	 * an in effect shrink TCP window size. To get more precise throttling,
	 * we would need to copy remainder of the unread buffer and reassemble
	 * when resuming reading. This is NYI.  */
484 485 486 487
	if (session_tasklist_get_len(session) >= task->ctx->worker->tcp_pipeline_max &&
	    !session_flags(session)->throttled && !session_flags(session)->closing) {
		session_stop_read(session);
		session_flags(session)->throttled = true;
488 489 490 491 492
	}

	return 0;
}

493
static void qr_task_complete(struct qr_task *task)
494
{
495
	struct request_ctx *ctx = task->ctx;
496

497
	/* Kill pending I/O requests */
498
	ioreq_kill_pending(task);
499 500
	assert(task->waiting.len == 0);
	assert(task->leading == false);
501

502 503
	struct session *s = ctx->source.session;
	if (s) {
504
		assert(!session_flags(s)->outgoing && session_waitinglist_is_empty(s));
505
		ctx->source.session = NULL;
506
		session_tasklist_del(s, task);
Grigorii Demidov's avatar
Grigorii Demidov committed
507
	}
508

509
	/* Release primary reference to task. */
510 511 512 513
	if (ctx->task == task) {
		ctx->task = NULL;
		qr_task_unref(task);
	}
514 515
}

516
/* This is called when we send subrequest / answer */
517
static int qr_task_on_send(struct qr_task *task, uv_handle_t *handle, int status)
518
{
519

520 521 522 523 524
	if (task->finished) {
		assert(task->leading == false);
		qr_task_complete(task);
	}

525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545
	if (!handle || handle->type != UV_TCP) {
		return status;
	}

	struct session* s = handle->data;
	assert(s);
	if (status != 0) {
		session_tasklist_del(s, task);
	}

	if (session_flags(s)->outgoing || session_flags(s)->closing) {
		return status;
	}

	struct worker_ctx *worker = task->ctx->worker;
	if (session_flags(s)->throttled &&
	    session_tasklist_get_len(s) < worker->tcp_pipeline_max/2) {
	   /* Start reading again if the session is throttled and
	    * the number of outgoing requests is below watermark. */
		session_start_read(s);
		session_flags(s)->throttled = false;
546
	}
547

548
	return status;
549 550
}

551 552 553
static void on_send(uv_udp_send_t *req, int status)
{
	struct qr_task *task = req->data;
554 555
	uv_handle_t *h = (uv_handle_t *)req->handle;
	qr_task_on_send(task, h, status);
556
	qr_task_unref(task);
557
	free(req);
558
}
559 560

static void on_write(uv_write_t *req, int status)
561 562
{
	struct qr_task *task = req->data;
563 564
	uv_handle_t *h = (uv_handle_t *)req->handle;
	qr_task_on_send(task, h, status);
565
	qr_task_unref(task);
566
	free(req);
567 568
}

569
static int qr_task_send(struct qr_task *task, struct session *session,
570
			struct sockaddr *addr, knot_pkt_t *pkt)
571
{
572 573
	if (!session) {
		return qr_task_on_send(task, NULL, kr_error(EIO));
574
	}
575

576
	int ret = 0;
577
	struct request_ctx *ctx = task->ctx;
578

579 580
	uv_handle_t *handle = session_get_handle(session);
	assert(handle && handle->data == session);
581 582 583
	const bool is_stream = handle->type == UV_TCP;
	if (!is_stream && handle->type != UV_UDP) abort();

584 585 586 587 588 589 590 591
	if (addr == NULL) {
		addr = session_get_peer(session);
	}

	if (pkt == NULL) {
		pkt = worker_task_get_pktbuf(task);
	}

592
	if (session_flags(session)->outgoing && handle->type == UV_TCP) {
593 594
		size_t try_limit = session_tasklist_get_len(session) + 1;
		uint16_t msg_id = knot_wire_get_id(pkt->wire);
595
		size_t try_count = 0;
596 597 598 599 600 601
		while (session_tasklist_find_msgid(session, msg_id) &&
		       try_count <= try_limit) {
			++msg_id;
			++try_count;
		}
		if (try_count > try_limit) {
602
			return kr_error(ENOENT);
603 604 605 606
		}
		worker_task_pkt_set_msgid(task, msg_id);
	}

607 608 609 610 611
	uv_handle_t *ioreq = malloc(is_stream ? sizeof(uv_write_t) : sizeof(uv_udp_send_t));
	if (!ioreq) {
		return qr_task_on_send(task, handle, kr_error(ENOMEM));
	}

612 613 614
	/* Pending ioreq on current task */
	qr_task_ref(task);

615
	struct worker_ctx *worker = ctx->worker;
616
	/* Send using given protocol */
617 618
	assert(!session_flags(session)->closing);
	if (session_flags(session)->has_tls) {
619 620
		uv_write_t *write_req = (uv_write_t *)ioreq;
		write_req->data = task;
621
		ret = tls_write(write_req, handle, pkt, &on_write);
622
	} else if (handle->type == UV_UDP) {
623
		uv_udp_send_t *send_req = (uv_udp_send_t *)ioreq;
624
		uv_buf_t buf = { (char *)pkt->wire, pkt->size };
625 626 627 628
		send_req->data = task;
		ret = uv_udp_send(send_req, (uv_udp_t *)handle, &buf, 1, addr, &on_send);
	} else if (handle->type == UV_TCP) {
		uv_write_t *write_req = (uv_write_t *)ioreq;
629 630 631 632 633
		uint16_t pkt_size = htons(pkt->size);
		uv_buf_t buf[2] = {
			{ (char *)&pkt_size, sizeof(pkt_size) },
			{ (char *)pkt->wire, pkt->size }
		};
634
		write_req->data = task;
635
		ret = uv_write(write_req, (uv_stream_t *)handle, buf, 2, &on_write);
636 637
	} else {
		assert(false);
638
	}
639

640
	if (ret == 0) {
641
		session_touch(session);
642 643 644
		if (session_flags(session)->outgoing) {
			session_tasklist_add(session, task);
		}
645 646
		if (worker->too_many_open &&
		    worker->stats.rconcurrent <
647
			worker->rconcurrent_highwatermark - 10) {
648 649
			worker->too_many_open = false;
		}
650
	} else {
651
		free(ioreq);
652
		qr_task_unref(task);
653 654 655
		if (ret == UV_EMFILE) {
			worker->too_many_open = true;
			worker->rconcurrent_highwatermark = worker->stats.rconcurrent;
656
			ret = kr_error(UV_EMFILE);
657
		}
658
	}
659

660
	/* Update statistics */
661 662
	if (session_flags(session)->outgoing && addr) {
		if (session_flags(session)->has_tls)
663 664
			worker->stats.tls += 1;
		else if (handle->type == UV_UDP)
665
			worker->stats.udp += 1;
666
		else
667
			worker->stats.tcp += 1;
668

669
		if (addr->sa_family == AF_INET6)
670
			worker->stats.ipv6 += 1;
671
		else if (addr->sa_family == AF_INET)
672
			worker->stats.ipv4 += 1;
673
	}
674
	return ret;
675 676
}

677 678
static int session_tls_hs_cb(struct session *session, int status)
{
679
	assert(session_flags(session)->outgoing);
680 681 682 683 684
	uv_handle_t *handle = session_get_handle(session);
	uv_loop_t *loop = handle->loop;
	struct worker_ctx *worker = loop->data;
	struct sockaddr *peer = session_get_peer(session);
	int deletion_res = worker_del_tcp_waiting(worker, peer);
685
	int ret = kr_ok();
686

687
	if (status) {
688
		kr_nsrep_update_rtt(NULL, peer, KR_NS_DEAD,
689 690
				    worker->engine->resolver.cache_rtt,
				    KR_NS_UPDATE_NORESET);
691 692 693 694
		return ret;
	}

	/* handshake was completed successfully */
695
	struct tls_client_ctx_t *tls_client_ctx = session_tls_get_client_ctx(session);
696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713
	struct tls_client_paramlist_entry *tls_params = tls_client_ctx->params;
	gnutls_session_t tls_session = tls_client_ctx->c.tls_session;
	if (gnutls_session_is_resumed(tls_session) != 0) {
		kr_log_verbose("[tls_client] TLS session has resumed\n");
	} else {
		kr_log_verbose("[tls_client] TLS session has not resumed\n");
		/* session wasn't resumed, delete old session data ... */
		if (tls_params->session_data.data != NULL) {
			gnutls_free(tls_params->session_data.data);
			tls_params->session_data.data = NULL;
			tls_params->session_data.size = 0;
		}
		/* ... and get the new session data */
		gnutls_datum_t tls_session_data = { NULL, 0 };
		ret = gnutls_session_get_data2(tls_session, &tls_session_data);
		if (ret == 0) {
			tls_params->session_data = tls_session_data;
		}
714
	}
715

716 717 718 719 720 721 722 723 724 725 726 727 728
	ret = kr_ok();
	if (deletion_res == kr_ok()) {
		/* peer was in the waiting list, add to the connected list. */
		ret = worker_add_tcp_connected(worker, peer, session);
	} else {
		/* peer wasn't in the waiting list.
		 * In this case it must be successful rehandshake.
		 * Peer must be already in the connected list. */
		const char *key = tcpsess_key(peer);
		assert(key);
		assert(map_contains(&worker->tcp_connected, key) != 0);
	}
	if (ret == kr_ok()) {
729 730 731 732 733 734 735 736
		while (!session_waitinglist_is_empty(session)) {
			struct qr_task *t = session_waitinglist_get(session);
			ret = qr_task_send(t, session, NULL, NULL);
			if (ret != 0) {
				break;
			}
			session_waitinglist_pop(session, true);
		}
737 738 739 740 741 742 743 744 745
	} else {
		ret = kr_error(EINVAL);
	}

	if (ret != kr_ok()) {
		/* Something went wrong.
		 * Session isn't in the list of waiting sessions,
		 * or addition to the list of connected sessions failed,
		 * or write to upstream failed. */
746
		worker_del_tcp_connected(worker, peer);
747
		session_waitinglist_finalize(session, KR_STATE_FAIL);
748
		assert(session_tasklist_is_empty(session));
749
		session_close(session);
750
	} else {
751
		session_timer_stop(session);
752
		session_timer_start(session, tcp_timeout_trigger,
753
				    MAX_TCP_INACTIVITY, MAX_TCP_INACTIVITY);
754 755 756 757
	}
	return kr_ok();
}

758

759 760 761
static struct kr_query *task_get_last_pending_query(struct qr_task *task)
{
	if (!task || task->ctx->req.rplan.pending.len == 0) {
762 763 764 765 766 767
		return NULL;
	}

	return array_tail(task->ctx->req.rplan.pending);
}

768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783
static int send_waiting(struct session *session)
{
	int ret = 0;
	while (!session_waitinglist_is_empty(session)) {
		struct qr_task *t = session_waitinglist_get(session);
		ret = qr_task_send(t, session, NULL, NULL);
		if (ret != 0) {
			session_waitinglist_finalize(session, KR_STATE_FAIL);
			session_tasklist_finalize(session, KR_STATE_FAIL);
			session_close(session);
			break;
		}
		session_waitinglist_pop(session, true);
	}
	return ret;
}
784

785
static void on_connect(uv_connect_t *req, int status)
786
{
787
	struct worker_ctx *worker = get_worker();
788
	uv_stream_t *handle = req->handle;
789
	struct session *session = handle->data;
790
	struct sockaddr *peer = session_get_peer(session);
791
	free(req);
792

793
	assert(session_flags(session)->outgoing);
794 795

	if (status == UV_ECANCELED) {
796
		worker_del_tcp_waiting(worker, peer);
797
		assert(session_is_empty(session) && session_flags(session)->closing);
798 799 800
		return;
	}

801
	if (session_flags(session)->closing) {
802 803
		worker_del_tcp_waiting(worker, peer);
		assert(session_is_empty(session));
804 805 806 807
		return;
	}

	if (status != 0) {
808 809
		worker_del_tcp_waiting(worker, peer);
		assert(session_tasklist_is_empty(session));
810
		session_waitinglist_retry(session, false);
811 812 813 814
		session_close(session);
		return;
	}

815
	if (!session_flags(session)->has_tls) {
816 817
		/* if there is a TLS, session still waiting for handshake,
		 * otherwise remove it from waiting list */
818
		if (worker_del_tcp_waiting(worker, peer) != 0) {
819 820
			/* session isn't in list of waiting queries, *
			 * something gone wrong */
821 822
			session_waitinglist_finalize(session, KR_STATE_FAIL);
			assert(session_tasklist_is_empty(session));
823 824 825 826 827
			session_close(session);
			return;
		}
	}

828
	struct qr_task *task = session_waitinglist_get(session);
829
	struct kr_query *qry = task_get_last_pending_query(task);
830
	WITH_VERBOSE (qry) {
831 832 833 834
		struct sockaddr *peer = session_get_peer(session);
		char peer_str[INET6_ADDRSTRLEN];
		inet_ntop(peer->sa_family, kr_inaddr(peer), peer_str, sizeof(peer_str));
		VERBOSE_MSG(qry, "=> connected to '%s'\n", peer_str);
835 836
	}

837
	session_flags(session)->connected = true;
838
	session_start_read(session);
839 840

	int ret = kr_ok();
841
	if (session_flags(session)->has_tls) {
842 843
		struct tls_client_ctx_t *tls_ctx = session_tls_get_client_ctx(session);
		ret = tls_client_connect_start(tls_ctx, session, session_tls_hs_cb);
844
		if (ret == kr_error(EAGAIN)) {
845 846
			session_timer_stop(session);
			session_timer_start(session, tcp_timeout_trigger,
847
					    MAX_TCP_INACTIVITY, MAX_TCP_INACTIVITY);
848 849
			return;
		}
850 851
	} else {
		worker_add_tcp_connected(worker, peer, session);
852
	}
853 854 855 856 857

	ret = send_waiting(session);
	if (ret != 0) {
		worker_del_tcp_connected(worker, peer);
		return;
858
	}
859

860 861
	session_timer_stop(session);
	session_timer_start(session, tcp_timeout_trigger,
862
			    MAX_TCP_INACTIVITY, MAX_TCP_INACTIVITY);
863 864
}

865
static void on_tcp_connect_timeout(uv_timer_t *timer)
866
{
867 868 869
	struct session *session = timer->data;

	uv_timer_stop(timer);
870
	struct worker_ctx *worker = get_worker();
871

872
	assert (session_tasklist_is_empty(session));
873

874 875
	struct sockaddr *peer = session_get_peer(session);
	worker_del_tcp_waiting(worker, peer);
876

877
	struct qr_task *task = session_waitinglist_get(session);
878
	struct kr_query *qry = task_get_last_pending_query(task);
879
	WITH_VERBOSE (qry) {
880 881 882
		char peer_str[INET6_ADDRSTRLEN];
		inet_ntop(peer->sa_family, kr_inaddr(peer), peer_str, sizeof(peer_str));
		VERBOSE_MSG(qry, "=> connection to '%s' failed\n", peer_str);
883
	}
884

885
	kr_nsrep_update_rtt(NULL, peer, KR_NS_DEAD,
886 887
			    worker->engine->resolver.cache_rtt,
			    KR_NS_UPDATE_NORESET);
888

889 890 891
	worker->stats.timeout += session_waitinglist_get_len(session);
	session_waitinglist_retry(session, true);
	assert (session_tasklist_is_empty(session));
892
	session_close(session);
893 894 895
}

/* This is called when I/O timeouts */
896
static void on_udp_timeout(uv_timer_t *timer)
897
{
898
	struct session *session = timer->data;
899 900 901
	assert(session_get_handle(session)->data == session);
	assert(session_tasklist_get_len(session) == 1);
	assert(session_waitinglist_is_empty(session));
902 903

	uv_timer_stop(timer);
904 905

	/* Penalize all tried nameservers with a timeout. */
906
	struct qr_task *task = session_tasklist_get_first(session);
907
	struct worker_ctx *worker = task->ctx->worker;
908
	if (task->leading && task->pending_count > 0) {
909
		struct kr_query *qry = array_tail(task->ctx->req.rplan.pending);
910 911 912
		struct sockaddr_in6 *addrlist = (struct sockaddr_in6 *)task->addrlist;
		for (uint16_t i = 0; i < MIN(task->pending_count, task->addrlist_count); ++i) {
			struct sockaddr *choice = (struct sockaddr *)(&addrlist[i]);
913
			WITH_VERBOSE(qry) {
914 915
				char addr_str[INET6_ADDRSTRLEN];
				inet_ntop(choice->sa_family, kr_inaddr(choice), addr_str, sizeof(addr_str));
916
				VERBOSE_MSG(qry, "=> server: '%s' flagged as 'bad'\n", addr_str);
917
			}
918
			kr_nsrep_update_rtt(&qry->ns, choice, KR_NS_DEAD,
919 920
					    worker->engine->resolver.cache_rtt,
					    KR_NS_UPDATE_NORESET);
921 922 923 924 925
		}
	}
	task->timeouts += 1;
	worker->stats.timeout += 1;
	qr_task_step(task, NULL, NULL);
926 927
}

928
static uv_handle_t *retransmit(struct qr_task *task)
929
{
930
	uv_handle_t *ret = NULL;
931
	if (task && task->addrlist && task->addrlist_count > 0) {
932
		struct sockaddr_in6 *choice = &((struct sockaddr_in6 *)task->addrlist)[task->addrlist_turn];
Grigorii Demidov's avatar
Grigorii Demidov committed
933 934 935
		if (!choice) {
			return ret;
		}
936 937 938
		if (task->pending_count >= MAX_PENDING) {
			return ret;
		}
939 940 941 942 943 944
		/* Checkout answer before sending it */
		struct request_ctx *ctx = task->ctx;
		if (kr_resolve_checkout(&ctx->req, NULL, (struct sockaddr *)choice, SOCK_DGRAM, task->pktbuf) != 0) {
			return ret;
		}
		ret = ioreq_spawn(ctx->worker, SOCK_DGRAM, choice->sin6_family);
945 946 947 948 949
		if (!ret) {
			return ret;
		}
		struct sockaddr *addr = (struct sockaddr *)choice;
		struct session *session = ret->data;
950
		struct sockaddr *peer = session_get_peer(session);
951
		assert (peer->sa_family == AF_UNSPEC && session_flags(session)->outgoing);
952
		memcpy(peer, addr, kr_sockaddr_len(addr));
953 954 955 956 957
		if (qr_task_send(task, session, (struct sockaddr *)choice,
				 task->pktbuf) != 0) {
			session_close(session);
			ret = NULL;
		} else {
958
			task->pending[task->pending_count] = session;
959
			task->pending_count += 1;
960 961
			task->addrlist_turn = (task->addrlist_turn + 1) %
					      task->addrlist_count; /* Round robin */
962
			session_start_read(session); /* Start reading answer */
963 964
		}
	}
965
	return ret;
966 967 968 969
}

static void on_retransmit(uv_timer_t *req)
{
970
	struct session *session = req->data;
971
	assert(session_tasklist_get_len(session) == 1);
972 973

	uv_timer_stop(req);
974
	struct qr_task *task = session_tasklist_get_first(session);
975
	if (retransmit(task) == NULL) {
976 977
		/* Not possible to spawn request, start timeout timer with remaining deadline. */
		uint64_t timeout = KR_CONN_RTT_MAX - task->pending_count * KR_CONN_RETRY;
978
		uv_timer_start(req, on_udp_timeout, timeout, 0);
979 980
	} else {
		uv_timer_start(req, on_retransmit, KR_CONN_RETRY, 0);
981
	}
982 983
}

984
static void subreq_finalize(struct qr_task *task, const struct sockaddr *packet_source, knot_pkt_t *pkt)
985
{
986 987 988
	if (!task || task->finished) {
		return;
	}
989
	/* Close pending timer */
990
	ioreq_kill_pending(task);
991
	/* Clear from outgoing table. */
992 993
	if (!task->leading)
		return;
994 995 996 997 998
	char key[SUBREQ_KEY_LEN];
	const int klen = subreq_key(key, task->pktbuf);
	if (klen > 0) {
		void *val_deleted;
		int ret = trie_del(task->ctx->worker->subreq_out, key, klen, &val_deleted);
999
		assert(ret == KNOT_EOK && val_deleted == task); (void)ret;
1000 1001
	}
	/* Notify waiting tasks. */
1002
	struct kr_query *leader_qry = array_tail(task->ctx->req.rplan.pending);
1003 1004
	for (size_t i = task->waiting.len; i > 0; i--) {
		struct qr_task *follower = task->waiting.at[i - 1];
1005
		/* Reuse MSGID and 0x20 secret */
1006 1007
		if (follower->ctx->req.rplan.pending.len > 0) {
			struct kr_query *qry = array_tail(follower->ctx->req.rplan.pending);
1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021
			qry->id = leader_qry->id;
			qry->secret = leader_qry->secret;
			leader_qry->secret = 0; /* Next will be already decoded */
		}
		qr_task_step(follower, packet_source, pkt);
		qr_task_unref(follower);
	}
	task->waiting.len = 0;
	task->leading = false;
}

static void subreq_lead(struct qr_task *task)
{
	assert(task);
1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032
	char key[SUBREQ_KEY_LEN];
	const int klen = subreq_key(key, task->pktbuf);
	if (klen < 0)
		return;
	struct qr_task **tvp = (struct qr_task **)
		trie_get_ins(task->ctx->worker->subreq_out, key, klen);
	if (unlikely(!tvp))
		return; /*ENOMEM*/
	if (unlikely(*tvp != NULL)) {
		assert(false);
		return;
1033
	}
1034 1035
	*tvp = task;
	task->leading = true;
1036 1037 1038 1039 1040
}

static bool subreq_enqueue(struct qr_task *task)
{
	assert(task);
1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055
	char key[SUBREQ_KEY_LEN];
	const int klen = subreq_key(key, task->pktbuf);
	if (klen < 0)
		return false;
	struct qr_task **leader = (struct qr_task **)
		trie_get_try(task->ctx->worker->subreq_out, key, klen);
	if (!leader /*ENOMEM*/ || !*leader)
		return false;
	/* Enqueue itself to leader for this subrequest. */
	int ret = array_push_mm((*leader)->waiting, task,
				kr_memreserve, &(*leader)->ctx->req.pool);
	if (unlikely(ret < 0)) /*ENOMEM*/
		return false;
	qr_task_ref(task);
	return true;
1056 1057 1058 1059 1060
}

static int qr_task_finalize(struct qr_task *task, int state)
{
	assert(task && task->leading == false);
1061 1062 1063
	if (task->finished) {
		return 0;
	}
1064
	struct request_ctx *ctx = task->ctx;
1065
	struct session *source_session = ctx->source.session;
1066
	kr_resolve_finish(&ctx->req, state);
1067

1068
	task->finished = true;
1069
	if (source_session == NULL) {
1070 1071 1072 1073
		(void) qr_task_on_send(task, NULL, kr_error(EIO));
		return state == KR_STATE_DONE ? 0 : kr_error(EIO);
	}

1074 1075 1076
	/* Reference task as the callback handler can close it */
	qr_task_ref(task);

1077
	/* Send back answer */
1078
	assert(!session_flags(source_session)->closing);
1079
	assert(ctx->source.addr.ip.sa_family != AF_UNSPEC);
1080
	int res = qr_task_send(task, source_session,
1081 1082 1083
			       (struct sockaddr *)&ctx->source.addr,
			        ctx->req.answer);
	if (res != kr_ok()) {
1084
		(void) qr_task_on_send(task, NULL, kr_error(EIO));
1085
		/* Since source session is erroneous detach all tasks. */
1086
		while (!session_tasklist_is_empty(source_session)) {
1087
			struct qr_task *t = session_tasklist_del_first(source_session, false);
1088 1089 1090 1091 1092 1093
			struct request_ctx *c = t->ctx;
			assert(c->source.session == source_session);
			c->source.session = NULL;
			/* Don't finalize them as there can be other tasks
			 * waiting for answer to this particular task.
			 * (ie. task->leading is true) */
1094
			worker_task_unref(t);
1095 1096
		}
		session_close(source_session);
1097
	}
1098

1099 1100
	qr_task_unref(task);

1101
	return state == KR_STATE_DONE ? 0 : kr_error(EIO);
1102
}
1103

1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305
static int udp_task_step(struct qr_task *task,
			 const struct sockaddr *packet_source, knot_pkt_t *packet)
{
	struct request_ctx *ctx = task->ctx;
	struct kr_request *req = &ctx->req;

	/* If there is already outgoing query, enqueue to it. */
	if (subreq_enqueue(task)) {
		return kr_ok(); /* Will be notified when outgoing query finishes. */
	}
	/* Start transmitting */
	uv_handle_t *handle = retransmit(task);
	if (handle == NULL) {
		subreq_finalize(task, packet_source, packet);
		return qr_task_finalize(task, KR_STATE_FAIL);
	}
	/* Check current query NSLIST */
	struct kr_query *qry = array_tail(req->rplan.pending);
	assert(qry != NULL);
	/* Retransmit at default interval, or more frequently if the mean
	 * RTT of the server is better. If the server is glued, use default rate. */
	size_t timeout = qry->ns.score;
	if (timeout > KR_NS_GLUED) {
		/* We don't have information about variance in RTT, expect +10ms */
		timeout = MIN(qry->ns.score + 10, KR_CONN_RETRY);
	} else {
		timeout = KR_CONN_RETRY;
	}
	/* Announce and start subrequest.
	 * @note Only UDP can lead I/O as it doesn't touch 'task->pktbuf' for reassembly.
	 */
	subreq_lead(task);
	struct session *session = handle->data;
	assert(session_get_handle(session) == handle && (handle->type == UV_UDP));
	int ret = session_timer_start(session, on_retransmit, timeout, 0);
	/* Start next step with timeout, fatal if can't start a timer. */
	if (ret != 0) {
		subreq_finalize(task, packet_source, packet);
		return qr_task_finalize(task, KR_STATE_FAIL);
	}
	return kr_ok();
}

static int tcp_task_waiting_connection(struct session *session, struct qr_task *task)
{
	assert(session_flags(session)->outgoing);
	if (session_flags(session)->closing) {
		/* Something went wrong. Better answer with KR_STATE_FAIL.
		 * TODO: normally should not happen,
		 * consider possibility to transform this into
		 * assert(!session_flags(session)->closing). */
		return kr_error(EINVAL);
	}
	/* Add task to the end of list of waiting tasks.
	 * It will be notified in on_connect() or qr_task_on_send(). */
	int ret = session_waitinglist_push(session, task);
	if (ret < 0) {
		return kr_error(EINVAL);
	}
	return kr_ok();
}

static int tcp_task_existing_connection(struct session *session, struct qr_task *task)
{
	assert(session_flags(session)->outgoing);
	struct request_ctx *ctx = task->ctx;
	struct worker_ctx *worker = ctx->worker;

	if (session_flags(session)->closing) {
		/* Something went wrong. Better answer with KR_STATE_FAIL.
		 * TODO: normally should not happen,
		 * consider possibility to transform this into
		 * assert(!session_flags(session)->closing). */
		return kr_error(EINVAL);
	}

	/* If there are any unsent queries, send it first. */
	int ret = send_waiting(session);
	if (ret != 0) {
		return kr_error(EINVAL);
	}

	/* No unsent queries at that point. */
	if (session_tasklist_get_len(session) >= worker->tcp_pipeline_max) {
		/* Too many outstanding queries, answer with SERFVAIL, */
		return kr_error(EINVAL);
	}

	/* Send query to upstream. */
	ret = qr_task_send(task, session, NULL, NULL);
	if (ret != 0) {
		/* Error, finalize task with SERVFAIL and
		 * close connection to upstream. */
		session_tasklist_finalize(session, KR_STATE_FAIL);
		session_close(session);
		return kr_error(EINVAL);
	}

	return kr_ok();
}

static int tcp_task_make_connection(struct session *session, struct qr_task *task,
				    const struct sockaddr *addr /* , knot_pkt_t *packet */)
{
	struct request_ctx *ctx = task->ctx;
	struct worker_ctx *worker = ctx->worker;

	uv_connect_t *conn = malloc(sizeof(uv_connect_t));
	if (!conn) {
		return kr_error(EINVAL);
	}
	uv_handle_t *client = ioreq_spawn(worker, SOCK_STREAM,
						  addr->sa_family);
	if (!client) {
		free(conn);
		return kr_error(EINVAL);
	}
	session = client->data;

	/* Add address to the waiting list.
	 * Now it "is waiting to be connected to." */
	int ret = worker_add_tcp_waiting(ctx->worker, addr, session);
	if (ret < 0) {
		free(conn);
		return kr_error(EINVAL);
	}

	/* Check if there must be TLS */
	struct engine *engine = ctx->worker->engine;
	struct network *net = &engine->net;
	const char *key = tcpsess_key(addr);
	struct tls_client_paramlist_entry *entry = map_get(&net->tls_client_params, key);
	if (entry) {
		/* Address is configured to be used with TLS.
		 * We need to allocate auxiliary data structure. */
		assert(session_tls_get_client_ctx(session) == NULL);
		struct tls_client_ctx_t *tls_ctx = tls_client_ctx_new(entry, worker);
		if (!tls_ctx) {
			worker_del_tcp_waiting(ctx->worker, addr);
			free(conn);
			return kr_error(EINVAL);
		}
		tls_client_ctx_set_session(tls_ctx, session);
		session_tls_set_client_ctx(session, tls_ctx);
		session_flags(session)->has_tls = true;
	}

	conn->data = session;
	/*  Store peer address for the session. */
	struct sockaddr *peer = session_get_peer(session);
	memcpy(peer, addr, kr_sockaddr_len(addr));

	/*  Start watchdog to catch eventual connection timeout. */
	ret = session_timer_start(session, on_tcp_connect_timeout,
				  KR_CONN_RTT_MAX,