1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511
|
/*
* Copyright (c) 2002-2016 Balabit
* Copyright (c) 2016 Viktor Juhasz <viktor.juhasz@balabit.com>
*
* This program is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License version 2 as published
* by the Free Software Foundation, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*
* As an additional exemption you are allowed to compile & link against the
* OpenSSL libraries as published by the OpenSSL project. See the file
* COPYING for details.
*
*/
#include "syslog-ng.h"
#include "qdisk.h"
#include "logpipe.h"
#include "logqueue-disk-reliable.h"
#include "messages.h"
#include "scratch-buffers.h"
/*pessimistic default for reliable disk queue 10000 x 16 kbyte*/
#define PESSIMISTIC_FLOW_CONTROL_WINDOW_BYTES 10000 * 16 *1024
#define ENTRIES_PER_MSG_IN_MEM_Q 3
static inline void
_push_to_memory_queue_tail(GQueue *queue, gint64 position, LogMessage *msg, const LogPathOptions *path_options)
{
gint64 *allocated_position = g_malloc(sizeof(gint64));
*allocated_position = position;
g_queue_push_tail(queue, allocated_position);
g_queue_push_tail(queue, msg);
g_queue_push_tail(queue, LOG_PATH_OPTIONS_TO_POINTER(path_options));
}
static inline void
_pop_from_memory_queue_head(GQueue *queue, gint64 *position, LogMessage **msg, LogPathOptions *path_options)
{
gint64 *allocated_position = g_queue_pop_head(queue);
*position = *allocated_position;
g_free(allocated_position);
*msg = g_queue_pop_head(queue);
POINTER_TO_LOG_PATH_OPTIONS(g_queue_pop_head(queue), path_options);
}
static inline gint64
_peek_memory_queue_head_position(GQueue *queue)
{
gint64 *position = g_queue_peek_head(queue);
return *position;
}
static gboolean
_start(LogQueueDisk *s)
{
return qdisk_start(s->qdisk, NULL, NULL, NULL);
}
static gboolean
_skip_message(LogQueueDisk *self)
{
if (!qdisk_started(self->qdisk))
return FALSE;
return qdisk_remove_head(self->qdisk);
}
static void
_empty_queue(LogQueueDiskReliable *self, GQueue *queue)
{
while (queue && queue->length > 0)
{
gint64 temppos;
LogMessage *msg;
LogPathOptions path_options = LOG_PATH_OPTIONS_INIT;
_pop_from_memory_queue_head(queue, &temppos, &msg, &path_options);
log_queue_memory_usage_sub(&self->super.super, log_msg_get_size(msg));
log_msg_drop(msg, &path_options, AT_PROCESSED);
}
}
static gint64
_get_length(LogQueue *s)
{
LogQueueDiskReliable *self = (LogQueueDiskReliable *)s;
if (!qdisk_started(self->super.qdisk))
return 0;
return qdisk_get_length(self->super.qdisk);
}
static void
_ack_backlog(LogQueue *s, gint num_msg_to_ack)
{
LogQueueDiskReliable *self = (LogQueueDiskReliable *)s;
LogMessage *msg;
LogPathOptions path_options = LOG_PATH_OPTIONS_INIT;
guint i;
g_mutex_lock(&s->lock);
for (i = 0; i < num_msg_to_ack; i++)
{
if (qdisk_get_backlog_head (self->super.qdisk) == qdisk_get_next_head_position(self->super.qdisk))
{
goto exit_reliable;
}
if (self->backlog->length > 0)
{
if (_peek_memory_queue_head_position(self->backlog) == qdisk_get_backlog_head(self->super.qdisk))
{
gint64 position;
_pop_from_memory_queue_head(self->backlog, &position, &msg, &path_options);
log_queue_memory_usage_sub(s, log_msg_get_size(msg));
log_msg_ack(msg, &path_options, AT_PROCESSED);
log_msg_unref(msg);
}
}
qdisk_ack_backlog(self->super.qdisk);
log_queue_disk_update_disk_related_counters(&self->super);
}
exit_reliable:
qdisk_reset_file_if_empty(self->super.qdisk);
g_mutex_unlock(&s->lock);
}
static gint
_find_pos_in_backlog(LogQueueDiskReliable *self, gint64 new_pos)
{
gint result = -1;
int i = 0;
GList *item = self->backlog->tail;
while (result == -1 && item != NULL)
{
GList *pos_i = item->prev->prev;
gint64 *pos = pos_i->data;
if (*pos == new_pos)
{
result = i;
}
item = pos_i->prev;
i++;
}
return result;
}
static void
_move_message_from_backlog_to_flow_control_window(LogQueueDiskReliable *self)
{
gpointer ptr_opt = g_queue_pop_tail(self->backlog);
gpointer ptr_msg = g_queue_pop_tail(self->backlog);
gpointer ptr_pos = g_queue_pop_tail(self->backlog);
g_queue_push_head(self->flow_control_window, ptr_opt);
g_queue_push_head(self->flow_control_window, ptr_msg);
g_queue_push_head(self->flow_control_window, ptr_pos);
}
static void
_rewind_from_backlog(LogQueueDiskReliable *self, gint64 new_pos)
{
gint i;
g_assert((self->backlog->length % 3) == 0);
gint rewind_backlog_queue = _find_pos_in_backlog(self, new_pos);
for (i = 0; i <= rewind_backlog_queue; i++)
{
_move_message_from_backlog_to_flow_control_window(self);
}
}
static void
_rewind_backlog(LogQueue *s, guint rewind_count)
{
LogQueueDiskReliable *self = (LogQueueDiskReliable *)s;
g_mutex_lock(&s->lock);
rewind_count = MIN(rewind_count, qdisk_get_backlog_count(self->super.qdisk));
qdisk_rewind_backlog(self->super.qdisk, rewind_count);
_rewind_from_backlog(self, qdisk_get_next_head_position(self->super.qdisk));
log_queue_queued_messages_add(s, rewind_count);
g_mutex_unlock(&s->lock);
}
static void
_rewind_backlog_all(LogQueue *s)
{
_rewind_backlog(s, G_MAXUINT);
}
static inline gboolean
_is_next_message_in_flow_control_window(LogQueueDiskReliable *self)
{
if (self->flow_control_window->length == 0)
return FALSE;
return _peek_memory_queue_head_position(self->flow_control_window) == qdisk_get_next_head_position(self->super.qdisk);
}
static inline gboolean
_is_next_message_in_front_cache(LogQueueDiskReliable *self)
{
if (self->front_cache->length == 0)
return FALSE;
return _peek_memory_queue_head_position(self->front_cache) == qdisk_get_next_head_position(self->super.qdisk);
}
static LogMessage *
_peek_head(LogQueue *s)
{
LogQueueDiskReliable *self = (LogQueueDiskReliable *)s;
LogMessage *msg = NULL;
g_mutex_lock(&s->lock);
if (_is_next_message_in_flow_control_window(self))
{
msg = g_queue_peek_nth(self->flow_control_window, 1);
goto exit;
}
if (_is_next_message_in_front_cache(self))
{
msg = g_queue_peek_nth(self->front_cache, 1);
goto exit;
}
msg = log_queue_disk_peek_message(&self->super);
exit:
g_mutex_unlock(&s->lock);
return msg;
}
static LogMessage *
_pop_head(LogQueue *s, LogPathOptions *path_options)
{
LogQueueDiskReliable *self = (LogQueueDiskReliable *)s;
LogMessage *msg = NULL;
gboolean qdisk_corrupt = FALSE;
g_mutex_lock(&s->lock);
if (_is_next_message_in_flow_control_window(self))
{
gint64 position;
_pop_from_memory_queue_head(self->flow_control_window, &position, &msg, path_options);
log_queue_memory_usage_sub(s, log_msg_get_size(msg));
if (!_skip_message(&self->super))
qdisk_corrupt = TRUE;
/* push to backlog */
log_msg_ref(msg);
_push_to_memory_queue_tail(self->backlog, position, msg, path_options);
log_queue_memory_usage_add(s, log_msg_get_size(msg));
goto exit;
}
if (_is_next_message_in_front_cache(self))
{
/*
* Fast path: use the message from the memory, saving a disk read and a deserialization.
*/
gint64 position;
_pop_from_memory_queue_head(self->front_cache, &position, &msg, path_options);
log_queue_memory_usage_sub(s, log_msg_get_size(msg));
if (!_skip_message(&self->super))
qdisk_corrupt = TRUE;
goto exit;
}
msg = log_queue_disk_read_message(&self->super, path_options);
exit:
if (!msg)
{
g_mutex_unlock(&s->lock);
return NULL;
}
log_queue_disk_update_disk_related_counters(&self->super);
log_queue_queued_messages_dec(s);
if (qdisk_corrupt)
log_queue_disk_restart_corrupted(&self->super);
g_mutex_unlock(&s->lock);
return msg;
}
static inline gboolean
_is_reserved_buffer_size_reached(LogQueueDiskReliable *self)
{
return qdisk_get_empty_space(self->super.qdisk) < qdisk_get_flow_control_window_bytes(self->super.qdisk);
}
static inline gboolean
_is_space_available_in_front_cache(LogQueueDiskReliable *self)
{
gint num_of_messages_in_front_cache = g_queue_get_length(self->front_cache) / ENTRIES_PER_MSG_IN_MEM_Q;
return num_of_messages_in_front_cache < self->front_cache_size;
}
static void
_push_tail(LogQueue *s, LogMessage *msg, const LogPathOptions *path_options)
{
LogQueueDiskReliable *self = (LogQueueDiskReliable *)s;
ScratchBuffersMarker marker;
GString *serialized_msg = scratch_buffers_alloc_and_mark(&marker);
if (!log_queue_disk_serialize_msg(&self->super, msg, serialized_msg))
{
msg_error("Failed to serialize message for reliable disk-buffer, dropping message",
evt_tag_str("filename", qdisk_get_filename(self->super.qdisk)),
evt_tag_str("persist_name", s->persist_name));
log_queue_disk_drop_message(&self->super, msg, path_options);
scratch_buffers_reclaim_marked(marker);
return;
}
g_mutex_lock(&s->lock);
gint64 message_position = qdisk_get_next_tail_position(self->super.qdisk);
if (!qdisk_push_tail(self->super.qdisk, serialized_msg))
{
EVTTAG *suggestion = NULL;
if (path_options->flow_control_requested)
{
suggestion = evt_tag_str("suggestion", "consider increasing flow-control-window-bytes() or decreasing "
"log-iw-size() values on the source side to avoid message loss");
}
/* we were not able to store the msg, warn */
msg_error("Destination reliable queue full, dropping message",
evt_tag_str("filename", qdisk_get_filename(self->super.qdisk)),
evt_tag_long("queue_len", log_queue_get_length(s)),
evt_tag_int("flow_control_window_bytes", qdisk_get_flow_control_window_bytes(self->super.qdisk)),
evt_tag_long("capacity_bytes", qdisk_get_maximum_size(self->super.qdisk)),
evt_tag_str("persist_name", s->persist_name),
suggestion);
log_queue_disk_drop_message(&self->super, msg, path_options);
scratch_buffers_reclaim_marked(marker);
g_mutex_unlock(&s->lock);
return;
}
log_queue_disk_update_disk_related_counters(&self->super);
scratch_buffers_reclaim_marked(marker);
if (_is_reserved_buffer_size_reached(self))
{
/*
* Keep the message in memory, and do not ack it, so flow-control can kick in.
*/
_push_to_memory_queue_tail(self->flow_control_window, message_position, msg, path_options);
log_queue_memory_usage_add(s, log_msg_get_size(msg));
goto exit;
}
log_msg_ack(msg, path_options, AT_PROCESSED);
if (_is_space_available_in_front_cache(self))
{
/*
* Keep the message in memory for fast-path.
* Set its ack_needed to FALSE, because we have already acked it.
*/
LogPathOptions local_path_options;
log_path_options_chain(&local_path_options, path_options);
local_path_options.ack_needed = FALSE;
_push_to_memory_queue_tail(self->front_cache, message_position, msg, &local_path_options);
log_queue_memory_usage_add(s, log_msg_get_size(msg));
goto exit;
}
log_msg_unref(msg);
exit:
log_queue_queued_messages_inc(s);
/* this releases the queue's lock for a short time, which may violate the
* consistency of the disk-buffer, so it must be the last call under lock in this function
*/
log_queue_push_notify(s);
g_mutex_unlock(&s->lock);
}
static void
_free(LogQueue *s)
{
LogQueueDiskReliable *self = (LogQueueDiskReliable *)s;
if (self->flow_control_window)
{
g_assert(g_queue_is_empty(self->flow_control_window));
g_queue_free(self->flow_control_window);
self->flow_control_window = NULL;
}
if (self->backlog)
{
g_assert(g_queue_is_empty(self->backlog));
g_queue_free(self->backlog);
self->backlog = NULL;
}
if (self->front_cache)
{
g_assert(g_queue_is_empty(self->front_cache));
g_queue_free(self->front_cache);
self->front_cache = NULL;
}
log_queue_disk_free_method(&self->super);
}
static gboolean
_stop(LogQueueDisk *s, gboolean *persistent)
{
LogQueueDiskReliable *self = (LogQueueDiskReliable *) s;
gboolean result = FALSE;
if (qdisk_stop(s->qdisk, NULL, NULL, NULL))
{
*persistent = TRUE;
result = TRUE;
}
_empty_queue(self, self->flow_control_window);
_empty_queue(self, self->front_cache);
_empty_queue(self, self->backlog);
return result;
}
static inline void
_set_logqueue_virtual_functions(LogQueue *s)
{
s->get_length = _get_length;
s->ack_backlog = _ack_backlog;
s->rewind_backlog = _rewind_backlog;
s->rewind_backlog_all = _rewind_backlog_all;
s->pop_head = _pop_head;
s->peek_head = _peek_head;
s->push_tail = _push_tail;
s->free_fn = _free;
}
static inline void
_set_logqueue_disk_virtual_functions(LogQueueDisk *s)
{
s->start = _start;
s->stop = _stop;
}
static inline void
_set_virtual_functions(LogQueueDiskReliable *self)
{
_set_logqueue_virtual_functions(&self->super.super);
_set_logqueue_disk_virtual_functions(&self->super);
}
LogQueue *
log_queue_disk_reliable_new(DiskQueueOptions *options, const gchar *filename, const gchar *persist_name,
gint stats_level, StatsClusterKeyBuilder *driver_sck_builder,
StatsClusterKeyBuilder *queue_sck_builder)
{
g_assert(options->reliable == TRUE);
LogQueueDiskReliable *self = g_new0(LogQueueDiskReliable, 1);
log_queue_disk_init_instance(&self->super, options, "SLRQ", filename, persist_name, stats_level,
driver_sck_builder, queue_sck_builder);
if (options->flow_control_window_bytes < 0)
{
options->flow_control_window_bytes = PESSIMISTIC_FLOW_CONTROL_WINDOW_BYTES;
}
self->flow_control_window = g_queue_new();
self->backlog = g_queue_new();
self->front_cache = g_queue_new();
self->front_cache_size = options->front_cache_size;
_set_virtual_functions(self);
return &self->super.super;
}
|