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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013
|
/*
ctdb lock handling
provide API to do non-blocking locks for single or all databases
Copyright (C) Amitay Isaacs 2012
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
#include "includes.h"
#include "include/ctdb_private.h"
#include "include/ctdb_protocol.h"
#include "tevent.h"
#include "tdb.h"
#include "db_wrap.h"
#include "system/filesys.h"
#include "lib/util/dlinklist.h"
/*
* Non-blocking Locking API
*
* 1. Create a child process to do blocking locks.
* 2. Once the locks are obtained, signal parent process via fd.
* 3. Invoke registered callback routine with locking status.
* 4. If the child process cannot get locks within certain time,
* execute an external script to debug.
*
* ctdb_lock_record() - get a lock on a record
* ctdb_lock_db() - get a lock on a DB
* ctdb_lock_alldb_prio() - get a lock on all DBs with given priority
* ctdb_lock_alldb() - get a lock on all DBs
*
* auto_mark - whether to mark/unmark DBs in before/after callback
*/
enum lock_type {
LOCK_RECORD,
LOCK_DB,
LOCK_ALLDB_PRIO,
LOCK_ALLDB,
};
static const char * const lock_type_str[] = {
"lock_record",
"lock_db",
"lock_alldb_prio",
"lock_alldb",
};
struct lock_request;
/* lock_context is the common part for a lock request */
struct lock_context {
struct lock_context *next, *prev;
enum lock_type type;
struct ctdb_context *ctdb;
struct ctdb_db_context *ctdb_db;
TDB_DATA key;
uint32_t priority;
bool auto_mark;
struct lock_request *request;
pid_t child;
int fd[2];
struct tevent_fd *tfd;
struct tevent_timer *ttimer;
struct timeval start_time;
uint32_t key_hash;
bool can_schedule;
};
/* lock_request is the client specific part for a lock request */
struct lock_request {
struct lock_context *lctx;
void (*callback)(void *, bool);
void *private_data;
};
/*
* Support samba 3.6.x (and older) versions which do not set db priority.
*
* By default, all databases are set to priority 1. So only when priority
* is set to 1, check for databases that need higher priority.
*/
static bool later_db(struct ctdb_context *ctdb, const char *name)
{
if (ctdb->tunable.samba3_hack == 0) {
return false;
}
if (strstr(name, "brlock") ||
strstr(name, "g_lock") ||
strstr(name, "notify_onelevel") ||
strstr(name, "serverid") ||
strstr(name, "xattr_tdb")) {
return true;
}
return false;
}
typedef int (*db_handler_t)(struct ctdb_db_context *ctdb_db,
uint32_t priority,
void *private_data);
static int ctdb_db_iterator(struct ctdb_context *ctdb, uint32_t priority,
db_handler_t handler, void *private_data)
{
struct ctdb_db_context *ctdb_db;
int ret;
for (ctdb_db = ctdb->db_list; ctdb_db; ctdb_db = ctdb_db->next) {
if (ctdb_db->priority != priority) {
continue;
}
if (later_db(ctdb, ctdb_db->db_name)) {
continue;
}
ret = handler(ctdb_db, priority, private_data);
if (ret != 0) {
return -1;
}
}
/* If priority != 1, later_db check is not required and can return */
if (priority != 1) {
return 0;
}
for (ctdb_db = ctdb->db_list; ctdb_db; ctdb_db = ctdb_db->next) {
if (!later_db(ctdb, ctdb_db->db_name)) {
continue;
}
ret = handler(ctdb_db, priority, private_data);
if (ret != 0) {
return -1;
}
}
return 0;
}
/*
* lock all databases - mark only
*/
static int db_lock_mark_handler(struct ctdb_db_context *ctdb_db, uint32_t priority,
void *private_data)
{
int tdb_transaction_write_lock_mark(struct tdb_context *);
DEBUG(DEBUG_INFO, ("marking locked database %s, priority:%u\n",
ctdb_db->db_name, priority));
if (tdb_transaction_write_lock_mark(ctdb_db->ltdb->tdb) != 0) {
DEBUG(DEBUG_ERR, ("Failed to mark (transaction lock) database %s\n",
ctdb_db->db_name));
return -1;
}
if (tdb_lockall_mark(ctdb_db->ltdb->tdb) != 0) {
DEBUG(DEBUG_ERR, ("Failed to mark (all lock) database %s\n",
ctdb_db->db_name));
return -1;
}
return 0;
}
int ctdb_lockall_mark_prio(struct ctdb_context *ctdb, uint32_t priority)
{
/*
* This function is only used by the main dameon during recovery.
* At this stage, the databases have already been locked, by a
* dedicated child process. The freeze_mode variable is used to track
* whether the actual locks are held by the child process or not.
*/
if (ctdb->freeze_mode[priority] != CTDB_FREEZE_FROZEN) {
DEBUG(DEBUG_ERR, ("Attempt to mark all databases locked when not frozen\n"));
return -1;
}
return ctdb_db_iterator(ctdb, priority, db_lock_mark_handler, NULL);
}
static int ctdb_lockall_mark(struct ctdb_context *ctdb)
{
uint32_t priority;
for (priority=1; priority<=NUM_DB_PRIORITIES; priority++) {
if (ctdb_db_iterator(ctdb, priority, db_lock_mark_handler, NULL) != 0) {
return -1;
}
}
return 0;
}
/*
* lock all databases - unmark only
*/
static int db_lock_unmark_handler(struct ctdb_db_context *ctdb_db, uint32_t priority,
void *private_data)
{
int tdb_transaction_write_lock_unmark(struct tdb_context *);
DEBUG(DEBUG_INFO, ("unmarking locked database %s, priority:%u\n",
ctdb_db->db_name, priority));
if (tdb_transaction_write_lock_unmark(ctdb_db->ltdb->tdb) != 0) {
DEBUG(DEBUG_ERR, ("Failed to unmark (transaction lock) database %s\n",
ctdb_db->db_name));
return -1;
}
if (tdb_lockall_unmark(ctdb_db->ltdb->tdb) != 0) {
DEBUG(DEBUG_ERR, ("Failed to unmark (all lock) database %s\n",
ctdb_db->db_name));
return -1;
}
return 0;
}
int ctdb_lockall_unmark_prio(struct ctdb_context *ctdb, uint32_t priority)
{
/*
* This function is only used by the main dameon during recovery.
* At this stage, the databases have already been locked, by a
* dedicated child process. The freeze_mode variable is used to track
* whether the actual locks are held by the child process or not.
*/
if (ctdb->freeze_mode[priority] != CTDB_FREEZE_FROZEN) {
DEBUG(DEBUG_ERR, ("Attempt to unmark all databases locked when not frozen\n"));
return -1;
}
return ctdb_db_iterator(ctdb, priority, db_lock_unmark_handler, NULL);
}
static int ctdb_lockall_unmark(struct ctdb_context *ctdb)
{
uint32_t priority;
for (priority=NUM_DB_PRIORITIES; priority>0; priority--) {
if (ctdb_db_iterator(ctdb, priority, db_lock_unmark_handler, NULL) != 0) {
return -1;
}
}
return 0;
}
static void ctdb_lock_schedule(struct ctdb_context *ctdb);
/*
* Destructor to kill the child locking process
*/
static int ctdb_lock_context_destructor(struct lock_context *lock_ctx)
{
if (lock_ctx->request) {
lock_ctx->request->lctx = NULL;
}
if (lock_ctx->child > 0) {
ctdb_kill(lock_ctx->ctdb, lock_ctx->child, SIGKILL);
if (lock_ctx->type == LOCK_RECORD) {
DLIST_REMOVE(lock_ctx->ctdb_db->lock_current, lock_ctx);
} else {
DLIST_REMOVE(lock_ctx->ctdb->lock_current, lock_ctx);
}
if (lock_ctx->ctdb_db) {
lock_ctx->ctdb_db->lock_num_current--;
}
CTDB_DECREMENT_STAT(lock_ctx->ctdb, locks.num_current);
if (lock_ctx->ctdb_db) {
CTDB_DECREMENT_DB_STAT(lock_ctx->ctdb_db, locks.num_current);
}
} else {
if (lock_ctx->type == LOCK_RECORD) {
DLIST_REMOVE(lock_ctx->ctdb_db->lock_pending, lock_ctx);
} else {
DLIST_REMOVE(lock_ctx->ctdb->lock_pending, lock_ctx);
}
CTDB_DECREMENT_STAT(lock_ctx->ctdb, locks.num_pending);
if (lock_ctx->ctdb_db) {
CTDB_DECREMENT_DB_STAT(lock_ctx->ctdb_db, locks.num_pending);
}
}
ctdb_lock_schedule(lock_ctx->ctdb);
return 0;
}
/*
* Destructor to remove lock request
*/
static int ctdb_lock_request_destructor(struct lock_request *lock_request)
{
TALLOC_FREE(lock_request->lctx);
return 0;
}
/*
* Process all the callbacks waiting for lock
*
* If lock has failed, callback is executed with locked=false
*/
static void process_callbacks(struct lock_context *lock_ctx, bool locked)
{
struct lock_request *request;
if (lock_ctx->auto_mark && locked) {
switch (lock_ctx->type) {
case LOCK_RECORD:
tdb_chainlock_mark(lock_ctx->ctdb_db->ltdb->tdb, lock_ctx->key);
break;
case LOCK_DB:
tdb_lockall_mark(lock_ctx->ctdb_db->ltdb->tdb);
break;
case LOCK_ALLDB_PRIO:
ctdb_lockall_mark_prio(lock_ctx->ctdb, lock_ctx->priority);
break;
case LOCK_ALLDB:
ctdb_lockall_mark(lock_ctx->ctdb);
break;
}
}
request = lock_ctx->request;
if (lock_ctx->auto_mark) {
/* Reset the destructor, so request is not removed from the list */
talloc_set_destructor(request, NULL);
}
request->callback(request->private_data, locked);
if (lock_ctx->auto_mark && locked) {
switch (lock_ctx->type) {
case LOCK_RECORD:
tdb_chainlock_unmark(lock_ctx->ctdb_db->ltdb->tdb, lock_ctx->key);
break;
case LOCK_DB:
tdb_lockall_unmark(lock_ctx->ctdb_db->ltdb->tdb);
break;
case LOCK_ALLDB_PRIO:
ctdb_lockall_unmark_prio(lock_ctx->ctdb, lock_ctx->priority);
break;
case LOCK_ALLDB:
ctdb_lockall_unmark(lock_ctx->ctdb);
break;
}
}
}
static int lock_bucket_id(double t)
{
double ms = 1.e-3, s = 1;
int id;
if (t < 1*ms) {
id = 0;
} else if (t < 10*ms) {
id = 1;
} else if (t < 100*ms) {
id = 2;
} else if (t < 1*s) {
id = 3;
} else if (t < 2*s) {
id = 4;
} else if (t < 4*s) {
id = 5;
} else if (t < 8*s) {
id = 6;
} else if (t < 16*s) {
id = 7;
} else if (t < 32*s) {
id = 8;
} else if (t < 64*s) {
id = 9;
} else {
id = 10;
}
return id;
}
/*
* Callback routine when the required locks are obtained.
* Called from parent context
*/
static void ctdb_lock_handler(struct tevent_context *ev,
struct tevent_fd *tfd,
uint16_t flags,
void *private_data)
{
struct lock_context *lock_ctx;
TALLOC_CTX *tmp_ctx = NULL;
char c;
bool locked;
double t;
int id;
lock_ctx = talloc_get_type_abort(private_data, struct lock_context);
/* cancel the timeout event */
TALLOC_FREE(lock_ctx->ttimer);
t = timeval_elapsed(&lock_ctx->start_time);
id = lock_bucket_id(t);
if (lock_ctx->auto_mark) {
tmp_ctx = talloc_new(ev);
talloc_steal(tmp_ctx, lock_ctx);
}
/* Read the status from the child process */
if (sys_read(lock_ctx->fd[0], &c, 1) != 1) {
locked = false;
} else {
locked = (c == 0 ? true : false);
}
/* Update statistics */
CTDB_INCREMENT_STAT(lock_ctx->ctdb, locks.num_calls);
if (lock_ctx->ctdb_db) {
CTDB_INCREMENT_DB_STAT(lock_ctx->ctdb_db, locks.num_calls);
}
if (locked) {
if (lock_ctx->ctdb_db) {
CTDB_INCREMENT_STAT(lock_ctx->ctdb, locks.buckets[id]);
CTDB_UPDATE_LATENCY(lock_ctx->ctdb, lock_ctx->ctdb_db,
lock_type_str[lock_ctx->type], locks.latency,
lock_ctx->start_time);
CTDB_UPDATE_DB_LATENCY(lock_ctx->ctdb_db, lock_type_str[lock_ctx->type], locks.latency, t);
CTDB_INCREMENT_DB_STAT(lock_ctx->ctdb_db, locks.buckets[id]);
}
} else {
CTDB_INCREMENT_STAT(lock_ctx->ctdb, locks.num_failed);
if (lock_ctx->ctdb_db) {
CTDB_INCREMENT_DB_STAT(lock_ctx->ctdb_db, locks.num_failed);
}
}
process_callbacks(lock_ctx, locked);
if (lock_ctx->auto_mark) {
talloc_free(tmp_ctx);
}
}
/*
* Callback routine when required locks are not obtained within timeout
* Called from parent context
*/
static void ctdb_lock_timeout_handler(struct tevent_context *ev,
struct tevent_timer *ttimer,
struct timeval current_time,
void *private_data)
{
static const char * debug_locks = NULL;
struct lock_context *lock_ctx;
struct ctdb_context *ctdb;
pid_t pid;
lock_ctx = talloc_get_type_abort(private_data, struct lock_context);
ctdb = lock_ctx->ctdb;
/* If a node stopped/banned, don't spam the logs */
if (ctdb->nodes[ctdb->pnn]->flags & NODE_FLAGS_INACTIVE) {
lock_ctx->ttimer = NULL;
return;
}
if (lock_ctx->ctdb_db) {
DEBUG(DEBUG_WARNING,
("Unable to get %s lock on database %s for %.0lf seconds\n",
(lock_ctx->type == LOCK_RECORD ? "RECORD" : "DB"),
lock_ctx->ctdb_db->db_name,
timeval_elapsed(&lock_ctx->start_time)));
} else {
DEBUG(DEBUG_WARNING,
("Unable to get ALLDB locks for %.0lf seconds\n",
timeval_elapsed(&lock_ctx->start_time)));
}
/* Fire a child process to find the blocking process. */
if (debug_locks == NULL) {
debug_locks = getenv("CTDB_DEBUG_LOCKS");
if (debug_locks == NULL) {
debug_locks = talloc_asprintf(ctdb,
"%s/debug_locks.sh",
getenv("CTDB_BASE"));
}
}
if (debug_locks != NULL) {
pid = vfork();
if (pid == 0) {
execl(debug_locks, debug_locks, NULL);
_exit(0);
}
ctdb_track_child(ctdb, pid);
} else {
DEBUG(DEBUG_WARNING,
(__location__
" Unable to setup lock debugging - no memory?\n"));
}
/* reset the timeout timer */
// talloc_free(lock_ctx->ttimer);
lock_ctx->ttimer = tevent_add_timer(ctdb->ev,
lock_ctx,
timeval_current_ofs(10, 0),
ctdb_lock_timeout_handler,
(void *)lock_ctx);
}
static int db_count_handler(struct ctdb_db_context *ctdb_db, uint32_t priority,
void *private_data)
{
int *count = (int *)private_data;
(*count)++;
return 0;
}
struct db_namelist {
const char **names;
int n;
};
static int db_name_handler(struct ctdb_db_context *ctdb_db, uint32_t priority,
void *private_data)
{
struct db_namelist *list = (struct db_namelist *)private_data;
list->names[list->n] = talloc_strdup(list->names, ctdb_db->db_path);
list->n++;
return 0;
}
static bool lock_helper_args(TALLOC_CTX *mem_ctx,
struct lock_context *lock_ctx, int fd,
int *argc, const char ***argv)
{
struct ctdb_context *ctdb = lock_ctx->ctdb;
const char **args = NULL;
int nargs, i;
int priority;
struct db_namelist list;
switch (lock_ctx->type) {
case LOCK_RECORD:
nargs = 5;
break;
case LOCK_DB:
nargs = 4;
break;
case LOCK_ALLDB_PRIO:
nargs = 3;
ctdb_db_iterator(ctdb, lock_ctx->priority, db_count_handler, &nargs);
break;
case LOCK_ALLDB:
nargs = 3;
for (priority=1; priority<NUM_DB_PRIORITIES; priority++) {
ctdb_db_iterator(ctdb, priority, db_count_handler, &nargs);
}
break;
}
/* Add extra argument for null termination */
nargs++;
args = talloc_array(mem_ctx, const char *, nargs);
if (args == NULL) {
return false;
}
args[0] = talloc_asprintf(args, "%d", getpid());
args[1] = talloc_asprintf(args, "%d", fd);
switch (lock_ctx->type) {
case LOCK_RECORD:
args[2] = talloc_strdup(args, "RECORD");
args[3] = talloc_strdup(args, lock_ctx->ctdb_db->db_path);
if (lock_ctx->key.dsize == 0) {
args[4] = talloc_strdup(args, "NULL");
} else {
args[4] = hex_encode_talloc(args, lock_ctx->key.dptr, lock_ctx->key.dsize);
}
break;
case LOCK_DB:
args[2] = talloc_strdup(args, "DB");
args[3] = talloc_strdup(args, lock_ctx->ctdb_db->db_path);
break;
case LOCK_ALLDB_PRIO:
args[2] = talloc_strdup(args, "DB");
list.names = args;
list.n = 3;
ctdb_db_iterator(ctdb, lock_ctx->priority, db_name_handler, &list);
break;
case LOCK_ALLDB:
args[2] = talloc_strdup(args, "DB");
list.names = args;
list.n = 3;
for (priority=1; priority<NUM_DB_PRIORITIES; priority++) {
ctdb_db_iterator(ctdb, priority, db_name_handler, &list);
}
break;
}
/* Make sure last argument is NULL */
args[nargs-1] = NULL;
for (i=0; i<nargs-1; i++) {
if (args[i] == NULL) {
talloc_free(args);
return false;
}
}
*argc = nargs;
*argv = args;
return true;
}
/*
* Find a lock request that can be scheduled
*/
struct lock_context *ctdb_find_lock_context(struct ctdb_context *ctdb)
{
struct lock_context *lock_ctx, *next_ctx;
struct ctdb_db_context *ctdb_db;
/* First check if there are database lock requests */
for (lock_ctx = ctdb->lock_pending; lock_ctx != NULL;
lock_ctx = next_ctx) {
if (lock_ctx->request != NULL) {
/* Found a lock context with a request */
return lock_ctx;
}
next_ctx = lock_ctx->next;
DEBUG(DEBUG_INFO, ("Removing lock context without lock "
"request\n"));
DLIST_REMOVE(ctdb->lock_pending, lock_ctx);
CTDB_DECREMENT_STAT(ctdb, locks.num_pending);
if (lock_ctx->ctdb_db) {
CTDB_DECREMENT_DB_STAT(lock_ctx->ctdb_db,
locks.num_pending);
}
talloc_free(lock_ctx);
}
/* Next check database queues */
for (ctdb_db = ctdb->db_list; ctdb_db; ctdb_db = ctdb_db->next) {
if (ctdb_db->lock_num_current ==
ctdb->tunable.lock_processes_per_db) {
continue;
}
for (lock_ctx = ctdb_db->lock_pending; lock_ctx != NULL;
lock_ctx = next_ctx) {
next_ctx = lock_ctx->next;
if (lock_ctx->request != NULL) {
return lock_ctx;
}
DEBUG(DEBUG_INFO, ("Removing lock context without "
"lock request\n"));
DLIST_REMOVE(ctdb_db->lock_pending, lock_ctx);
CTDB_DECREMENT_STAT(ctdb, locks.num_pending);
CTDB_DECREMENT_DB_STAT(ctdb_db, locks.num_pending);
talloc_free(lock_ctx);
}
}
return NULL;
}
/*
* Schedule a new lock child process
* Set up callback handler and timeout handler
*/
static void ctdb_lock_schedule(struct ctdb_context *ctdb)
{
struct lock_context *lock_ctx;
int ret, argc;
TALLOC_CTX *tmp_ctx;
const char *helper = BINDIR "/ctdb_lock_helper";
static const char *prog = NULL;
const char **args;
if (prog == NULL) {
const char *t;
t = getenv("CTDB_LOCK_HELPER");
if (t != NULL) {
prog = talloc_strdup(ctdb, t);
} else {
prog = talloc_strdup(ctdb, helper);
}
CTDB_NO_MEMORY_VOID(ctdb, prog);
}
/* Find a lock context with requests */
lock_ctx = ctdb_find_lock_context(ctdb);
if (lock_ctx == NULL) {
return;
}
lock_ctx->child = -1;
ret = pipe(lock_ctx->fd);
if (ret != 0) {
DEBUG(DEBUG_ERR, ("Failed to create pipe in ctdb_lock_schedule\n"));
return;
}
set_close_on_exec(lock_ctx->fd[0]);
/* Create data for child process */
tmp_ctx = talloc_new(lock_ctx);
if (tmp_ctx == NULL) {
DEBUG(DEBUG_ERR, ("Failed to allocate memory for helper args\n"));
close(lock_ctx->fd[0]);
close(lock_ctx->fd[1]);
return;
}
/* Create arguments for lock helper */
if (!lock_helper_args(tmp_ctx, lock_ctx, lock_ctx->fd[1],
&argc, &args)) {
DEBUG(DEBUG_ERR, ("Failed to create lock helper args\n"));
close(lock_ctx->fd[0]);
close(lock_ctx->fd[1]);
talloc_free(tmp_ctx);
return;
}
if (!ctdb_vfork_with_logging(lock_ctx, ctdb, "lock_helper",
prog, argc, (const char **)args,
NULL, NULL, &lock_ctx->child)) {
DEBUG(DEBUG_ERR, ("Failed to create a child in ctdb_lock_schedule\n"));
close(lock_ctx->fd[0]);
close(lock_ctx->fd[1]);
talloc_free(tmp_ctx);
return;
}
/* Parent process */
close(lock_ctx->fd[1]);
talloc_set_destructor(lock_ctx, ctdb_lock_context_destructor);
talloc_free(tmp_ctx);
/* Set up timeout handler */
lock_ctx->ttimer = tevent_add_timer(ctdb->ev,
lock_ctx,
timeval_current_ofs(10, 0),
ctdb_lock_timeout_handler,
(void *)lock_ctx);
if (lock_ctx->ttimer == NULL) {
ctdb_kill(ctdb, lock_ctx->child, SIGKILL);
lock_ctx->child = -1;
talloc_set_destructor(lock_ctx, NULL);
close(lock_ctx->fd[0]);
return;
}
/* Set up callback */
lock_ctx->tfd = tevent_add_fd(ctdb->ev,
lock_ctx,
lock_ctx->fd[0],
EVENT_FD_READ,
ctdb_lock_handler,
(void *)lock_ctx);
if (lock_ctx->tfd == NULL) {
TALLOC_FREE(lock_ctx->ttimer);
ctdb_kill(ctdb, lock_ctx->child, SIGKILL);
lock_ctx->child = -1;
talloc_set_destructor(lock_ctx, NULL);
close(lock_ctx->fd[0]);
return;
}
tevent_fd_set_auto_close(lock_ctx->tfd);
/* Move the context from pending to current */
if (lock_ctx->type == LOCK_RECORD) {
DLIST_REMOVE(lock_ctx->ctdb_db->lock_pending, lock_ctx);
DLIST_ADD_END(lock_ctx->ctdb_db->lock_current, lock_ctx, NULL);
} else {
DLIST_REMOVE(ctdb->lock_pending, lock_ctx);
DLIST_ADD_END(ctdb->lock_current, lock_ctx, NULL);
}
CTDB_DECREMENT_STAT(lock_ctx->ctdb, locks.num_pending);
CTDB_INCREMENT_STAT(lock_ctx->ctdb, locks.num_current);
if (lock_ctx->ctdb_db) {
lock_ctx->ctdb_db->lock_num_current++;
CTDB_DECREMENT_DB_STAT(lock_ctx->ctdb_db, locks.num_pending);
CTDB_INCREMENT_DB_STAT(lock_ctx->ctdb_db, locks.num_current);
}
}
/*
* Lock record / db depending on type
*/
static struct lock_request *ctdb_lock_internal(TALLOC_CTX *mem_ctx,
struct ctdb_context *ctdb,
struct ctdb_db_context *ctdb_db,
TDB_DATA key,
uint32_t priority,
void (*callback)(void *, bool),
void *private_data,
enum lock_type type,
bool auto_mark)
{
struct lock_context *lock_ctx = NULL;
struct lock_request *request;
if (callback == NULL) {
DEBUG(DEBUG_WARNING, ("No callback function specified, not locking\n"));
return NULL;
}
lock_ctx = talloc_zero(ctdb, struct lock_context);
if (lock_ctx == NULL) {
DEBUG(DEBUG_ERR, ("Failed to create a new lock context\n"));
return NULL;
}
if ((request = talloc_zero(mem_ctx, struct lock_request)) == NULL) {
talloc_free(lock_ctx);
return NULL;
}
lock_ctx->type = type;
lock_ctx->ctdb = ctdb;
lock_ctx->ctdb_db = ctdb_db;
lock_ctx->key.dsize = key.dsize;
if (key.dsize > 0) {
lock_ctx->key.dptr = talloc_memdup(lock_ctx, key.dptr, key.dsize);
if (lock_ctx->key.dptr == NULL) {
DEBUG(DEBUG_ERR, (__location__ "Memory allocation error\n"));
talloc_free(lock_ctx);
return NULL;
}
lock_ctx->key_hash = ctdb_hash(&key);
} else {
lock_ctx->key.dptr = NULL;
}
lock_ctx->priority = priority;
lock_ctx->auto_mark = auto_mark;
lock_ctx->request = request;
lock_ctx->child = -1;
/* Non-record locks are required by recovery and should be scheduled
* immediately, so keep them at the head of the pending queue.
*/
if (lock_ctx->type == LOCK_RECORD) {
DLIST_ADD_END(ctdb_db->lock_pending, lock_ctx, NULL);
} else {
DLIST_ADD_END(ctdb->lock_pending, lock_ctx, NULL);
}
CTDB_INCREMENT_STAT(ctdb, locks.num_pending);
if (ctdb_db) {
CTDB_INCREMENT_DB_STAT(ctdb_db, locks.num_pending);
}
/* Start the timer when we activate the context */
lock_ctx->start_time = timeval_current();
request->lctx = lock_ctx;
request->callback = callback;
request->private_data = private_data;
talloc_set_destructor(request, ctdb_lock_request_destructor);
ctdb_lock_schedule(ctdb);
return request;
}
/*
* obtain a lock on a record in a database
*/
struct lock_request *ctdb_lock_record(TALLOC_CTX *mem_ctx,
struct ctdb_db_context *ctdb_db,
TDB_DATA key,
bool auto_mark,
void (*callback)(void *, bool),
void *private_data)
{
return ctdb_lock_internal(mem_ctx,
ctdb_db->ctdb,
ctdb_db,
key,
0,
callback,
private_data,
LOCK_RECORD,
auto_mark);
}
/*
* obtain a lock on a database
*/
struct lock_request *ctdb_lock_db(TALLOC_CTX *mem_ctx,
struct ctdb_db_context *ctdb_db,
bool auto_mark,
void (*callback)(void *, bool),
void *private_data)
{
return ctdb_lock_internal(mem_ctx,
ctdb_db->ctdb,
ctdb_db,
tdb_null,
0,
callback,
private_data,
LOCK_DB,
auto_mark);
}
/*
* obtain locks on all databases of specified priority
*/
struct lock_request *ctdb_lock_alldb_prio(TALLOC_CTX *mem_ctx,
struct ctdb_context *ctdb,
uint32_t priority,
bool auto_mark,
void (*callback)(void *, bool),
void *private_data)
{
if (priority < 1 || priority > NUM_DB_PRIORITIES) {
DEBUG(DEBUG_ERR, ("Invalid db priority: %u\n", priority));
return NULL;
}
return ctdb_lock_internal(mem_ctx,
ctdb,
NULL,
tdb_null,
priority,
callback,
private_data,
LOCK_ALLDB_PRIO,
auto_mark);
}
/*
* obtain locks on all databases
*/
struct lock_request *ctdb_lock_alldb(TALLOC_CTX *mem_ctx,
struct ctdb_context *ctdb,
bool auto_mark,
void (*callback)(void *, bool),
void *private_data)
{
return ctdb_lock_internal(mem_ctx,
ctdb,
NULL,
tdb_null,
0,
callback,
private_data,
LOCK_ALLDB,
auto_mark);
}
|