?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
lib: Fix blank line endings
[Samba.git] / ctdb / server / ctdb_call.c
blob1830b4a7c5b9b0f22ff76bb35dd88f3dcf75a434
1 /*
2 ctdb_call protocol code
3
4 Copyright (C) Andrew Tridgell 2006
5
6 This program is free software; you can redistribute it and/or modify
7 it under the terms of the GNU General Public License as published by
8 the Free Software Foundation; either version 3 of the License, or
9 (at your option) any later version.
10
11 This program is distributed in the hope that it will be useful,
12 but WITHOUT ANY WARRANTY; without even the implied warranty of
13 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 GNU General Public License for more details.
15
16 You should have received a copy of the GNU General Public License
17 along with this program; if not, see <http://www.gnu.org/licenses/>.
18 */
19 /*
20 see http://wiki.samba.org/index.php/Samba_%26_Clustering for
21 protocol design and packet details
22 */
23 #include "includes.h"
24 #include "tdb.h"
25 #include "lib/util/dlinklist.h"
26 #include "system/network.h"
27 #include "system/filesys.h"
28 #include "../include/ctdb_private.h"
29 #include "../common/rb_tree.h"
30
31 struct ctdb_sticky_record {
32 struct ctdb_context *ctdb;
33 struct ctdb_db_context *ctdb_db;
34 TDB_CONTEXT *pindown;
35 };
36
37 /*
38 find the ctdb_db from a db index
39 */
40 struct ctdb_db_context *find_ctdb_db(struct ctdb_context *ctdb, uint32_t id)
41 {
42 struct ctdb_db_context *ctdb_db;
43
44 for (ctdb_db=ctdb->db_list; ctdb_db; ctdb_db=ctdb_db->next) {
45 if (ctdb_db->db_id == id) {
46 break;
47 }
48 }
49 return ctdb_db;
50 }
51
52 /*
53 a varient of input packet that can be used in lock requeue
54 */
55 static void ctdb_call_input_pkt(void *p, struct ctdb_req_header *hdr)
56 {
57 struct ctdb_context *ctdb = talloc_get_type(p, struct ctdb_context);
58 ctdb_input_pkt(ctdb, hdr);
59 }
60
61
62 /*
63 send an error reply
64 */
65 static void ctdb_send_error(struct ctdb_context *ctdb,
66 struct ctdb_req_header *hdr, uint32_t status,
67 const char *fmt, ...) PRINTF_ATTRIBUTE(4,5);
68 static void ctdb_send_error(struct ctdb_context *ctdb,
69 struct ctdb_req_header *hdr, uint32_t status,
70 const char *fmt, ...)
71 {
72 va_list ap;
73 struct ctdb_reply_error *r;
74 char *msg;
75 int msglen, len;
76
77 if (ctdb->methods == NULL) {
78 DEBUG(DEBUG_INFO,(__location__ " Failed to send error. Transport is DOWN\n"));
79 return;
80 }
81
82 va_start(ap, fmt);
83 msg = talloc_vasprintf(ctdb, fmt, ap);
84 if (msg == NULL) {
85 ctdb_fatal(ctdb, "Unable to allocate error in ctdb_send_error\n");
86 }
87 va_end(ap);
88
89 msglen = strlen(msg)+1;
90 len = offsetof(struct ctdb_reply_error, msg);
91 r = ctdb_transport_allocate(ctdb, msg, CTDB_REPLY_ERROR, len + msglen,
92 struct ctdb_reply_error);
93 CTDB_NO_MEMORY_FATAL(ctdb, r);
94
95 r->hdr.destnode = hdr->srcnode;
96 r->hdr.reqid = hdr->reqid;
97 r->status = status;
98 r->msglen = msglen;
99 memcpy(&r->msg[0], msg, msglen);
100
101 ctdb_queue_packet(ctdb, &r->hdr);
102
103 talloc_free(msg);
104 }
105
106
107 /**
108 * send a redirect reply
109 *
110 * The logic behind this function is this:
111 *
112 * A client wants to grab a record and sends a CTDB_REQ_CALL packet
113 * to its local ctdb (ctdb_request_call). If the node is not itself
114 * the record's DMASTER, it first redirects the packet to the
115 * record's LMASTER. The LMASTER then redirects the call packet to
116 * the current DMASTER. Note that this works because of this: When
117 * a record is migrated off a node, then the new DMASTER is stored
118 * in the record's copy on the former DMASTER.
119 */
120 static void ctdb_call_send_redirect(struct ctdb_context *ctdb,
121 struct ctdb_db_context *ctdb_db,
122 TDB_DATA key,
123 struct ctdb_req_call *c,
124 struct ctdb_ltdb_header *header)
125 {
126 uint32_t lmaster = ctdb_lmaster(ctdb, &key);
127
128 c->hdr.destnode = lmaster;
129 if (ctdb->pnn == lmaster) {
130 c->hdr.destnode = header->dmaster;
131 }
132 c->hopcount++;
133
134 if (c->hopcount%100 > 95) {
135 DEBUG(DEBUG_WARNING,("High hopcount %d dbid:%s "
136 "key:0x%08x reqid=%08x pnn:%d src:%d lmaster:%d "
137 "header->dmaster:%d dst:%d\n",
138 c->hopcount, ctdb_db->db_name, ctdb_hash(&key),
139 c->hdr.reqid, ctdb->pnn, c->hdr.srcnode, lmaster,
140 header->dmaster, c->hdr.destnode));
141 }
142
143 ctdb_queue_packet(ctdb, &c->hdr);
144 }
145
146
147 /*
148 send a dmaster reply
149
150 caller must have the chainlock before calling this routine. Caller must be
151 the lmaster
152 */
153 static void ctdb_send_dmaster_reply(struct ctdb_db_context *ctdb_db,
154 struct ctdb_ltdb_header *header,
155 TDB_DATA key, TDB_DATA data,
156 uint32_t new_dmaster,
157 uint32_t reqid)
158 {
159 struct ctdb_context *ctdb = ctdb_db->ctdb;
160 struct ctdb_reply_dmaster *r;
161 int ret, len;
162 TALLOC_CTX *tmp_ctx;
163
164 if (ctdb->pnn != ctdb_lmaster(ctdb, &key)) {
165 DEBUG(DEBUG_ALERT,(__location__ " Caller is not lmaster!\n"));
166 return;
167 }
168
169 header->dmaster = new_dmaster;
170 ret = ctdb_ltdb_store(ctdb_db, key, header, data);
171 if (ret != 0) {
172 ctdb_fatal(ctdb, "ctdb_send_dmaster_reply unable to update dmaster");
173 return;
174 }
175
176 if (ctdb->methods == NULL) {
177 ctdb_fatal(ctdb, "ctdb_send_dmaster_reply cant update dmaster since transport is down");
178 return;
179 }
180
181 /* put the packet on a temporary context, allowing us to safely free
182 it below even if ctdb_reply_dmaster() has freed it already */
183 tmp_ctx = talloc_new(ctdb);
184
185 /* send the CTDB_REPLY_DMASTER */
186 len = offsetof(struct ctdb_reply_dmaster, data) + key.dsize + data.dsize + sizeof(uint32_t);
187 r = ctdb_transport_allocate(ctdb, tmp_ctx, CTDB_REPLY_DMASTER, len,
188 struct ctdb_reply_dmaster);
189 CTDB_NO_MEMORY_FATAL(ctdb, r);
190
191 r->hdr.destnode = new_dmaster;
192 r->hdr.reqid = reqid;
193 r->rsn = header->rsn;
194 r->keylen = key.dsize;
195 r->datalen = data.dsize;
196 r->db_id = ctdb_db->db_id;
197 memcpy(&r->data[0], key.dptr, key.dsize);
198 memcpy(&r->data[key.dsize], data.dptr, data.dsize);
199 memcpy(&r->data[key.dsize+data.dsize], &header->flags, sizeof(uint32_t));
200
201 ctdb_queue_packet(ctdb, &r->hdr);
202
203 talloc_free(tmp_ctx);
204 }
205
206 /*
207 send a dmaster request (give another node the dmaster for a record)
208
209 This is always sent to the lmaster, which ensures that the lmaster
210 always knows who the dmaster is. The lmaster will then send a
211 CTDB_REPLY_DMASTER to the new dmaster
212 */
213 static void ctdb_call_send_dmaster(struct ctdb_db_context *ctdb_db,
214 struct ctdb_req_call *c,
215 struct ctdb_ltdb_header *header,
216 TDB_DATA *key, TDB_DATA *data)
217 {
218 struct ctdb_req_dmaster *r;
219 struct ctdb_context *ctdb = ctdb_db->ctdb;
220 int len;
221 uint32_t lmaster = ctdb_lmaster(ctdb, key);
222
223 if (ctdb->methods == NULL) {
224 ctdb_fatal(ctdb, "Failed ctdb_call_send_dmaster since transport is down");
225 return;
226 }
227
228 if (data->dsize != 0) {
229 header->flags |= CTDB_REC_FLAG_MIGRATED_WITH_DATA;
230 }
231
232 if (lmaster == ctdb->pnn) {
233 ctdb_send_dmaster_reply(ctdb_db, header, *key, *data,
234 c->hdr.srcnode, c->hdr.reqid);
235 return;
236 }
237
238 len = offsetof(struct ctdb_req_dmaster, data) + key->dsize + data->dsize
239 + sizeof(uint32_t);
240 r = ctdb_transport_allocate(ctdb, ctdb, CTDB_REQ_DMASTER, len,
241 struct ctdb_req_dmaster);
242 CTDB_NO_MEMORY_FATAL(ctdb, r);
243 r->hdr.destnode = lmaster;
244 r->hdr.reqid = c->hdr.reqid;
245 r->db_id = c->db_id;
246 r->rsn = header->rsn;
247 r->dmaster = c->hdr.srcnode;
248 r->keylen = key->dsize;
249 r->datalen = data->dsize;
250 memcpy(&r->data[0], key->dptr, key->dsize);
251 memcpy(&r->data[key->dsize], data->dptr, data->dsize);
252 memcpy(&r->data[key->dsize + data->dsize], &header->flags, sizeof(uint32_t));
253
254 header->dmaster = c->hdr.srcnode;
255 if (ctdb_ltdb_store(ctdb_db, *key, header, *data) != 0) {
256 ctdb_fatal(ctdb, "Failed to store record in ctdb_call_send_dmaster");
257 }
258
259 ctdb_queue_packet(ctdb, &r->hdr);
260
261 talloc_free(r);
262 }
263
264 static void ctdb_sticky_pindown_timeout(struct event_context *ev, struct timed_event *te,
265 struct timeval t, void *private_data)
266 {
267 struct ctdb_sticky_record *sr = talloc_get_type(private_data,
268 struct ctdb_sticky_record);
269
270 DEBUG(DEBUG_ERR,("Pindown timeout db:%s unstick record\n", sr->ctdb_db->db_name));
271 if (sr->pindown != NULL) {
272 talloc_free(sr->pindown);
273 sr->pindown = NULL;
274 }
275 }
276
277 static int
278 ctdb_set_sticky_pindown(struct ctdb_context *ctdb, struct ctdb_db_context *ctdb_db, TDB_DATA key)
279 {
280 TALLOC_CTX *tmp_ctx = talloc_new(NULL);
281 uint32_t *k;
282 struct ctdb_sticky_record *sr;
283
284 k = talloc_zero_size(tmp_ctx, ((key.dsize + 3) & 0xfffffffc) + 4);
285 if (k == NULL) {
286 DEBUG(DEBUG_ERR,("Failed to allocate key for sticky record\n"));
287 talloc_free(tmp_ctx);
288 return -1;
289 }
290
291 k[0] = (key.dsize + 3) / 4 + 1;
292 memcpy(&k[1], key.dptr, key.dsize);
293
294 sr = trbt_lookuparray32(ctdb_db->sticky_records, k[0], &k[0]);
295 if (sr == NULL) {
296 talloc_free(tmp_ctx);
297 return 0;
298 }
299
300 talloc_free(tmp_ctx);
301
302 if (sr->pindown == NULL) {
303 DEBUG(DEBUG_ERR,("Pinning down record in %s for %d ms\n", ctdb_db->db_name, ctdb->tunable.sticky_pindown));
304 sr->pindown = talloc_new(sr);
305 if (sr->pindown == NULL) {
306 DEBUG(DEBUG_ERR,("Failed to allocate pindown context for sticky record\n"));
307 return -1;
308 }
309 event_add_timed(ctdb->ev, sr->pindown, timeval_current_ofs(ctdb->tunable.sticky_pindown / 1000, (ctdb->tunable.sticky_pindown * 1000) % 1000000), ctdb_sticky_pindown_timeout, sr);
310 }
311
312 return 0;
313 }
314
315 /*
316 called when a CTDB_REPLY_DMASTER packet comes in, or when the lmaster
317 gets a CTDB_REQUEST_DMASTER for itself. We become the dmaster.
318
319 must be called with the chainlock held. This function releases the chainlock
320 */
321 static void ctdb_become_dmaster(struct ctdb_db_context *ctdb_db,
322 struct ctdb_req_header *hdr,
323 TDB_DATA key, TDB_DATA data,
324 uint64_t rsn, uint32_t record_flags)
325 {
326 struct ctdb_call_state *state;
327 struct ctdb_context *ctdb = ctdb_db->ctdb;
328 struct ctdb_ltdb_header header;
329 int ret;
330
331 DEBUG(DEBUG_DEBUG,("pnn %u dmaster response %08x\n", ctdb->pnn, ctdb_hash(&key)));
332
333 ZERO_STRUCT(header);
334 header.rsn = rsn;
335 header.dmaster = ctdb->pnn;
336 header.flags = record_flags;
337
338 state = ctdb_reqid_find(ctdb, hdr->reqid, struct ctdb_call_state);
339
340 if (state) {
341 if (state->call->flags & CTDB_CALL_FLAG_VACUUM_MIGRATION) {
342 /*
343 * We temporarily add the VACUUM_MIGRATED flag to
344 * the record flags, so that ctdb_ltdb_store can
345 * decide whether the record should be stored or
346 * deleted.
347 */
348 header.flags |= CTDB_REC_FLAG_VACUUM_MIGRATED;
349 }
350 }
351
352 if (ctdb_ltdb_store(ctdb_db, key, &header, data) != 0) {
353 ctdb_fatal(ctdb, "ctdb_reply_dmaster store failed\n");
354
355 ret = ctdb_ltdb_unlock(ctdb_db, key);
356 if (ret != 0) {
357 DEBUG(DEBUG_ERR,(__location__ " ctdb_ltdb_unlock() failed with error %d\n", ret));
358 }
359 return;
360 }
361
362 /* we just became DMASTER and this database is "sticky",
363 see if the record is flagged as "hot" and set up a pin-down
364 context to stop migrations for a little while if so
365 */
366 if (ctdb_db->sticky) {
367 ctdb_set_sticky_pindown(ctdb, ctdb_db, key);
368 }
369
370 if (state == NULL) {
371 DEBUG(DEBUG_ERR,("pnn %u Invalid reqid %u in ctdb_become_dmaster from node %u\n",
372 ctdb->pnn, hdr->reqid, hdr->srcnode));
373
374 ret = ctdb_ltdb_unlock(ctdb_db, key);
375 if (ret != 0) {
376 DEBUG(DEBUG_ERR,(__location__ " ctdb_ltdb_unlock() failed with error %d\n", ret));
377 }
378 return;
379 }
380
381 if (key.dsize != state->call->key.dsize || memcmp(key.dptr, state->call->key.dptr, key.dsize)) {
382 DEBUG(DEBUG_ERR, ("Got bogus DMASTER packet reqid:%u from node %u. Key does not match key held in matching idr.\n", hdr->reqid, hdr->srcnode));
383
384 ret = ctdb_ltdb_unlock(ctdb_db, key);
385 if (ret != 0) {
386 DEBUG(DEBUG_ERR,(__location__ " ctdb_ltdb_unlock() failed with error %d\n", ret));
387 }
388 return;
389 }
390
391 if (hdr->reqid != state->reqid) {
392 /* we found a record but it was the wrong one */
393 DEBUG(DEBUG_ERR, ("Dropped orphan in ctdb_become_dmaster with reqid:%u\n from node %u", hdr->reqid, hdr->srcnode));
394
395 ret = ctdb_ltdb_unlock(ctdb_db, key);
396 if (ret != 0) {
397 DEBUG(DEBUG_ERR,(__location__ " ctdb_ltdb_unlock() failed with error %d\n", ret));
398 }
399 return;
400 }
401
402 ctdb_call_local(ctdb_db, state->call, &header, state, &data, true);
403
404 ret = ctdb_ltdb_unlock(ctdb_db, state->call->key);
405 if (ret != 0) {
406 DEBUG(DEBUG_ERR,(__location__ " ctdb_ltdb_unlock() failed with error %d\n", ret));
407 }
408
409 state->state = CTDB_CALL_DONE;
410 if (state->async.fn) {
411 state->async.fn(state);
412 }
413 }
414
415
416
417 /*
418 called when a CTDB_REQ_DMASTER packet comes in
419
420 this comes into the lmaster for a record when the current dmaster
421 wants to give up the dmaster role and give it to someone else
422 */
423 void ctdb_request_dmaster(struct ctdb_context *ctdb, struct ctdb_req_header *hdr)
424 {
425 struct ctdb_req_dmaster *c = (struct ctdb_req_dmaster *)hdr;
426 TDB_DATA key, data, data2;
427 struct ctdb_ltdb_header header;
428 struct ctdb_db_context *ctdb_db;
429 uint32_t record_flags = 0;
430 size_t len;
431 int ret;
432
433 key.dptr = c->data;
434 key.dsize = c->keylen;
435 data.dptr = c->data + c->keylen;
436 data.dsize = c->datalen;
437 len = offsetof(struct ctdb_req_dmaster, data) + key.dsize + data.dsize
438 + sizeof(uint32_t);
439 if (len <= c->hdr.length) {
440 record_flags = *(uint32_t *)&c->data[c->keylen + c->datalen];
441 }
442
443 ctdb_db = find_ctdb_db(ctdb, c->db_id);
444 if (!ctdb_db) {
445 ctdb_send_error(ctdb, hdr, -1,
446 "Unknown database in request. db_id==0x%08x",
447 c->db_id);
448 return;
449 }
450
451 /* fetch the current record */
452 ret = ctdb_ltdb_lock_fetch_requeue(ctdb_db, key, &header, hdr, &data2,
453 ctdb_call_input_pkt, ctdb, false);
454 if (ret == -1) {
455 ctdb_fatal(ctdb, "ctdb_req_dmaster failed to fetch record");
456 return;
457 }
458 if (ret == -2) {
459 DEBUG(DEBUG_INFO,(__location__ " deferring ctdb_request_dmaster\n"));
460 return;
461 }
462
463 if (ctdb_lmaster(ctdb, &key) != ctdb->pnn) {
464 DEBUG(DEBUG_ALERT,("pnn %u dmaster request to non-lmaster lmaster=%u gen=%u curgen=%u\n",
465 ctdb->pnn, ctdb_lmaster(ctdb, &key),
466 hdr->generation, ctdb->vnn_map->generation));
467 ctdb_fatal(ctdb, "ctdb_req_dmaster to non-lmaster");
468 }
469
470 DEBUG(DEBUG_DEBUG,("pnn %u dmaster request on %08x for %u from %u\n",
471 ctdb->pnn, ctdb_hash(&key), c->dmaster, c->hdr.srcnode));
472
473 /* its a protocol error if the sending node is not the current dmaster */
474 if (header.dmaster != hdr->srcnode) {
475 DEBUG(DEBUG_ALERT,("pnn %u dmaster request for new-dmaster %u from non-master %u real-dmaster=%u key %08x dbid 0x%08x gen=%u curgen=%u c->rsn=%llu header.rsn=%llu reqid=%u keyval=0x%08x\n",
476 ctdb->pnn, c->dmaster, hdr->srcnode, header.dmaster, ctdb_hash(&key),
477 ctdb_db->db_id, hdr->generation, ctdb->vnn_map->generation,
478 (unsigned long long)c->rsn, (unsigned long long)header.rsn, c->hdr.reqid,
479 (key.dsize >= 4)?(*(uint32_t *)key.dptr):0));
480 if (header.rsn != 0 || header.dmaster != ctdb->pnn) {
481 DEBUG(DEBUG_ERR,("ctdb_req_dmaster from non-master. Force a recovery.\n"));
482
483 ctdb->recovery_mode = CTDB_RECOVERY_ACTIVE;
484 ctdb_ltdb_unlock(ctdb_db, key);
485 return;
486 }
487 }
488
489 if (header.rsn > c->rsn) {
490 DEBUG(DEBUG_ALERT,("pnn %u dmaster request with older RSN new-dmaster %u from %u real-dmaster=%u key %08x dbid 0x%08x gen=%u curgen=%u c->rsn=%llu header.rsn=%llu reqid=%u\n",
491 ctdb->pnn, c->dmaster, hdr->srcnode, header.dmaster, ctdb_hash(&key),
492 ctdb_db->db_id, hdr->generation, ctdb->vnn_map->generation,
493 (unsigned long long)c->rsn, (unsigned long long)header.rsn, c->hdr.reqid));
494 }
495
496 /* use the rsn from the sending node */
497 header.rsn = c->rsn;
498
499 /* store the record flags from the sending node */
500 header.flags = record_flags;
501
502 /* check if the new dmaster is the lmaster, in which case we
503 skip the dmaster reply */
504 if (c->dmaster == ctdb->pnn) {
505 ctdb_become_dmaster(ctdb_db, hdr, key, data, c->rsn, record_flags);
506 } else {
507 ctdb_send_dmaster_reply(ctdb_db, &header, key, data, c->dmaster, hdr->reqid);
508
509 ret = ctdb_ltdb_unlock(ctdb_db, key);
510 if (ret != 0) {
511 DEBUG(DEBUG_ERR,(__location__ " ctdb_ltdb_unlock() failed with error %d\n", ret));
512 }
513 }
514 }
515
516 static void ctdb_sticky_record_timeout(struct event_context *ev, struct timed_event *te,
517 struct timeval t, void *private_data)
518 {
519 struct ctdb_sticky_record *sr = talloc_get_type(private_data,
520 struct ctdb_sticky_record);
521 talloc_free(sr);
522 }
523
524 static void *ctdb_make_sticky_record_callback(void *parm, void *data)
525 {
526 if (data) {
527 DEBUG(DEBUG_ERR,("Already have sticky record registered. Free old %p and create new %p\n", data, parm));
528 talloc_free(data);
529 }
530 return parm;
531 }
532
533 static int
534 ctdb_make_record_sticky(struct ctdb_context *ctdb, struct ctdb_db_context *ctdb_db, TDB_DATA key)
535 {
536 TALLOC_CTX *tmp_ctx = talloc_new(NULL);
537 uint32_t *k;
538 struct ctdb_sticky_record *sr;
539
540 k = talloc_zero_size(tmp_ctx, ((key.dsize + 3) & 0xfffffffc) + 4);
541 if (k == NULL) {
542 DEBUG(DEBUG_ERR,("Failed to allocate key for sticky record\n"));
543 talloc_free(tmp_ctx);
544 return -1;
545 }
546
547 k[0] = (key.dsize + 3) / 4 + 1;
548 memcpy(&k[1], key.dptr, key.dsize);
549
550 sr = trbt_lookuparray32(ctdb_db->sticky_records, k[0], &k[0]);
551 if (sr != NULL) {
552 talloc_free(tmp_ctx);
553 return 0;
554 }
555
556 sr = talloc(ctdb_db->sticky_records, struct ctdb_sticky_record);
557 if (sr == NULL) {
558 talloc_free(tmp_ctx);
559 DEBUG(DEBUG_ERR,("Failed to allocate sticky record structure\n"));
560 return -1;
561 }
562
563 sr->ctdb = ctdb;
564 sr->ctdb_db = ctdb_db;
565 sr->pindown = NULL;
566
567 DEBUG(DEBUG_ERR,("Make record sticky for %d seconds in db %s key:0x%08x.\n",
568 ctdb->tunable.sticky_duration,
569 ctdb_db->db_name, ctdb_hash(&key)));
570
571 trbt_insertarray32_callback(ctdb_db->sticky_records, k[0], &k[0], ctdb_make_sticky_record_callback, sr);
572
573 event_add_timed(ctdb->ev, sr, timeval_current_ofs(ctdb->tunable.sticky_duration, 0), ctdb_sticky_record_timeout, sr);
574
575 talloc_free(tmp_ctx);
576 return 0;
577 }
578
579 struct pinned_down_requeue_handle {
580 struct ctdb_context *ctdb;
581 struct ctdb_req_header *hdr;
582 };
583
584 struct pinned_down_deferred_call {
585 struct ctdb_context *ctdb;
586 struct ctdb_req_header *hdr;
587 };
588
589 static void pinned_down_requeue(struct event_context *ev, struct timed_event *te,
590 struct timeval t, void *private_data)
591 {
592 struct pinned_down_requeue_handle *handle = talloc_get_type(private_data, struct pinned_down_requeue_handle);
593 struct ctdb_context *ctdb = handle->ctdb;
594
595 talloc_steal(ctdb, handle->hdr);
596 ctdb_call_input_pkt(ctdb, handle->hdr);
597
598 talloc_free(handle);
599 }
600
601 static int pinned_down_destructor(struct pinned_down_deferred_call *pinned_down)
602 {
603 struct ctdb_context *ctdb = pinned_down->ctdb;
604 struct pinned_down_requeue_handle *handle = talloc(ctdb, struct pinned_down_requeue_handle);
605
606 handle->ctdb = pinned_down->ctdb;
607 handle->hdr = pinned_down->hdr;
608 talloc_steal(handle, handle->hdr);
609
610 event_add_timed(ctdb->ev, handle, timeval_zero(), pinned_down_requeue, handle);
611
612 return 0;
613 }
614
615 static int
616 ctdb_defer_pinned_down_request(struct ctdb_context *ctdb, struct ctdb_db_context *ctdb_db, TDB_DATA key, struct ctdb_req_header *hdr)
617 {
618 TALLOC_CTX *tmp_ctx = talloc_new(NULL);
619 uint32_t *k;
620 struct ctdb_sticky_record *sr;
621 struct pinned_down_deferred_call *pinned_down;
622
623 k = talloc_zero_size(tmp_ctx, ((key.dsize + 3) & 0xfffffffc) + 4);
624 if (k == NULL) {
625 DEBUG(DEBUG_ERR,("Failed to allocate key for sticky record\n"));
626 talloc_free(tmp_ctx);
627 return -1;
628 }
629
630 k[0] = (key.dsize + 3) / 4 + 1;
631 memcpy(&k[1], key.dptr, key.dsize);
632
633 sr = trbt_lookuparray32(ctdb_db->sticky_records, k[0], &k[0]);
634 if (sr == NULL) {
635 talloc_free(tmp_ctx);
636 return -1;
637 }
638
639 talloc_free(tmp_ctx);
640
641 if (sr->pindown == NULL) {
642 return -1;
643 }
644
645 pinned_down = talloc(sr->pindown, struct pinned_down_deferred_call);
646 if (pinned_down == NULL) {
647 DEBUG(DEBUG_ERR,("Failed to allocate structure for deferred pinned down request\n"));
648 return -1;
649 }
650
651 pinned_down->ctdb = ctdb;
652 pinned_down->hdr = hdr;
653
654 talloc_set_destructor(pinned_down, pinned_down_destructor);
655 talloc_steal(pinned_down, hdr);
656
657 return 0;
658 }
659
660 static void
661 ctdb_update_db_stat_hot_keys(struct ctdb_db_context *ctdb_db, TDB_DATA key, int hopcount)
662 {
663 int i, id;
664
665 /* smallest value is always at index 0 */
666 if (hopcount <= ctdb_db->statistics.hot_keys[0].count) {
667 return;
668 }
669
670 /* see if we already know this key */
671 for (i = 0; i < MAX_HOT_KEYS; i++) {
672 if (key.dsize != ctdb_db->statistics.hot_keys[i].key.dsize) {
673 continue;
674 }
675 if (memcmp(key.dptr, ctdb_db->statistics.hot_keys[i].key.dptr, key.dsize)) {
676 continue;
677 }
678 /* found an entry for this key */
679 if (hopcount <= ctdb_db->statistics.hot_keys[i].count) {
680 return;
681 }
682 ctdb_db->statistics.hot_keys[i].count = hopcount;
683 goto sort_keys;
684 }
685
686 if (ctdb_db->statistics.num_hot_keys < MAX_HOT_KEYS) {
687 id = ctdb_db->statistics.num_hot_keys;
688 ctdb_db->statistics.num_hot_keys++;
689 } else {
690 id = 0;
691 }
692
693 if (ctdb_db->statistics.hot_keys[id].key.dptr != NULL) {
694 talloc_free(ctdb_db->statistics.hot_keys[id].key.dptr);
695 }
696 ctdb_db->statistics.hot_keys[id].key.dsize = key.dsize;
697 ctdb_db->statistics.hot_keys[id].key.dptr = talloc_memdup(ctdb_db, key.dptr, key.dsize);
698 ctdb_db->statistics.hot_keys[id].count = hopcount;
699 DEBUG(DEBUG_NOTICE,("Updated hot key database=%s key=0x%08x id=%d hop_count=%d\n",
700 ctdb_db->db_name, ctdb_hash(&key), id, hopcount));
701
702 sort_keys:
703 for (i = 1; i < MAX_HOT_KEYS; i++) {
704 if (ctdb_db->statistics.hot_keys[i].count == 0) {
705 continue;
706 }
707 if (ctdb_db->statistics.hot_keys[i].count < ctdb_db->statistics.hot_keys[0].count) {
708 hopcount = ctdb_db->statistics.hot_keys[i].count;
709 ctdb_db->statistics.hot_keys[i].count = ctdb_db->statistics.hot_keys[0].count;
710 ctdb_db->statistics.hot_keys[0].count = hopcount;
711
712 key = ctdb_db->statistics.hot_keys[i].key;
713 ctdb_db->statistics.hot_keys[i].key = ctdb_db->statistics.hot_keys[0].key;
714 ctdb_db->statistics.hot_keys[0].key = key;
715 }
716 }
717 }
718
719 /*
720 called when a CTDB_REQ_CALL packet comes in
721 */
722 void ctdb_request_call(struct ctdb_context *ctdb, struct ctdb_req_header *hdr)
723 {
724 struct ctdb_req_call *c = (struct ctdb_req_call *)hdr;
725 TDB_DATA data;
726 struct ctdb_reply_call *r;
727 int ret, len;
728 struct ctdb_ltdb_header header;
729 struct ctdb_call *call;
730 struct ctdb_db_context *ctdb_db;
731 int tmp_count, bucket;
732
733 if (ctdb->methods == NULL) {
734 DEBUG(DEBUG_INFO,(__location__ " Failed ctdb_request_call. Transport is DOWN\n"));
735 return;
736 }
737
738
739 ctdb_db = find_ctdb_db(ctdb, c->db_id);
740 if (!ctdb_db) {
741 ctdb_send_error(ctdb, hdr, -1,
742 "Unknown database in request. db_id==0x%08x",
743 c->db_id);
744 return;
745 }
746
747 call = talloc(hdr, struct ctdb_call);
748 CTDB_NO_MEMORY_FATAL(ctdb, call);
749
750 call->call_id = c->callid;
751 call->key.dptr = c->data;
752 call->key.dsize = c->keylen;
753 call->call_data.dptr = c->data + c->keylen;
754 call->call_data.dsize = c->calldatalen;
755 call->reply_data.dptr = NULL;
756 call->reply_data.dsize = 0;
757
758
759 /* If this record is pinned down we should defer the
760 request until the pindown times out
761 */
762 if (ctdb_db->sticky) {
763 if (ctdb_defer_pinned_down_request(ctdb, ctdb_db, call->key, hdr) == 0) {
764 DEBUG(DEBUG_WARNING,
765 ("Defer request for pinned down record in %s\n", ctdb_db->db_name));
766 talloc_free(call);
767 return;
768 }
769 }
770
771
772 /* determine if we are the dmaster for this key. This also
773 fetches the record data (if any), thus avoiding a 2nd fetch of the data
774 if the call will be answered locally */
775
776 ret = ctdb_ltdb_lock_fetch_requeue(ctdb_db, call->key, &header, hdr, &data,
777 ctdb_call_input_pkt, ctdb, false);
778 if (ret == -1) {
779 ctdb_send_error(ctdb, hdr, ret, "ltdb fetch failed in ctdb_request_call");
780 talloc_free(call);
781 return;
782 }
783 if (ret == -2) {
784 DEBUG(DEBUG_INFO,(__location__ " deferred ctdb_request_call\n"));
785 talloc_free(call);
786 return;
787 }
788
789 /* Dont do READONLY if we dont have a tracking database */
790 if ((c->flags & CTDB_WANT_READONLY) && !ctdb_db->readonly) {
791 c->flags &= ~CTDB_WANT_READONLY;
792 }
793
794 if (header.flags & CTDB_REC_RO_REVOKE_COMPLETE) {
795 header.flags &= ~CTDB_REC_RO_FLAGS;
796 CTDB_INCREMENT_STAT(ctdb, total_ro_revokes);
797 CTDB_INCREMENT_DB_STAT(ctdb_db, db_ro_revokes);
798 if (ctdb_ltdb_store(ctdb_db, call->key, &header, data) != 0) {
799 ctdb_fatal(ctdb, "Failed to write header with cleared REVOKE flag");
800 }
801 /* and clear out the tracking data */
802 if (tdb_delete(ctdb_db->rottdb, call->key) != 0) {
803 DEBUG(DEBUG_ERR,(__location__ " Failed to clear out trackingdb record\n"));
804 }
805 }
806
807 /* if we are revoking, we must defer all other calls until the revoke
808 * had completed.
809 */
810 if (header.flags & CTDB_REC_RO_REVOKING_READONLY) {
811 talloc_free(data.dptr);
812 ret = ctdb_ltdb_unlock(ctdb_db, call->key);
813
814 if (ctdb_add_revoke_deferred_call(ctdb, ctdb_db, call->key, hdr, ctdb_call_input_pkt, ctdb) != 0) {
815 ctdb_fatal(ctdb, "Failed to add deferred call for revoke child");
816 }
817 talloc_free(call);
818 return;
819 }
820
821 /*
822 * If we are not the dmaster and are not hosting any delegations,
823 * then we redirect the request to the node than can answer it
824 * (the lmaster or the dmaster).
825 */
826 if ((header.dmaster != ctdb->pnn)
827 && (!(header.flags & CTDB_REC_RO_HAVE_DELEGATIONS)) ) {
828 talloc_free(data.dptr);
829 ctdb_call_send_redirect(ctdb, ctdb_db, call->key, c, &header);
830
831 ret = ctdb_ltdb_unlock(ctdb_db, call->key);
832 if (ret != 0) {
833 DEBUG(DEBUG_ERR,(__location__ " ctdb_ltdb_unlock() failed with error %d\n", ret));
834 }
835 talloc_free(call);
836 return;
837 }
838
839 if ( (!(c->flags & CTDB_WANT_READONLY))
840 && (header.flags & (CTDB_REC_RO_HAVE_DELEGATIONS|CTDB_REC_RO_HAVE_READONLY)) ) {
841 header.flags |= CTDB_REC_RO_REVOKING_READONLY;
842 if (ctdb_ltdb_store(ctdb_db, call->key, &header, data) != 0) {
843 ctdb_fatal(ctdb, "Failed to store record with HAVE_DELEGATIONS set");
844 }
845 ret = ctdb_ltdb_unlock(ctdb_db, call->key);
846
847 if (ctdb_start_revoke_ro_record(ctdb, ctdb_db, call->key, &header, data) != 0) {
848 ctdb_fatal(ctdb, "Failed to start record revoke");
849 }
850 talloc_free(data.dptr);
851
852 if (ctdb_add_revoke_deferred_call(ctdb, ctdb_db, call->key, hdr, ctdb_call_input_pkt, ctdb) != 0) {
853 ctdb_fatal(ctdb, "Failed to add deferred call for revoke child");
854 }
855 talloc_free(call);
856
857 return;
858 }
859
860 /* If this is the first request for delegation. bump rsn and set
861 * the delegations flag
862 */
863 if ((c->flags & CTDB_WANT_READONLY)
864 && (c->callid == CTDB_FETCH_WITH_HEADER_FUNC)
865 && (!(header.flags & CTDB_REC_RO_HAVE_DELEGATIONS))) {
866 header.rsn += 3;
867 header.flags |= CTDB_REC_RO_HAVE_DELEGATIONS;
868 if (ctdb_ltdb_store(ctdb_db, call->key, &header, data) != 0) {
869 ctdb_fatal(ctdb, "Failed to store record with HAVE_DELEGATIONS set");
870 }
871 }
872 if ((c->flags & CTDB_WANT_READONLY)
873 && (call->call_id == CTDB_FETCH_WITH_HEADER_FUNC)) {
874 TDB_DATA tdata;
875
876 tdata = tdb_fetch(ctdb_db->rottdb, call->key);
877 if (ctdb_trackingdb_add_pnn(ctdb, &tdata, c->hdr.srcnode) != 0) {
878 ctdb_fatal(ctdb, "Failed to add node to trackingdb");
879 }
880 if (tdb_store(ctdb_db->rottdb, call->key, tdata, TDB_REPLACE) != 0) {
881 ctdb_fatal(ctdb, "Failed to store trackingdb data");
882 }
883 free(tdata.dptr);
884
885 ret = ctdb_ltdb_unlock(ctdb_db, call->key);
886 if (ret != 0) {
887 DEBUG(DEBUG_ERR,(__location__ " ctdb_ltdb_unlock() failed with error %d\n", ret));
888 }
889
890 len = offsetof(struct ctdb_reply_call, data) + data.dsize + sizeof(struct ctdb_ltdb_header);
891 r = ctdb_transport_allocate(ctdb, ctdb, CTDB_REPLY_CALL, len,
892 struct ctdb_reply_call);
893 CTDB_NO_MEMORY_FATAL(ctdb, r);
894 r->hdr.destnode = c->hdr.srcnode;
895 r->hdr.reqid = c->hdr.reqid;
896 r->status = 0;
897 r->datalen = data.dsize + sizeof(struct ctdb_ltdb_header);
898 header.rsn -= 2;
899 header.flags |= CTDB_REC_RO_HAVE_READONLY;
900 header.flags &= ~CTDB_REC_RO_HAVE_DELEGATIONS;
901 memcpy(&r->data[0], &header, sizeof(struct ctdb_ltdb_header));
902
903 if (data.dsize) {
904 memcpy(&r->data[sizeof(struct ctdb_ltdb_header)], data.dptr, data.dsize);
905 }
906
907 ctdb_queue_packet(ctdb, &r->hdr);
908 CTDB_INCREMENT_STAT(ctdb, total_ro_delegations);
909 CTDB_INCREMENT_DB_STAT(ctdb_db, db_ro_delegations);
910
911 talloc_free(r);
912 talloc_free(call);
913 return;
914 }
915
916 CTDB_UPDATE_STAT(ctdb, max_hop_count, c->hopcount);
917 tmp_count = c->hopcount;
918 bucket = 0;
919 while (tmp_count) {
920 tmp_count >>= 2;
921 bucket++;
922 }
923 if (bucket >= MAX_COUNT_BUCKETS) {
924 bucket = MAX_COUNT_BUCKETS - 1;
925 }
926 CTDB_INCREMENT_STAT(ctdb, hop_count_bucket[bucket]);
927 CTDB_INCREMENT_DB_STAT(ctdb_db, hop_count_bucket[bucket]);
928 ctdb_update_db_stat_hot_keys(ctdb_db, call->key, c->hopcount);
929
930 /* If this database supports sticky records, then check if the
931 hopcount is big. If it is it means the record is hot and we
932 should make it sticky.
933 */
934 if (ctdb_db->sticky && c->hopcount >= ctdb->tunable.hopcount_make_sticky) {
935 ctdb_make_record_sticky(ctdb, ctdb_db, call->key);
936 }
937
938
939 /* Try if possible to migrate the record off to the caller node.
940 * From the clients perspective a fetch of the data is just as
941 * expensive as a migration.
942 */
943 if (c->hdr.srcnode != ctdb->pnn) {
944 if (ctdb_db->persistent_state) {
945 DEBUG(DEBUG_INFO, (__location__ " refusing migration"
946 " of key %s while transaction is active\n",
947 (char *)call->key.dptr));
948 } else {
949 DEBUG(DEBUG_DEBUG,("pnn %u starting migration of %08x to %u\n",
950 ctdb->pnn, ctdb_hash(&(call->key)), c->hdr.srcnode));
951 ctdb_call_send_dmaster(ctdb_db, c, &header, &(call->key), &data);
952 talloc_free(data.dptr);
953
954 ret = ctdb_ltdb_unlock(ctdb_db, call->key);
955 if (ret != 0) {
956 DEBUG(DEBUG_ERR,(__location__ " ctdb_ltdb_unlock() failed with error %d\n", ret));
957 }
958 }
959 talloc_free(call);
960 return;
961 }
962
963 ret = ctdb_call_local(ctdb_db, call, &header, hdr, &data, true);
964 if (ret != 0) {
965 DEBUG(DEBUG_ERR,(__location__ " ctdb_call_local failed\n"));
966 call->status = -1;
967 }
968
969 ret = ctdb_ltdb_unlock(ctdb_db, call->key);
970 if (ret != 0) {
971 DEBUG(DEBUG_ERR,(__location__ " ctdb_ltdb_unlock() failed with error %d\n", ret));
972 }
973
974 len = offsetof(struct ctdb_reply_call, data) + call->reply_data.dsize;
975 r = ctdb_transport_allocate(ctdb, ctdb, CTDB_REPLY_CALL, len,
976 struct ctdb_reply_call);
977 CTDB_NO_MEMORY_FATAL(ctdb, r);
978 r->hdr.destnode = hdr->srcnode;
979 r->hdr.reqid = hdr->reqid;
980 r->status = call->status;
981 r->datalen = call->reply_data.dsize;
982 if (call->reply_data.dsize) {
983 memcpy(&r->data[0], call->reply_data.dptr, call->reply_data.dsize);
984 }
985
986 ctdb_queue_packet(ctdb, &r->hdr);
987
988 talloc_free(r);
989 talloc_free(call);
990 }
991
992 /**
993 * called when a CTDB_REPLY_CALL packet comes in
994 *
995 * This packet comes in response to a CTDB_REQ_CALL request packet. It
996 * contains any reply data from the call
997 */
998 void ctdb_reply_call(struct ctdb_context *ctdb, struct ctdb_req_header *hdr)
999 {
1000 struct ctdb_reply_call *c = (struct ctdb_reply_call *)hdr;
1001 struct ctdb_call_state *state;
1002
1003 state = ctdb_reqid_find(ctdb, hdr->reqid, struct ctdb_call_state);
1004 if (state == NULL) {
1005 DEBUG(DEBUG_ERR, (__location__ " reqid %u not found\n", hdr->reqid));
1006 return;
1007 }
1008
1009 if (hdr->reqid != state->reqid) {
1010 /* we found a record but it was the wrong one */
1011 DEBUG(DEBUG_ERR, ("Dropped orphaned call reply with reqid:%u\n",hdr->reqid));
1012 return;
1013 }
1014
1015
1016 /* read only delegation processing */
1017 /* If we got a FETCH_WITH_HEADER we should check if this is a ro
1018 * delegation since we may need to update the record header
1019 */
1020 if (state->c->callid == CTDB_FETCH_WITH_HEADER_FUNC) {
1021 struct ctdb_db_context *ctdb_db = state->ctdb_db;
1022 struct ctdb_ltdb_header *header = (struct ctdb_ltdb_header *)&c->data[0];
1023 struct ctdb_ltdb_header oldheader;
1024 TDB_DATA key, data, olddata;
1025 int ret;
1026
1027 if (!(header->flags & CTDB_REC_RO_HAVE_READONLY)) {
1028 goto finished_ro;
1029 return;
1030 }
1031
1032 key.dsize = state->c->keylen;
1033 key.dptr = state->c->data;
1034 ret = ctdb_ltdb_lock_requeue(ctdb_db, key, hdr,
1035 ctdb_call_input_pkt, ctdb, false);
1036 if (ret == -2) {
1037 return;
1038 }
1039 if (ret != 0) {
1040 DEBUG(DEBUG_ERR,(__location__ " Failed to get lock in ctdb_reply_call\n"));
1041 return;
1042 }
1043
1044 ret = ctdb_ltdb_fetch(ctdb_db, key, &oldheader, state, &olddata);
1045 if (ret != 0) {
1046 DEBUG(DEBUG_ERR, ("Failed to fetch old record in ctdb_reply_call\n"));
1047 ctdb_ltdb_unlock(ctdb_db, key);
1048 goto finished_ro;
1049 }
1050
1051 if (header->rsn <= oldheader.rsn) {
1052 ctdb_ltdb_unlock(ctdb_db, key);
1053 goto finished_ro;
1054 }
1055
1056 if (c->datalen < sizeof(struct ctdb_ltdb_header)) {
1057 DEBUG(DEBUG_ERR,(__location__ " Got FETCH_WITH_HEADER reply with too little data: %d bytes\n", c->datalen));
1058 ctdb_ltdb_unlock(ctdb_db, key);
1059 goto finished_ro;
1060 }
1061
1062 data.dsize = c->datalen - sizeof(struct ctdb_ltdb_header);
1063 data.dptr = &c->data[sizeof(struct ctdb_ltdb_header)];
1064 ret = ctdb_ltdb_store(ctdb_db, key, header, data);
1065 if (ret != 0) {
1066 DEBUG(DEBUG_ERR, ("Failed to store new record in ctdb_reply_call\n"));
1067 ctdb_ltdb_unlock(ctdb_db, key);
1068 goto finished_ro;
1069 }
1070
1071 ctdb_ltdb_unlock(ctdb_db, key);
1072 }
1073 finished_ro:
1074
1075 state->call->reply_data.dptr = c->data;
1076 state->call->reply_data.dsize = c->datalen;
1077 state->call->status = c->status;
1078
1079 talloc_steal(state, c);
1080
1081 state->state = CTDB_CALL_DONE;
1082 if (state->async.fn) {
1083 state->async.fn(state);
1084 }
1085 }
1086
1087
1088 /**
1089 * called when a CTDB_REPLY_DMASTER packet comes in
1090 *
1091 * This packet comes in from the lmaster in response to a CTDB_REQ_CALL
1092 * request packet. It means that the current dmaster wants to give us
1093 * the dmaster role.
1094 */
1095 void ctdb_reply_dmaster(struct ctdb_context *ctdb, struct ctdb_req_header *hdr)
1096 {
1097 struct ctdb_reply_dmaster *c = (struct ctdb_reply_dmaster *)hdr;
1098 struct ctdb_db_context *ctdb_db;
1099 TDB_DATA key, data;
1100 uint32_t record_flags = 0;
1101 size_t len;
1102 int ret;
1103
1104 ctdb_db = find_ctdb_db(ctdb, c->db_id);
1105 if (ctdb_db == NULL) {
1106 DEBUG(DEBUG_ERR,("Unknown db_id 0x%x in ctdb_reply_dmaster\n", c->db_id));
1107 return;
1108 }
1109
1110 key.dptr = c->data;
1111 key.dsize = c->keylen;
1112 data.dptr = &c->data[key.dsize];
1113 data.dsize = c->datalen;
1114 len = offsetof(struct ctdb_reply_dmaster, data) + key.dsize + data.dsize
1115 + sizeof(uint32_t);
1116 if (len <= c->hdr.length) {
1117 record_flags = *(uint32_t *)&c->data[c->keylen + c->datalen];
1118 }
1119
1120 ret = ctdb_ltdb_lock_requeue(ctdb_db, key, hdr,
1121 ctdb_call_input_pkt, ctdb, false);
1122 if (ret == -2) {
1123 return;
1124 }
1125 if (ret != 0) {
1126 DEBUG(DEBUG_ERR,(__location__ " Failed to get lock in ctdb_reply_dmaster\n"));
1127 return;
1128 }
1129
1130 ctdb_become_dmaster(ctdb_db, hdr, key, data, c->rsn, record_flags);
1131 }
1132
1133
1134 /*
1135 called when a CTDB_REPLY_ERROR packet comes in
1136 */
1137 void ctdb_reply_error(struct ctdb_context *ctdb, struct ctdb_req_header *hdr)
1138 {
1139 struct ctdb_reply_error *c = (struct ctdb_reply_error *)hdr;
1140 struct ctdb_call_state *state;
1141
1142 state = ctdb_reqid_find(ctdb, hdr->reqid, struct ctdb_call_state);
1143 if (state == NULL) {
1144 DEBUG(DEBUG_ERR,("pnn %u Invalid reqid %u in ctdb_reply_error\n",
1145 ctdb->pnn, hdr->reqid));
1146 return;
1147 }
1148
1149 if (hdr->reqid != state->reqid) {
1150 /* we found a record but it was the wrong one */
1151 DEBUG(DEBUG_ERR, ("Dropped orphaned error reply with reqid:%u\n",hdr->reqid));
1152 return;
1153 }
1154
1155 talloc_steal(state, c);
1156
1157 state->state = CTDB_CALL_ERROR;
1158 state->errmsg = (char *)c->msg;
1159 if (state->async.fn) {
1160 state->async.fn(state);
1161 }
1162 }
1163
1164
1165 /*
1166 destroy a ctdb_call
1167 */
1168 static int ctdb_call_destructor(struct ctdb_call_state *state)
1169 {
1170 DLIST_REMOVE(state->ctdb_db->ctdb->pending_calls, state);
1171 ctdb_reqid_remove(state->ctdb_db->ctdb, state->reqid);
1172 return 0;
1173 }
1174
1175
1176 /*
1177 called when a ctdb_call needs to be resent after a reconfigure event
1178 */
1179 static void ctdb_call_resend(struct ctdb_call_state *state)
1180 {
1181 struct ctdb_context *ctdb = state->ctdb_db->ctdb;
1182
1183 state->generation = ctdb->vnn_map->generation;
1184
1185 /* use a new reqid, in case the old reply does eventually come in */
1186 ctdb_reqid_remove(ctdb, state->reqid);
1187 state->reqid = ctdb_reqid_new(ctdb, state);
1188 state->c->hdr.reqid = state->reqid;
1189
1190 /* update the generation count for this request, so its valid with the new vnn_map */
1191 state->c->hdr.generation = state->generation;
1192
1193 /* send the packet to ourselves, it will be redirected appropriately */
1194 state->c->hdr.destnode = ctdb->pnn;
1195
1196 ctdb_queue_packet(ctdb, &state->c->hdr);
1197 DEBUG(DEBUG_NOTICE,("resent ctdb_call\n"));
1198 }
1199
1200 /*
1201 resend all pending calls on recovery
1202 */
1203 void ctdb_call_resend_all(struct ctdb_context *ctdb)
1204 {
1205 struct ctdb_call_state *state, *next;
1206 for (state=ctdb->pending_calls;state;state=next) {
1207 next = state->next;
1208 ctdb_call_resend(state);
1209 }
1210 }
1211
1212 /*
1213 this allows the caller to setup a async.fn
1214 */
1215 static void call_local_trigger(struct event_context *ev, struct timed_event *te,
1216 struct timeval t, void *private_data)
1217 {
1218 struct ctdb_call_state *state = talloc_get_type(private_data, struct ctdb_call_state);
1219 if (state->async.fn) {
1220 state->async.fn(state);
1221 }
1222 }
1223
1224
1225 /*
1226 construct an event driven local ctdb_call
1227
1228 this is used so that locally processed ctdb_call requests are processed
1229 in an event driven manner
1230 */
1231 struct ctdb_call_state *ctdb_call_local_send(struct ctdb_db_context *ctdb_db,
1232 struct ctdb_call *call,
1233 struct ctdb_ltdb_header *header,
1234 TDB_DATA *data)
1235 {
1236 struct ctdb_call_state *state;
1237 struct ctdb_context *ctdb = ctdb_db->ctdb;
1238 int ret;
1239
1240 state = talloc_zero(ctdb_db, struct ctdb_call_state);
1241 CTDB_NO_MEMORY_NULL(ctdb, state);
1242
1243 talloc_steal(state, data->dptr);
1244
1245 state->state = CTDB_CALL_DONE;
1246 state->call = talloc(state, struct ctdb_call);
1247 CTDB_NO_MEMORY_NULL(ctdb, state->call);
1248 *(state->call) = *call;
1249 state->ctdb_db = ctdb_db;
1250
1251 ret = ctdb_call_local(ctdb_db, state->call, header, state, data, true);
1252 if (ret != 0) {
1253 DEBUG(DEBUG_DEBUG,("ctdb_call_local() failed, ignoring return code %d\n", ret));
1254 }
1255
1256 event_add_timed(ctdb->ev, state, timeval_zero(), call_local_trigger, state);
1257
1258 return state;
1259 }
1260
1261
1262 /*
1263 make a remote ctdb call - async send. Called in daemon context.
1264
1265 This constructs a ctdb_call request and queues it for processing.
1266 This call never blocks.
1267 */
1268 struct ctdb_call_state *ctdb_daemon_call_send_remote(struct ctdb_db_context *ctdb_db,
1269 struct ctdb_call *call,
1270 struct ctdb_ltdb_header *header)
1271 {
1272 uint32_t len;
1273 struct ctdb_call_state *state;
1274 struct ctdb_context *ctdb = ctdb_db->ctdb;
1275
1276 if (ctdb->methods == NULL) {
1277 DEBUG(DEBUG_INFO,(__location__ " Failed send packet. Transport is down\n"));
1278 return NULL;
1279 }
1280
1281 state = talloc_zero(ctdb_db, struct ctdb_call_state);
1282 CTDB_NO_MEMORY_NULL(ctdb, state);
1283 state->call = talloc(state, struct ctdb_call);
1284 CTDB_NO_MEMORY_NULL(ctdb, state->call);
1285
1286 state->reqid = ctdb_reqid_new(ctdb, state);
1287 state->ctdb_db = ctdb_db;
1288 talloc_set_destructor(state, ctdb_call_destructor);
1289
1290 len = offsetof(struct ctdb_req_call, data) + call->key.dsize + call->call_data.dsize;
1291 state->c = ctdb_transport_allocate(ctdb, state, CTDB_REQ_CALL, len,
1292 struct ctdb_req_call);
1293 CTDB_NO_MEMORY_NULL(ctdb, state->c);
1294 state->c->hdr.destnode = header->dmaster;
1295
1296 /* this limits us to 16k outstanding messages - not unreasonable */
1297 state->c->hdr.reqid = state->reqid;
1298 state->c->flags = call->flags;
1299 state->c->db_id = ctdb_db->db_id;
1300 state->c->callid = call->call_id;
1301 state->c->hopcount = 0;
1302 state->c->keylen = call->key.dsize;
1303 state->c->calldatalen = call->call_data.dsize;
1304 memcpy(&state->c->data[0], call->key.dptr, call->key.dsize);
1305 memcpy(&state->c->data[call->key.dsize],
1306 call->call_data.dptr, call->call_data.dsize);
1307 *(state->call) = *call;
1308 state->call->call_data.dptr = &state->c->data[call->key.dsize];
1309 state->call->key.dptr = &state->c->data[0];
1310
1311 state->state = CTDB_CALL_WAIT;
1312 state->generation = ctdb->vnn_map->generation;
1313
1314 DLIST_ADD(ctdb->pending_calls, state);
1315
1316 ctdb_queue_packet(ctdb, &state->c->hdr);
1317
1318 return state;
1319 }
1320
1321 /*
1322 make a remote ctdb call - async recv - called in daemon context
1323
1324 This is called when the program wants to wait for a ctdb_call to complete and get the
1325 results. This call will block unless the call has already completed.
1326 */
1327 int ctdb_daemon_call_recv(struct ctdb_call_state *state, struct ctdb_call *call)
1328 {
1329 while (state->state < CTDB_CALL_DONE) {
1330 event_loop_once(state->ctdb_db->ctdb->ev);
1331 }
1332 if (state->state != CTDB_CALL_DONE) {
1333 ctdb_set_error(state->ctdb_db->ctdb, "%s", state->errmsg);
1334 talloc_free(state);
1335 return -1;
1336 }
1337
1338 if (state->call->reply_data.dsize) {
1339 call->reply_data.dptr = talloc_memdup(call,
1340 state->call->reply_data.dptr,
1341 state->call->reply_data.dsize);
1342 call->reply_data.dsize = state->call->reply_data.dsize;
1343 } else {
1344 call->reply_data.dptr = NULL;
1345 call->reply_data.dsize = 0;
1346 }
1347 call->status = state->call->status;
1348 talloc_free(state);
1349 return 0;
1350 }
1351
1352
1353 /*
1354 send a keepalive packet to the other node
1355 */
1356 void ctdb_send_keepalive(struct ctdb_context *ctdb, uint32_t destnode)
1357 {
1358 struct ctdb_req_keepalive *r;
1359
1360 if (ctdb->methods == NULL) {
1361 DEBUG(DEBUG_INFO,(__location__ " Failed to send keepalive. Transport is DOWN\n"));
1362 return;
1363 }
1364
1365 r = ctdb_transport_allocate(ctdb, ctdb, CTDB_REQ_KEEPALIVE,
1366 sizeof(struct ctdb_req_keepalive),
1367 struct ctdb_req_keepalive);
1368 CTDB_NO_MEMORY_FATAL(ctdb, r);
1369 r->hdr.destnode = destnode;
1370 r->hdr.reqid = 0;
1371
1372 CTDB_INCREMENT_STAT(ctdb, keepalive_packets_sent);
1373
1374 ctdb_queue_packet(ctdb, &r->hdr);
1375
1376 talloc_free(r);
1377 }
1378
1379
1380
1381 struct revokechild_deferred_call {
1382 struct ctdb_context *ctdb;
1383 struct ctdb_req_header *hdr;
1384 deferred_requeue_fn fn;
1385 void *ctx;
1386 };
1387
1388 struct revokechild_handle {
1389 struct revokechild_handle *next, *prev;
1390 struct ctdb_context *ctdb;
1391 struct ctdb_db_context *ctdb_db;
1392 struct fd_event *fde;
1393 int status;
1394 int fd[2];
1395 pid_t child;
1396 TDB_DATA key;
1397 };
1398
1399 struct revokechild_requeue_handle {
1400 struct ctdb_context *ctdb;
1401 struct ctdb_req_header *hdr;
1402 deferred_requeue_fn fn;
1403 void *ctx;
1404 };
1405
1406 static void deferred_call_requeue(struct event_context *ev, struct timed_event *te,
1407 struct timeval t, void *private_data)
1408 {
1409 struct revokechild_requeue_handle *requeue_handle = talloc_get_type(private_data, struct revokechild_requeue_handle);
1410
1411 requeue_handle->fn(requeue_handle->ctx, requeue_handle->hdr);
1412 talloc_free(requeue_handle);
1413 }
1414
1415 static int deferred_call_destructor(struct revokechild_deferred_call *deferred_call)
1416 {
1417 struct ctdb_context *ctdb = deferred_call->ctdb;
1418 struct revokechild_requeue_handle *requeue_handle = talloc(ctdb, struct revokechild_requeue_handle);
1419 struct ctdb_req_call *c = (struct ctdb_req_call *)deferred_call->hdr;
1420
1421 requeue_handle->ctdb = ctdb;
1422 requeue_handle->hdr = deferred_call->hdr;
1423 requeue_handle->fn = deferred_call->fn;
1424 requeue_handle->ctx = deferred_call->ctx;
1425 talloc_steal(requeue_handle, requeue_handle->hdr);
1426
1427 /* when revoking, any READONLY requests have 1 second grace to let read/write finish first */
1428 event_add_timed(ctdb->ev, requeue_handle, timeval_current_ofs(c->flags & CTDB_WANT_READONLY ? 1 : 0, 0), deferred_call_requeue, requeue_handle);
1429
1430 return 0;
1431 }
1432
1433
1434 static int revokechild_destructor(struct revokechild_handle *rc)
1435 {
1436 if (rc->fde != NULL) {
1437 talloc_free(rc->fde);
1438 }
1439
1440 if (rc->fd[0] != -1) {
1441 close(rc->fd[0]);
1442 }
1443 if (rc->fd[1] != -1) {
1444 close(rc->fd[1]);
1445 }
1446 ctdb_kill(rc->ctdb, rc->child, SIGKILL);
1447
1448 DLIST_REMOVE(rc->ctdb_db->revokechild_active, rc);
1449 return 0;
1450 }
1451
1452 static void revokechild_handler(struct event_context *ev, struct fd_event *fde,
1453 uint16_t flags, void *private_data)
1454 {
1455 struct revokechild_handle *rc = talloc_get_type(private_data,
1456 struct revokechild_handle);
1457 int ret;
1458 char c;
1459
1460 ret = read(rc->fd[0], &c, 1);
1461 if (ret != 1) {
1462 DEBUG(DEBUG_ERR,("Failed to read status from revokechild. errno:%d\n", errno));
1463 rc->status = -1;
1464 talloc_free(rc);
1465 return;
1466 }
1467 if (c != 0) {
1468 DEBUG(DEBUG_ERR,("revokechild returned failure. status:%d\n", c));
1469 rc->status = -1;
1470 talloc_free(rc);
1471 return;
1472 }
1473
1474 talloc_free(rc);
1475 }
1476
1477 struct ctdb_revoke_state {
1478 struct ctdb_db_context *ctdb_db;
1479 TDB_DATA key;
1480 struct ctdb_ltdb_header *header;
1481 TDB_DATA data;
1482 int count;
1483 int status;
1484 int finished;
1485 };
1486
1487 static void update_record_cb(struct ctdb_client_control_state *state)
1488 {
1489 struct ctdb_revoke_state *revoke_state;
1490 int ret;
1491 int32_t res;
1492
1493 if (state == NULL) {
1494 return;
1495 }
1496 revoke_state = state->async.private_data;
1497
1498 state->async.fn = NULL;
1499 ret = ctdb_control_recv(state->ctdb, state, state, NULL, &res, NULL);
1500 if ((ret != 0) || (res != 0)) {
1501 DEBUG(DEBUG_ERR,("Recv for revoke update record failed ret:%d res:%d\n", ret, res));
1502 revoke_state->status = -1;
1503 }
1504
1505 revoke_state->count--;
1506 if (revoke_state->count <= 0) {
1507 revoke_state->finished = 1;
1508 }
1509 }
1510
1511 static void revoke_send_cb(struct ctdb_context *ctdb, uint32_t pnn, void *private_data)
1512 {
1513 struct ctdb_revoke_state *revoke_state = private_data;
1514 struct ctdb_client_control_state *state;
1515
1516 state = ctdb_ctrl_updaterecord_send(ctdb, revoke_state, timeval_current_ofs(ctdb->tunable.control_timeout,0), pnn, revoke_state->ctdb_db, revoke_state->key, revoke_state->header, revoke_state->data);
1517 if (state == NULL) {
1518 DEBUG(DEBUG_ERR,("Failure to send update record to revoke readonly delegation\n"));
1519 revoke_state->status = -1;
1520 return;
1521 }
1522 state->async.fn = update_record_cb;
1523 state->async.private_data = revoke_state;
1524
1525 revoke_state->count++;
1526
1527 }
1528
1529 static void ctdb_revoke_timeout_handler(struct event_context *ev, struct timed_event *te,
1530 struct timeval yt, void *private_data)
1531 {
1532 struct ctdb_revoke_state *state = private_data;
1533
1534 DEBUG(DEBUG_ERR,("Timed out waiting for revoke to finish\n"));
1535 state->finished = 1;
1536 state->status = -1;
1537 }
1538
1539 static int ctdb_revoke_all_delegations(struct ctdb_context *ctdb, struct ctdb_db_context *ctdb_db, TDB_DATA tdata, TDB_DATA key, struct ctdb_ltdb_header *header, TDB_DATA data)
1540 {
1541 struct ctdb_revoke_state *state = talloc_zero(ctdb, struct ctdb_revoke_state);
1542 int status;
1543
1544 state->ctdb_db = ctdb_db;
1545 state->key = key;
1546 state->header = header;
1547 state->data = data;
1548
1549 ctdb_trackingdb_traverse(ctdb, tdata, revoke_send_cb, state);
1550
1551 event_add_timed(ctdb->ev, state, timeval_current_ofs(ctdb->tunable.control_timeout, 0), ctdb_revoke_timeout_handler, state);
1552
1553 while (state->finished == 0) {
1554 event_loop_once(ctdb->ev);
1555 }
1556
1557 status = state->status;
1558
1559 if (status == 0) {
1560 struct ctdb_ltdb_header new_header;
1561 TDB_DATA new_data;
1562
1563 if (ctdb_ltdb_lock(ctdb_db, key) != 0) {
1564 DEBUG(DEBUG_ERR,("Failed to chainlock the database in revokechild\n"));
1565 talloc_free(state);
1566 return -1;
1567 }
1568 if (ctdb_ltdb_fetch(ctdb_db, key, &new_header, state, &new_data) != 0) {
1569 ctdb_ltdb_unlock(ctdb_db, key);
1570 DEBUG(DEBUG_ERR,("Failed for fetch tdb record in revokechild\n"));
1571 talloc_free(state);
1572 return -1;
1573 }
1574 header->rsn++;
1575 if (new_header.rsn > header->rsn) {
1576 ctdb_ltdb_unlock(ctdb_db, key);
1577 DEBUG(DEBUG_ERR,("RSN too high in tdb record in revokechild\n"));
1578 talloc_free(state);
1579 return -1;
1580 }
1581 if ( (new_header.flags & (CTDB_REC_RO_REVOKING_READONLY|CTDB_REC_RO_HAVE_DELEGATIONS)) != (CTDB_REC_RO_REVOKING_READONLY|CTDB_REC_RO_HAVE_DELEGATIONS) ) {
1582 ctdb_ltdb_unlock(ctdb_db, key);
1583 DEBUG(DEBUG_ERR,("Flags are wrong in tdb record in revokechild\n"));
1584 talloc_free(state);
1585 return -1;
1586 }
1587 new_header.rsn++;
1588 new_header.flags |= CTDB_REC_RO_REVOKE_COMPLETE;
1589 if (ctdb_ltdb_store(ctdb_db, key, &new_header, new_data) != 0) {
1590 ctdb_ltdb_unlock(ctdb_db, key);
1591 DEBUG(DEBUG_ERR,("Failed to write new record in revokechild\n"));
1592 talloc_free(state);
1593 return -1;
1594 }
1595 ctdb_ltdb_unlock(ctdb_db, key);
1596 }
1597
1598 talloc_free(state);
1599 return status;
1600 }
1601
1602
1603 int ctdb_start_revoke_ro_record(struct ctdb_context *ctdb, struct ctdb_db_context *ctdb_db, TDB_DATA key, struct ctdb_ltdb_header *header, TDB_DATA data)
1604 {
1605 TDB_DATA tdata;
1606 struct revokechild_handle *rc;
1607 pid_t parent = getpid();
1608 int ret;
1609
1610 header->flags &= ~(CTDB_REC_RO_REVOKING_READONLY|CTDB_REC_RO_HAVE_DELEGATIONS|CTDB_REC_RO_HAVE_READONLY);
1611 header->flags |= CTDB_REC_FLAG_MIGRATED_WITH_DATA;
1612 header->rsn -= 1;
1613
1614 if ((rc = talloc_zero(ctdb_db, struct revokechild_handle)) == NULL) {
1615 DEBUG(DEBUG_ERR,("Failed to allocate revokechild_handle\n"));
1616 return -1;
1617 }
1618
1619 tdata = tdb_fetch(ctdb_db->rottdb, key);
1620 if (tdata.dsize > 0) {
1621 uint8_t *tmp;
1622
1623 tmp = tdata.dptr;
1624 tdata.dptr = talloc_memdup(rc, tdata.dptr, tdata.dsize);
1625 free(tmp);
1626 }
1627
1628 rc->status = 0;
1629 rc->ctdb = ctdb;
1630 rc->ctdb_db = ctdb_db;
1631 rc->fd[0] = -1;
1632 rc->fd[1] = -1;
1633
1634 talloc_set_destructor(rc, revokechild_destructor);
1635
1636 rc->key.dsize = key.dsize;
1637 rc->key.dptr = talloc_memdup(rc, key.dptr, key.dsize);
1638 if (rc->key.dptr == NULL) {
1639 DEBUG(DEBUG_ERR,("Failed to allocate key for revokechild_handle\n"));
1640 talloc_free(rc);
1641 return -1;
1642 }
1643
1644 ret = pipe(rc->fd);
1645 if (ret != 0) {
1646 DEBUG(DEBUG_ERR,("Failed to allocate key for revokechild_handle\n"));
1647 talloc_free(rc);
1648 return -1;
1649 }
1650
1651
1652 rc->child = ctdb_fork(ctdb);
1653 if (rc->child == (pid_t)-1) {
1654 DEBUG(DEBUG_ERR,("Failed to fork child for revokechild\n"));
1655 talloc_free(rc);
1656 return -1;
1657 }
1658
1659 if (rc->child == 0) {
1660 char c = 0;
1661 close(rc->fd[0]);
1662 debug_extra = talloc_asprintf(NULL, "revokechild-%s:", ctdb_db->db_name);
1663
1664 ctdb_set_process_name("ctdb_revokechild");
1665 if (switch_from_server_to_client(ctdb, "revokechild-%s", ctdb_db->db_name) != 0) {
1666 DEBUG(DEBUG_ERR,("Failed to switch from server to client for revokechild process\n"));
1667 c = 1;
1668 goto child_finished;
1669 }
1670
1671 c = ctdb_revoke_all_delegations(ctdb, ctdb_db, tdata, key, header, data);
1672
1673 child_finished:
1674 write(rc->fd[1], &c, 1);
1675 /* make sure we die when our parent dies */
1676 while (ctdb_kill(ctdb, parent, 0) == 0 || errno != ESRCH) {
1677 sleep(5);
1678 }
1679 _exit(0);
1680 }
1681
1682 close(rc->fd[1]);
1683 rc->fd[1] = -1;
1684 set_close_on_exec(rc->fd[0]);
1685
1686 /* This is an active revokechild child process */
1687 DLIST_ADD_END(ctdb_db->revokechild_active, rc, NULL);
1688
1689 rc->fde = event_add_fd(ctdb->ev, rc, rc->fd[0],
1690 EVENT_FD_READ, revokechild_handler,
1691 (void *)rc);
1692 if (rc->fde == NULL) {
1693 DEBUG(DEBUG_ERR,("Failed to set up fd event for revokechild process\n"));
1694 talloc_free(rc);
1695 }
1696 tevent_fd_set_auto_close(rc->fde);
1697
1698 return 0;
1699 }
1700
1701 int ctdb_add_revoke_deferred_call(struct ctdb_context *ctdb, struct ctdb_db_context *ctdb_db, TDB_DATA key, struct ctdb_req_header *hdr, deferred_requeue_fn fn, void *call_context)
1702 {
1703 struct revokechild_handle *rc;
1704 struct revokechild_deferred_call *deferred_call;
1705
1706 for (rc = ctdb_db->revokechild_active; rc; rc = rc->next) {
1707 if (rc->key.dsize == 0) {
1708 continue;
1709 }
1710 if (rc->key.dsize != key.dsize) {
1711 continue;
1712 }
1713 if (!memcmp(rc->key.dptr, key.dptr, key.dsize)) {
1714 break;
1715 }
1716 }
1717
1718 if (rc == NULL) {
1719 DEBUG(DEBUG_ERR,("Failed to add deferred call to revoke list. revoke structure not found\n"));
1720 return -1;
1721 }
1722
1723 deferred_call = talloc(rc, struct revokechild_deferred_call);
1724 if (deferred_call == NULL) {
1725 DEBUG(DEBUG_ERR,("Failed to allocate deferred call structure for revoking record\n"));
1726 return -1;
1727 }
1728
1729 deferred_call->ctdb = ctdb;
1730 deferred_call->hdr = hdr;
1731 deferred_call->fn = fn;
1732 deferred_call->ctx = call_context;
1733
1734 talloc_set_destructor(deferred_call, deferred_call_destructor);
1735 talloc_steal(deferred_call, hdr);
1736
1737 return 0;
1738 }
Samba GIT mirror