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
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
|
// SPDX-License-Identifier: GPL-3.0-or-later
#include "sqlite_metadata.h"
// SQL statements
#define SQL_STORE_CLAIM_ID "INSERT INTO node_instance " \
"(host_id, claim_id, date_created) VALUES (@host_id, @claim_id, unixepoch()) " \
"ON CONFLICT(host_id) DO UPDATE SET claim_id = excluded.claim_id;"
#define SQL_DELETE_HOST_LABELS "DELETE FROM host_label WHERE host_id = @uuid;"
#define STORE_HOST_LABEL \
"INSERT INTO host_label (host_id, source_type, label_key, label_value, date_created) VALUES "
#define STORE_CHART_LABEL \
"INSERT INTO chart_label (chart_id, source_type, label_key, label_value, date_created) VALUES "
#define STORE_HOST_OR_CHART_LABEL_VALUE "(u2h('%s'), %d,'%s','%s', unixepoch())"
#define DELETE_DIMENSION_UUID "DELETE FROM dimension WHERE dim_id = @uuid;"
#define SQL_STORE_HOST_INFO \
"INSERT OR REPLACE INTO host (host_id, hostname, registry_hostname, update_every, os, timezone, tags, hops, " \
"memory_mode, abbrev_timezone, utc_offset, program_name, program_version, entries, health_enabled, last_connected) " \
"VALUES (@host_id, @hostname, @registry_hostname, @update_every, @os, @timezone, @tags, @hops, " \
"@memory_mode, @abbrev_tz, @utc_offset, @prog_name, @prog_version, @entries, @health_enabled, @last_connected);"
#define SQL_STORE_CHART \
"INSERT INTO chart (chart_id, host_id, type, id, name, family, context, title, unit, plugin, module, priority, " \
"update_every, chart_type, memory_mode, history_entries) " \
"values (@chart_id, @host_id, @type, @id, @name, @family, @context, @title, @unit, @plugin, @module, @priority, " \
"@update_every, @chart_type, @memory_mode, @history_entries) " \
"ON CONFLICT(chart_id) DO UPDATE SET type=excluded.type, id=excluded.id, name=excluded.name, " \
"family=excluded.family, context=excluded.context, title=excluded.title, unit=excluded.unit, " \
"plugin=excluded.plugin, module=excluded.module, priority=excluded.priority, update_every=excluded.update_every, " \
"chart_type=excluded.chart_type, memory_mode = excluded.memory_mode, history_entries = excluded.history_entries"
#define SQL_STORE_DIMENSION \
"INSERT INTO dimension (dim_id, chart_id, id, name, multiplier, divisor , algorithm, options) " \
"VALUES (@dim_id, @chart_id, @id, @name, @multiplier, @divisor, @algorithm, @options) " \
"ON CONFLICT(dim_id) DO UPDATE SET id=excluded.id, name=excluded.name, multiplier=excluded.multiplier, " \
"divisor=excluded.divisor, algorithm=excluded.algorithm, options=excluded.options"
#define SELECT_DIMENSION_LIST "SELECT dim_id, rowid FROM dimension WHERE rowid > @row_id"
#define SELECT_CHART_LIST "SELECT chart_id, rowid FROM chart WHERE rowid > @row_id"
#define SELECT_CHART_LABEL_LIST "SELECT chart_id, rowid FROM chart_label WHERE rowid > @row_id"
#define SQL_STORE_HOST_SYSTEM_INFO_VALUES \
"INSERT OR REPLACE INTO host_info (host_id, system_key, system_value, date_created) VALUES " \
"(@uuid, @name, @value, UNIXEPOCH())"
#define MIGRATE_LOCALHOST_TO_NEW_MACHINE_GUID \
"UPDATE chart SET host_id = @host_id WHERE host_id in (SELECT host_id FROM host where host_id <> @host_id and hops = 0);"
#define DELETE_NON_EXISTING_LOCALHOST "DELETE FROM host WHERE hops = 0 AND host_id <> @host_id;"
#define DELETE_MISSING_NODE_INSTANCES "DELETE FROM node_instance WHERE host_id NOT IN (SELECT host_id FROM host);"
#define METADATA_CMD_Q_MAX_SIZE (2048) // Max queue size; callers will block until there is room
#define METADATA_MAINTENANCE_FIRST_CHECK (1800) // Maintenance first run after agent startup in seconds
#define METADATA_MAINTENANCE_REPEAT (60) // Repeat if last run for dimensions, charts, labels needs more work
#define METADATA_HEALTH_LOG_INTERVAL (3600) // Repeat maintenance for health
#define METADATA_DIM_CHECK_INTERVAL (3600) // Repeat maintenance for dimensions
#define METADATA_CHART_CHECK_INTERVAL (3600) // Repeat maintenance for charts
#define METADATA_LABEL_CHECK_INTERVAL (3600) // Repeat maintenance for labels
#define METADATA_RUNTIME_THRESHOLD (5) // Run time threshold for cleanup task
#define METADATA_HOST_CHECK_FIRST_CHECK (5) // First check for pending metadata
#define METADATA_HOST_CHECK_INTERVAL (30) // Repeat check for pending metadata
#define METADATA_HOST_CHECK_IMMEDIATE (5) // Repeat immediate run because we have more metadata to write
#define MAX_METADATA_CLEANUP (500) // Maximum metadata write operations (e.g deletes before retrying)
#define METADATA_MAX_BATCH_SIZE (512) // Maximum commands to execute before running the event loop
#define DATABASE_FREE_PAGES_THRESHOLD_PC (5) // Percentage of free pages to trigger vacuum
#define DATABASE_FREE_PAGES_VACUUM_PC (10) // Percentage of free pages to vacuum
enum metadata_opcode {
METADATA_DATABASE_NOOP = 0,
METADATA_DATABASE_TIMER,
METADATA_DEL_DIMENSION,
METADATA_STORE_CLAIM_ID,
METADATA_ADD_HOST_INFO,
METADATA_SCAN_HOSTS,
METADATA_LOAD_HOST_CONTEXT,
METADATA_MAINTENANCE,
METADATA_SYNC_SHUTDOWN,
METADATA_UNITTEST,
METADATA_ML_LOAD_MODELS,
// leave this last
// we need it to check for worker utilization
METADATA_MAX_ENUMERATIONS_DEFINED
};
#define MAX_PARAM_LIST (2)
struct metadata_cmd {
enum metadata_opcode opcode;
struct completion *completion;
const void *param[MAX_PARAM_LIST];
struct metadata_cmd *prev, *next;
};
struct metadata_database_cmdqueue {
struct metadata_cmd *cmd_base;
};
typedef enum {
METADATA_FLAG_PROCESSING = (1 << 0), // store or cleanup
METADATA_FLAG_SHUTDOWN = (1 << 1), // Shutting down
METADATA_FLAG_ML_LOADING = (1 << 2), // ML model load in progress
} METADATA_FLAG;
struct metadata_wc {
uv_thread_t thread;
uv_loop_t *loop;
uv_async_t async;
uv_timer_t timer_req;
time_t metadata_check_after;
volatile unsigned queue_size;
METADATA_FLAG flags;
struct completion init_complete;
struct completion *scan_complete;
/* FIFO command queue */
uv_mutex_t cmd_mutex;
struct metadata_database_cmdqueue cmd_queue;
};
#define metadata_flag_check(target_flags, flag) (__atomic_load_n(&((target_flags)->flags), __ATOMIC_SEQ_CST) & (flag))
#define metadata_flag_set(target_flags, flag) __atomic_or_fetch(&((target_flags)->flags), (flag), __ATOMIC_SEQ_CST)
#define metadata_flag_clear(target_flags, flag) __atomic_and_fetch(&((target_flags)->flags), ~(flag), __ATOMIC_SEQ_CST)
//
// For unittest
//
struct thread_unittest {
int join;
unsigned added;
unsigned processed;
unsigned *done;
};
// Metadata functions
struct query_build {
BUFFER *sql;
int count;
char uuid_str[UUID_STR_LEN];
};
static int host_label_store_to_sql_callback(const char *name, const char *value, RRDLABEL_SRC ls, void *data) {
struct query_build *lb = data;
if (unlikely(!lb->count))
buffer_sprintf(lb->sql, STORE_HOST_LABEL);
else
buffer_strcat(lb->sql, ", ");
buffer_sprintf(lb->sql, STORE_HOST_OR_CHART_LABEL_VALUE, lb->uuid_str, (int) (ls & ~(RRDLABEL_FLAG_INTERNAL)), name, value);
lb->count++;
return 1;
}
static int chart_label_store_to_sql_callback(const char *name, const char *value, RRDLABEL_SRC ls, void *data) {
struct query_build *lb = data;
if (unlikely(!lb->count))
buffer_sprintf(lb->sql, STORE_CHART_LABEL);
else
buffer_strcat(lb->sql, ", ");
buffer_sprintf(lb->sql, STORE_HOST_OR_CHART_LABEL_VALUE, lb->uuid_str, (int) (ls & ~(RRDLABEL_FLAG_INTERNAL)), name, value);
lb->count++;
return 1;
}
#define SQL_DELETE_CHART_LABEL "DELETE FROM chart_label WHERE chart_id = @chart_id;"
#define SQL_DELETE_CHART_LABEL_HISTORY "DELETE FROM chart_label WHERE date_created < %ld AND chart_id = @chart_id;"
static void clean_old_chart_labels(RRDSET *st)
{
char sql[512];
time_t first_time_s = rrdset_first_entry_s(st);
if (unlikely(!first_time_s))
snprintfz(sql, 511,SQL_DELETE_CHART_LABEL);
else
snprintfz(sql, 511,SQL_DELETE_CHART_LABEL_HISTORY, first_time_s);
int rc = exec_statement_with_uuid(sql, &st->chart_uuid);
if (unlikely(rc))
error_report("METADATA: 'host:%s' Failed to clean old labels for chart %s", rrdhost_hostname(st->rrdhost), rrdset_name(st));
}
static int check_and_update_chart_labels(RRDSET *st, BUFFER *work_buffer, size_t *query_counter)
{
size_t old_version = st->rrdlabels_last_saved_version;
size_t new_version = rrdlabels_version(st->rrdlabels);
if (new_version == old_version)
return 0;
struct query_build tmp = {.sql = work_buffer, .count = 0};
uuid_unparse_lower(st->chart_uuid, tmp.uuid_str);
rrdlabels_walkthrough_read(st->rrdlabels, chart_label_store_to_sql_callback, &tmp);
buffer_strcat(work_buffer, " ON CONFLICT (chart_id, label_key) DO UPDATE SET source_type = excluded.source_type, label_value=excluded.label_value, date_created=UNIXEPOCH()");
int rc = db_execute(db_meta, buffer_tostring(work_buffer));
if (likely(!rc)) {
st->rrdlabels_last_saved_version = new_version;
(*query_counter)++;
}
clean_old_chart_labels(st);
return rc;
}
// Migrate all hosts with hops zero to this host_uuid
void migrate_localhost(uuid_t *host_uuid)
{
int rc;
rc = exec_statement_with_uuid(MIGRATE_LOCALHOST_TO_NEW_MACHINE_GUID, host_uuid);
if (!rc)
rc = exec_statement_with_uuid(DELETE_NON_EXISTING_LOCALHOST, host_uuid);
if (!rc) {
if (unlikely(db_execute(db_meta, DELETE_MISSING_NODE_INSTANCES)))
error_report("Failed to remove deleted hosts from node instances");
}
}
static int store_claim_id(uuid_t *host_id, uuid_t *claim_id)
{
sqlite3_stmt *res = NULL;
int rc;
if (unlikely(!db_meta)) {
if (default_rrd_memory_mode == RRD_MEMORY_MODE_DBENGINE)
error_report("Database has not been initialized");
return 1;
}
rc = sqlite3_prepare_v2(db_meta, SQL_STORE_CLAIM_ID, -1, &res, 0);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to prepare statement to store host claim id");
return 1;
}
rc = sqlite3_bind_blob(res, 1, host_id, sizeof(*host_id), SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to bind host_id parameter to store claim id");
goto failed;
}
if (claim_id)
rc = sqlite3_bind_blob(res, 2, claim_id, sizeof(*claim_id), SQLITE_STATIC);
else
rc = sqlite3_bind_null(res, 2);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to bind claim_id parameter to host claim id");
goto failed;
}
rc = execute_insert(res);
if (unlikely(rc != SQLITE_DONE))
error_report("Failed to store host claim id rc = %d", rc);
failed:
if (unlikely(sqlite3_finalize(res) != SQLITE_OK))
error_report("Failed to finalize the prepared statement when storing a host claim id");
return rc != SQLITE_DONE;
}
static void delete_dimension_uuid(uuid_t *dimension_uuid, sqlite3_stmt **action_res __maybe_unused, bool flag __maybe_unused)
{
static __thread sqlite3_stmt *res = NULL;
int rc;
if (unlikely(!res)) {
rc = prepare_statement(db_meta, DELETE_DIMENSION_UUID, &res);
if (rc != SQLITE_OK) {
error_report("Failed to prepare statement to delete a dimension uuid");
return;
}
}
rc = sqlite3_bind_blob(res, 1, dimension_uuid, sizeof(*dimension_uuid), SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto skip_execution;
rc = sqlite3_step_monitored(res);
if (unlikely(rc != SQLITE_DONE))
error_report("Failed to delete dimension uuid, rc = %d", rc);
skip_execution:
rc = sqlite3_reset(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to reset statement when deleting dimension UUID, rc = %d", rc);
}
//
// Store host and host system info information in the database
static int store_host_metadata(RRDHOST *host)
{
static __thread sqlite3_stmt *res = NULL;
int rc, param = 0;
if (unlikely((!res))) {
rc = prepare_statement(db_meta, SQL_STORE_HOST_INFO, &res);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to prepare statement to store host, rc = %d", rc);
return 1;
}
}
rc = sqlite3_bind_blob(res, ++param, &host->host_uuid, sizeof(host->host_uuid), SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = bind_text_null(res, ++param, rrdhost_hostname(host), 0);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = bind_text_null(res, ++param, rrdhost_registry_hostname(host), 1);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, host->rrd_update_every);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = bind_text_null(res, ++param, rrdhost_os(host), 1);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = bind_text_null(res, ++param, rrdhost_timezone(host), 1);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = bind_text_null(res, ++param, rrdhost_tags(host), 1);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, host->system_info ? host->system_info->hops : 0);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, host->rrd_memory_mode);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = bind_text_null(res, ++param, rrdhost_abbrev_timezone(host), 1);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, host->utc_offset);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = bind_text_null(res, ++param, rrdhost_program_name(host), 1);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = bind_text_null(res, ++param, rrdhost_program_version(host), 1);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int64(res, ++param, host->rrd_history_entries);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, (int ) host->health.health_enabled);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int64(res, ++param, (sqlite3_int64) host->last_connected);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
int store_rc = sqlite3_step_monitored(res);
if (unlikely(store_rc != SQLITE_DONE))
error_report("Failed to store host %s, rc = %d", rrdhost_hostname(host), rc);
rc = sqlite3_reset(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to reset statement to store host %s, rc = %d", rrdhost_hostname(host), rc);
return store_rc != SQLITE_DONE;
bind_fail:
error_report("Failed to bind %d parameter to store host %s, rc = %d", param, rrdhost_hostname(host), rc);
rc = sqlite3_reset(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to reset statement to store host %s, rc = %d", rrdhost_hostname(host), rc);
return 1;
}
static int add_host_sysinfo_key_value(const char *name, const char *value, uuid_t *uuid)
{
static __thread sqlite3_stmt *res = NULL;
int rc, param = 0;
if (unlikely(!db_meta)) {
if (default_rrd_memory_mode != RRD_MEMORY_MODE_DBENGINE)
return 0;
error_report("Database has not been initialized");
return 0;
}
if (unlikely((!res))) {
rc = prepare_statement(db_meta, SQL_STORE_HOST_SYSTEM_INFO_VALUES, &res);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to prepare statement to store host info values, rc = %d", rc);
return 0;
}
}
rc = sqlite3_bind_blob(res, ++param, uuid, sizeof(*uuid), SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = bind_text_null(res, ++param, name, 0);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = bind_text_null(res, ++param, value ? value : "unknown", 0);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
int store_rc = sqlite3_step_monitored(res);
if (unlikely(store_rc != SQLITE_DONE))
error_report("Failed to store host info value %s, rc = %d", name, rc);
rc = sqlite3_reset(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to reset statement to store host info value %s, rc = %d", name, rc);
return store_rc == SQLITE_DONE;
bind_fail:
error_report("Failed to bind %d parameter to store host info values %s, rc = %d", param, name, rc);
rc = sqlite3_reset(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to reset statement to store host info values %s, rc = %d", name, rc);
return 0;
}
static bool store_host_systeminfo(RRDHOST *host)
{
struct rrdhost_system_info *system_info = host->system_info;
if (unlikely(!system_info))
return false;
int ret = 0;
ret += add_host_sysinfo_key_value("NETDATA_CONTAINER_OS_NAME", system_info->container_os_name, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_CONTAINER_OS_ID", system_info->container_os_id, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_CONTAINER_OS_ID_LIKE", system_info->container_os_id_like, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_CONTAINER_OS_VERSION", system_info->container_os_version, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_CONTAINER_OS_VERSION_ID", system_info->container_os_version_id, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_CONTAINER_OS_DETECTION", system_info->host_os_detection, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_HOST_OS_NAME", system_info->host_os_name, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_HOST_OS_ID", system_info->host_os_id, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_HOST_OS_ID_LIKE", system_info->host_os_id_like, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_HOST_OS_VERSION", system_info->host_os_version, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_HOST_OS_VERSION_ID", system_info->host_os_version_id, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_HOST_OS_DETECTION", system_info->host_os_detection, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_KERNEL_NAME", system_info->kernel_name, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_CPU_LOGICAL_CPU_COUNT", system_info->host_cores, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_CPU_FREQ", system_info->host_cpu_freq, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_TOTAL_RAM", system_info->host_ram_total, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_TOTAL_DISK_SIZE", system_info->host_disk_space, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_KERNEL_VERSION", system_info->kernel_version, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_ARCHITECTURE", system_info->architecture, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_VIRTUALIZATION", system_info->virtualization, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_VIRT_DETECTION", system_info->virt_detection, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_CONTAINER", system_info->container, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_SYSTEM_CONTAINER_DETECTION", system_info->container_detection, &host->host_uuid);
ret += add_host_sysinfo_key_value("NETDATA_HOST_IS_K8S_NODE", system_info->is_k8s_node, &host->host_uuid);
return !(24 == ret);
}
/*
* Store a chart in the database
*/
static int store_chart_metadata(RRDSET *st)
{
static __thread sqlite3_stmt *res = NULL;
int rc, param = 0, store_rc = 0;
if (unlikely(!res)) {
rc = prepare_statement(db_meta, SQL_STORE_CHART, &res);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to prepare statement to store chart, rc = %d", rc);
return 1;
}
}
rc = sqlite3_bind_blob(res, ++param, &st->chart_uuid, sizeof(st->chart_uuid), SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_blob(res, ++param, &st->rrdhost->host_uuid, sizeof(st->rrdhost->host_uuid), SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_text(res, ++param, string2str(st->parts.type), -1, SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_text(res, ++param, string2str(st->parts.id), -1, SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
const char *name = string2str(st->parts.name);
if (name && *name)
rc = sqlite3_bind_text(res, ++param, name, -1, SQLITE_STATIC);
else
rc = sqlite3_bind_null(res, ++param);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_text(res, ++param, rrdset_family(st), -1, SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_text(res, ++param, rrdset_context(st), -1, SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_text(res, ++param, rrdset_title(st), -1, SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_text(res, ++param, rrdset_units(st), -1, SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_text(res, ++param, rrdset_plugin_name(st), -1, SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_text(res, ++param, rrdset_module_name(st), -1, SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, (int) st->priority);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, st->update_every);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, st->chart_type);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, st->rrd_memory_mode);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, (int) st->db.entries);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
store_rc = execute_insert(res);
if (unlikely(store_rc != SQLITE_DONE))
error_report("Failed to store chart, rc = %d", store_rc);
rc = sqlite3_reset(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to reset statement in chart store function, rc = %d", rc);
return store_rc != SQLITE_DONE;
bind_fail:
error_report("Failed to bind parameter %d to store chart, rc = %d", param, rc);
rc = sqlite3_reset(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to reset statement in chart store function, rc = %d", rc);
return 1;
}
/*
* Store a dimension
*/
static int store_dimension_metadata(RRDDIM *rd)
{
static __thread sqlite3_stmt *res = NULL;
int rc, param = 0;
if (unlikely(!res)) {
rc = prepare_statement(db_meta, SQL_STORE_DIMENSION, &res);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to prepare statement to store dimension, rc = %d", rc);
return 1;
}
}
rc = sqlite3_bind_blob(res, ++param, &rd->metric_uuid, sizeof(rd->metric_uuid), SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_blob(res, ++param, &rd->rrdset->chart_uuid, sizeof(rd->rrdset->chart_uuid), SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_text(res, ++param, string2str(rd->id), -1, SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_text(res, ++param, string2str(rd->name), -1, SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, (int) rd->multiplier);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, (int ) rd->divisor);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = sqlite3_bind_int(res, ++param, rd->algorithm);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
if (rrddim_option_check(rd, RRDDIM_OPTION_HIDDEN))
rc = sqlite3_bind_text(res, ++param, "hidden", -1, SQLITE_STATIC);
else
rc = sqlite3_bind_null(res, ++param);
if (unlikely(rc != SQLITE_OK))
goto bind_fail;
rc = execute_insert(res);
if (unlikely(rc != SQLITE_DONE))
error_report("Failed to store dimension, rc = %d", rc);
rc = sqlite3_reset(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to reset statement in store dimension, rc = %d", rc);
return 0;
bind_fail:
error_report("Failed to bind parameter %d to store dimension, rc = %d", param, rc);
rc = sqlite3_reset(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to reset statement in store dimension, rc = %d", rc);
return 1;
}
static bool dimension_can_be_deleted(uuid_t *dim_uuid __maybe_unused, sqlite3_stmt **res __maybe_unused, bool flag __maybe_unused)
{
#ifdef ENABLE_DBENGINE
if(dbengine_enabled) {
bool no_retention = true;
for (size_t tier = 0; tier < storage_tiers; tier++) {
if (!multidb_ctx[tier])
continue;
time_t first_time_t = 0, last_time_t = 0;
if (rrdeng_metric_retention_by_uuid((void *) multidb_ctx[tier], dim_uuid, &first_time_t, &last_time_t)) {
if (first_time_t > 0) {
no_retention = false;
break;
}
}
}
return no_retention;
}
else
return false;
#else
return false;
#endif
}
int get_pragma_value(sqlite3 *database, const char *sql)
{
sqlite3_stmt *res = NULL;
int rc = sqlite3_prepare_v2(database, sql, -1, &res, 0);
if (unlikely(rc != SQLITE_OK))
return -1;
int result = -1;
rc = sqlite3_step_monitored(res);
if (likely(rc == SQLITE_ROW))
result = sqlite3_column_int(res, 0);
rc = sqlite3_finalize(res);
(void) rc;
return result;
}
int get_free_page_count(sqlite3 *database)
{
return get_pragma_value(database, "PRAGMA freelist_count");
}
int get_database_page_count(sqlite3 *database)
{
return get_pragma_value(database, "PRAGMA page_count");
}
static bool run_cleanup_loop(
sqlite3_stmt *res,
struct metadata_wc *wc,
bool (*check_cb)(uuid_t *, sqlite3_stmt **, bool),
void (*action_cb)(uuid_t *, sqlite3_stmt **, bool),
uint32_t *total_checked,
uint32_t *total_deleted,
uint64_t *row_id,
sqlite3_stmt **check_stmt,
sqlite3_stmt **action_stmt,
bool check_flag,
bool action_flag)
{
if (unlikely(metadata_flag_check(wc, METADATA_FLAG_SHUTDOWN)))
return true;
int rc = sqlite3_bind_int64(res, 1, (sqlite3_int64) *row_id);
if (unlikely(rc != SQLITE_OK))
return true;
time_t start_running = now_monotonic_sec();
bool time_expired = false;
while (!time_expired && sqlite3_step_monitored(res) == SQLITE_ROW &&
(*total_deleted < MAX_METADATA_CLEANUP && *total_checked < MAX_METADATA_CLEANUP)) {
if (unlikely(metadata_flag_check(wc, METADATA_FLAG_SHUTDOWN)))
break;
*row_id = sqlite3_column_int64(res, 1);
rc = check_cb((uuid_t *)sqlite3_column_blob(res, 0), check_stmt, check_flag);
if (rc == true) {
action_cb((uuid_t *)sqlite3_column_blob(res, 0), action_stmt, action_flag);
(*total_deleted)++;
}
(*total_checked)++;
time_expired = ((now_monotonic_sec() - start_running) > METADATA_RUNTIME_THRESHOLD);
}
return time_expired || (*total_checked == MAX_METADATA_CLEANUP) || (*total_deleted == MAX_METADATA_CLEANUP);
}
#define SQL_CHECK_CHART_EXISTENCE_IN_DIMENSION "SELECT count(1) FROM dimension WHERE chart_id = @chart_id"
#define SQL_CHECK_CHART_EXISTENCE_IN_CHART "SELECT count(1) FROM chart WHERE chart_id = @chart_id"
static bool chart_can_be_deleted(uuid_t *chart_uuid, sqlite3_stmt **check_res, bool check_in_dimension)
{
int rc, result = 1;
sqlite3_stmt *res = check_res ? *check_res : NULL;
if (!res) {
if (check_in_dimension)
rc = sqlite3_prepare_v2(db_meta, SQL_CHECK_CHART_EXISTENCE_IN_DIMENSION, -1, &res, 0);
else
rc = sqlite3_prepare_v2(db_meta, SQL_CHECK_CHART_EXISTENCE_IN_CHART, -1, &res, 0);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to prepare statement to check for chart existence, rc = %d", rc);
return 0;
}
if (check_res)
*check_res = res;
}
rc = sqlite3_bind_blob(res, 1, chart_uuid, sizeof(*chart_uuid), SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to bind chart uuid parameter, rc = %d", rc);
goto skip;
}
rc = sqlite3_step_monitored(res);
if (likely(rc == SQLITE_ROW))
result = sqlite3_column_int(res, 0);
skip:
if (check_res)
rc = sqlite3_reset(res);
else
rc = sqlite3_finalize(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to %s statement that checks chart uuid existence rc = %d", check_res ? "reset" : "finalize", rc);
return result == 0;
}
#define SQL_DELETE_CHART_BY_UUID "DELETE FROM chart WHERE chart_id = @chart_id"
#define SQL_DELETE_CHART_LABEL_BY_UUID "DELETE FROM chart_label WHERE chart_id = @chart_id"
static void delete_chart_uuid(uuid_t *chart_uuid, sqlite3_stmt **action_res, bool label_only)
{
int rc;
sqlite3_stmt *res = action_res ? *action_res : NULL;
if (!res) {
if (label_only)
rc = sqlite3_prepare_v2(db_meta, SQL_DELETE_CHART_LABEL_BY_UUID, -1, &res, 0);
else
rc = sqlite3_prepare_v2(db_meta, SQL_DELETE_CHART_BY_UUID, -1, &res, 0);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to prepare statement to check for chart existence, rc = %d", rc);
return;
}
if (action_res)
*action_res = res;
}
rc = sqlite3_bind_blob(res, 1, chart_uuid, sizeof(*chart_uuid), SQLITE_STATIC);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to bind chart uuid parameter, rc = %d", rc);
goto skip;
}
rc = sqlite3_step_monitored(res);
if (unlikely(rc != SQLITE_DONE))
error_report("Failed to delete a chart uuid from the %s table, rc = %d", label_only ? "labels" : "chart", rc);
skip:
if (action_res)
rc = sqlite3_reset(res);
else
rc = sqlite3_finalize(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to %s statement that deletes a chart uuid rc = %d", action_res ? "reset" : "finalize", rc);
}
static void check_dimension_metadata(struct metadata_wc *wc)
{
static time_t next_execution_t = 0;
static uint64_t last_row_id = 0;
time_t now = now_realtime_sec();
if (!next_execution_t)
next_execution_t = now + METADATA_MAINTENANCE_FIRST_CHECK;
if (next_execution_t && next_execution_t > now)
return;
int rc;
sqlite3_stmt *res = NULL;
rc = sqlite3_prepare_v2(db_meta, SELECT_DIMENSION_LIST, -1, &res, 0);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to prepare statement to fetch host dimensions");
return;
}
uint32_t total_checked = 0;
uint32_t total_deleted = 0;
internal_error(true, "METADATA: Checking dimensions starting after row %"PRIu64, last_row_id);
bool more_to_do = run_cleanup_loop(
res,
wc,
dimension_can_be_deleted,
delete_dimension_uuid,
&total_checked,
&total_deleted,
&last_row_id,
NULL,
NULL,
false,
false);
now = now_realtime_sec();
if (more_to_do)
next_execution_t = now + METADATA_MAINTENANCE_REPEAT;
else {
last_row_id = 0;
next_execution_t = now + METADATA_DIM_CHECK_INTERVAL;
}
netdata_log_info(
"METADATA: Dimensions checked %u, deleted %u. Checks will %s in %lld seconds",
total_checked,
total_deleted,
last_row_id ? "resume" : "restart",
(long long)(next_execution_t - now));
rc = sqlite3_finalize(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to finalize the prepared statement to check dimensions");
}
static void check_chart_metadata(struct metadata_wc *wc)
{
static time_t next_execution_t = 0;
static uint64_t last_row_id = 0;
time_t now = now_realtime_sec();
if (!next_execution_t)
next_execution_t = now + METADATA_MAINTENANCE_FIRST_CHECK;
if (next_execution_t && next_execution_t > now)
return;
sqlite3_stmt *res = NULL;
int rc = sqlite3_prepare_v2(db_meta, SELECT_CHART_LIST, -1, &res, 0);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to prepare statement to fetch charts");
return;
}
uint32_t total_checked = 0;
uint32_t total_deleted = 0;
internal_error(true, "METADATA: Checking charts starting after row %"PRIu64, last_row_id);
sqlite3_stmt *check_res = NULL;
sqlite3_stmt *action_res = NULL;
bool more_to_do = run_cleanup_loop(
res,
wc,
chart_can_be_deleted,
delete_chart_uuid,
&total_checked,
&total_deleted,
&last_row_id,
&check_res,
&action_res,
true,
false);
if (check_res)
sqlite3_finalize(check_res);
if (action_res)
sqlite3_finalize(action_res);
now = now_realtime_sec();
if (more_to_do)
next_execution_t = now + METADATA_MAINTENANCE_REPEAT;
else {
last_row_id = 0;
next_execution_t = now + METADATA_CHART_CHECK_INTERVAL;
}
netdata_log_info(
"METADATA: Charts checked %u, deleted %u. Checks will %s in %lld seconds",
total_checked,
total_deleted,
last_row_id ? "resume" : "restart",
(long long)(next_execution_t - now));
rc = sqlite3_finalize(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to finalize the prepared statement when reading charts");
}
static void check_label_metadata(struct metadata_wc *wc)
{
static time_t next_execution_t = 0;
static uint64_t last_row_id = 0;
time_t now = now_realtime_sec();
if (!next_execution_t)
next_execution_t = now + METADATA_MAINTENANCE_FIRST_CHECK;
if (next_execution_t && next_execution_t > now)
return;
int rc;
sqlite3_stmt *res = NULL;
rc = sqlite3_prepare_v2(db_meta, SELECT_CHART_LABEL_LIST, -1, &res, 0);
if (unlikely(rc != SQLITE_OK)) {
error_report("Failed to prepare statement to fetch charts");
return;
}
uint32_t total_checked = 0;
uint32_t total_deleted = 0;
internal_error(true,"METADATA: Checking charts labels starting after row %"PRIu64, last_row_id);
sqlite3_stmt *check_res = NULL;
sqlite3_stmt *action_res = NULL;
bool more_to_do = run_cleanup_loop(
res,
wc,
chart_can_be_deleted,
delete_chart_uuid,
&total_checked,
&total_deleted,
&last_row_id,
&check_res,
&action_res,
false,
true);
if (check_res)
sqlite3_finalize(check_res);
if (action_res)
sqlite3_finalize(action_res);
now = now_realtime_sec();
if (more_to_do)
next_execution_t = now + METADATA_MAINTENANCE_REPEAT;
else {
last_row_id = 0;
next_execution_t = now + METADATA_LABEL_CHECK_INTERVAL;
}
netdata_log_info(
"METADATA: Chart labels checked %u, deleted %u. Checks will %s in %lld seconds",
total_checked,
total_deleted,
last_row_id ? "resume" : "restart",
(long long)(next_execution_t - now));
rc = sqlite3_finalize(res);
if (unlikely(rc != SQLITE_OK))
error_report("Failed to finalize the prepared statement when checking charts");
}
static void cleanup_health_log(struct metadata_wc *wc)
{
static time_t next_execution_t = 0;
time_t now = now_realtime_sec();
if (!next_execution_t)
next_execution_t = now + METADATA_MAINTENANCE_FIRST_CHECK;
if (next_execution_t && next_execution_t > now)
return;
next_execution_t = now + METADATA_HEALTH_LOG_INTERVAL;
RRDHOST *host;
bool is_claimed = claimed();
dfe_start_reentrant(rrdhost_root_index, host){
if (rrdhost_flag_check(host, RRDHOST_FLAG_ARCHIVED))
continue;
sql_health_alarm_log_cleanup(host, is_claimed);
if (unlikely(metadata_flag_check(wc, METADATA_FLAG_SHUTDOWN)))
break;
}
dfe_done(host);
if (unlikely(metadata_flag_check(wc, METADATA_FLAG_SHUTDOWN)))
return;
(void) db_execute(db_meta,"DELETE FROM health_log WHERE host_id NOT IN (SELECT host_id FROM host)");
(void) db_execute(db_meta,"DELETE FROM health_log_detail WHERE health_log_id NOT IN (SELECT health_log_id FROM health_log)");
}
//
// EVENT LOOP STARTS HERE
//
static void metadata_init_cmd_queue(struct metadata_wc *wc)
{
wc->cmd_queue.cmd_base = NULL;
fatal_assert(0 == uv_mutex_init(&wc->cmd_mutex));
}
static void metadata_free_cmd_queue(struct metadata_wc *wc)
{
uv_mutex_lock(&wc->cmd_mutex);
while(wc->cmd_queue.cmd_base) {
struct metadata_cmd *t = wc->cmd_queue.cmd_base;
DOUBLE_LINKED_LIST_REMOVE_ITEM_UNSAFE(wc->cmd_queue.cmd_base, t, prev, next);
freez(t);
}
uv_mutex_unlock(&wc->cmd_mutex);
}
static void metadata_enq_cmd(struct metadata_wc *wc, struct metadata_cmd *cmd)
{
if (cmd->opcode == METADATA_SYNC_SHUTDOWN) {
metadata_flag_set(wc, METADATA_FLAG_SHUTDOWN);
goto wakeup_event_loop;
}
if (unlikely(metadata_flag_check(wc, METADATA_FLAG_SHUTDOWN)))
goto wakeup_event_loop;
struct metadata_cmd *t = mallocz(sizeof(*t));
*t = *cmd;
t->prev = t->next = NULL;
uv_mutex_lock(&wc->cmd_mutex);
DOUBLE_LINKED_LIST_APPEND_ITEM_UNSAFE(wc->cmd_queue.cmd_base, t, prev, next);
uv_mutex_unlock(&wc->cmd_mutex);
wakeup_event_loop:
(void) uv_async_send(&wc->async);
}
static struct metadata_cmd metadata_deq_cmd(struct metadata_wc *wc)
{
struct metadata_cmd ret;
uv_mutex_lock(&wc->cmd_mutex);
if(wc->cmd_queue.cmd_base) {
struct metadata_cmd *t = wc->cmd_queue.cmd_base;
DOUBLE_LINKED_LIST_REMOVE_ITEM_UNSAFE(wc->cmd_queue.cmd_base, t, prev, next);
ret = *t;
freez(t);
}
else {
ret.opcode = METADATA_DATABASE_NOOP;
ret.completion = NULL;
}
uv_mutex_unlock(&wc->cmd_mutex);
return ret;
}
static void async_cb(uv_async_t *handle)
{
uv_stop(handle->loop);
uv_update_time(handle->loop);
}
#define TIMER_INITIAL_PERIOD_MS (1000)
#define TIMER_REPEAT_PERIOD_MS (1000)
static void timer_cb(uv_timer_t* handle)
{
uv_stop(handle->loop);
uv_update_time(handle->loop);
struct metadata_wc *wc = handle->data;
struct metadata_cmd cmd;
memset(&cmd, 0, sizeof(cmd));
time_t now = now_realtime_sec();
if (wc->metadata_check_after && wc->metadata_check_after < now) {
cmd.opcode = METADATA_SCAN_HOSTS;
metadata_enq_cmd(wc, &cmd);
}
}
void vacuum_database(sqlite3 *database, const char *db_alias, int threshold, int vacuum_pc)
{
int free_pages = get_free_page_count(database);
int total_pages = get_database_page_count(database);
if (!threshold)
threshold = DATABASE_FREE_PAGES_THRESHOLD_PC;
if (!vacuum_pc)
vacuum_pc = DATABASE_FREE_PAGES_VACUUM_PC;
if (free_pages > (total_pages * threshold / 100)) {
int do_free_pages = (int) (free_pages * vacuum_pc / 100);
netdata_log_info("%s: Freeing %d database pages", db_alias, do_free_pages);
char sql[128];
snprintfz(sql, 127, "PRAGMA incremental_vacuum(%d)", do_free_pages);
(void) db_execute(database, sql);
}
}
void run_metadata_cleanup(struct metadata_wc *wc)
{
if (unlikely(metadata_flag_check(wc, METADATA_FLAG_SHUTDOWN)))
return;
check_dimension_metadata(wc);
check_chart_metadata(wc);
check_label_metadata(wc);
cleanup_health_log(wc);
if (unlikely(metadata_flag_check(wc, METADATA_FLAG_SHUTDOWN)))
return;
vacuum_database(db_meta, "METADATA", DATABASE_FREE_PAGES_THRESHOLD_PC, DATABASE_FREE_PAGES_VACUUM_PC);
(void) sqlite3_wal_checkpoint(db_meta, NULL);
}
struct ml_model_payload {
uv_work_t request;
struct metadata_wc *wc;
Pvoid_t JudyL;
size_t count;
};
struct scan_metadata_payload {
uv_work_t request;
struct metadata_wc *wc;
BUFFER *work_buffer;
uint32_t max_count;
};
struct host_context_load_thread {
uv_thread_t thread;
RRDHOST *host;
bool busy;
bool finished;
};
static void restore_host_context(void *arg)
{
struct host_context_load_thread *hclt = arg;
RRDHOST *host = hclt->host;
usec_t started_ut = now_monotonic_usec(); (void)started_ut;
rrdhost_load_rrdcontext_data(host);
usec_t ended_ut = now_monotonic_usec(); (void)ended_ut;
rrdhost_flag_clear(host, RRDHOST_FLAG_PENDING_CONTEXT_LOAD | RRDHOST_FLAG_CONTEXT_LOAD_IN_PROGRESS);
#ifdef ENABLE_ACLK
aclk_queue_node_info(host, false);
#endif
internal_error(true, "METADATA: 'host:%s' context load in %0.2f ms", rrdhost_hostname(host),
(double)(ended_ut - started_ut) / USEC_PER_MS);
__atomic_store_n(&hclt->finished, true, __ATOMIC_RELEASE);
}
// Callback after scan of hosts is done
static void after_start_host_load_context(uv_work_t *req, int status __maybe_unused)
{
struct scan_metadata_payload *data = req->data;
freez(data);
}
#define MAX_FIND_THREAD_RETRIES (10)
static void cleanup_finished_threads(struct host_context_load_thread *hclt, size_t max_thread_slots, bool wait)
{
for (size_t index = 0; index < max_thread_slots; index++) {
if (__atomic_load_n(&(hclt[index].finished), __ATOMIC_RELAXED)
|| (wait && __atomic_load_n(&(hclt[index].busy), __ATOMIC_ACQUIRE))) {
int rc = uv_thread_join(&(hclt[index].thread));
if (rc)
netdata_log_error("Failed to join thread, rc = %d",rc);
__atomic_store_n(&(hclt[index].busy), false, __ATOMIC_RELEASE);
__atomic_store_n(&(hclt[index].finished), false, __ATOMIC_RELEASE);
}
}
}
static size_t find_available_thread_slot(struct host_context_load_thread *hclt, size_t max_thread_slots, size_t *found_index)
{
size_t retries = MAX_FIND_THREAD_RETRIES;
while (retries--) {
size_t index = 0;
while (index < max_thread_slots) {
if (false == __atomic_load_n(&(hclt[index].busy), __ATOMIC_ACQUIRE)) {
*found_index = index;
return true;
}
index++;
}
sleep_usec(10 * USEC_PER_MS);
}
return false;
}
static void start_all_host_load_context(uv_work_t *req __maybe_unused)
{
register_libuv_worker_jobs();
struct scan_metadata_payload *data = req->data;
UNUSED(data);
worker_is_busy(UV_EVENT_HOST_CONTEXT_LOAD);
usec_t started_ut = now_monotonic_usec(); (void)started_ut;
RRDHOST *host;
size_t max_threads = MIN(get_netdata_cpus() / 2, 6);
struct host_context_load_thread *hclt = callocz(max_threads, sizeof(*hclt));
size_t thread_index;
dfe_start_reentrant(rrdhost_root_index, host) {
if (rrdhost_flag_check(host, RRDHOST_FLAG_CONTEXT_LOAD_IN_PROGRESS) ||
!rrdhost_flag_check(host, RRDHOST_FLAG_PENDING_CONTEXT_LOAD))
continue;
rrdhost_flag_set(host, RRDHOST_FLAG_CONTEXT_LOAD_IN_PROGRESS);
internal_error(true, "METADATA: 'host:%s' loading context", rrdhost_hostname(host));
cleanup_finished_threads(hclt, max_threads, false);
bool found_slot = find_available_thread_slot(hclt, max_threads, &thread_index);
if (unlikely(!found_slot)) {
struct host_context_load_thread hclt_sync = {.host = host};
restore_host_context(&hclt_sync);
}
else {
__atomic_store_n(&hclt[thread_index].busy, true, __ATOMIC_RELAXED);
hclt[thread_index].host = host;
assert(0 == uv_thread_create(&hclt[thread_index].thread, restore_host_context, &hclt[thread_index]));
}
}
dfe_done(host);
cleanup_finished_threads(hclt, max_threads, true);
freez(hclt);
usec_t ended_ut = now_monotonic_usec(); (void)ended_ut;
internal_error(true, "METADATA: 'host:ALL' contexts loaded in %0.2f ms", (double)(ended_ut - started_ut) / USEC_PER_MS);
worker_is_idle();
}
// Callback after scan of hosts is done
static void after_metadata_hosts(uv_work_t *req, int status __maybe_unused)
{
struct scan_metadata_payload *data = req->data;
struct metadata_wc *wc = data->wc;
metadata_flag_clear(wc, METADATA_FLAG_PROCESSING);
internal_error(true, "METADATA: scanning hosts complete");
if (unlikely(wc->scan_complete)) {
completion_mark_complete(wc->scan_complete);
internal_error(true, "METADATA: Sending completion done");
}
freez(data);
}
static bool metadata_scan_host(RRDHOST *host, uint32_t max_count, bool use_transaction, BUFFER *work_buffer, size_t *query_counter) {
RRDSET *st;
int rc;
bool more_to_do = false;
uint32_t scan_count = 1;
if (use_transaction)
(void)db_execute(db_meta, "BEGIN TRANSACTION");
rrdset_foreach_reentrant(st, host) {
if (scan_count == max_count) {
more_to_do = true;
break;
}
if(rrdset_flag_check(st, RRDSET_FLAG_METADATA_UPDATE)) {
(*query_counter)++;
rrdset_flag_clear(st, RRDSET_FLAG_METADATA_UPDATE);
scan_count++;
buffer_flush(work_buffer);
rc = check_and_update_chart_labels(st, work_buffer, query_counter);
if (unlikely(rc))
error_report("METADATA: 'host:%s': Failed to update labels for chart %s", rrdhost_hostname(host), rrdset_name(st));
else
(*query_counter)++;
rc = store_chart_metadata(st);
if (unlikely(rc))
error_report("METADATA: 'host:%s': Failed to store metadata for chart %s", rrdhost_hostname(host), rrdset_name(st));
}
RRDDIM *rd;
rrddim_foreach_read(rd, st) {
if(rrddim_flag_check(rd, RRDDIM_FLAG_METADATA_UPDATE)) {
(*query_counter)++;
rrddim_flag_clear(rd, RRDDIM_FLAG_METADATA_UPDATE);
if (rrddim_option_check(rd, RRDDIM_OPTION_HIDDEN))
rrddim_flag_set(rd, RRDDIM_FLAG_META_HIDDEN);
else
rrddim_flag_clear(rd, RRDDIM_FLAG_META_HIDDEN);
rc = store_dimension_metadata(rd);
if (unlikely(rc))
error_report("METADATA: 'host:%s': Failed to dimension metadata for chart %s. dimension %s",
rrdhost_hostname(host), rrdset_name(st),
rrddim_name(rd));
}
}
rrddim_foreach_done(rd);
}
rrdset_foreach_done(st);
if (use_transaction)
(void)db_execute(db_meta, "COMMIT TRANSACTION");
return more_to_do;
}
static void store_host_and_system_info(RRDHOST *host, size_t *query_counter)
{
if (unlikely(store_host_systeminfo(host))) {
error_report("METADATA: 'host:%s': Failed to store host updated system information in the database", rrdhost_hostname(host));
rrdhost_flag_set(host, RRDHOST_FLAG_METADATA_INFO | RRDHOST_FLAG_METADATA_UPDATE);
}
else {
if (likely(query_counter))
(*query_counter)++;
}
if (unlikely(store_host_metadata(host))) {
error_report("METADATA: 'host:%s': Failed to store host info in the database", rrdhost_hostname(host));
rrdhost_flag_set(host, RRDHOST_FLAG_METADATA_INFO | RRDHOST_FLAG_METADATA_UPDATE);
}
else {
if (likely(query_counter))
(*query_counter)++;
}
}
// Worker thread to scan hosts for pending metadata to store
static void start_metadata_hosts(uv_work_t *req __maybe_unused)
{
register_libuv_worker_jobs();
RRDHOST *host;
int transaction_started = 0;
struct scan_metadata_payload *data = req->data;
struct metadata_wc *wc = data->wc;
BUFFER *work_buffer = data->work_buffer;
usec_t all_started_ut = now_monotonic_usec(); (void)all_started_ut;
internal_error(true, "METADATA: checking all hosts...");
usec_t started_ut = now_monotonic_usec(); (void)started_ut;
bool run_again = false;
worker_is_busy(UV_EVENT_METADATA_STORE);
if (!data->max_count)
transaction_started = !db_execute(db_meta, "BEGIN TRANSACTION;");
dfe_start_reentrant(rrdhost_root_index, host) {
if (rrdhost_flag_check(host, RRDHOST_FLAG_ARCHIVED) || !rrdhost_flag_check(host, RRDHOST_FLAG_METADATA_UPDATE))
continue;
size_t query_counter = 0; (void)query_counter;
rrdhost_flag_clear(host,RRDHOST_FLAG_METADATA_UPDATE);
if (unlikely(rrdhost_flag_check(host, RRDHOST_FLAG_METADATA_LABELS))) {
rrdhost_flag_clear(host, RRDHOST_FLAG_METADATA_LABELS);
int rc = exec_statement_with_uuid(SQL_DELETE_HOST_LABELS, &host->host_uuid);
if (likely(!rc)) {
query_counter++;
buffer_flush(work_buffer);
struct query_build tmp = {.sql = work_buffer, .count = 0};
uuid_unparse_lower(host->host_uuid, tmp.uuid_str);
rrdlabels_walkthrough_read(host->rrdlabels, host_label_store_to_sql_callback, &tmp);
buffer_strcat(work_buffer, " ON CONFLICT (host_id, label_key) DO UPDATE SET source_type = excluded.source_type, label_value=excluded.label_value, date_created=UNIXEPOCH()");
rc = db_execute(db_meta, buffer_tostring(work_buffer));
if (unlikely(rc)) {
error_report("METADATA: 'host:%s': failed to update metadata host labels", rrdhost_hostname(host));
rrdhost_flag_set(host, RRDHOST_FLAG_METADATA_LABELS | RRDHOST_FLAG_METADATA_UPDATE);
}
else
query_counter++;
} else {
error_report("METADATA: 'host:%s': failed to delete old host labels", rrdhost_hostname(host));
rrdhost_flag_set(host, RRDHOST_FLAG_METADATA_LABELS | RRDHOST_FLAG_METADATA_UPDATE);
}
}
if (unlikely(rrdhost_flag_check(host, RRDHOST_FLAG_METADATA_CLAIMID))) {
rrdhost_flag_clear(host, RRDHOST_FLAG_METADATA_CLAIMID);
uuid_t uuid;
int rc;
if (likely(host->aclk_state.claimed_id && !uuid_parse(host->aclk_state.claimed_id, uuid)))
rc = store_claim_id(&host->host_uuid, &uuid);
else
rc = store_claim_id(&host->host_uuid, NULL);
if (unlikely(rc))
rrdhost_flag_set(host, RRDHOST_FLAG_METADATA_CLAIMID | RRDHOST_FLAG_METADATA_UPDATE);
else
query_counter++;
}
if (unlikely(rrdhost_flag_check(host, RRDHOST_FLAG_METADATA_INFO))) {
rrdhost_flag_clear(host, RRDHOST_FLAG_METADATA_INFO);
store_host_and_system_info(host, &query_counter);
}
// For clarity
bool use_transaction = data->max_count;
if (unlikely(metadata_scan_host(host, data->max_count, use_transaction, work_buffer, &query_counter))) {
run_again = true;
rrdhost_flag_set(host,RRDHOST_FLAG_METADATA_UPDATE);
internal_error(true,"METADATA: 'host:%s': scheduling another run, more charts to store", rrdhost_hostname(host));
}
usec_t ended_ut = now_monotonic_usec(); (void)ended_ut;
internal_error(true, "METADATA: 'host:%s': saved metadata with %zu SQL statements, in %0.2f ms",
rrdhost_hostname(host), query_counter,
(double)(ended_ut - started_ut) / USEC_PER_MS);
}
dfe_done(host);
if (!data->max_count && transaction_started)
transaction_started = db_execute(db_meta, "COMMIT TRANSACTION;");
usec_t all_ended_ut = now_monotonic_usec(); (void)all_ended_ut;
internal_error(true, "METADATA: checking all hosts completed in %0.2f ms",
(double)(all_ended_ut - all_started_ut) / USEC_PER_MS);
if (unlikely(run_again))
wc->metadata_check_after = now_realtime_sec() + METADATA_HOST_CHECK_IMMEDIATE;
else {
wc->metadata_check_after = now_realtime_sec() + METADATA_HOST_CHECK_INTERVAL;
run_metadata_cleanup(wc);
}
worker_is_idle();
}
// Callback after scan of hosts is done
static void after_start_ml_model_load(uv_work_t *req, int status __maybe_unused)
{
struct ml_model_payload *ml_data = req->data;
struct metadata_wc *wc = ml_data->wc;
metadata_flag_clear(wc, METADATA_FLAG_ML_LOADING);
JudyLFreeArray(&ml_data->JudyL, PJE0);
freez(ml_data);
}
static void start_ml_model_load(uv_work_t *req __maybe_unused)
{
register_libuv_worker_jobs();
struct ml_model_payload *ml_data = req->data;
worker_is_busy(UV_EVENT_METADATA_ML_LOAD);
Pvoid_t *PValue;
Word_t Index = 0;
bool first = true;
RRDDIM *rd;
RRDDIM_ACQUIRED *rda;
internal_error(true, "Batch ML load loader, %zu items", ml_data->count);
while((PValue = JudyLFirstThenNext(ml_data->JudyL, &Index, &first))) {
UNUSED(PValue);
rda = (RRDDIM_ACQUIRED *) Index;
rd = rrddim_acquired_to_rrddim(rda);
ml_dimension_load_models(rd);
rrddim_acquired_release(rda);
}
worker_is_idle();
}
static void metadata_event_loop(void *arg)
{
worker_register("METASYNC");
worker_register_job_name(METADATA_DATABASE_NOOP, "noop");
worker_register_job_name(METADATA_DATABASE_TIMER, "timer");
worker_register_job_name(METADATA_DEL_DIMENSION, "delete dimension");
worker_register_job_name(METADATA_STORE_CLAIM_ID, "add claim id");
worker_register_job_name(METADATA_ADD_HOST_INFO, "add host info");
worker_register_job_name(METADATA_MAINTENANCE, "maintenance");
worker_register_job_name(METADATA_ML_LOAD_MODELS, "ml load models");
int ret;
uv_loop_t *loop;
unsigned cmd_batch_size;
struct metadata_wc *wc = arg;
enum metadata_opcode opcode;
uv_thread_set_name_np(wc->thread, "METASYNC");
loop = wc->loop = mallocz(sizeof(uv_loop_t));
ret = uv_loop_init(loop);
if (ret) {
netdata_log_error("uv_loop_init(): %s", uv_strerror(ret));
goto error_after_loop_init;
}
loop->data = wc;
ret = uv_async_init(wc->loop, &wc->async, async_cb);
if (ret) {
netdata_log_error("uv_async_init(): %s", uv_strerror(ret));
goto error_after_async_init;
}
wc->async.data = wc;
ret = uv_timer_init(loop, &wc->timer_req);
if (ret) {
netdata_log_error("uv_timer_init(): %s", uv_strerror(ret));
goto error_after_timer_init;
}
wc->timer_req.data = wc;
fatal_assert(0 == uv_timer_start(&wc->timer_req, timer_cb, TIMER_INITIAL_PERIOD_MS, TIMER_REPEAT_PERIOD_MS));
netdata_log_info("Starting metadata sync thread with %d entries command queue", METADATA_CMD_Q_MAX_SIZE);
struct metadata_cmd cmd;
memset(&cmd, 0, sizeof(cmd));
metadata_flag_clear(wc, METADATA_FLAG_PROCESSING);
wc->metadata_check_after = now_realtime_sec() + METADATA_HOST_CHECK_FIRST_CHECK;
int shutdown = 0;
completion_mark_complete(&wc->init_complete);
BUFFER *work_buffer = buffer_create(1024, &netdata_buffers_statistics.buffers_sqlite);
struct scan_metadata_payload *data;
struct ml_model_payload *ml_data = NULL;
while (shutdown == 0 || (wc->flags & METADATA_FLAG_PROCESSING)) {
uuid_t *uuid;
RRDHOST *host = NULL;
worker_is_idle();
uv_run(loop, UV_RUN_DEFAULT);
/* wait for commands */
cmd_batch_size = 0;
do {
if (unlikely(cmd_batch_size >= METADATA_MAX_BATCH_SIZE))
break;
cmd = metadata_deq_cmd(wc);
opcode = cmd.opcode;
if (unlikely(opcode == METADATA_DATABASE_NOOP && metadata_flag_check(wc, METADATA_FLAG_SHUTDOWN))) {
shutdown = 1;
continue;
}
++cmd_batch_size;
if (likely(opcode != METADATA_DATABASE_NOOP))
worker_is_busy(opcode);
// Have pending ML models to load?
if (opcode != METADATA_ML_LOAD_MODELS && ml_data && ml_data->count) {
static usec_t ml_submit_last = 0;
usec_t now = now_monotonic_usec();
if (!ml_submit_last)
ml_submit_last = now;
if (!metadata_flag_check(wc, METADATA_FLAG_ML_LOADING) && (now - ml_submit_last > 150 * USEC_PER_MS)) {
metadata_flag_set(wc, METADATA_FLAG_ML_LOADING);
if (unlikely(uv_queue_work(loop, &ml_data->request, start_ml_model_load, after_start_ml_model_load)))
metadata_flag_clear(wc, METADATA_FLAG_ML_LOADING);
else {
ml_submit_last = now;
ml_data = NULL;
}
}
}
switch (opcode) {
case METADATA_DATABASE_NOOP:
case METADATA_DATABASE_TIMER:
break;
case METADATA_ML_LOAD_MODELS: {
RRDDIM *rd = (RRDDIM *) cmd.param[0];
RRDDIM_ACQUIRED *rda = rrddim_find_and_acquire(rd->rrdset, rrddim_id(rd));
if (likely(rda)) {
if (!ml_data) {
ml_data = callocz(1,sizeof(*ml_data));
ml_data->request.data = ml_data;
ml_data->wc = wc;
}
JudyLIns(&ml_data->JudyL, (Word_t)rda, PJE0);
ml_data->count++;
}
break;
}
case METADATA_DEL_DIMENSION:
uuid = (uuid_t *) cmd.param[0];
if (likely(dimension_can_be_deleted(uuid, NULL, false)))
delete_dimension_uuid(uuid, NULL, false);
freez(uuid);
break;
case METADATA_STORE_CLAIM_ID:
store_claim_id((uuid_t *) cmd.param[0], (uuid_t *) cmd.param[1]);
freez((void *) cmd.param[0]);
freez((void *) cmd.param[1]);
break;
case METADATA_ADD_HOST_INFO:
host = (RRDHOST *) cmd.param[0];
store_host_and_system_info(host, NULL);
break;
case METADATA_SCAN_HOSTS:
if (unlikely(metadata_flag_check(wc, METADATA_FLAG_PROCESSING)))
break;
if (unittest_running)
break;
data = mallocz(sizeof(*data));
data->request.data = data;
data->wc = wc;
data->work_buffer = work_buffer;
if (unlikely(cmd.completion)) {
data->max_count = 0; // 0 will process all pending updates
cmd.completion = NULL; // Do not complete after launching worker (worker will do)
}
else
data->max_count = 5000;
metadata_flag_set(wc, METADATA_FLAG_PROCESSING);
if (unlikely(
uv_queue_work(loop,&data->request,
start_metadata_hosts,
after_metadata_hosts))) {
// Failed to launch worker -- let the event loop handle completion
cmd.completion = wc->scan_complete;
freez(data);
metadata_flag_clear(wc, METADATA_FLAG_PROCESSING);
}
break;
case METADATA_LOAD_HOST_CONTEXT:;
if (unittest_running)
break;
data = callocz(1,sizeof(*data));
data->request.data = data;
data->wc = wc;
if (unlikely(
uv_queue_work(loop,&data->request, start_all_host_load_context,
after_start_host_load_context))) {
freez(data);
}
break;
case METADATA_UNITTEST:;
struct thread_unittest *tu = (struct thread_unittest *) cmd.param[0];
sleep_usec(1000); // processing takes 1ms
__atomic_fetch_add(&tu->processed, 1, __ATOMIC_SEQ_CST);
break;
default:
break;
}
if (cmd.completion)
completion_mark_complete(cmd.completion);
} while (opcode != METADATA_DATABASE_NOOP);
}
if (!uv_timer_stop(&wc->timer_req))
uv_close((uv_handle_t *)&wc->timer_req, NULL);
uv_close((uv_handle_t *)&wc->async, NULL);
int rc;
do {
rc = uv_loop_close(loop);
} while (rc != UV_EBUSY);
buffer_free(work_buffer);
freez(loop);
worker_unregister();
netdata_log_info("METADATA: Shutting down event loop");
completion_mark_complete(&wc->init_complete);
completion_destroy(wc->scan_complete);
freez(wc->scan_complete);
metadata_free_cmd_queue(wc);
return;
error_after_timer_init:
uv_close((uv_handle_t *)&wc->async, NULL);
error_after_async_init:
fatal_assert(0 == uv_loop_close(loop));
error_after_loop_init:
freez(loop);
worker_unregister();
}
struct metadata_wc metasync_worker = {.loop = NULL};
void metadata_sync_shutdown(void)
{
completion_init(&metasync_worker.init_complete);
struct metadata_cmd cmd;
memset(&cmd, 0, sizeof(cmd));
netdata_log_info("METADATA: Sending a shutdown command");
cmd.opcode = METADATA_SYNC_SHUTDOWN;
metadata_enq_cmd(&metasync_worker, &cmd);
/* wait for metadata thread to shut down */
netdata_log_info("METADATA: Waiting for shutdown ACK");
completion_wait_for(&metasync_worker.init_complete);
completion_destroy(&metasync_worker.init_complete);
netdata_log_info("METADATA: Shutdown complete");
}
void metadata_sync_shutdown_prepare(void)
{
if (unlikely(!metasync_worker.loop))
return;
struct metadata_cmd cmd;
memset(&cmd, 0, sizeof(cmd));
struct metadata_wc *wc = &metasync_worker;
struct completion *compl = mallocz(sizeof(*compl));
completion_init(compl);
__atomic_store_n(&wc->scan_complete, compl, __ATOMIC_RELAXED);
netdata_log_info("METADATA: Sending a scan host command");
uint32_t max_wait_iterations = 2000;
while (unlikely(metadata_flag_check(&metasync_worker, METADATA_FLAG_PROCESSING)) && max_wait_iterations--) {
if (max_wait_iterations == 1999)
netdata_log_info("METADATA: Current worker is running; waiting to finish");
sleep_usec(1000);
}
cmd.opcode = METADATA_SCAN_HOSTS;
metadata_enq_cmd(&metasync_worker, &cmd);
netdata_log_info("METADATA: Waiting for host scan completion");
completion_wait_for(wc->scan_complete);
netdata_log_info("METADATA: Host scan complete; can continue with shutdown");
}
// -------------------------------------------------------------
// Init function called on agent startup
void metadata_sync_init(void)
{
struct metadata_wc *wc = &metasync_worker;
memset(wc, 0, sizeof(*wc));
metadata_init_cmd_queue(wc);
completion_init(&wc->init_complete);
fatal_assert(0 == uv_thread_create(&(wc->thread), metadata_event_loop, wc));
completion_wait_for(&wc->init_complete);
completion_destroy(&wc->init_complete);
netdata_log_info("SQLite metadata sync initialization complete");
}
// Helpers
static inline void queue_metadata_cmd(enum metadata_opcode opcode, const void *param0, const void *param1)
{
struct metadata_cmd cmd;
cmd.opcode = opcode;
cmd.param[0] = param0;
cmd.param[1] = param1;
cmd.completion = NULL;
metadata_enq_cmd(&metasync_worker, &cmd);
}
// Public
void metaqueue_delete_dimension_uuid(uuid_t *uuid)
{
if (unlikely(!metasync_worker.loop))
return;
uuid_t *use_uuid = mallocz(sizeof(*uuid));
uuid_copy(*use_uuid, *uuid);
queue_metadata_cmd(METADATA_DEL_DIMENSION, use_uuid, NULL);
}
void metaqueue_store_claim_id(uuid_t *host_uuid, uuid_t *claim_uuid)
{
if (unlikely(!host_uuid))
return;
uuid_t *local_host_uuid = mallocz(sizeof(*host_uuid));
uuid_t *local_claim_uuid = NULL;
uuid_copy(*local_host_uuid, *host_uuid);
if (likely(claim_uuid)) {
local_claim_uuid = mallocz(sizeof(*claim_uuid));
uuid_copy(*local_claim_uuid, *claim_uuid);
}
queue_metadata_cmd(METADATA_STORE_CLAIM_ID, local_host_uuid, local_claim_uuid);
}
void metaqueue_host_update_info(RRDHOST *host)
{
if (unlikely(!metasync_worker.loop))
return;
queue_metadata_cmd(METADATA_ADD_HOST_INFO, host, NULL);
}
void metaqueue_ml_load_models(RRDDIM *rd)
{
if (unlikely(!metasync_worker.loop))
return;
queue_metadata_cmd(METADATA_ML_LOAD_MODELS, rd, NULL);
}
void metadata_queue_load_host_context(RRDHOST *host)
{
if (unlikely(!metasync_worker.loop))
return;
queue_metadata_cmd(METADATA_LOAD_HOST_CONTEXT, host, NULL);
}
//
// unitests
//
static void *unittest_queue_metadata(void *arg) {
struct thread_unittest *tu = arg;
struct metadata_cmd cmd;
cmd.opcode = METADATA_UNITTEST;
cmd.param[0] = tu;
cmd.param[1] = NULL;
cmd.completion = NULL;
metadata_enq_cmd(&metasync_worker, &cmd);
do {
__atomic_fetch_add(&tu->added, 1, __ATOMIC_SEQ_CST);
metadata_enq_cmd(&metasync_worker, &cmd);
sleep_usec(10000);
} while (!__atomic_load_n(&tu->join, __ATOMIC_RELAXED));
return arg;
}
static void *metadata_unittest_threads(void)
{
unsigned done;
struct thread_unittest tu = {
.join = 0,
.added = 0,
.processed = 0,
.done = &done,
};
// Queue messages / Time it
time_t seconds_to_run = 5;
int threads_to_create = 4;
fprintf(
stderr,
"\nChecking metadata queue using %d threads for %lld seconds...\n",
threads_to_create,
(long long)seconds_to_run);
netdata_thread_t threads[threads_to_create];
tu.join = 0;
for (int i = 0; i < threads_to_create; i++) {
char buf[100 + 1];
snprintf(buf, 100, "META[%d]", i);
netdata_thread_create(
&threads[i],
buf,
NETDATA_THREAD_OPTION_DONT_LOG | NETDATA_THREAD_OPTION_JOINABLE,
unittest_queue_metadata,
&tu);
}
(void) uv_async_send(&metasync_worker.async);
sleep_usec(seconds_to_run * USEC_PER_SEC);
__atomic_store_n(&tu.join, 1, __ATOMIC_RELAXED);
for (int i = 0; i < threads_to_create; i++) {
void *retval;
netdata_thread_join(threads[i], &retval);
}
sleep_usec(5 * USEC_PER_SEC);
fprintf(stderr, "Added %u elements, processed %u\n", tu.added, tu.processed);
return 0;
}
int metadata_unittest(void)
{
metadata_sync_init();
// Queue items for a specific period of time
metadata_unittest_threads();
metadata_sync_shutdown();
return 0;
}
|