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
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
|
/*
* Copyright (C) 2004, 2005, 2007 Internet Systems Consortium, Inc. ("ISC")
* Copyright (C) 1999-2002 Internet Software Consortium.
*
* Permission to use, copy, modify, and/or distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND ISC DISCLAIMS ALL WARRANTIES WITH
* REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY
* AND FITNESS. IN NO EVENT SHALL ISC BE LIABLE FOR ANY SPECIAL, DIRECT,
* INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
* LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE
* OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
* PERFORMANCE OF THIS SOFTWARE.
*/
/* $Id: journal.c,v 1.86.18.12 2007/09/07 05:21:41 marka Exp $ */
#include <config.h>
#include <stdlib.h>
#include <unistd.h>
#include <errno.h>
#include <isc/file.h>
#include <isc/mem.h>
#include <isc/stdio.h>
#include <isc/string.h>
#include <isc/util.h>
#include <dns/compress.h>
#include <dns/db.h>
#include <dns/dbiterator.h>
#include <dns/diff.h>
#include <dns/fixedname.h>
#include <dns/journal.h>
#include <dns/log.h>
#include <dns/rdataset.h>
#include <dns/rdatasetiter.h>
#include <dns/result.h>
#include <dns/soa.h>
/*! \file
* \brief Journalling.
*
* A journal file consists of
*
* \li A fixed-size header of type journal_rawheader_t.
*
* \li The index. This is an unordered array of index entries
* of type journal_rawpos_t giving the locations
* of some arbitrary subset of the journal's addressable
* transactions. The index entries are used as hints to
* speed up the process of locating a transaction with a given
* serial number. Unused index entries have an "offset"
* field of zero. The size of the index can vary between
* journal files, but does not change during the lifetime
* of a file. The size can be zero.
*
* \li The journal data. This consists of one or more transactions.
* Each transaction begins with a transaction header of type
* journal_rawxhdr_t. The transaction header is followed by a
* sequence of RRs, similar in structure to an IXFR difference
* sequence (RFC1995). That is, the pre-transaction SOA,
* zero or more other deleted RRs, the post-transaction SOA,
* and zero or more other added RRs. Unlike in IXFR, each RR
* is prefixed with a 32-bit length.
*
* The journal data part grows as new transactions are
* appended to the file. Only those transactions
* whose serial number is current-(2^31-1) to current
* are considered "addressable" and may be pointed
* to from the header or index. They may be preceded
* by old transactions that are no longer addressable,
* and they may be followed by transactions that were
* appended to the journal but never committed by updating
* the "end" position in the header. The latter will
* be overwritten when new transactions are added.
*/
/*%
* When true, accept IXFR difference sequences where the
* SOA serial number does not change (BIND 8 sends such
* sequences).
*/
static isc_boolean_t bind8_compat = ISC_TRUE; /* XXX config */
/**************************************************************************/
/*
* Miscellaneous utilities.
*/
#define JOURNAL_COMMON_LOGARGS \
dns_lctx, DNS_LOGCATEGORY_GENERAL, DNS_LOGMODULE_JOURNAL
#define JOURNAL_DEBUG_LOGARGS(n) \
JOURNAL_COMMON_LOGARGS, ISC_LOG_DEBUG(n)
/*%
* It would be non-sensical (or at least obtuse) to use FAIL() with an
* ISC_R_SUCCESS code, but the test is there to keep the Solaris compiler
* from complaining about "end-of-loop code not reached".
*/
#define FAIL(code) \
do { result = (code); \
if (result != ISC_R_SUCCESS) goto failure; \
} while (0)
#define CHECK(op) \
do { result = (op); \
if (result != ISC_R_SUCCESS) goto failure; \
} while (0)
static isc_result_t index_to_disk(dns_journal_t *);
static inline isc_uint32_t
decode_uint32(unsigned char *p) {
return ((p[0] << 24) +
(p[1] << 16) +
(p[2] << 8) +
(p[3] << 0));
}
static inline void
encode_uint32(isc_uint32_t val, unsigned char *p) {
p[0] = (isc_uint8_t)(val >> 24);
p[1] = (isc_uint8_t)(val >> 16);
p[2] = (isc_uint8_t)(val >> 8);
p[3] = (isc_uint8_t)(val >> 0);
}
isc_result_t
dns_db_createsoatuple(dns_db_t *db, dns_dbversion_t *ver, isc_mem_t *mctx,
dns_diffop_t op, dns_difftuple_t **tp)
{
isc_result_t result;
dns_dbnode_t *node;
dns_rdataset_t rdataset;
dns_rdata_t rdata = DNS_RDATA_INIT;
dns_name_t *zonename;
zonename = dns_db_origin(db);
node = NULL;
result = dns_db_findnode(db, zonename, ISC_FALSE, &node);
if (result != ISC_R_SUCCESS)
goto nonode;
dns_rdataset_init(&rdataset);
result = dns_db_findrdataset(db, node, ver, dns_rdatatype_soa, 0,
(isc_stdtime_t)0, &rdataset, NULL);
if (result != ISC_R_SUCCESS)
goto freenode;
result = dns_rdataset_first(&rdataset);
if (result != ISC_R_SUCCESS)
goto freenode;
dns_rdataset_current(&rdataset, &rdata);
result = dns_difftuple_create(mctx, op, zonename, rdataset.ttl,
&rdata, tp);
dns_rdataset_disassociate(&rdataset);
dns_db_detachnode(db, &node);
return (ISC_R_SUCCESS);
freenode:
dns_db_detachnode(db, &node);
nonode:
UNEXPECTED_ERROR(__FILE__, __LINE__, "missing SOA");
return (result);
}
/* Journalling */
/*%
* On-disk representation of a "pointer" to a journal entry.
* These are used in the journal header to locate the beginning
* and end of the journal, and in the journal index to locate
* other transactions.
*/
typedef struct {
unsigned char serial[4]; /*%< SOA serial before update. */
/*
* XXXRTH Should offset be 8 bytes?
* XXXDCL ... probably, since isc_offset_t is 8 bytes on many OSs.
* XXXAG ... but we will not be able to seek >2G anyway on many
* platforms as long as we are using fseek() rather
* than lseek().
*/
unsigned char offset[4]; /*%< Offset from beginning of file. */
} journal_rawpos_t;
/*%
* The header is of a fixed size, with some spare room for future
* extensions.
*/
#define JOURNAL_HEADER_SIZE 64 /* Bytes. */
/*%
* The on-disk representation of the journal header.
* All numbers are stored in big-endian order.
*/
typedef union {
struct {
/*% File format version ID. */
unsigned char format[16];
/*% Position of the first addressable transaction */
journal_rawpos_t begin;
/*% Position of the next (yet nonexistent) transaction. */
journal_rawpos_t end;
/*% Number of index entries following the header. */
unsigned char index_size[4];
} h;
/* Pad the header to a fixed size. */
unsigned char pad[JOURNAL_HEADER_SIZE];
} journal_rawheader_t;
/*%
* The on-disk representation of the transaction header.
* There is one of these at the beginning of each transaction.
*/
typedef struct {
unsigned char size[4]; /*%< In bytes, excluding header. */
unsigned char serial0[4]; /*%< SOA serial before update. */
unsigned char serial1[4]; /*%< SOA serial after update. */
} journal_rawxhdr_t;
/*%
* The on-disk representation of the RR header.
* There is one of these at the beginning of each RR.
*/
typedef struct {
unsigned char size[4]; /*%< In bytes, excluding header. */
} journal_rawrrhdr_t;
/*%
* The in-core representation of the journal header.
*/
typedef struct {
isc_uint32_t serial;
isc_offset_t offset;
} journal_pos_t;
#define POS_VALID(pos) ((pos).offset != 0)
#define POS_INVALIDATE(pos) ((pos).offset = 0, (pos).serial = 0)
typedef struct {
unsigned char format[16];
journal_pos_t begin;
journal_pos_t end;
isc_uint32_t index_size;
} journal_header_t;
/*%
* The in-core representation of the transaction header.
*/
typedef struct {
isc_uint32_t size;
isc_uint32_t serial0;
isc_uint32_t serial1;
} journal_xhdr_t;
/*%
* The in-core representation of the RR header.
*/
typedef struct {
isc_uint32_t size;
} journal_rrhdr_t;
/*%
* Initial contents to store in the header of a newly created
* journal file.
*
* The header starts with the magic string ";BIND LOG V9\n"
* to identify the file as a BIND 9 journal file. An ASCII
* identification string is used rather than a binary magic
* number to be consistent with BIND 8 (BIND 8 journal files
* are ASCII text files).
*/
static journal_header_t
initial_journal_header = { ";BIND LOG V9\n", { 0, 0 }, { 0, 0 }, 0 };
#define JOURNAL_EMPTY(h) ((h)->begin.offset == (h)->end.offset)
typedef enum {
JOURNAL_STATE_INVALID,
JOURNAL_STATE_READ,
JOURNAL_STATE_WRITE,
JOURNAL_STATE_TRANSACTION
} journal_state_t;
struct dns_journal {
unsigned int magic; /*%< JOUR */
isc_mem_t *mctx; /*%< Memory context */
journal_state_t state;
const char *filename; /*%< Journal file name */
FILE * fp; /*%< File handle */
isc_offset_t offset; /*%< Current file offset */
journal_header_t header; /*%< In-core journal header */
unsigned char *rawindex; /*%< In-core buffer for journal index in on-disk format */
journal_pos_t *index; /*%< In-core journal index */
/*% Current transaction state (when writing). */
struct {
unsigned int n_soa; /*%< Number of SOAs seen */
journal_pos_t pos[2]; /*%< Begin/end position */
} x;
/*% Iteration state (when reading). */
struct {
/* These define the part of the journal we iterate over. */
journal_pos_t bpos; /*%< Position before first, */
journal_pos_t epos; /*%< and after last transaction */
/* The rest is iterator state. */
isc_uint32_t current_serial; /*%< Current SOA serial */
isc_buffer_t source; /*%< Data from disk */
isc_buffer_t target; /*%< Data from _fromwire check */
dns_decompress_t dctx; /*%< Dummy decompression ctx */
dns_name_t name; /*%< Current domain name */
dns_rdata_t rdata; /*%< Current rdata */
isc_uint32_t ttl; /*%< Current TTL */
unsigned int xsize; /*%< Size of transaction data */
unsigned int xpos; /*%< Current position in it */
isc_result_t result; /*%< Result of last call */
} it;
};
#define DNS_JOURNAL_MAGIC ISC_MAGIC('J', 'O', 'U', 'R')
#define DNS_JOURNAL_VALID(t) ISC_MAGIC_VALID(t, DNS_JOURNAL_MAGIC)
static void
journal_pos_decode(journal_rawpos_t *raw, journal_pos_t *cooked) {
cooked->serial = decode_uint32(raw->serial);
cooked->offset = decode_uint32(raw->offset);
}
static void
journal_pos_encode(journal_rawpos_t *raw, journal_pos_t *cooked) {
encode_uint32(cooked->serial, raw->serial);
encode_uint32(cooked->offset, raw->offset);
}
static void
journal_header_decode(journal_rawheader_t *raw, journal_header_t *cooked) {
INSIST(sizeof(cooked->format) == sizeof(raw->h.format));
memcpy(cooked->format, raw->h.format, sizeof(cooked->format));
journal_pos_decode(&raw->h.begin, &cooked->begin);
journal_pos_decode(&raw->h.end, &cooked->end);
cooked->index_size = decode_uint32(raw->h.index_size);
}
static void
journal_header_encode(journal_header_t *cooked, journal_rawheader_t *raw) {
INSIST(sizeof(cooked->format) == sizeof(raw->h.format));
memset(raw->pad, 0, sizeof(raw->pad));
memcpy(raw->h.format, cooked->format, sizeof(raw->h.format));
journal_pos_encode(&raw->h.begin, &cooked->begin);
journal_pos_encode(&raw->h.end, &cooked->end);
encode_uint32(cooked->index_size, raw->h.index_size);
}
/*
* Journal file I/O subroutines, with error checking and reporting.
*/
static isc_result_t
journal_seek(dns_journal_t *j, isc_uint32_t offset) {
isc_result_t result;
result = isc_stdio_seek(j->fp, (long)offset, SEEK_SET);
if (result != ISC_R_SUCCESS) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: seek: %s", j->filename,
isc_result_totext(result));
return (ISC_R_UNEXPECTED);
}
j->offset = offset;
return (ISC_R_SUCCESS);
}
static isc_result_t
journal_read(dns_journal_t *j, void *mem, size_t nbytes) {
isc_result_t result;
result = isc_stdio_read(mem, 1, nbytes, j->fp, NULL);
if (result != ISC_R_SUCCESS) {
if (result == ISC_R_EOF)
return (ISC_R_NOMORE);
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: read: %s",
j->filename, isc_result_totext(result));
return (ISC_R_UNEXPECTED);
}
j->offset += nbytes;
return (ISC_R_SUCCESS);
}
static isc_result_t
journal_write(dns_journal_t *j, void *mem, size_t nbytes) {
isc_result_t result;
result = isc_stdio_write(mem, 1, nbytes, j->fp, NULL);
if (result != ISC_R_SUCCESS) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: write: %s",
j->filename, isc_result_totext(result));
return (ISC_R_UNEXPECTED);
}
j->offset += nbytes;
return (ISC_R_SUCCESS);
}
static isc_result_t
journal_fsync(dns_journal_t *j) {
isc_result_t result;
result = isc_stdio_flush(j->fp);
if (result != ISC_R_SUCCESS) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: flush: %s",
j->filename, isc_result_totext(result));
return (ISC_R_UNEXPECTED);
}
result = isc_stdio_sync(j->fp);
if (result != ISC_R_SUCCESS) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: fsync: %s",
j->filename, isc_result_totext(result));
return (ISC_R_UNEXPECTED);
}
return (ISC_R_SUCCESS);
}
/*
* Read/write a transaction header at the current file position.
*/
static isc_result_t
journal_read_xhdr(dns_journal_t *j, journal_xhdr_t *xhdr) {
journal_rawxhdr_t raw;
isc_result_t result;
result = journal_read(j, &raw, sizeof(raw));
if (result != ISC_R_SUCCESS)
return (result);
xhdr->size = decode_uint32(raw.size);
xhdr->serial0 = decode_uint32(raw.serial0);
xhdr->serial1 = decode_uint32(raw.serial1);
return (ISC_R_SUCCESS);
}
static isc_result_t
journal_write_xhdr(dns_journal_t *j, isc_uint32_t size,
isc_uint32_t serial0, isc_uint32_t serial1)
{
journal_rawxhdr_t raw;
encode_uint32(size, raw.size);
encode_uint32(serial0, raw.serial0);
encode_uint32(serial1, raw.serial1);
return (journal_write(j, &raw, sizeof(raw)));
}
/*
* Read an RR header at the current file position.
*/
static isc_result_t
journal_read_rrhdr(dns_journal_t *j, journal_rrhdr_t *rrhdr) {
journal_rawrrhdr_t raw;
isc_result_t result;
result = journal_read(j, &raw, sizeof(raw));
if (result != ISC_R_SUCCESS)
return (result);
rrhdr->size = decode_uint32(raw.size);
return (ISC_R_SUCCESS);
}
static isc_result_t
journal_file_create(isc_mem_t *mctx, const char *filename) {
FILE *fp = NULL;
isc_result_t result;
journal_header_t header;
journal_rawheader_t rawheader;
int index_size = 56; /* XXX configurable */
int size;
void *mem; /* Memory for temporary index image. */
INSIST(sizeof(journal_rawheader_t) == JOURNAL_HEADER_SIZE);
result = isc_stdio_open(filename, "wb", &fp);
if (result != ISC_R_SUCCESS) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: create: %s",
filename, isc_result_totext(result));
return (ISC_R_UNEXPECTED);
}
header = initial_journal_header;
header.index_size = index_size;
journal_header_encode(&header, &rawheader);
size = sizeof(journal_rawheader_t) +
index_size * sizeof(journal_rawpos_t);
mem = isc_mem_get(mctx, size);
if (mem == NULL) {
(void)isc_stdio_close(fp);
(void)isc_file_remove(filename);
return (ISC_R_NOMEMORY);
}
memset(mem, 0, size);
memcpy(mem, &rawheader, sizeof(rawheader));
result = isc_stdio_write(mem, 1, (size_t) size, fp, NULL);
if (result != ISC_R_SUCCESS) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: write: %s",
filename, isc_result_totext(result));
(void)isc_stdio_close(fp);
(void)isc_file_remove(filename);
isc_mem_put(mctx, mem, size);
return (ISC_R_UNEXPECTED);
}
isc_mem_put(mctx, mem, size);
result = isc_stdio_close(fp);
if (result != ISC_R_SUCCESS) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: close: %s",
filename, isc_result_totext(result));
(void)isc_file_remove(filename);
return (ISC_R_UNEXPECTED);
}
return (ISC_R_SUCCESS);
}
static isc_result_t
journal_open(isc_mem_t *mctx, const char *filename, isc_boolean_t write,
isc_boolean_t create, dns_journal_t **journalp) {
FILE *fp = NULL;
isc_result_t result;
journal_rawheader_t rawheader;
dns_journal_t *j;
INSIST(journalp != NULL && *journalp == NULL);
j = isc_mem_get(mctx, sizeof(*j));
if (j == NULL)
return (ISC_R_NOMEMORY);
j->mctx = mctx;
j->state = JOURNAL_STATE_INVALID;
j->fp = NULL;
j->filename = filename;
j->index = NULL;
j->rawindex = NULL;
result = isc_stdio_open(j->filename, write ? "rb+" : "rb", &fp);
if (result == ISC_R_FILENOTFOUND) {
if (create) {
isc_log_write(JOURNAL_COMMON_LOGARGS,
ISC_LOG_INFO,
"journal file %s does not exist, "
"creating it",
j->filename);
CHECK(journal_file_create(mctx, filename));
/*
* Retry.
*/
result = isc_stdio_open(j->filename, "rb+", &fp);
} else {
FAIL(ISC_R_NOTFOUND);
}
}
if (result != ISC_R_SUCCESS) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: open: %s",
j->filename, isc_result_totext(result));
FAIL(ISC_R_UNEXPECTED);
}
j->fp = fp;
/*
* Set magic early so that seek/read can succeed.
*/
j->magic = DNS_JOURNAL_MAGIC;
CHECK(journal_seek(j, 0));
CHECK(journal_read(j, &rawheader, sizeof(rawheader)));
if (memcmp(rawheader.h.format, initial_journal_header.format,
sizeof(initial_journal_header.format)) != 0) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: journal format not recognized",
j->filename);
FAIL(ISC_R_UNEXPECTED);
}
journal_header_decode(&rawheader, &j->header);
/*
* If there is an index, read the raw index into a dynamically
* allocated buffer and then convert it into a cooked index.
*/
if (j->header.index_size != 0) {
unsigned int i;
unsigned int rawbytes;
unsigned char *p;
rawbytes = j->header.index_size * sizeof(journal_rawpos_t);
j->rawindex = isc_mem_get(mctx, rawbytes);
if (j->rawindex == NULL)
FAIL(ISC_R_NOMEMORY);
CHECK(journal_read(j, j->rawindex, rawbytes));
j->index = isc_mem_get(mctx, j->header.index_size *
sizeof(journal_pos_t));
if (j->index == NULL)
FAIL(ISC_R_NOMEMORY);
p = j->rawindex;
for (i = 0; i < j->header.index_size; i++) {
j->index[i].serial = decode_uint32(p);
p += 4;
j->index[i].offset = decode_uint32(p);
p += 4;
}
INSIST(p == j->rawindex + rawbytes);
}
j->offset = -1; /* Invalid, must seek explicitly. */
/*
* Initialize the iterator.
*/
dns_name_init(&j->it.name, NULL);
dns_rdata_init(&j->it.rdata);
/*
* Set up empty initial buffers for uncheched and checked
* wire format RR data. They will be reallocated
* later.
*/
isc_buffer_init(&j->it.source, NULL, 0);
isc_buffer_init(&j->it.target, NULL, 0);
dns_decompress_init(&j->it.dctx, -1, DNS_DECOMPRESS_NONE);
j->state =
write ? JOURNAL_STATE_WRITE : JOURNAL_STATE_READ;
*journalp = j;
return (ISC_R_SUCCESS);
failure:
j->magic = 0;
if (j->index != NULL) {
isc_mem_put(j->mctx, j->index, j->header.index_size *
sizeof(journal_rawpos_t));
j->index = NULL;
}
if (j->fp != NULL)
(void)isc_stdio_close(j->fp);
isc_mem_put(j->mctx, j, sizeof(*j));
return (result);
}
isc_result_t
dns_journal_open(isc_mem_t *mctx, const char *filename, isc_boolean_t write,
dns_journal_t **journalp) {
isc_result_t result;
int namelen;
char backup[1024];
result = journal_open(mctx, filename, write, write, journalp);
if (result == ISC_R_NOTFOUND) {
namelen = strlen(filename);
if (namelen > 4 && strcmp(filename + namelen - 4, ".jnl") == 0)
namelen -= 4;
result = isc_string_printf(backup, sizeof(backup), "%.*s.jbk",
namelen, filename);
if (result != ISC_R_SUCCESS)
return (result);
result = journal_open(mctx, backup, write, write, journalp);
}
return (result);
}
/*
* A comparison function defining the sorting order for
* entries in the IXFR-style journal file.
*
* The IXFR format requires that deletions are sorted before
* additions, and within either one, SOA records are sorted
* before others.
*
* Also sort the non-SOA records by type as a courtesy to the
* server receiving the IXFR - it may help reduce the amount of
* rdataset merging it has to do.
*/
static int
ixfr_order(const void *av, const void *bv) {
dns_difftuple_t const * const *ap = av;
dns_difftuple_t const * const *bp = bv;
dns_difftuple_t const *a = *ap;
dns_difftuple_t const *b = *bp;
int r;
r = (b->op == DNS_DIFFOP_DEL) - (a->op == DNS_DIFFOP_DEL);
if (r != 0)
return (r);
r = (b->rdata.type == dns_rdatatype_soa) -
(a->rdata.type == dns_rdatatype_soa);
if (r != 0)
return (r);
r = (a->rdata.type - b->rdata.type);
return (r);
}
/*
* Advance '*pos' to the next journal transaction.
*
* Requires:
* *pos refers to a valid journal transaction.
*
* Ensures:
* When ISC_R_SUCCESS is returned,
* *pos refers to the next journal transaction.
*
* Returns one of:
*
* ISC_R_SUCCESS
* ISC_R_NOMORE *pos pointed at the last transaction
* Other results due to file errors are possible.
*/
static isc_result_t
journal_next(dns_journal_t *j, journal_pos_t *pos) {
isc_result_t result;
journal_xhdr_t xhdr;
REQUIRE(DNS_JOURNAL_VALID(j));
result = journal_seek(j, pos->offset);
if (result != ISC_R_SUCCESS)
return (result);
if (pos->serial == j->header.end.serial)
return (ISC_R_NOMORE);
/*
* Read the header of the current transaction.
* This will return ISC_R_NOMORE if we are at EOF.
*/
result = journal_read_xhdr(j, &xhdr);
if (result != ISC_R_SUCCESS)
return (result);
/*
* Check serial number consistency.
*/
if (xhdr.serial0 != pos->serial) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: journal file corrupt: "
"expected serial %u, got %u",
j->filename, pos->serial, xhdr.serial0);
return (ISC_R_UNEXPECTED);
}
/*
* Check for offset wraparound.
*/
if ((isc_offset_t)(pos->offset + sizeof(journal_rawxhdr_t) + xhdr.size)
< pos->offset) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: offset too large", j->filename);
return (ISC_R_UNEXPECTED);
}
pos->offset += sizeof(journal_rawxhdr_t) + xhdr.size;
pos->serial = xhdr.serial1;
return (ISC_R_SUCCESS);
}
/*
* If the index of the journal 'j' contains an entry "better"
* than '*best_guess', replace '*best_guess' with it.
*
* "Better" means having a serial number closer to 'serial'
* but not greater than 'serial'.
*/
static void
index_find(dns_journal_t *j, isc_uint32_t serial, journal_pos_t *best_guess) {
unsigned int i;
if (j->index == NULL)
return;
for (i = 0; i < j->header.index_size; i++) {
if (POS_VALID(j->index[i]) &&
DNS_SERIAL_GE(serial, j->index[i].serial) &&
DNS_SERIAL_GT(j->index[i].serial, best_guess->serial))
*best_guess = j->index[i];
}
}
/*
* Add a new index entry. If there is no room, make room by removing
* the odd-numbered entries and compacting the others into the first
* half of the index. This decimates old index entries exponentially
* over time, so that the index always contains a much larger fraction
* of recent serial numbers than of old ones. This is deliberate -
* most index searches are for outgoing IXFR, and IXFR tends to request
* recent versions more often than old ones.
*/
static void
index_add(dns_journal_t *j, journal_pos_t *pos) {
unsigned int i;
if (j->index == NULL)
return;
/*
* Search for a vacant position.
*/
for (i = 0; i < j->header.index_size; i++) {
if (! POS_VALID(j->index[i]))
break;
}
if (i == j->header.index_size) {
unsigned int k = 0;
/*
* Found no vacant position. Make some room.
*/
for (i = 0; i < j->header.index_size; i += 2) {
j->index[k++] = j->index[i];
}
i = k; /* 'i' identifies the first vacant position. */
while (k < j->header.index_size) {
POS_INVALIDATE(j->index[k]);
k++;
}
}
INSIST(i < j->header.index_size);
INSIST(! POS_VALID(j->index[i]));
/*
* Store the new index entry.
*/
j->index[i] = *pos;
}
/*
* Invalidate any existing index entries that could become
* ambiguous when a new transaction with number 'serial' is added.
*/
static void
index_invalidate(dns_journal_t *j, isc_uint32_t serial) {
unsigned int i;
if (j->index == NULL)
return;
for (i = 0; i < j->header.index_size; i++) {
if (! DNS_SERIAL_GT(serial, j->index[i].serial))
POS_INVALIDATE(j->index[i]);
}
}
/*
* Try to find a transaction with initial serial number 'serial'
* in the journal 'j'.
*
* If found, store its position at '*pos' and return ISC_R_SUCCESS.
*
* If 'serial' is current (= the ending serial number of the
* last transaction in the journal), set '*pos' to
* the position immediately following the last transaction and
* return ISC_R_SUCCESS.
*
* If 'serial' is within the range of addressable serial numbers
* covered by the journal but that particular serial number is missing
* (from the journal, not just from the index), return ISC_R_NOTFOUND.
*
* If 'serial' is outside the range of addressable serial numbers
* covered by the journal, return ISC_R_RANGE.
*
*/
static isc_result_t
journal_find(dns_journal_t *j, isc_uint32_t serial, journal_pos_t *pos) {
isc_result_t result;
journal_pos_t current_pos;
REQUIRE(DNS_JOURNAL_VALID(j));
if (DNS_SERIAL_GT(j->header.begin.serial, serial))
return (ISC_R_RANGE);
if (DNS_SERIAL_GT(serial, j->header.end.serial))
return (ISC_R_RANGE);
if (serial == j->header.end.serial) {
*pos = j->header.end;
return (ISC_R_SUCCESS);
}
current_pos = j->header.begin;
index_find(j, serial, ¤t_pos);
while (current_pos.serial != serial) {
if (DNS_SERIAL_GT(current_pos.serial, serial))
return (ISC_R_NOTFOUND);
result = journal_next(j, ¤t_pos);
if (result != ISC_R_SUCCESS)
return (result);
}
*pos = current_pos;
return (ISC_R_SUCCESS);
}
isc_result_t
dns_journal_begin_transaction(dns_journal_t *j) {
isc_uint32_t offset;
isc_result_t result;
journal_rawxhdr_t hdr;
REQUIRE(DNS_JOURNAL_VALID(j));
REQUIRE(j->state == JOURNAL_STATE_WRITE);
/*
* Find the file offset where the new transaction should
* be written, and seek there.
*/
if (JOURNAL_EMPTY(&j->header)) {
offset = sizeof(journal_rawheader_t) +
j->header.index_size * sizeof(journal_rawpos_t);
} else {
offset = j->header.end.offset;
}
j->x.pos[0].offset = offset;
j->x.pos[1].offset = offset; /* Initial value, will be incremented. */
j->x.n_soa = 0;
CHECK(journal_seek(j, offset));
/*
* Write a dummy transaction header of all zeroes to reserve
* space. It will be filled in when the transaction is
* finished.
*/
memset(&hdr, 0, sizeof(hdr));
CHECK(journal_write(j, &hdr, sizeof(hdr)));
j->x.pos[1].offset = j->offset;
j->state = JOURNAL_STATE_TRANSACTION;
result = ISC_R_SUCCESS;
failure:
return (result);
}
isc_result_t
dns_journal_writediff(dns_journal_t *j, dns_diff_t *diff) {
dns_difftuple_t *t;
isc_buffer_t buffer;
void *mem = NULL;
unsigned int size;
isc_result_t result;
isc_region_t used;
REQUIRE(DNS_DIFF_VALID(diff));
REQUIRE(j->state == JOURNAL_STATE_TRANSACTION);
isc_log_write(JOURNAL_DEBUG_LOGARGS(3), "writing to journal");
(void)dns_diff_print(diff, NULL);
/*
* Pass 1: determine the buffer size needed, and
* keep track of SOA serial numbers.
*/
size = 0;
for (t = ISC_LIST_HEAD(diff->tuples); t != NULL;
t = ISC_LIST_NEXT(t, link))
{
if (t->rdata.type == dns_rdatatype_soa) {
if (j->x.n_soa < 2)
j->x.pos[j->x.n_soa].serial =
dns_soa_getserial(&t->rdata);
j->x.n_soa++;
}
size += sizeof(journal_rawrrhdr_t);
size += t->name.length; /* XXX should have access macro? */
size += 10;
size += t->rdata.length;
}
mem = isc_mem_get(j->mctx, size);
if (mem == NULL)
return (ISC_R_NOMEMORY);
isc_buffer_init(&buffer, mem, size);
/*
* Pass 2. Write RRs to buffer.
*/
for (t = ISC_LIST_HEAD(diff->tuples); t != NULL;
t = ISC_LIST_NEXT(t, link))
{
/*
* Write the RR header.
*/
isc_buffer_putuint32(&buffer, t->name.length + 10 +
t->rdata.length);
/*
* Write the owner name, RR header, and RR data.
*/
isc_buffer_putmem(&buffer, t->name.ndata, t->name.length);
isc_buffer_putuint16(&buffer, t->rdata.type);
isc_buffer_putuint16(&buffer, t->rdata.rdclass);
isc_buffer_putuint32(&buffer, t->ttl);
INSIST(t->rdata.length < 65536);
isc_buffer_putuint16(&buffer, (isc_uint16_t)t->rdata.length);
INSIST(isc_buffer_availablelength(&buffer) >= t->rdata.length);
isc_buffer_putmem(&buffer, t->rdata.data, t->rdata.length);
}
isc_buffer_usedregion(&buffer, &used);
INSIST(used.length == size);
j->x.pos[1].offset += used.length;
/*
* Write the buffer contents to the journal file.
*/
CHECK(journal_write(j, used.base, used.length));
result = ISC_R_SUCCESS;
failure:
if (mem != NULL)
isc_mem_put(j->mctx, mem, size);
return (result);
}
isc_result_t
dns_journal_commit(dns_journal_t *j) {
isc_result_t result;
journal_rawheader_t rawheader;
REQUIRE(DNS_JOURNAL_VALID(j));
REQUIRE(j->state == JOURNAL_STATE_TRANSACTION);
/*
* Perform some basic consistency checks.
*/
if (j->x.n_soa != 2) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: malformed transaction: %d SOAs",
j->filename, j->x.n_soa);
return (ISC_R_UNEXPECTED);
}
if (! (DNS_SERIAL_GT(j->x.pos[1].serial, j->x.pos[0].serial) ||
(bind8_compat &&
j->x.pos[1].serial == j->x.pos[0].serial)))
{
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: malformed transaction: serial number "
"would decrease", j->filename);
return (ISC_R_UNEXPECTED);
}
if (! JOURNAL_EMPTY(&j->header)) {
if (j->x.pos[0].serial != j->header.end.serial) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"malformed transaction: "
"%s last serial %u != "
"transaction first serial %u",
j->filename,
j->header.end.serial,
j->x.pos[0].serial);
return (ISC_R_UNEXPECTED);
}
}
/*
* Some old journal entries may become non-addressable
* when we increment the current serial number. Purge them
* by stepping header.begin forward to the first addressable
* transaction. Also purge them from the index.
*/
if (! JOURNAL_EMPTY(&j->header)) {
while (! DNS_SERIAL_GT(j->x.pos[1].serial,
j->header.begin.serial)) {
CHECK(journal_next(j, &j->header.begin));
}
index_invalidate(j, j->x.pos[1].serial);
}
#ifdef notyet
if (DNS_SERIAL_GT(last_dumped_serial, j->x.pos[1].serial)) {
force_dump(...);
}
#endif
/*
* Commit the transaction data to stable storage.
*/
CHECK(journal_fsync(j));
/*
* Update the transaction header.
*/
CHECK(journal_seek(j, j->x.pos[0].offset));
CHECK(journal_write_xhdr(j, (j->x.pos[1].offset - j->x.pos[0].offset) -
sizeof(journal_rawxhdr_t),
j->x.pos[0].serial, j->x.pos[1].serial));
/*
* Update the journal header.
*/
if (JOURNAL_EMPTY(&j->header)) {
j->header.begin = j->x.pos[0];
}
j->header.end = j->x.pos[1];
journal_header_encode(&j->header, &rawheader);
CHECK(journal_seek(j, 0));
CHECK(journal_write(j, &rawheader, sizeof(rawheader)));
/*
* Update the index.
*/
index_add(j, &j->x.pos[0]);
/*
* Convert the index into on-disk format and write
* it to disk.
*/
CHECK(index_to_disk(j));
/*
* Commit the header to stable storage.
*/
CHECK(journal_fsync(j));
/*
* We no longer have a transaction open.
*/
j->state = JOURNAL_STATE_WRITE;
result = ISC_R_SUCCESS;
failure:
return (result);
}
isc_result_t
dns_journal_write_transaction(dns_journal_t *j, dns_diff_t *diff) {
isc_result_t result;
CHECK(dns_diff_sort(diff, ixfr_order));
CHECK(dns_journal_begin_transaction(j));
CHECK(dns_journal_writediff(j, diff));
CHECK(dns_journal_commit(j));
result = ISC_R_SUCCESS;
failure:
return (result);
}
void
dns_journal_destroy(dns_journal_t **journalp) {
dns_journal_t *j = *journalp;
REQUIRE(DNS_JOURNAL_VALID(j));
j->it.result = ISC_R_FAILURE;
dns_name_invalidate(&j->it.name);
dns_decompress_invalidate(&j->it.dctx);
if (j->rawindex != NULL)
isc_mem_put(j->mctx, j->rawindex, j->header.index_size *
sizeof(journal_rawpos_t));
if (j->index != NULL)
isc_mem_put(j->mctx, j->index, j->header.index_size *
sizeof(journal_pos_t));
if (j->it.target.base != NULL)
isc_mem_put(j->mctx, j->it.target.base, j->it.target.length);
if (j->it.source.base != NULL)
isc_mem_put(j->mctx, j->it.source.base, j->it.source.length);
if (j->fp != NULL)
(void)isc_stdio_close(j->fp);
j->magic = 0;
isc_mem_put(j->mctx, j, sizeof(*j));
*journalp = NULL;
}
/*
* Roll the open journal 'j' into the database 'db'.
* A new database version will be created.
*/
/* XXX Share code with incoming IXFR? */
static isc_result_t
roll_forward(dns_journal_t *j, dns_db_t *db) {
isc_buffer_t source; /* Transaction data from disk */
isc_buffer_t target; /* Ditto after _fromwire check */
isc_uint32_t db_serial; /* Database SOA serial */
isc_uint32_t end_serial; /* Last journal SOA serial */
isc_result_t result;
dns_dbversion_t *ver = NULL;
journal_pos_t pos;
dns_diff_t diff;
unsigned int n_soa = 0;
unsigned int n_put = 0;
REQUIRE(DNS_JOURNAL_VALID(j));
REQUIRE(DNS_DB_VALID(db));
dns_diff_init(j->mctx, &diff);
/*
* Set up empty initial buffers for uncheched and checked
* wire format transaction data. They will be reallocated
* later.
*/
isc_buffer_init(&source, NULL, 0);
isc_buffer_init(&target, NULL, 0);
/*
* Create the new database version.
*/
CHECK(dns_db_newversion(db, &ver));
/*
* Get the current database SOA serial number.
*/
CHECK(dns_db_getsoaserial(db, ver, &db_serial));
/*
* Locate a journal entry for the current database serial.
*/
CHECK(journal_find(j, db_serial, &pos));
/*
* XXX do more drastic things, like marking zone stale,
* if this fails?
*/
/*
* XXXRTH The zone code should probably mark the zone as bad and
* scream loudly into the log if this is a dynamic update
* log reply that failed.
*/
end_serial = dns_journal_last_serial(j);
if (db_serial == end_serial)
CHECK(DNS_R_UPTODATE);
CHECK(dns_journal_iter_init(j, db_serial, end_serial));
for (result = dns_journal_first_rr(j);
result == ISC_R_SUCCESS;
result = dns_journal_next_rr(j))
{
dns_name_t *name;
isc_uint32_t ttl;
dns_rdata_t *rdata;
dns_difftuple_t *tuple = NULL;
name = NULL;
rdata = NULL;
dns_journal_current_rr(j, &name, &ttl, &rdata);
if (rdata->type == dns_rdatatype_soa) {
n_soa++;
if (n_soa == 2)
db_serial = j->it.current_serial;
}
if (n_soa == 3)
n_soa = 1;
if (n_soa == 0) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: journal file corrupt: missing "
"initial SOA", j->filename);
FAIL(ISC_R_UNEXPECTED);
}
CHECK(dns_difftuple_create(diff.mctx, n_soa == 1 ?
DNS_DIFFOP_DEL : DNS_DIFFOP_ADD,
name, ttl, rdata, &tuple));
dns_diff_append(&diff, &tuple);
if (++n_put > 100) {
isc_log_write(JOURNAL_DEBUG_LOGARGS(3),
"%s: applying diff to database (%u)",
j->filename, db_serial);
(void)dns_diff_print(&diff, NULL);
CHECK(dns_diff_apply(&diff, db, ver));
dns_diff_clear(&diff);
n_put = 0;
}
}
if (result == ISC_R_NOMORE)
result = ISC_R_SUCCESS;
CHECK(result);
if (n_put != 0) {
isc_log_write(JOURNAL_DEBUG_LOGARGS(3),
"%s: applying final diff to database (%u)",
j->filename, db_serial);
(void)dns_diff_print(&diff, NULL);
CHECK(dns_diff_apply(&diff, db, ver));
dns_diff_clear(&diff);
}
failure:
if (ver != NULL)
dns_db_closeversion(db, &ver, result == ISC_R_SUCCESS ?
ISC_TRUE : ISC_FALSE);
if (source.base != NULL)
isc_mem_put(j->mctx, source.base, source.length);
if (target.base != NULL)
isc_mem_put(j->mctx, target.base, target.length);
dns_diff_clear(&diff);
return (result);
}
isc_result_t
dns_journal_rollforward(isc_mem_t *mctx, dns_db_t *db, const char *filename) {
dns_journal_t *j;
isc_result_t result;
REQUIRE(DNS_DB_VALID(db));
REQUIRE(filename != NULL);
j = NULL;
result = dns_journal_open(mctx, filename, ISC_FALSE, &j);
if (result == ISC_R_NOTFOUND) {
isc_log_write(JOURNAL_DEBUG_LOGARGS(3),
"no journal file, but that's OK");
return (DNS_R_NOJOURNAL);
}
if (result != ISC_R_SUCCESS)
return (result);
if (JOURNAL_EMPTY(&j->header))
result = DNS_R_UPTODATE;
else
result = roll_forward(j, db);
dns_journal_destroy(&j);
return (result);
}
isc_result_t
dns_journal_print(isc_mem_t *mctx, const char *filename, FILE *file) {
dns_journal_t *j;
isc_buffer_t source; /* Transaction data from disk */
isc_buffer_t target; /* Ditto after _fromwire check */
isc_uint32_t start_serial; /* Database SOA serial */
isc_uint32_t end_serial; /* Last journal SOA serial */
isc_result_t result;
dns_diff_t diff;
unsigned int n_soa = 0;
unsigned int n_put = 0;
REQUIRE(filename != NULL);
j = NULL;
result = dns_journal_open(mctx, filename, ISC_FALSE, &j);
if (result == ISC_R_NOTFOUND) {
isc_log_write(JOURNAL_DEBUG_LOGARGS(3), "no journal file");
return (DNS_R_NOJOURNAL);
}
if (result != ISC_R_SUCCESS) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"journal open failure: %s: %s",
isc_result_totext(result), j->filename);
return (result);
}
dns_diff_init(j->mctx, &diff);
/*
* Set up empty initial buffers for uncheched and checked
* wire format transaction data. They will be reallocated
* later.
*/
isc_buffer_init(&source, NULL, 0);
isc_buffer_init(&target, NULL, 0);
start_serial = dns_journal_first_serial(j);
end_serial = dns_journal_last_serial(j);
CHECK(dns_journal_iter_init(j, start_serial, end_serial));
for (result = dns_journal_first_rr(j);
result == ISC_R_SUCCESS;
result = dns_journal_next_rr(j))
{
dns_name_t *name;
isc_uint32_t ttl;
dns_rdata_t *rdata;
dns_difftuple_t *tuple = NULL;
name = NULL;
rdata = NULL;
dns_journal_current_rr(j, &name, &ttl, &rdata);
if (rdata->type == dns_rdatatype_soa)
n_soa++;
if (n_soa == 3)
n_soa = 1;
if (n_soa == 0) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: journal file corrupt: missing "
"initial SOA", j->filename);
FAIL(ISC_R_UNEXPECTED);
}
CHECK(dns_difftuple_create(diff.mctx, n_soa == 1 ?
DNS_DIFFOP_DEL : DNS_DIFFOP_ADD,
name, ttl, rdata, &tuple));
dns_diff_append(&diff, &tuple);
if (++n_put > 100) {
result = dns_diff_print(&diff, file);
dns_diff_clear(&diff);
n_put = 0;
if (result != ISC_R_SUCCESS)
break;
}
}
if (result == ISC_R_NOMORE)
result = ISC_R_SUCCESS;
CHECK(result);
if (n_put != 0) {
result = dns_diff_print(&diff, file);
dns_diff_clear(&diff);
}
goto cleanup;
failure:
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: cannot print: journal file corrupt", j->filename);
cleanup:
if (source.base != NULL)
isc_mem_put(j->mctx, source.base, source.length);
if (target.base != NULL)
isc_mem_put(j->mctx, target.base, target.length);
dns_diff_clear(&diff);
dns_journal_destroy(&j);
return (result);
}
/**************************************************************************/
/*
* Miscellaneous accessors.
*/
isc_uint32_t dns_journal_first_serial(dns_journal_t *j) {
return (j->header.begin.serial);
}
isc_uint32_t dns_journal_last_serial(dns_journal_t *j) {
return (j->header.end.serial);
}
/**************************************************************************/
/*
* Iteration support.
*
* When serving an outgoing IXFR, we transmit a part the journal starting
* at the serial number in the IXFR request and ending at the serial
* number that is current when the IXFR request arrives. The ending
* serial number is not necessarily at the end of the journal:
* the journal may grow while the IXFR is in progress, but we stop
* when we reach the serial number that was current when the IXFR started.
*/
static isc_result_t read_one_rr(dns_journal_t *j);
/*
* Make sure the buffer 'b' is has at least 'size' bytes
* allocated, and clear it.
*
* Requires:
* Either b->base is NULL, or it points to b->length bytes of memory
* previously allocated by isc_mem_get().
*/
static isc_result_t
size_buffer(isc_mem_t *mctx, isc_buffer_t *b, unsigned size) {
if (b->length < size) {
void *mem = isc_mem_get(mctx, size);
if (mem == NULL)
return (ISC_R_NOMEMORY);
if (b->base != NULL)
isc_mem_put(mctx, b->base, b->length);
b->base = mem;
b->length = size;
}
isc_buffer_clear(b);
return (ISC_R_SUCCESS);
}
isc_result_t
dns_journal_iter_init(dns_journal_t *j,
isc_uint32_t begin_serial, isc_uint32_t end_serial)
{
isc_result_t result;
CHECK(journal_find(j, begin_serial, &j->it.bpos));
INSIST(j->it.bpos.serial == begin_serial);
CHECK(journal_find(j, end_serial, &j->it.epos));
INSIST(j->it.epos.serial == end_serial);
result = ISC_R_SUCCESS;
failure:
j->it.result = result;
return (j->it.result);
}
isc_result_t
dns_journal_first_rr(dns_journal_t *j) {
isc_result_t result;
/*
* Seek to the beginning of the first transaction we are
* interested in.
*/
CHECK(journal_seek(j, j->it.bpos.offset));
j->it.current_serial = j->it.bpos.serial;
j->it.xsize = 0; /* We have no transaction data yet... */
j->it.xpos = 0; /* ...and haven't used any of it. */
return (read_one_rr(j));
failure:
return (result);
}
static isc_result_t
read_one_rr(dns_journal_t *j) {
isc_result_t result;
dns_rdatatype_t rdtype;
dns_rdataclass_t rdclass;
unsigned int rdlen;
isc_uint32_t ttl;
journal_xhdr_t xhdr;
journal_rrhdr_t rrhdr;
INSIST(j->offset <= j->it.epos.offset);
if (j->offset == j->it.epos.offset)
return (ISC_R_NOMORE);
if (j->it.xpos == j->it.xsize) {
/*
* We are at a transaction boundary.
* Read another transaction header.
*/
CHECK(journal_read_xhdr(j, &xhdr));
if (xhdr.size == 0) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: journal corrupt: empty transaction",
j->filename);
FAIL(ISC_R_UNEXPECTED);
}
if (xhdr.serial0 != j->it.current_serial) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: journal file corrupt: "
"expected serial %u, got %u",
j->filename,
j->it.current_serial, xhdr.serial0);
FAIL(ISC_R_UNEXPECTED);
}
j->it.xsize = xhdr.size;
j->it.xpos = 0;
}
/*
* Read an RR.
*/
CHECK(journal_read_rrhdr(j, &rrhdr));
/*
* Perform a sanity check on the journal RR size.
* The smallest possible RR has a 1-byte owner name
* and a 10-byte header. The largest possible
* RR has 65535 bytes of data, a header, and a maximum-
* size owner name, well below 70 k total.
*/
if (rrhdr.size < 1+10 || rrhdr.size > 70000) {
isc_log_write(JOURNAL_COMMON_LOGARGS, ISC_LOG_ERROR,
"%s: journal corrupt: impossible RR size "
"(%d bytes)", j->filename, rrhdr.size);
FAIL(ISC_R_UNEXPECTED);
}
CHECK(size_buffer(j->mctx, &j->it.source, rrhdr.size));
CHECK(journal_read(j, j->it.source.base, rrhdr.size));
isc_buffer_add(&j->it.source, rrhdr.size);
/*
* The target buffer is made the same size
* as the source buffer, with the assumption that when
* no compression in present, the output of dns_*_fromwire()
* is no larger than the input.
*/
CHECK(size_buffer(j->mctx, &j->it.target, rrhdr.size));
/*
* Parse the owner name. We don't know where it
* ends yet, so we make the entire "remaining"
* part of the buffer "active".
*/
isc_buffer_setactive(&j->it.source,
j->it.source.used - j->it.source.current);
CHECK(dns_name_fromwire(&j->it.name, &j->it.source,
&j->it.dctx, 0, &j->it.target));
/*
* Check that the RR header is there, and parse it.
*/
if (isc_buffer_remaininglength(&j->it.source) < 10)
FAIL(DNS_R_FORMERR);
rdtype = isc_buffer_getuint16(&j->it.source);
rdclass = isc_buffer_getuint16(&j->it.source);
ttl = isc_buffer_getuint32(&j->it.source);
rdlen = isc_buffer_getuint16(&j->it.source);
/*
* Parse the rdata.
*/
if (isc_buffer_remaininglength(&j->it.source) != rdlen)
FAIL(DNS_R_FORMERR);
isc_buffer_setactive(&j->it.source, rdlen);
dns_rdata_reset(&j->it.rdata);
CHECK(dns_rdata_fromwire(&j->it.rdata, rdclass,
rdtype, &j->it.source, &j->it.dctx,
0, &j->it.target));
j->it.ttl = ttl;
j->it.xpos += sizeof(journal_rawrrhdr_t) + rrhdr.size;
if (rdtype == dns_rdatatype_soa) {
/* XXX could do additional consistency checks here */
j->it.current_serial = dns_soa_getserial(&j->it.rdata);
}
result = ISC_R_SUCCESS;
failure:
j->it.result = result;
return (result);
}
isc_result_t
dns_journal_next_rr(dns_journal_t *j) {
j->it.result = read_one_rr(j);
return (j->it.result);
}
void
dns_journal_current_rr(dns_journal_t *j, dns_name_t **name, isc_uint32_t *ttl,
dns_rdata_t **rdata)
{
REQUIRE(j->it.result == ISC_R_SUCCESS);
*name = &j->it.name;
*ttl = j->it.ttl;
*rdata = &j->it.rdata;
}
/**************************************************************************/
/*
* Generating diffs from databases
*/
/*
* Construct a diff containing all the RRs at the current name of the
* database iterator 'dbit' in database 'db', version 'ver'.
* Set '*name' to the current name, and append the diff to 'diff'.
* All new tuples will have the operation 'op'.
*
* Requires: 'name' must have buffer large enough to hold the name.
* Typically, a dns_fixedname_t would be used.
*/
static isc_result_t
get_name_diff(dns_db_t *db, dns_dbversion_t *ver, isc_stdtime_t now,
dns_dbiterator_t *dbit, dns_name_t *name, dns_diffop_t op,
dns_diff_t *diff)
{
isc_result_t result;
dns_dbnode_t *node = NULL;
dns_rdatasetiter_t *rdsiter = NULL;
dns_difftuple_t *tuple = NULL;
result = dns_dbiterator_current(dbit, &node, name);
if (result != ISC_R_SUCCESS)
return (result);
result = dns_db_allrdatasets(db, node, ver, now, &rdsiter);
if (result != ISC_R_SUCCESS)
goto cleanup_node;
for (result = dns_rdatasetiter_first(rdsiter);
result == ISC_R_SUCCESS;
result = dns_rdatasetiter_next(rdsiter))
{
dns_rdataset_t rdataset;
dns_rdataset_init(&rdataset);
dns_rdatasetiter_current(rdsiter, &rdataset);
for (result = dns_rdataset_first(&rdataset);
result == ISC_R_SUCCESS;
result = dns_rdataset_next(&rdataset))
{
dns_rdata_t rdata = DNS_RDATA_INIT;
dns_rdataset_current(&rdataset, &rdata);
result = dns_difftuple_create(diff->mctx, op, name,
rdataset.ttl, &rdata,
&tuple);
if (result != ISC_R_SUCCESS) {
dns_rdataset_disassociate(&rdataset);
goto cleanup_iterator;
}
dns_diff_append(diff, &tuple);
}
dns_rdataset_disassociate(&rdataset);
if (result != ISC_R_NOMORE)
goto cleanup_iterator;
}
if (result != ISC_R_NOMORE)
goto cleanup_iterator;
result = ISC_R_SUCCESS;
cleanup_iterator:
dns_rdatasetiter_destroy(&rdsiter);
cleanup_node:
dns_db_detachnode(db, &node);
return (result);
}
/*
* Comparison function for use by dns_diff_subtract when sorting
* the diffs to be subtracted. The sort keys are the rdata type
* and the rdata itself. The owner name is ignored, because
* it is known to be the same for all tuples.
*/
static int
rdata_order(const void *av, const void *bv) {
dns_difftuple_t const * const *ap = av;
dns_difftuple_t const * const *bp = bv;
dns_difftuple_t const *a = *ap;
dns_difftuple_t const *b = *bp;
int r;
r = (b->rdata.type - a->rdata.type);
if (r != 0)
return (r);
r = dns_rdata_compare(&a->rdata, &b->rdata);
return (r);
}
static isc_result_t
dns_diff_subtract(dns_diff_t diff[2], dns_diff_t *r) {
isc_result_t result;
dns_difftuple_t *p[2];
int i, t;
isc_boolean_t append;
CHECK(dns_diff_sort(&diff[0], rdata_order));
CHECK(dns_diff_sort(&diff[1], rdata_order));
for (;;) {
p[0] = ISC_LIST_HEAD(diff[0].tuples);
p[1] = ISC_LIST_HEAD(diff[1].tuples);
if (p[0] == NULL && p[1] == NULL)
break;
for (i = 0; i < 2; i++)
if (p[!i] == NULL) {
ISC_LIST_UNLINK(diff[i].tuples, p[i], link);
ISC_LIST_APPEND(r->tuples, p[i], link);
goto next;
}
t = rdata_order(&p[0], &p[1]);
if (t < 0) {
ISC_LIST_UNLINK(diff[0].tuples, p[0], link);
ISC_LIST_APPEND(r->tuples, p[0], link);
goto next;
}
if (t > 0) {
ISC_LIST_UNLINK(diff[1].tuples, p[1], link);
ISC_LIST_APPEND(r->tuples, p[1], link);
goto next;
}
INSIST(t == 0);
/*
* Identical RRs in both databases; skip them both
* if the ttl differs.
*/
append = ISC_TF(p[0]->ttl != p[1]->ttl);
for (i = 0; i < 2; i++) {
ISC_LIST_UNLINK(diff[i].tuples, p[i], link);
if (append) {
ISC_LIST_APPEND(r->tuples, p[i], link);
} else {
dns_difftuple_free(&p[i]);
}
}
next: ;
}
result = ISC_R_SUCCESS;
failure:
return (result);
}
/*
* Compare the databases 'dba' and 'dbb' and generate a journal
* entry containing the changes to make 'dba' from 'dbb' (note
* the order). This journal entry will consist of a single,
* possibly very large transaction.
*/
isc_result_t
dns_db_diff(isc_mem_t *mctx,
dns_db_t *dba, dns_dbversion_t *dbvera,
dns_db_t *dbb, dns_dbversion_t *dbverb,
const char *journal_filename)
{
dns_db_t *db[2];
dns_dbversion_t *ver[2];
dns_dbiterator_t *dbit[2] = { NULL, NULL };
isc_boolean_t have[2] = { ISC_FALSE, ISC_FALSE };
dns_fixedname_t fixname[2];
isc_result_t result, itresult[2];
dns_diff_t diff[2], resultdiff;
int i, t;
dns_journal_t *journal = NULL;
db[0] = dba, db[1] = dbb;
ver[0] = dbvera, ver[1] = dbverb;
dns_diff_init(mctx, &diff[0]);
dns_diff_init(mctx, &diff[1]);
dns_diff_init(mctx, &resultdiff);
dns_fixedname_init(&fixname[0]);
dns_fixedname_init(&fixname[1]);
result = dns_journal_open(mctx, journal_filename, ISC_TRUE, &journal);
if (result != ISC_R_SUCCESS)
return (result);
result = dns_db_createiterator(db[0], ISC_FALSE, &dbit[0]);
if (result != ISC_R_SUCCESS)
goto cleanup_journal;
result = dns_db_createiterator(db[1], ISC_FALSE, &dbit[1]);
if (result != ISC_R_SUCCESS)
goto cleanup_interator0;
itresult[0] = dns_dbiterator_first(dbit[0]);
itresult[1] = dns_dbiterator_first(dbit[1]);
for (;;) {
for (i = 0; i < 2; i++) {
if (! have[i] && itresult[i] == ISC_R_SUCCESS) {
CHECK(get_name_diff(db[i], ver[i], 0, dbit[i],
dns_fixedname_name(&fixname[i]),
i == 0 ?
DNS_DIFFOP_ADD :
DNS_DIFFOP_DEL,
&diff[i]));
itresult[i] = dns_dbiterator_next(dbit[i]);
have[i] = ISC_TRUE;
}
}
if (! have[0] && ! have[1]) {
INSIST(ISC_LIST_EMPTY(diff[0].tuples));
INSIST(ISC_LIST_EMPTY(diff[1].tuples));
break;
}
for (i = 0; i < 2; i++) {
if (! have[!i]) {
ISC_LIST_APPENDLIST(resultdiff.tuples,
diff[i].tuples, link);
INSIST(ISC_LIST_EMPTY(diff[i].tuples));
have[i] = ISC_FALSE;
goto next;
}
}
t = dns_name_compare(dns_fixedname_name(&fixname[0]),
dns_fixedname_name(&fixname[1]));
if (t < 0) {
ISC_LIST_APPENDLIST(resultdiff.tuples,
diff[0].tuples, link);
INSIST(ISC_LIST_EMPTY(diff[0].tuples));
have[0] = ISC_FALSE;
continue;
}
if (t > 0) {
ISC_LIST_APPENDLIST(resultdiff.tuples,
diff[1].tuples, link);
INSIST(ISC_LIST_EMPTY(diff[1].tuples));
have[1] = ISC_FALSE;
continue;
}
INSIST(t == 0);
CHECK(dns_diff_subtract(diff, &resultdiff));
INSIST(ISC_LIST_EMPTY(diff[0].tuples));
INSIST(ISC_LIST_EMPTY(diff[1].tuples));
have[0] = have[1] = ISC_FALSE;
next: ;
}
if (itresult[0] != ISC_R_NOMORE)
FAIL(itresult[0]);
if (itresult[1] != ISC_R_NOMORE)
FAIL(itresult[1]);
if (ISC_LIST_EMPTY(resultdiff.tuples)) {
isc_log_write(JOURNAL_DEBUG_LOGARGS(3), "no changes");
} else {
CHECK(dns_journal_write_transaction(journal, &resultdiff));
}
INSIST(ISC_LIST_EMPTY(diff[0].tuples));
INSIST(ISC_LIST_EMPTY(diff[1].tuples));
failure:
dns_diff_clear(&resultdiff);
dns_dbiterator_destroy(&dbit[1]);
cleanup_interator0:
dns_dbiterator_destroy(&dbit[0]);
cleanup_journal:
dns_journal_destroy(&journal);
return (result);
}
isc_result_t
dns_journal_compact(isc_mem_t *mctx, char *filename, isc_uint32_t serial,
isc_uint32_t target_size)
{
unsigned int i;
journal_pos_t best_guess;
journal_pos_t current_pos;
dns_journal_t *j = NULL;
dns_journal_t *new = NULL;
journal_rawheader_t rawheader;
unsigned int copy_length;
int namelen;
char *buf = NULL;
unsigned int size = 0;
isc_result_t result;
unsigned int indexend;
char newname[1024];
char backup[1024];
isc_boolean_t is_backup = ISC_FALSE;
namelen = strlen(filename);
if (namelen > 4 && strcmp(filename + namelen - 4, ".jnl") == 0)
namelen -= 4;
result = isc_string_printf(newname, sizeof(newname), "%.*s.jnw",
namelen, filename);
if (result != ISC_R_SUCCESS)
return (result);
result = isc_string_printf(backup, sizeof(backup), "%.*s.jbk",
namelen, filename);
if (result != ISC_R_SUCCESS)
return (result);
result = journal_open(mctx, filename, ISC_FALSE, ISC_FALSE, &j);
if (result == ISC_R_NOTFOUND) {
is_backup = ISC_TRUE;
result = journal_open(mctx, backup, ISC_FALSE, ISC_FALSE, &j);
}
if (result != ISC_R_SUCCESS)
return (result);
if (JOURNAL_EMPTY(&j->header)) {
dns_journal_destroy(&j);
return (ISC_R_SUCCESS);
}
if (DNS_SERIAL_GT(j->header.begin.serial, serial) ||
DNS_SERIAL_GT(serial, j->header.end.serial)) {
dns_journal_destroy(&j);
return (ISC_R_RANGE);
}
/*
* Cope with very small target sizes.
*/
indexend = sizeof(journal_rawheader_t) +
j->header.index_size * sizeof(journal_rawpos_t);
if (target_size < indexend * 2)
target_size = target_size/2 + indexend;
/*
* See if there is any work to do.
*/
if ((isc_uint32_t) j->header.end.offset < target_size) {
dns_journal_destroy(&j);
return (ISC_R_SUCCESS);
}
CHECK(journal_open(mctx, newname, ISC_TRUE, ISC_TRUE, &new));
/*
* Remove overhead so space test below can succeed.
*/
if (target_size >= indexend)
target_size -= indexend;
/*
* Find if we can create enough free space.
*/
best_guess = j->header.begin;
for (i = 0; i < j->header.index_size; i++) {
if (POS_VALID(j->index[i]) &&
DNS_SERIAL_GE(serial, j->index[i].serial) &&
((isc_uint32_t)(j->header.end.offset - j->index[i].offset)
>= target_size / 2) &&
j->index[i].offset > best_guess.offset)
best_guess = j->index[i];
}
current_pos = best_guess;
while (current_pos.serial != serial) {
CHECK(journal_next(j, ¤t_pos));
if (current_pos.serial == j->header.end.serial)
break;
if (DNS_SERIAL_GE(serial, current_pos.serial) &&
((isc_uint32_t)(j->header.end.offset - current_pos.offset)
>= (target_size / 2)) &&
current_pos.offset > best_guess.offset)
best_guess = current_pos;
else
break;
}
INSIST(best_guess.serial != j->header.end.serial);
if (best_guess.serial != serial)
CHECK(journal_next(j, &best_guess));
/*
* We should now be roughly half target_size provided
* we did not reach 'serial'. If not we will just copy
* all uncommitted deltas regardless of the size.
*/
copy_length = j->header.end.offset - best_guess.offset;
if (copy_length != 0) {
/*
* Copy best_guess to end into space just freed.
*/
size = 64*1024;
if (copy_length < size)
size = copy_length;
buf = isc_mem_get(mctx, size);
if (buf == NULL) {
result = ISC_R_NOMEMORY;
goto failure;
}
CHECK(journal_seek(j, best_guess.offset));
CHECK(journal_seek(new, indexend));
for (i = 0; i < copy_length; i += size) {
unsigned int len = (copy_length - i) > size ? size :
(copy_length - i);
CHECK(journal_read(j, buf, len));
CHECK(journal_write(new, buf, len));
}
CHECK(journal_fsync(new));
/*
* Compute new header.
*/
new->header.begin.serial = best_guess.serial;
new->header.begin.offset = indexend;
new->header.end.serial = j->header.end.serial;
new->header.end.offset = indexend + copy_length;
/*
* Update the journal header.
*/
journal_header_encode(&new->header, &rawheader);
CHECK(journal_seek(new, 0));
CHECK(journal_write(new, &rawheader, sizeof(rawheader)));
CHECK(journal_fsync(new));
/*
* Build new index.
*/
current_pos = new->header.begin;
while (current_pos.serial != new->header.end.serial) {
index_add(new, ¤t_pos);
CHECK(journal_next(new, ¤t_pos));
}
/*
* Write index.
*/
CHECK(index_to_disk(new));
CHECK(journal_fsync(new));
indexend = new->header.end.offset;
}
dns_journal_destroy(&new);
/*
* With a UFS file system this should just succeed and be atomic.
* Any IXFR outs will just continue and the old journal will be
* removed on final close.
*
* With MSDOS / NTFS we need to do a two stage rename triggered
* bu EEXISTS. Hopefully all IXFR's that were active at the last
* rename are now complete.
*/
if (rename(newname, filename) == -1) {
if (errno == EACCES && !is_backup) {
result = isc_file_remove(backup);
if (result != ISC_R_SUCCESS &&
result != ISC_R_FILENOTFOUND)
goto failure;
if (rename(filename, backup) == -1)
goto maperrno;
if (rename(newname, filename) == -1)
goto maperrno;
(void)isc_file_remove(backup);
} else {
maperrno:
result = ISC_R_FAILURE;
goto failure;
}
}
dns_journal_destroy(&j);
result = ISC_R_SUCCESS;
failure:
(void)isc_file_remove(newname);
if (buf != NULL)
isc_mem_put(mctx, buf, size);
if (j != NULL)
dns_journal_destroy(&j);
if (new != NULL)
dns_journal_destroy(&new);
return (result);
}
static isc_result_t
index_to_disk(dns_journal_t *j) {
isc_result_t result = ISC_R_SUCCESS;
if (j->header.index_size != 0) {
unsigned int i;
unsigned char *p;
unsigned int rawbytes;
rawbytes = j->header.index_size * sizeof(journal_rawpos_t);
p = j->rawindex;
for (i = 0; i < j->header.index_size; i++) {
encode_uint32(j->index[i].serial, p);
p += 4;
encode_uint32(j->index[i].offset, p);
p += 4;
}
INSIST(p == j->rawindex + rawbytes);
CHECK(journal_seek(j, sizeof(journal_rawheader_t)));
CHECK(journal_write(j, j->rawindex, rawbytes));
}
failure:
return (result);
}
|