rustfs-ecstore 1.0.1

Erasure coding storage backend for RustFS, providing efficient data storage and retrieval with redundancy.
Documentation
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
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
// Copyright 2024 RustFS Team
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

use super::*;
use crate::bucket::{
    metadata::{BUCKET_TABLE_RESERVED_PREFIX, table_bucket_catalog_metadata_prefix},
    utils::is_meta_bucketname,
};
use crate::error::is_err_bucket_not_found;
use crate::runtime::sources as runtime_sources;
use crate::set_disk::{BucketInfoQuorum, get_lock_acquire_timeout};
use crate::storage_api_contracts::bucket::{BUCKET_LIFECYCLE_LOCK_OBJECT, SRBucketDeleteOp};
use crate::storage_api_contracts::namespace::NamespaceLocking as _;
use futures::stream::{self, StreamExt};
use rustfs_policy::policy::BucketPolicy;
use std::collections::BTreeMap;
use std::future::Future;

const DELETED_BUCKETS_PREFIX: &str = ".deleted";
const SCANNER_BUCKET_LIST_SET_CONCURRENCY: usize = 4;
const EVENT_BUCKET_DELETE_BLOCKED: &str = "bucket_delete_blocked";
const EVENT_BUCKET_DELETE_ROLLBACK_FAILED: &str = "bucket_delete_rollback_failed";

/// Record why `DeleteBucket` refused, and at a level that matches who can act
/// on it.
///
/// `DeleteBucket` answers from a raw per-disk residue scan, not from a listing,
/// so the server can refuse for a reason the client has no way to observe: the
/// caller drained every version the S3 API will show and still gets
/// `BucketNotEmpty`, with nothing to go on. This event is the only record of
/// which residue blocked it and where — and it used to be emitted at `debug`,
/// below both the `error` default log level and the `info` the CI s3-tests lane
/// runs at, so in practice it was never written down. An intermittent
/// `BucketNotEmpty` in that lane left a server log with no trace of the refusal
/// at all, which is not a diagnosable state.
///
/// A blocker the client can still see and delete is ordinary — the 409 already
/// says everything useful — so it stays at `warn`. Residue the client cannot
/// reach through the API is a server-side integrity problem and is reported at
/// `error`, which is what makes it survive a default deployment's filter.
fn record_bucket_delete_blocker(bucket: &str, kind: BucketDeleteBlockerKind, residue: &BucketMetadataLessResidue) {
    metrics::counter!("rustfs_bucket_delete_blockers_total", "kind" => kind.as_str()).increment(1);
    if kind.is_client_visible() {
        warn!(
            event = EVENT_BUCKET_DELETE_BLOCKED,
            component = "ecstore",
            subsystem = "bucket",
            bucket,
            blocker = kind.as_str(),
            files = residue.files,
            uuid_data_dirs = residue.uuid_data_dirs,
            entries_scanned = residue.entries_scanned,
            diagnostic_bytes_read = residue.diagnostic_bytes_read,
            diagnostic_truncated = residue.diagnostic_truncated,
            sample = residue.sample.as_deref().unwrap_or("<none>"),
            "Bucket deletion was blocked by durable local state"
        );
        return;
    }

    error!(
        event = EVENT_BUCKET_DELETE_BLOCKED,
        component = "ecstore",
        subsystem = "bucket",
        bucket,
        blocker = kind.as_str(),
        files = residue.files,
        uuid_data_dirs = residue.uuid_data_dirs,
        entries_scanned = residue.entries_scanned,
        diagnostic_bytes_read = residue.diagnostic_bytes_read,
        diagnostic_truncated = residue.diagnostic_truncated,
        sample = residue.sample.as_deref().unwrap_or("<none>"),
        "Bucket deletion was blocked by residue the client cannot reach through the S3 API"
    );
}

fn bucket_list_set_concurrency(set_count: usize, max_concurrency: usize) -> usize {
    set_count.clamp(1, max_concurrency.max(1))
}

fn should_override_created_from_metadata(created: OffsetDateTime) -> bool {
    created != OffsetDateTime::UNIX_EPOCH
}

fn validate_table_bucket_delete_allowed(
    bucket: &str,
    table_bucket_enabled: bool,
    table_catalog_metadata_exists: bool,
) -> Result<()> {
    if table_bucket_enabled && table_catalog_metadata_exists {
        return Err(StorageError::BucketNotEmpty(bucket.to_string()));
    }

    Ok(())
}

async fn table_catalog_metadata_exists(ctx: &crate::runtime::instance::InstanceContext, bucket: &str) -> Result<bool> {
    let local_disks = runtime_sources::local_disks_in(ctx).await;
    for disk in local_disks.iter() {
        let Some(bucket_path) = disk.get_bucket_path_for_io_if_local(bucket) else {
            continue;
        };
        let catalog_path = bucket_path?.join(BUCKET_TABLE_RESERVED_PREFIX);
        if has_xlmeta_files(&catalog_path).await? {
            return Ok(true);
        }
    }

    Ok(false)
}

async fn validate_table_bucket_delete_guard(ctx: &crate::runtime::instance::InstanceContext, bucket: &str) -> Result<()> {
    let table_bucket_enabled = metadata_sys::get_in(ctx, bucket)
        .await
        .is_ok_and(|metadata| metadata.table_bucket_enabled());
    if table_bucket_enabled {
        validate_table_bucket_delete_allowed(bucket, true, table_catalog_metadata_exists(ctx, bucket).await?)?;
    }

    Ok(())
}

fn bucket_delete_metadata_cleanup_prefixes(bucket: &str) -> [String; 2] {
    [
        table_bucket_catalog_metadata_prefix(bucket),
        format!("{BUCKET_META_PREFIX}/{bucket}"),
    ]
}

fn bucket_deleted_marker_prefix(bucket: &str) -> String {
    format!("{BUCKET_META_PREFIX}/{DELETED_BUCKETS_PREFIX}/{bucket}")
}

fn bucket_deleted_marker_volume(bucket: &str) -> String {
    format!("{RUSTFS_META_BUCKET}/{}", bucket_deleted_marker_prefix(bucket))
}

pub(crate) async fn await_bucket_namespace_operation<T, F>(
    guard: Option<&rustfs_lock::NamespaceLockGuard>,
    bucket: &str,
    operation: &'static str,
    future: F,
) -> Result<T>
where
    F: Future<Output = Result<T>>,
{
    let Some(guard) = guard else {
        return future.await;
    };
    if guard.is_lock_lost() {
        return Err(StorageError::other(format!(
            "bucket namespace lock was lost before {operation}: {bucket}"
        )));
    }
    tokio::select! {
        biased;
        _ = guard.lock_lost_notified() => Err(StorageError::other(format!(
            "bucket namespace lock was lost during {operation}: {bucket}"
        ))),
        result = future => result,
    }
}

async fn await_bucket_lifecycle_operation<T, F>(
    lifecycle_guard: Option<&rustfs_lock::NamespaceLockGuard>,
    namespace_guard: Option<&rustfs_lock::NamespaceLockGuard>,
    bucket: &str,
    operation: &'static str,
    future: F,
) -> Result<T>
where
    F: Future<Output = Result<T>>,
{
    await_bucket_namespace_operation(
        lifecycle_guard,
        bucket,
        operation,
        await_bucket_namespace_operation(namespace_guard, bucket, operation, future),
    )
    .await
}

async fn run_bucket_usage_cleanup<F>(guard: Option<&rustfs_lock::NamespaceLockGuard>, bucket: &str, future: F) -> Result<()>
where
    F: Future<Output = Result<()>>,
{
    await_bucket_namespace_operation(guard, bucket, "bucket usage cleanup", future).await
}

async fn run_physical_bucket_deletion<F>(guard: Option<&rustfs_lock::NamespaceLockGuard>, bucket: &str, future: F) -> Result<()>
where
    F: Future<Output = Result<()>>,
{
    // Fence before polling deletion: the physical namespace may become
    // invisible at any await point inside the storage operation.
    list_objects::observe_scanner_namespace_mutations(bucket, 1);
    await_bucket_namespace_operation(guard, bucket, "physical bucket deletion", future).await
}

async fn bucket_delete_local_blocker(
    ctx: &crate::runtime::instance::InstanceContext,
    bucket: &str,
    budget: &mut BucketDeleteDiagnosticBudget,
) -> Result<Option<StorageError>> {
    let local_disks = runtime_sources::local_disks_in(ctx).await;
    let mut residue = BucketMetadataLessResidue::default();

    for disk in local_disks.iter() {
        let Some(bucket_path) = disk.get_bucket_path_for_io_if_local(bucket) else {
            continue;
        };
        let scan = scan_metadata_less_residue_with_budget(&bucket_path?, budget).await?;
        if scan.xlmeta_found {
            record_bucket_delete_blocker(bucket, scan.xlmeta_blocker.unwrap_or(BucketDeleteBlockerKind::UnknownXlMeta), &scan);
            return Ok(Some(StorageError::BucketNotEmpty(bucket.to_string())));
        }
        residue.files = residue.files.saturating_add(scan.files);
        residue.uuid_data_dirs = residue.uuid_data_dirs.saturating_add(scan.uuid_data_dirs);
        residue.entries_scanned = residue.entries_scanned.saturating_add(scan.entries_scanned);
        residue.diagnostic_bytes_read = residue.diagnostic_bytes_read.saturating_add(scan.diagnostic_bytes_read);
        if residue.sample.is_none() {
            residue.sample = scan.sample.clone();
        }
        if scan.diagnostic_truncated {
            residue.diagnostic_truncated = true;
            if residue.files == 0 {
                // A zero-evidence timeout is inconclusive. Preflight still uses
                // non-recursive `force_if_empty`; post-failure keeps the physical error.
                return Ok(None);
            }
            record_bucket_delete_blocker(bucket, BucketDeleteBlockerKind::DiagnosticBudgetExceeded, &residue);
            return Ok(Some(StorageError::BucketNotEmptyWithDetails {
                bucket: bucket.to_string(),
                details: residue.describe(),
            }));
        }
    }

    if residue.has_residue_without_xlmeta() {
        record_bucket_delete_blocker(bucket, BucketDeleteBlockerKind::OrphanDirectory, &residue);
        return Ok(Some(StorageError::BucketNotEmptyWithDetails {
            bucket: bucket.to_string(),
            details: residue.describe(),
        }));
    }

    Ok(None)
}

impl ECStore {
    fn bucket_sets(&self) -> impl Iterator<Item = (usize, usize, Arc<crate::set_disk::SetDisks>)> + '_ {
        self.pools.iter().flat_map(|pool| {
            pool.disk_set
                .iter()
                .map(|set| (set.pool_index, set.set_index, Arc::clone(set)))
        })
    }

    pub async fn get_bucket_metadata(&self, bucket: &str) -> Result<Arc<BucketMetadata>> {
        let sys = metadata_sys::require_bucket_metadata_sys_in(&self.ctx)?;
        sys.read().await.get(bucket).await
    }

    pub async fn get_bucket_policy(&self, bucket: &str) -> Result<(BucketPolicy, OffsetDateTime)> {
        let sys = metadata_sys::require_bucket_metadata_sys_in(&self.ctx)?;
        sys.read().await.get_bucket_policy(bucket).await
    }

    pub async fn get_bucket_policy_raw(&self, bucket: &str) -> Result<(String, OffsetDateTime)> {
        let sys = metadata_sys::require_bucket_metadata_sys_in(&self.ctx)?;
        sys.read().await.get_bucket_policy_raw(bucket).await
    }

    pub async fn restricts_public_bucket_access(&self, bucket: &str) -> Result<bool> {
        let sys = metadata_sys::require_bucket_metadata_sys_in(&self.ctx)?;
        let (config, _) = sys.read().await.get_public_access_block_config(bucket).await?;
        Ok(config.restrict_public_buckets.unwrap_or(false))
    }

    pub async fn update_bucket_metadata_config(&self, bucket: &str, config_file: &str, data: Vec<u8>) -> Result<OffsetDateTime> {
        metadata_sys::update_in(&self.ctx, bucket, config_file, data).await
    }

    pub async fn bucket_incarnation_id(&self, bucket: &str) -> Result<Uuid> {
        metadata_sys::get_cached_bucket_incarnation_id_in(&self.ctx, bucket).await
    }

    pub async fn bucket_incarnation_id_from_disk(&self, bucket: &str) -> Result<Uuid> {
        metadata_sys::get_bucket_incarnation_id_in(&self.ctx, bucket).await
    }

    /// The object commit path acquires this sentinel before bucket metadata and
    /// exact-object namespace locks.
    pub(crate) async fn acquire_bucket_lifecycle_read_lock(&self, bucket: &str) -> Result<rustfs_lock::NamespaceLockGuard> {
        let lock = self.new_ns_lock(bucket, BUCKET_LIFECYCLE_LOCK_OBJECT).await?;
        lock.get_read_lock(get_lock_acquire_timeout()).await.map_err(|err| match err {
            rustfs_lock::error::LockError::QuorumNotReached { required, achieved } => {
                StorageError::NamespaceLockQuorumUnavailable {
                    mode: "bucket_lifecycle_read",
                    bucket: bucket.to_string(),
                    object: BUCKET_LIFECYCLE_LOCK_OBJECT.to_string(),
                    required,
                    achieved,
                }
            }
            other => StorageError::Lock(other),
        })
    }

    /// Acquire the bucket lifecycle read lock and validate the bucket
    /// incarnation against `expected`, memoizing the validation while this
    /// node keeps continuous read-lock coverage (see [`super::bucket_fence`]).
    ///
    /// Semantics are identical to the pre-existing per-PUT
    /// `acquire_bucket_lifecycle_read_lock` + from-disk
    /// `validate_bucket_incarnation` pair: the first PUT in a coverage window
    /// performs exactly that authoritative disk validation; overlapping PUTs
    /// reuse its result, which is sound because bucket deletion/recreation
    /// requires the lifecycle WRITE lock and therefore cannot have run while
    /// any read guard was continuously held.
    pub async fn acquire_bucket_incarnation_fence(
        &self,
        bucket: &str,
        expected: uuid::Uuid,
    ) -> Result<super::BucketIncarnationFenceGuard> {
        let inner = self.acquire_bucket_lifecycle_read_lock(bucket).await?;
        let pieces = super::bucket_fence::FencePieces {
            registry: self.bucket_fence_registry.clone(),
            inner,
        };
        let registration = pieces.enter(bucket);
        let current = match registration.memoized {
            Some(current) => current,
            None => match metadata_sys::get_bucket_incarnation_id_in(&self.ctx, bucket).await {
                Ok(current) => {
                    // Never memoize under lost coverage: a granted lifecycle
                    // write lock could already have changed the incarnation.
                    if !pieces.lock_lost() {
                        pieces.memoize(bucket, current);
                    }
                    current
                }
                Err(err) => {
                    pieces.abandon(bucket, registration.token);
                    return Err(err);
                }
            },
        };
        if current != expected {
            pieces.abandon(bucket, registration.token);
            return Err(StorageError::BucketNotFound(bucket.to_string()));
        }
        Ok(pieces.into_guard(bucket, registration.token))
    }

    /// Hold this guard through recursive-delete authorization and mutation so
    /// writers cannot introduce an unchecked object into the deletion scope.
    pub async fn lock_bucket_for_recursive_delete(&self, bucket: &str) -> Result<rustfs_lock::NamespaceLockGuard> {
        if self.ctx.lock_manager().is_disabled() {
            return Err(StorageError::InvalidArgument(
                bucket.to_owned(),
                String::new(),
                "Recursive deletion requires namespace locking".to_owned(),
            ));
        }
        self.acquire_bucket_lifecycle_write_lock(bucket).await
    }

    pub(crate) async fn acquire_bucket_lifecycle_write_lock(&self, bucket: &str) -> Result<rustfs_lock::NamespaceLockGuard> {
        let lock = self.new_ns_lock(bucket, BUCKET_LIFECYCLE_LOCK_OBJECT).await?;
        lock.get_write_lock(get_lock_acquire_timeout())
            .await
            .map_err(|err| match err {
                rustfs_lock::error::LockError::QuorumNotReached { required, achieved } => {
                    StorageError::NamespaceLockQuorumUnavailable {
                        mode: "bucket_lifecycle_write",
                        bucket: bucket.to_string(),
                        object: BUCKET_LIFECYCLE_LOCK_OBJECT.to_string(),
                        required,
                        achieved,
                    }
                }
                other => StorageError::Lock(other),
            })
    }

    async fn mark_bucket_deleted(&self, bucket: &str) -> Result<()> {
        let marker_volume = bucket_deleted_marker_volume(bucket);

        self.make_bucket_on_sets(
            marker_volume.as_str(),
            &MakeBucketOptions {
                force_create: true,
                ..Default::default()
            },
        )
        .await
        .map_err(|err| to_object_err(err, vec![bucket]))?;

        Ok(())
    }

    async fn make_bucket_on_sets(&self, bucket: &str, opts: &MakeBucketOptions) -> Result<()> {
        let results = futures::future::join_all(
            self.bucket_sets()
                .map(|(_, _, set)| async move { set.make_bucket(bucket, opts).await }),
        )
        .await;
        if results.is_empty() {
            return Err(StorageError::ErasureWriteQuorum);
        }

        let mut bucket_exists_error = None;
        let mut first_hard_error = None;
        for result in results {
            let Err(err) = result else {
                continue;
            };
            if matches!(err, StorageError::VolumeExists) || is_err_bucket_exists(&err) {
                if bucket_exists_error.is_none() {
                    bucket_exists_error = Some(err);
                }
            } else if first_hard_error.is_none() {
                first_hard_error = Some(err);
            }
        }

        match first_hard_error.or(bucket_exists_error) {
            Some(err) => Err(err),
            None => Ok(()),
        }
    }

    async fn delete_bucket_on_sets(&self, bucket: &str, opts: &DeleteBucketOptions) -> Result<()> {
        let results = futures::future::join_all(
            self.bucket_sets()
                .map(|(_, _, set)| async move { set.delete_bucket(bucket, opts).await }),
        )
        .await;
        if results.is_empty() {
            return Err(StorageError::ErasureWriteQuorum);
        }

        let mut deleted = false;
        let mut first_error = None;
        for result in results {
            match result {
                Ok(()) => deleted = true,
                Err(err) if is_err_strict_volume_not_found(&err) => {}
                Err(err) if first_error.is_none() => first_error = Some(err),
                Err(_) => {}
            }
        }

        if let Some(delete_error) = first_error {
            if !opts.no_recreate {
                let rollback_opts = MakeBucketOptions {
                    force_create: true,
                    no_lock: true,
                    ..Default::default()
                };
                if let Err(rollback_error) = self.make_bucket_on_sets(bucket, &rollback_opts).await {
                    warn!(
                        event = EVENT_BUCKET_DELETE_ROLLBACK_FAILED,
                        component = "ecstore",
                        subsystem = "bucket",
                        bucket,
                        deletion_error = ?delete_error,
                        rollback_error = ?rollback_error,
                        "Bucket deletion rollback could not restore every bucket volume"
                    );
                }
            }
            return Err(delete_error);
        }

        if deleted { Ok(()) } else { Err(StorageError::VolumeNotFound) }
    }

    async fn cleanup_deleted_bucket_metadata(
        &self,
        bucket: &str,
        include_deleted_marker: bool,
        guard: Option<&rustfs_lock::NamespaceLockGuard>,
    ) -> Result<()> {
        for prefix in bucket_delete_metadata_cleanup_prefixes(bucket) {
            await_bucket_namespace_operation(
                guard,
                bucket,
                "deleted bucket metadata cleanup",
                self.delete_all(RUSTFS_META_BUCKET, prefix.as_str()),
            )
            .await?;
        }

        if include_deleted_marker {
            let marker_prefix = bucket_deleted_marker_prefix(bucket);
            await_bucket_namespace_operation(
                guard,
                bucket,
                "deleted bucket marker cleanup",
                self.delete_all(RUSTFS_META_BUCKET, marker_prefix.as_str()),
            )
            .await?;
        }

        await_bucket_namespace_operation(
            guard,
            bucket,
            "deleted bucket metadata cache cleanup",
            metadata_sys::remove_bucket_metadata_in(&self.ctx, bucket),
        )
        .await?;
        runtime_sources::delete_bucket_monitor_entry(bucket);
        Ok(())
    }

    async fn cleanup_bucket_usage(&self, bucket: &str, guard: Option<&rustfs_lock::NamespaceLockGuard>) -> Result<()> {
        run_bucket_usage_cleanup(guard, bucket, async {
            crate::data_usage::prepare_bucket_usage_for_namespace_change(bucket, guard).await?;
            crate::data_usage::remove_bucket_usage_from_backend_with_guard_fenced(self, bucket, guard).await
        })
        .await
    }

    async fn cleanup_bucket_usage_best_effort(&self, bucket: &str, guard: Option<&rustfs_lock::NamespaceLockGuard>) {
        if let Err(err) = self.cleanup_bucket_usage(bucket, guard).await {
            warn!(
                bucket = %bucket,
                error = ?err,
                "bucket data usage cleanup deferred to scanner reconciliation"
            );
        }
    }

    async fn rollback_failed_bucket_creation(
        &self,
        bucket: &str,
        lifecycle_guard: Option<&rustfs_lock::NamespaceLockGuard>,
        namespace_guard: Option<&rustfs_lock::NamespaceLockGuard>,
    ) {
        let rollback_opts = DeleteBucketOptions {
            no_lock: true,
            no_recreate: true,
            ..Default::default()
        };
        if let Err(err) =
            await_bucket_lifecycle_operation(lifecycle_guard, namespace_guard, bucket, "failed bucket creation rollback", async {
                self.delete_bucket_on_sets(bucket, &rollback_opts)
                    .await
                    .map_err(|rollback_err| to_object_err(rollback_err, vec![bucket]))
            })
            .await
        {
            warn!(
                bucket = %bucket,
                error = ?err,
                "failed bucket creation rollback did not remove every physical bucket volume"
            );
            return;
        }
        if let Err(err) = self.cleanup_deleted_bucket_metadata(bucket, false, namespace_guard).await {
            warn!(
                bucket = %bucket,
                error = ?err,
                "failed bucket creation rollback left internal bucket metadata"
            );
        }
    }

    #[instrument(skip(self))]
    pub(super) async fn handle_make_bucket(&self, bucket: &str, opts: &MakeBucketOptions) -> Result<()> {
        if !is_meta_bucketname(bucket)
            && let Err(err) = check_valid_bucket_name_strict(bucket)
        {
            return Err(StorageError::BucketNameInvalid(err.to_string()));
        }

        // Lock order: bucket lifecycle -> metadata transaction -> exact bucket namespace.
        let bucket_lifecycle_guard = if !opts.no_lock {
            Some(self.acquire_bucket_lifecycle_write_lock(bucket).await?)
        } else {
            None
        };
        let metadata_transaction_guard = if !opts.no_lock && !is_meta_bucketname(bucket) {
            Some(metadata_sys::acquire_bucket_metadata_transaction_lock_in(&self.ctx, bucket).await?)
        } else {
            None
        };
        let ns_guard = if !opts.no_lock {
            let ns_lock = self.new_ns_lock(bucket, bucket).await?;
            Some(
                await_bucket_namespace_operation(
                    bucket_lifecycle_guard.as_ref(),
                    bucket,
                    "bucket namespace lock acquisition",
                    async {
                        ns_lock.get_write_lock(get_lock_acquire_timeout()).await.map_err(|e| match e {
                            rustfs_lock::error::LockError::QuorumNotReached { required, achieved } => {
                                StorageError::NamespaceLockQuorumUnavailable {
                                    mode: "write",
                                    bucket: bucket.to_string(),
                                    object: bucket.to_string(),
                                    required,
                                    achieved,
                                }
                            }
                            other => StorageError::Lock(other),
                        })
                    },
                )
                .await?,
            )
        } else {
            None
        };

        let existing_bucket_info = match self.get_bucket_info_from_sets(bucket, &BucketOptions::default()).await {
            Ok(info) => Some(info),
            Err(err) => {
                if is_err_bucket_not_found(&err) {
                    None
                } else {
                    return Err(to_object_err(err, vec![bucket]));
                }
            }
        };
        let confirmed_missing = existing_bucket_info.is_none();
        let existing_metadata = if opts.force_create && !confirmed_missing && !is_meta_bucketname(bucket) {
            let (mut metadata, persisted) = metadata_sys::get_config_from_disk_with_presence_in(&self.ctx, bucket).await?;
            if !persisted {
                metadata = BucketMetadata::new(bucket);
                metadata.created = existing_bucket_info
                    .as_ref()
                    .and_then(|info| info.created)
                    .unwrap_or(OffsetDateTime::UNIX_EPOCH);
                // A sidecar left by an interrupted legacy migration is the
                // bucket's published incarnation; keep it rather than minting
                // one that would invalidate fences taken on the stored value.
                // A retired incarnation is residue of a deleted bucket and
                // must not come back.
                let store = metadata_sys::object_store_in(&self.ctx).await?;
                if let Some(stored) = crate::bucket::metadata::load_bucket_incarnation(store.clone(), bucket).await?
                    && !crate::bucket::retirement::is_retired(store, bucket, stored).await?
                {
                    metadata.bucket_incarnation_id = stored;
                }
            } else if !metadata.bucket_incarnation_sidecar && !metadata.bucket_incarnation_id.is_nil() {
                return Err(Error::other(format!(
                    "bucket incarnation sidecar is missing for new-format metadata: {bucket}"
                )));
            } else if !metadata.bucket_incarnation_sidecar {
                metadata.bucket_incarnation_id = Uuid::new_v4();
            }
            Some(metadata)
        } else {
            None
        };

        let mut meta = existing_metadata.unwrap_or_else(|| {
            if confirmed_missing && !is_meta_bucketname(bucket) {
                BucketMetadata::new_with_default_durability(bucket)
            } else {
                BucketMetadata::new(bucket)
            }
        });
        let existing_incarnation_is_authoritative = meta.bucket_incarnation_sidecar;
        if confirmed_missing || is_meta_bucketname(bucket) {
            meta.set_created(opts.created_at);

            if opts.versioning_enabled {
                meta.versioning_config_xml =
                    crate::bucket::utils::serialize::<VersioningConfiguration>(&ENABLED_VERSIONING_CONFIG)?;
            }
        }

        // Object Lock enable is one-way, and it must apply to an existing bucket
        // too. Site replication replays make-with-versioning carrying the
        // source's lockEnabled against a destination bucket that already exists;
        // gating this on `confirmed_missing` returned success while leaving the
        // replica unlocked, so replicated versions could be deleted without the
        // retention the source enforces.
        let lock_newly_enabled = opts.lock_enabled && !meta.lock_enabled;
        if opts.lock_enabled {
            meta.lock_enabled = true;
            meta.object_lock_config_xml = crate::bucket::utils::serialize(&*crate::store::ENABLED_OBJECT_LOCK_CONFIG)?;
            meta.versioning_config_xml = crate::bucket::utils::serialize::<VersioningConfiguration>(&ENABLED_VERSIONING_CONFIG)?;
        }

        let metadata_persisted_before_physical = confirmed_missing && !is_meta_bucketname(bucket) && opts.lock_enabled;
        if metadata_persisted_before_physical {
            metadata_sys::set_new_bucket_metadata_intent_in(&self.ctx, meta.clone()).await?;
            if bucket_lifecycle_guard.as_ref().is_some_and(|guard| guard.is_lock_lost())
                || metadata_transaction_guard.as_ref().is_some_and(|guard| guard.is_lock_lost())
                || ns_guard.as_ref().is_some_and(|guard| guard.is_lock_lost())
            {
                return Err(Error::other(format!("bucket metadata creation intent lock was lost: {bucket}")));
            }
        }

        if confirmed_missing && !is_meta_bucketname(bucket) {
            // Fence every scanner cycle that could have observed the namespace
            // before physical creation. Creation may become visible even when a
            // later metadata write or namespace-lock check fails.
            crate::store::list_objects::observe_scanner_namespace_mutations(bucket, 1);
            self.cleanup_bucket_usage(bucket, ns_guard.as_ref()).await?;
        }

        if let Err(err) = await_bucket_lifecycle_operation(
            bucket_lifecycle_guard.as_ref(),
            ns_guard.as_ref(),
            bucket,
            "physical bucket creation",
            await_bucket_namespace_operation(
                metadata_transaction_guard.as_ref(),
                bucket,
                "bucket creation metadata transaction",
                async {
                    self.make_bucket_on_sets(bucket, opts)
                        .await
                        .map_err(|err| to_object_err(err, vec![bucket]))
                },
            ),
        )
        .await
        {
            if is_err_bucket_exists(&err)
                && let Err(heal_err) = self
                    .handle_heal_bucket(
                        bucket,
                        &HealOpts {
                            recreate: true,
                            ..Default::default()
                        },
                    )
                    .await
            {
                warn!("best-effort bucket heal after BucketExists failed: {heal_err}");
            }
            if confirmed_missing && !is_err_bucket_exists(&err) && ns_guard.as_ref().is_none_or(|guard| !guard.is_lock_lost()) {
                error!("make bucket failed: {err}");
                self.rollback_failed_bucket_creation(bucket, bucket_lifecycle_guard.as_ref(), ns_guard.as_ref())
                    .await;
            }
            return Err(err);
        };

        let metadata_result = async {
            if metadata_persisted_before_physical {
                return metadata_sys::commit_bucket_metadata_in(&self.ctx, meta).await;
            }
            if is_meta_bucketname(bucket) {
                metadata_sys::set_bucket_metadata_in(&self.ctx, meta).await
            } else if existing_incarnation_is_authoritative && !lock_newly_enabled {
                metadata_sys::commit_bucket_metadata_in(&self.ctx, meta).await
            } else {
                metadata_sys::set_new_bucket_metadata_in(&self.ctx, meta).await
            }
        }
        .await;
        let metadata_lock_lost = bucket_lifecycle_guard.as_ref().is_some_and(|guard| guard.is_lock_lost())
            || metadata_transaction_guard.as_ref().is_some_and(|guard| guard.is_lock_lost())
            || ns_guard.as_ref().is_some_and(|guard| guard.is_lock_lost());
        if let Err(err) = metadata_result {
            if !metadata_lock_lost {
                self.rollback_failed_bucket_creation(bucket, bucket_lifecycle_guard.as_ref(), ns_guard.as_ref())
                    .await;
            }
            return Err(err);
        }
        if metadata_lock_lost {
            return Err(Error::other(format!("bucket metadata initialization lock was lost: {bucket}")));
        }

        if confirmed_missing && !is_meta_bucketname(bucket) {
            // A scanner may have sampled the first fence before the bucket
            // became visible. Fence again after metadata initialization so
            // that snapshot cannot publish as complete.
            crate::store::list_objects::observe_scanner_namespace_mutations(bucket, 1);
        }

        Ok(())
    }

    /// Prove a live bucket generation before repairing missing expansion volumes.
    /// Unlike request validation, repair only needs one erasure set to confirm
    /// existence; an incomplete expansion set is precisely what repair fixes.
    /// Callers must hold the bucket namespace lock through the subsequent heal.
    pub(crate) async fn bucket_exists_for_heal(&self, bucket: &str) -> Result<bool> {
        let results = futures::future::join_all(
            self.bucket_sets()
                .map(|(_, _, set)| async move { set.get_bucket_info(bucket, &BucketOptions::default()).await }),
        )
        .await;
        let mut first_error = None;
        for result in results {
            match result {
                Ok(_) => return Ok(true),
                Err(err) if is_err_strict_volume_not_found(&err) => {}
                Err(err) if first_error.is_none() => first_error = Some(err),
                Err(_) => {}
            }
        }
        match first_error {
            Some(err) => Err(err),
            None => Ok(false),
        }
    }

    #[instrument(skip(self))]
    pub(crate) async fn get_bucket_info_from_sets(&self, bucket: &str, opts: &BucketOptions) -> Result<BucketInfo> {
        self.get_bucket_info_from_sets_with_quorum(bucket, opts, BucketInfoQuorum::Write)
            .await
    }

    pub(crate) async fn get_bucket_info_from_sets_at_read_quorum(
        &self,
        bucket: &str,
        opts: &BucketOptions,
    ) -> Result<BucketInfo> {
        self.get_bucket_info_from_sets_with_quorum(bucket, opts, BucketInfoQuorum::Read)
            .await
    }

    async fn get_bucket_info_from_sets_with_quorum(
        &self,
        bucket: &str,
        opts: &BucketOptions,
        quorum: BucketInfoQuorum,
    ) -> Result<BucketInfo> {
        // One host may participate in several pools after expansion. Resolve the
        // namespace against each erasure set so disks from different pools can
        // never be combined into one bucket quorum.
        // Bucket validation is request-path IO. Keep the previous peer fanout's
        // latency shape by probing every set concurrently; scanner listings use
        // a separate bounded path below because they run continuously.
        let mut scoped_results = futures::future::join_all(self.bucket_sets().map(|(pool_index, set_index, set)| async move {
            let result = match quorum {
                BucketInfoQuorum::Read => set.stat_bucket_with_quorum(bucket, quorum).await,
                BucketInfoQuorum::Write => set.get_bucket_info(bucket, opts).await,
            };
            (pool_index, set_index, result)
        }))
        .await;
        scoped_results.sort_unstable_by_key(|(pool_index, set_index, _)| (*pool_index, *set_index));

        let mut first_info = None;
        let mut first_error = None;
        for (_, _, result) in scoped_results {
            match result {
                Ok(info) if first_info.is_none() => first_info = Some(info),
                Ok(_) => {}
                Err(err) if is_err_strict_volume_not_found(&err) => {}
                Err(err) if first_error.is_none() => first_error = Some(err),
                Err(_) => {}
            }
        }

        if let Some(err) = first_error {
            return Err(err);
        }

        first_info.ok_or(Error::VolumeNotFound)
    }

    #[instrument(skip(self))]
    pub(super) async fn handle_get_bucket_info(&self, bucket: &str, opts: &BucketOptions) -> Result<BucketInfo> {
        let mut info = match self.get_bucket_info_from_sets(bucket, opts).await {
            Ok(info) => info,
            Err(Error::ErasureWriteQuorum) => return self.get_bucket_info_at_read_quorum(bucket, opts).await,
            Err(err) => return Err(err),
        };

        if let Ok(sys) = metadata_sys::get_in(&self.ctx, bucket).await {
            if should_override_created_from_metadata(sys.created) {
                info.created = Some(sys.created);
            }
            info.versioning = sys.versioning();
            info.object_locking = sys.object_locking();
        }

        Ok(info)
    }

    async fn get_bucket_info_at_read_quorum(&self, bucket: &str, opts: &BucketOptions) -> Result<BucketInfo> {
        // Lock order: bucket lifecycle -> internal metadata object read locks.
        // Keep create/delete from changing the namespace while a read quorum
        // confirms both physical presence and persisted bucket metadata.
        let guard = self.acquire_bucket_lifecycle_read_lock(bucket).await?;
        await_bucket_namespace_operation(Some(&guard), bucket, "bucket read quorum validation", async {
            let mut info = self
                .get_bucket_info_from_sets_with_quorum(bucket, opts, BucketInfoQuorum::Read)
                .await?;
            let (metadata, persisted) = metadata_sys::get_config_from_disk_with_presence_in(&self.ctx, bucket).await?;
            if !persisted {
                // A minority of directories left by failed creation is not an
                // authoritative bucket. Never turn fabricated defaults into
                // permission to serve degraded reads.
                return Err(Error::ErasureReadQuorum);
            }
            if metadata.lock_enabled && !metadata.bucket_creation_committed {
                return Err(Error::ErasureWriteQuorum);
            }
            if metadata.name != bucket {
                return Err(Error::FileCorrupt);
            }
            if should_override_created_from_metadata(metadata.created) {
                info.created = Some(metadata.created);
            }
            info.versioning = metadata.versioning();
            info.object_locking = metadata.object_locking();
            Ok(info)
        })
        .await
    }

    #[instrument(skip(self))]
    pub(super) async fn handle_list_bucket(&self, opts: &BucketOptions) -> Result<Vec<BucketInfo>> {
        // TODO(backlog): support cached bucket listing via opts.cached
        Ok(self.list_bucket_from_sets(opts, usize::MAX).await?.buckets)
    }

    pub async fn list_bucket_for_scanner(&self, opts: &BucketOptions) -> Result<crate::cluster::rpc::ScannerBucketListing> {
        self.list_bucket_from_sets(opts, SCANNER_BUCKET_LIST_SET_CONCURRENCY).await
    }

    async fn list_bucket_from_sets(
        &self,
        opts: &BucketOptions,
        max_concurrency: usize,
    ) -> Result<crate::cluster::rpc::ScannerBucketListing> {
        let set_count = self.pools.iter().map(|pool| pool.disk_set.len()).sum();
        let concurrency = bucket_list_set_concurrency(set_count, max_concurrency);
        let deleted = opts.deleted;
        let cached = opts.cached;
        let no_metadata = opts.no_metadata;
        let mut set_listings = stream::iter(self.bucket_sets().map(move |(pool_index, set_index, set)| {
            let opts = BucketOptions {
                deleted,
                cached,
                no_metadata,
            };
            async move {
                set.list_bucket_for_scanner(&opts)
                    .await
                    .map(|(buckets, complete)| (pool_index, set_index, buckets, complete))
            }
        }))
        .buffer_unordered(concurrency);
        let mut topology_complete = set_count != 0;
        let mut bucket_map = BTreeMap::<String, BucketInfo>::new();
        let mut scoped_buckets = Vec::with_capacity(set_count);
        while let Some(set_listing) = set_listings.next().await {
            let (pool_index, set_index, buckets, set_complete) = set_listing?;
            topology_complete &= set_complete;
            for bucket in &buckets {
                bucket_map.entry(bucket.name.clone()).or_insert_with(|| bucket.clone());
            }
            scoped_buckets.push(crate::cluster::rpc::ScannerSetBucketListing {
                pool_index,
                set_index,
                buckets,
            });
        }
        scoped_buckets.sort_unstable_by_key(|scope| (scope.pool_index, scope.set_index));
        let mut listing = crate::cluster::rpc::ScannerBucketListing {
            buckets: bucket_map.into_values().collect(),
            set_buckets: scoped_buckets,
            topology_complete,
        };

        if !opts.no_metadata {
            for bucket in &mut listing.buckets {
                if let Ok(created) = metadata_sys::created_at_in(&self.ctx, &bucket.name).await
                    && should_override_created_from_metadata(created)
                {
                    bucket.created = Some(created);
                }
            }
            let created_by_bucket = listing
                .buckets
                .iter()
                .map(|bucket| (bucket.name.as_str(), bucket.created))
                .collect::<BTreeMap<_, _>>();
            for scope in &mut listing.set_buckets {
                for bucket in &mut scope.buckets {
                    if let Some(created) = created_by_bucket.get(bucket.name.as_str()) {
                        bucket.created = *created;
                    }
                }
            }
        }

        Ok(listing)
    }

    #[instrument(skip(self))]
    pub(super) async fn handle_delete_bucket(&self, bucket: &str, opts: &DeleteBucketOptions) -> Result<()> {
        self.handle_delete_bucket_with_diagnostic_budget(bucket, opts, BucketDeleteDiagnosticBudget::new())
            .await
    }

    async fn handle_delete_bucket_with_diagnostic_budget(
        &self,
        bucket: &str,
        opts: &DeleteBucketOptions,
        mut diagnostic_budget: BucketDeleteDiagnosticBudget,
    ) -> Result<()> {
        if is_meta_bucketname(bucket) {
            return Err(StorageError::BucketNameInvalid(bucket.to_string()));
        }

        if let Err(err) = check_valid_bucket_name(bucket) {
            return Err(StorageError::BucketNameInvalid(err.to_string()));
        }

        let bucket_lifecycle_guard = if !opts.no_lock {
            Some(self.acquire_bucket_lifecycle_write_lock(bucket).await?)
        } else {
            None
        };
        let ns_guard = if !opts.no_lock {
            let ns_lock = self.new_ns_lock(bucket, bucket).await?;
            Some(
                await_bucket_namespace_operation(
                    bucket_lifecycle_guard.as_ref(),
                    bucket,
                    "bucket namespace lock acquisition",
                    async {
                        ns_lock.get_write_lock(get_lock_acquire_timeout()).await.map_err(|e| match e {
                            rustfs_lock::error::LockError::QuorumNotReached { required, achieved } => {
                                StorageError::NamespaceLockQuorumUnavailable {
                                    mode: "write",
                                    bucket: bucket.to_string(),
                                    object: bucket.to_string(),
                                    required,
                                    achieved,
                                }
                            }
                            other => StorageError::Lock(other),
                        })
                    },
                )
                .await?,
            )
        } else {
            None
        };

        let sr_mark_delete = opts.srdelete_op == SRBucketDeleteOp::MarkDelete;
        let sr_purge = opts.srdelete_op == SRBucketDeleteOp::Purge;
        let sr_delete = sr_mark_delete || sr_purge;
        let mut delete_opts = opts.clone();
        let bucket_exists = match self.get_bucket_info_from_sets(bucket, &BucketOptions::default()).await {
            Ok(_) => true,
            Err(err) => {
                if is_err_strict_volume_not_found(&err) && sr_delete {
                    false
                } else if is_err_strict_volume_not_found(&err) {
                    return Err(StorageError::BucketNotFound(bucket.to_string()));
                } else {
                    return Err(to_object_err(err, vec![bucket]));
                }
            }
        };
        if bucket_exists {
            validate_table_bucket_delete_guard(&self.ctx, bucket).await?;

            if !opts.force {
                if let Some(blocker) = bucket_delete_local_blocker(&self.ctx, bucket, &mut diagnostic_budget).await? {
                    return Err(blocker);
                }
                delete_opts.force_if_empty = true;
            }
        }

        if sr_delete && !bucket_exists {
            delete_opts.force_if_empty = true;
        }

        #[cfg(test)]
        crate::cluster::rpc::peer_s3_client::pause_after_delete_bucket_empty_scan().await;

        if sr_mark_delete {
            await_bucket_lifecycle_operation(
                bucket_lifecycle_guard.as_ref(),
                ns_guard.as_ref(),
                bucket,
                "bucket delete marker creation",
                self.mark_bucket_deleted(bucket),
            )
            .await?;
        }

        // Capture the authoritative old identity before its namespace disappears.
        // Legacy buckets without a stamp cannot produce retirement authority.
        let retirement = if bucket_exists && bucket_lifecycle_guard.is_some() && !is_meta_bucketname(bucket) {
            if let Some(store) = metadata_sys::object_store_if_initialized_in(&self.ctx).await {
                crate::bucket::metadata::load_bucket_incarnation(store.clone(), bucket)
                    .await?
                    .map(|incarnation| (store, incarnation))
            } else {
                None
            }
        } else {
            None
        };

        let delete_result = await_bucket_namespace_operation(
            bucket_lifecycle_guard.as_ref(),
            bucket,
            "physical bucket deletion",
            run_physical_bucket_deletion(ns_guard.as_ref(), bucket, async {
                self.delete_bucket_on_sets(bucket, &delete_opts)
                    .await
                    .map_err(|err| to_object_err(err, vec![bucket]))
            }),
        )
        .await;
        if let Err(err) = delete_result
            && (!sr_delete || !is_err_strict_volume_not_found(&err))
        {
            if delete_opts.force_if_empty && matches!(&err, StorageError::BucketNotEmpty(_)) {
                let mut diagnostic_budget = BucketDeleteDiagnosticBudget::new();
                if let Some(blocker) = bucket_delete_local_blocker(&self.ctx, bucket, &mut diagnostic_budget).await? {
                    return Err(blocker);
                }
            }
            return Err(err);
        }

        if let Some((store, incarnation)) = retirement {
            let mut record_opts = ObjectOptions {
                max_parity: true,
                ..Default::default()
            };
            if let Some(guard) = bucket_lifecycle_guard.as_ref() {
                record_opts.add_bucket_lifecycle_lock_guard(guard);
            }
            if let Some(guard) = ns_guard.as_ref() {
                record_opts.add_namespace_lock_guard(guard);
            }
            await_bucket_lifecycle_operation(
                bucket_lifecycle_guard.as_ref(),
                ns_guard.as_ref(),
                bucket,
                "bucket retirement publication",
                crate::bucket::retirement::commit_retirement(store, bucket, incarnation, &record_opts),
            )
            .await?;
        }

        self.cleanup_bucket_usage_best_effort(bucket, ns_guard.as_ref()).await;

        if let Err(err) = self
            .cleanup_deleted_bucket_metadata(bucket, sr_purge, ns_guard.as_ref())
            .await
        {
            warn!(
                bucket = %bucket,
                error = ?err,
                "physical bucket deletion succeeded but metadata cleanup remains pending"
            );
        }
        // A scanner may have sampled the first fence before the physical
        // namespace disappeared. The completion fence invalidates that scan.
        crate::store::list_objects::observe_scanner_namespace_mutations(bucket, 1);
        Ok(())
    }
}

#[cfg(test)]
mod tests {
    use super::{
        BUCKET_DELETE_DIAGNOSTIC_MAX_ELAPSED, BUCKET_DELETE_DIAGNOSTIC_MAX_ENTRIES, BUCKET_DELETE_XLMETA_DIAGNOSTIC_MAX_BYTES,
        BucketDeleteBlockerKind, BucketDeleteDiagnosticBudget, BucketMetadataLessResidue, SCANNER_BUCKET_LIST_SET_CONCURRENCY,
        await_bucket_namespace_operation, bucket_delete_metadata_cleanup_prefixes, bucket_deleted_marker_prefix,
        bucket_deleted_marker_volume, bucket_list_set_concurrency, record_bucket_delete_blocker, run_bucket_usage_cleanup,
        run_physical_bucket_deletion, scan_metadata_less_residue, scan_metadata_less_residue_with_budget,
        should_override_created_from_metadata, validate_table_bucket_delete_allowed,
    };
    use crate::bucket::metadata::{BucketMetadata, table_bucket_catalog_metadata_prefix};
    use crate::bucket::metadata_sys;
    use crate::cluster::rpc::peer_s3_client::install_delete_bucket_empty_scan_barrier;
    use crate::disk::{BUCKET_META_PREFIX, DiskAPI, RUSTFS_META_BUCKET, STORAGE_FORMAT_FILE};
    use crate::error::StorageError;
    use crate::object_api::{ObjectOptions, PutObjReader};
    use crate::runtime::instance::InstanceContext;
    use crate::storage_api_contracts::{
        bucket::{BucketOperations as _, BucketOptions, DeleteBucketOptions, MakeBucketOptions, SRBucketDeleteOp},
        list::ListOperations as _,
        namespace::NamespaceLocking as _,
        object::{ObjectIO as _, ObjectOperations as _},
    };
    use crate::store::{ECStore, init_local_disks_with_instance_ctx};
    use crate::{
        disk::endpoint::Endpoint,
        layout::endpoints::{EndpointServerPools, Endpoints, PoolEndpoints},
    };
    use rustfs_data_usage::{BucketUsageInfo, DATA_USAGE_OBJECT_NAME, DataUsageInfo};
    use rustfs_filemeta::{FileInfo, FileMeta, TRANSITION_COMPLETE};
    use rustfs_lock::{LocalClient, LockRequest, LockType, NamespaceLock, ObjectKey};
    use serial_test::serial;
    use std::path::{Path, PathBuf};
    use std::sync::Arc;
    use std::sync::atomic::{AtomicBool, Ordering};
    use std::time::{Duration, SystemTime};
    use time::OffsetDateTime;
    use tokio::io::AsyncReadExt;
    use tokio::sync::{Notify, OnceCell};
    use tokio_util::sync::CancellationToken;
    use uuid::Uuid;

    static BUCKET_DELETE_TEST_ENV: OnceCell<(Vec<PathBuf>, Arc<ECStore>)> = OnceCell::const_new();

    #[tokio::test(start_paused = true)]
    async fn bucket_delete_diagnostic_budget_starts_with_first_scan_io_and_latches_once() {
        let mut budget = BucketDeleteDiagnosticBudget::with_limits(8, Duration::from_millis(100));
        tokio::time::advance(Duration::from_secs(10)).await;

        let first_polled = Arc::new(AtomicBool::new(false));
        let first_polled_for_io = first_polled.clone();
        let first = budget
            .run_io(async move {
                first_polled_for_io.store(true, Ordering::SeqCst);
                Ok::<_, std::io::Error>(7_u8)
            })
            .await
            .expect("the first diagnostic IO should succeed");
        assert_eq!(first, Some(7));
        assert!(first_polled.load(Ordering::SeqCst));

        tokio::time::advance(Duration::from_millis(101)).await;
        let expired_polled = Arc::new(AtomicBool::new(false));
        let expired_polled_for_io = expired_polled.clone();
        let expired = budget
            .run_io(async move {
                expired_polled_for_io.store(true, Ordering::SeqCst);
                Ok::<_, std::io::Error>(9_u8)
            })
            .await
            .expect("an expired diagnostic budget should not become an IO error");
        assert_eq!(expired, None);
        assert!(
            !expired_polled.load(Ordering::SeqCst),
            "the deadline must remain latched after the first scan IO"
        );
    }

    #[tokio::test(start_paused = true)]
    async fn bucket_delete_diagnostic_budget_times_out_its_first_pending_io() {
        let mut budget = BucketDeleteDiagnosticBudget::with_limits(8, Duration::from_millis(100));
        let io_polled = Arc::new(AtomicBool::new(false));
        let io_polled_for_future = io_polled.clone();

        let result = budget
            .run_io(std::future::poll_fn(move |_cx| {
                io_polled_for_future.store(true, Ordering::SeqCst);
                std::task::Poll::<std::io::Result<()>>::Pending
            }))
            .await
            .expect("a diagnostic timeout should fail closed without an IO error");

        assert_eq!(result, None);
        assert!(
            io_polled.load(Ordering::SeqCst),
            "the first diagnostic IO must be polled before its timeout"
        );
    }

    #[tokio::test(start_paused = true)]
    async fn delayed_metadata_less_scans_still_detect_orphans_and_xlmeta() {
        let root = tempfile::tempdir().expect("temporary delayed-scan roots should be created");
        let orphan_root = root.path().join("orphan-root");
        let xlmeta_root = root.path().join("xlmeta-root");
        std::fs::create_dir_all(&orphan_root).expect("orphan root should be created");
        std::fs::create_dir_all(&xlmeta_root).expect("xlmeta root should be created");
        std::fs::write(orphan_root.join("orphan-part"), b"orphan").expect("orphan fixture should be written");
        std::fs::write(xlmeta_root.join(STORAGE_FORMAT_FILE), b"invalid-xlmeta").expect("xl.meta fixture should be written");

        let mut orphan_budget = BucketDeleteDiagnosticBudget::with_limits(16, Duration::from_secs(5));
        let mut xlmeta_budget = BucketDeleteDiagnosticBudget::with_limits(16, Duration::from_secs(5));
        tokio::time::advance(Duration::from_secs(60)).await;

        let orphan = scan_metadata_less_residue_with_budget(&orphan_root, &mut orphan_budget)
            .await
            .expect("delayed orphan scan should complete");
        assert!(orphan.has_residue_without_xlmeta());
        assert!(!orphan.diagnostic_truncated);

        let xlmeta = scan_metadata_less_residue_with_budget(&xlmeta_root, &mut xlmeta_budget)
            .await
            .expect("delayed xl.meta scan should complete");
        assert!(xlmeta.xlmeta_found);
        assert!(!xlmeta.diagnostic_truncated);
    }

    #[tokio::test(start_paused = true)]
    async fn bucket_namespace_operation_fails_closed_after_lease_expiry() {
        let ttl = Duration::from_millis(20);
        let lock = NamespaceLock::new("bucket-operation-test".to_string(), Arc::new(LocalClient::new()));
        let request = LockRequest::new(ObjectKey::new("bucket", ""), LockType::Exclusive, "test-owner")
            .with_acquire_timeout(Duration::from_secs(1))
            .with_ttl(ttl)
            .with_refresh_interval(ttl);
        let guard = lock
            .acquire_guard(&request)
            .await
            .expect("namespace lock acquisition should not fail")
            .expect("namespace lock should be acquired");
        tokio::time::advance(ttl + Duration::from_millis(1)).await;

        let operation_ran = Arc::new(AtomicBool::new(false));
        let operation_ran_for_future = operation_ran.clone();
        let result = await_bucket_namespace_operation(Some(&guard), "bucket", "test operation", async move {
            operation_ran_for_future.store(true, Ordering::SeqCst);
            Ok(())
        })
        .await;

        assert!(result.is_err(), "an expired namespace lease must fence the operation");
        assert!(
            !operation_ran.load(Ordering::SeqCst),
            "a fenced namespace operation must not poll its mutation future"
        );
    }

    #[tokio::test]
    #[serial]
    async fn physical_bucket_delete_fences_scanner_before_polling_storage() {
        let generation_before = crate::store::list_objects::scanner_namespace_mutation_generation();
        let storage_polled = Arc::new(AtomicBool::new(false));
        let storage_polled_for_future = storage_polled.clone();

        run_physical_bucket_deletion(None, "generation-order-bucket", async move {
            assert!(
                crate::store::list_objects::scanner_namespace_mutation_generation() > generation_before,
                "scanner generation must advance before physical deletion is polled"
            );
            storage_polled_for_future.store(true, Ordering::SeqCst);
            Ok(())
        })
        .await
        .expect("synthetic physical deletion should succeed");

        assert!(storage_polled.load(Ordering::SeqCst));
    }

    #[tokio::test]
    async fn bucket_usage_cleanup_stops_after_parent_cancellation() {
        let started = Arc::new(Notify::new());
        let started_wait = started.notified();
        let release = Arc::new(Notify::new());
        let completed = Arc::new(AtomicBool::new(false));
        let started_for_cleanup = started.clone();
        let release_for_cleanup = release.clone();
        let completed_for_cleanup = completed.clone();
        let parent = tokio::spawn(run_bucket_usage_cleanup(None, "bucket", async move {
            started_for_cleanup.notify_one();
            release_for_cleanup.notified().await;
            completed_for_cleanup.store(true, Ordering::SeqCst);
            Ok(())
        }));
        started_wait.await;

        parent.abort();
        let _ = parent.await;
        tokio::task::yield_now().await;
        release.notify_waiters();
        tokio::task::yield_now().await;
        assert!(
            !completed.load(Ordering::SeqCst),
            "a cancelled cleanup future must not continue in a detached task"
        );
    }

    #[tokio::test(start_paused = true)]
    #[serial]
    async fn bucket_namespace_operation_stops_in_flight_work_after_lock_loss() {
        let ttl = Duration::from_millis(20);
        let lock = NamespaceLock::new("bucket-operation-in-flight-loss-test".to_string(), Arc::new(LocalClient::new()));
        let request = LockRequest::new(ObjectKey::new("bucket", ""), LockType::Exclusive, "test-owner")
            .with_acquire_timeout(Duration::from_secs(1))
            .with_ttl(ttl)
            .with_refresh_interval(ttl);
        let guard = lock
            .acquire_guard(&request)
            .await
            .expect("namespace lock acquisition should not fail")
            .expect("namespace lock should be acquired");

        let operation_started = Arc::new(Notify::new());
        let started_wait = operation_started.notified();
        let operation_release = Arc::new(Notify::new());
        let operation_completed = Arc::new(AtomicBool::new(false));
        let started_for_operation = operation_started.clone();
        let release_for_operation = operation_release.clone();
        let completed_for_operation = operation_completed.clone();
        let task = tokio::spawn(async move {
            await_bucket_namespace_operation(Some(&guard), "bucket", "test operation", async move {
                started_for_operation.notify_one();
                release_for_operation.notified().await;
                completed_for_operation.store(true, Ordering::SeqCst);
                Ok(())
            })
            .await
        });
        started_wait.await;

        tokio::time::advance(ttl + Duration::from_millis(1)).await;
        let err = task
            .await
            .expect("operation task should join")
            .expect_err("an operation still running after lease loss must be fenced");
        assert!(err.to_string().contains("namespace lock was lost during test operation"));
        operation_release.notify_waiters();
        tokio::task::yield_now().await;
        assert!(
            !operation_completed.load(Ordering::SeqCst),
            "lock loss must stop the old owner before a successor can acquire the namespace"
        );
    }

    async fn setup_bucket_delete_test_env() -> (Vec<PathBuf>, Arc<ECStore>) {
        BUCKET_DELETE_TEST_ENV
            .get_or_init(|| async {
                let temp_dir = std::env::temp_dir().join(format!("rustfs_bucket_delete_test_{}", Uuid::new_v4()));
                tokio::fs::create_dir_all(&temp_dir)
                    .await
                    .expect("test base directory should be created");

                let disk_paths = (0..4)
                    .map(|disk_idx| temp_dir.join(format!("disk{disk_idx}")))
                    .collect::<Vec<_>>();

                for disk_path in &disk_paths {
                    tokio::fs::create_dir_all(disk_path)
                        .await
                        .expect("disk directory should be created");
                }

                let mut endpoints = Vec::with_capacity(disk_paths.len());
                for (disk_idx, disk_path) in disk_paths.iter().enumerate() {
                    let mut endpoint =
                        Endpoint::try_from(disk_path.to_str().expect("disk path should be utf8")).expect("endpoint should parse");
                    endpoint.set_pool_index(0);
                    endpoint.set_set_index(0);
                    endpoint.set_disk_index(disk_idx);
                    endpoints.push(endpoint);
                }

                let endpoint_pools = EndpointServerPools(vec![PoolEndpoints {
                    legacy: false,
                    set_count: 1,
                    drives_per_set: 4,
                    endpoints: Endpoints::from(endpoints),
                    cmd_line: "bucket-delete-test".to_string(),
                    platform: format!("OS: {} | Arch: {}", std::env::consts::OS, std::env::consts::ARCH),
                }]);

                let instance_ctx = Arc::new(InstanceContext::new());
                init_local_disks_with_instance_ctx(&instance_ctx, endpoint_pools.clone())
                    .await
                    .expect("local disks should initialize");
                let ecstore = ECStore::new_with_instance_ctx(
                    "127.0.0.1:0".parse().expect("test address"),
                    endpoint_pools,
                    CancellationToken::new(),
                    instance_ctx,
                )
                .await
                .expect("ECStore should initialize");
                let storage_class = crate::config::storageclass::lookup_config_for_pools_without_env(
                    &rustfs_config::server_config::KVS::new(),
                    &[4],
                )
                .expect("bucket test storage class should match its four-disk pool");
                for pool in &ecstore.pools {
                    for set in &pool.disk_set {
                        set.set_test_storage_class_config(storage_class.clone());
                    }
                }

                metadata_sys::init_bucket_metadata_sys(ecstore.clone(), Vec::new()).await;

                (disk_paths, ecstore)
            })
            .await
            .clone()
    }

    async fn setup_multi_pool_bucket_test_env() -> (tempfile::TempDir, Arc<ECStore>) {
        setup_bucket_quorum_test_env(&[4, 4], None).await
    }

    async fn setup_bucket_quorum_test_env(
        drives_per_pool: &[usize],
        standard_parity: Option<usize>,
    ) -> (tempfile::TempDir, Arc<ECStore>) {
        let temp_dir = tempfile::tempdir().expect("multi-pool bucket test directory should be created");
        let mut pools = Vec::new();
        for (pool_index, &drive_count) in drives_per_pool.iter().enumerate() {
            let mut endpoints = Vec::new();
            for disk_index in 0..drive_count {
                let disk_path = temp_dir.path().join(format!("pool{pool_index}-disk{disk_index}"));
                tokio::fs::create_dir_all(&disk_path)
                    .await
                    .expect("multi-pool bucket test disk should be created");
                let mut endpoint =
                    Endpoint::try_from(disk_path.to_str().expect("disk path should be utf8")).expect("endpoint should parse");
                endpoint.set_pool_index(pool_index);
                endpoint.set_set_index(0);
                endpoint.set_disk_index(disk_index);
                endpoints.push(endpoint);
            }
            pools.push(PoolEndpoints {
                legacy: false,
                set_count: 1,
                drives_per_set: drive_count,
                endpoints: Endpoints::from(endpoints),
                cmd_line: format!("bucket-test-pool-{pool_index}"),
                platform: format!("OS: {} | Arch: {}", std::env::consts::OS, std::env::consts::ARCH),
            });
        }

        let endpoint_pools = EndpointServerPools(pools);
        let instance_ctx = Arc::new(InstanceContext::new());
        instance_ctx.set_endpoints(endpoint_pools.clone());
        init_local_disks_with_instance_ctx(&instance_ctx, endpoint_pools.clone())
            .await
            .expect("multi-pool local disks should initialize");
        let ecstore = ECStore::new_with_instance_ctx(
            "127.0.0.1:0".parse().expect("test address"),
            endpoint_pools,
            CancellationToken::new(),
            instance_ctx,
        )
        .await
        .expect("multi-pool ECStore should initialize");
        let mut storage_class_kvs = rustfs_config::server_config::KVS::new();
        if let Some(parity) = standard_parity {
            storage_class_kvs.insert(crate::config::storageclass::CLASS_STANDARD.to_string(), format!("EC:{parity}"));
        }
        let storage_class = crate::config::storageclass::lookup_config_for_pools_without_env(&storage_class_kvs, drives_per_pool)
            .expect("storage class should match every test erasure set");
        for pool in &ecstore.pools {
            for set in &pool.disk_set {
                set.set_test_storage_class_config(storage_class.clone());
            }
        }

        (temp_dir, ecstore)
    }

    async fn take_set_disks_offline(
        ecstore: &ECStore,
        set: &Arc<crate::set_disk::SetDisks>,
        disk_indexes: &[usize],
    ) -> Vec<(usize, crate::disk::DiskStore)> {
        let offline = {
            let disks = set.disks.read().await;
            disk_indexes
                .iter()
                .map(|index| (*index, disks[*index].clone().expect("fault-injection disk should start online")))
                .collect::<Vec<_>>()
        };
        for (_, disk) in &offline {
            disk.close().await.expect("fault injection should stop per-disk monitoring");
            disk.force_offline_for_test();
        }
        // Empty slots can be renewed from the still-present disk paths by the endpoint monitor.
        // Keep the original handles and prove reconnect cannot clear their injected IO failure.
        set.connect_disks().await;
        {
            let disks = set.disks.read().await;
            for (index, disk) in &offline {
                let current = disks[*index].as_ref().expect("fault-injection slot should remain populated");
                assert!(Arc::ptr_eq(current, disk), "reconnect must preserve the fault-injection handle");
                assert_eq!(disk.runtime_state(), crate::disk::health_state::RuntimeDriveHealthState::Offline);
            }
        }
        let local_disk_map = ecstore.ctx.local_disk_map();
        let mut local_disks = local_disk_map.write().await;
        for (_, disk) in &offline {
            local_disks.insert(disk.endpoint().to_string(), None);
        }
        offline
    }

    async fn restore_set_disks(
        ecstore: &ECStore,
        set: &Arc<crate::set_disk::SetDisks>,
        offline: Vec<(usize, crate::disk::DiskStore)>,
    ) {
        {
            let local_disk_map = ecstore.ctx.local_disk_map();
            let mut local_disks = local_disk_map.write().await;
            for (_, disk) in &offline {
                local_disks.insert(disk.endpoint().to_string(), Some(Arc::clone(disk)));
            }
        }
        let disks = set.disks.read().await;
        for (index, disk) in offline {
            let current = disks[index].as_ref().expect("fault-injection slot should remain populated");
            assert!(Arc::ptr_eq(current, &disk), "fault-injection disk handle should remain unchanged");
            assert_eq!(disk.runtime_state(), crate::disk::health_state::RuntimeDriveHealthState::Offline);
            disk.reset_health_for_store_init_retry();
        }
    }

    #[tokio::test]
    async fn request_metadata_methods_fail_closed_before_instance_initialization() {
        let (_temp_dir, store) = setup_multi_pool_bucket_test_env().await;

        let expected = "bucket metadata sys not initialized for this instance";
        let errors = [
            store.get_bucket_metadata("bucket").await.unwrap_err(),
            store.get_bucket_policy("bucket").await.unwrap_err(),
            store.get_bucket_policy_raw("bucket").await.unwrap_err(),
            store.restricts_public_bucket_access("bucket").await.unwrap_err(),
            store
                .update_bucket_metadata_config("bucket", crate::bucket::metadata::BUCKET_POLICY_CONFIG, Vec::new())
                .await
                .unwrap_err(),
        ];
        for error in errors {
            assert_eq!(error.to_string(), format!("Io error: {expected}"));
        }
    }

    async fn create_bucket_with_object(ecstore: &Arc<ECStore>, bucket: &str, object: &str) {
        let generation_before_make = ecstore.scanner_namespace_mutation_generation();
        ecstore
            .make_bucket(bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created");
        assert_eq!(
            ecstore.scanner_namespace_mutation_generation(),
            generation_before_make.saturating_add(2),
            "successful bucket creation should fence scanner namespace activity before and after creation"
        );

        let generation_before_put = ecstore.scanner_namespace_mutation_generation();
        let mut reader = PutObjReader::from_vec(b"delete bucket semantics".to_vec());
        ecstore
            .put_object(bucket, object, &mut reader, &ObjectOptions::default())
            .await
            .expect("object should be written");
        let lock = ecstore.pools[0].disk_set[0]
            .new_ns_lock(bucket, object)
            .await
            .expect("fixture namespace lock should be created");
        drop(
            lock.get_write_lock(Duration::from_secs(30))
                .await
                .expect("fixture rename tail should finish before checking its generation"),
        );
        assert_eq!(
            ecstore.scanner_namespace_mutation_generation(),
            generation_before_put.saturating_add(3),
            "successful object creation must observe the logical mutation and both fanout boundaries"
        );
        ecstore
            .get_object_info(bucket, object, &ObjectOptions::default())
            .await
            .expect("object should be readable before bucket delete");
    }

    async fn any_disk_path_exists(disk_paths: &[PathBuf], relative_path: impl AsRef<Path>) -> bool {
        for disk_path in disk_paths {
            if tokio::fs::try_exists(disk_path.join(relative_path.as_ref()))
                .await
                .expect("test disk path should be stat-able")
            {
                return true;
            }
        }
        false
    }

    async fn any_disk_has_object_metadata(disk_paths: &[PathBuf], bucket: &str) -> bool {
        for disk_path in disk_paths {
            if super::has_xlmeta_files(&disk_path.join(bucket))
                .await
                .expect("object metadata scan should succeed")
            {
                return true;
            }
        }
        false
    }

    async fn write_metadata_less_part_on_all_disks(disk_paths: &[PathBuf], bucket: &str, object: &str) {
        for disk_path in disk_paths {
            let data_dir = Uuid::new_v4();
            let part_path = disk_path.join(bucket).join(object).join(data_dir.to_string()).join("part.1");
            tokio::fs::create_dir_all(part_path.parent().expect("part path should have a parent"))
                .await
                .expect("metadata-less data dir should be created");
            tokio::fs::write(part_path, b"orphan shard")
                .await
                .expect("metadata-less part should be written");
        }
    }

    async fn write_bucket_metadata_marker(disk_paths: &[PathBuf], metadata_prefix: &str) {
        for disk_path in disk_paths {
            let marker_path = disk_path.join(metadata_prefix).join("config.json");
            tokio::fs::create_dir_all(marker_path.parent().expect("metadata marker path should have a parent"))
                .await
                .expect("metadata marker parent should be created");
            tokio::fs::write(marker_path, b"bucket metadata")
                .await
                .expect("metadata marker should be written");
        }
    }

    #[test]
    fn should_not_override_when_metadata_created_is_unix_epoch() {
        assert!(!should_override_created_from_metadata(OffsetDateTime::UNIX_EPOCH));
    }

    #[test]
    fn should_override_when_metadata_created_is_valid_time() {
        let created = OffsetDateTime::from_unix_timestamp(1704067200).expect("valid timestamp");
        assert!(should_override_created_from_metadata(created));
    }

    #[test]
    fn table_bucket_delete_guard_rejects_remaining_catalog_metadata() {
        let err = validate_table_bucket_delete_allowed("table-bucket", true, true).unwrap_err();

        assert!(matches!(err, StorageError::BucketNotEmpty(bucket) if bucket == "table-bucket"));
        assert!(validate_table_bucket_delete_allowed("table-bucket", true, false).is_ok());
        assert!(validate_table_bucket_delete_allowed("regular-bucket", false, true).is_ok());
    }

    #[test]
    fn bucket_delete_metadata_cleanup_removes_internal_table_catalog_prefix() {
        let prefixes = bucket_delete_metadata_cleanup_prefixes("analytics");

        assert!(prefixes.contains(&table_bucket_catalog_metadata_prefix("analytics")));
        assert!(prefixes.contains(&"buckets/analytics".to_string()));
    }

    #[test]
    fn bucket_delete_marker_path_uses_internal_deleted_bucket_metadata_prefix() {
        assert_eq!(bucket_deleted_marker_prefix("analytics"), "buckets/.deleted/analytics");
        assert_eq!(
            bucket_deleted_marker_volume("analytics"),
            format!("{RUSTFS_META_BUCKET}/{BUCKET_META_PREFIX}/.deleted/analytics")
        );
    }

    #[test]
    fn bucket_listing_selects_request_and_scanner_fanout() {
        assert_eq!(bucket_list_set_concurrency(0, SCANNER_BUCKET_LIST_SET_CONCURRENCY), 1);
        assert_eq!(bucket_list_set_concurrency(2, SCANNER_BUCKET_LIST_SET_CONCURRENCY), 2);
        assert_eq!(
            bucket_list_set_concurrency(100, SCANNER_BUCKET_LIST_SET_CONCURRENCY),
            SCANNER_BUCKET_LIST_SET_CONCURRENCY
        );
        assert_eq!(bucket_list_set_concurrency(100, usize::MAX), 100);
    }

    #[tokio::test]
    async fn metadata_less_residue_scan_ignores_empty_dirs_and_reports_uuid_data() {
        let root = tempfile::tempdir().expect("temporary bucket root should be created");
        let bucket_path = root.path().join("bucket");
        tokio::fs::create_dir_all(bucket_path.join("empty/child"))
            .await
            .expect("empty directory residue should be created");

        let empty = scan_metadata_less_residue(&bucket_path)
            .await
            .expect("empty residue scan should succeed");
        assert!(!empty.has_residue_without_xlmeta());

        let data_dir = Uuid::new_v4();
        let part_path = bucket_path.join("object").join(data_dir.to_string()).join("part.1");
        tokio::fs::create_dir_all(part_path.parent().expect("part path should have a parent"))
            .await
            .expect("data dir should be created");
        tokio::fs::write(&part_path, b"orphan shard")
            .await
            .expect("part file should be written");

        let residue = scan_metadata_less_residue(&bucket_path)
            .await
            .expect("metadata-less residue scan should succeed");
        assert!(residue.has_residue_without_xlmeta());
        assert_eq!(residue.files, 1);
        assert_eq!(residue.uuid_data_dirs, 1);
        let sample = residue.sample.as_deref().expect("part sample should be recorded");
        assert!(sample.starts_with("object/"));
        assert!(sample.ends_with("/part.1"));
    }

    #[tokio::test]
    async fn metadata_less_residue_scan_shares_one_entry_budget_across_roots() {
        let root = tempfile::tempdir().expect("temporary diagnostic roots should be created");
        let first_root = root.path().join("disk-a");
        let second_root = root.path().join("disk-b");
        tokio::fs::create_dir_all(&first_root)
            .await
            .expect("first diagnostic root should be created");
        tokio::fs::create_dir_all(&second_root)
            .await
            .expect("second diagnostic root should be created");
        for index in 0..3 {
            std::fs::write(first_root.join(format!("first-{index}")), b"").expect("first-root fixture should be written");
            std::fs::write(second_root.join(format!("second-{index}")), b"").expect("second-root fixture should be written");
        }

        let mut budget = BucketDeleteDiagnosticBudget::with_limits(4, Duration::from_secs(5));
        let first = scan_metadata_less_residue_with_budget(&first_root, &mut budget)
            .await
            .expect("first root should fit the shared budget");
        assert!(!first.diagnostic_truncated);
        let second = scan_metadata_less_residue_with_budget(&second_root, &mut budget)
            .await
            .expect("second root should stop at the remaining shared budget");
        assert!(second.diagnostic_truncated);
        assert!(first.entries_scanned + second.entries_scanned <= 4);
    }

    #[tokio::test]
    async fn metadata_less_residue_scan_honors_an_expired_request_deadline() {
        let root = tempfile::tempdir().expect("temporary diagnostic root should be created");
        std::fs::write(root.path().join("orphan"), b"").expect("deadline fixture should be written");
        let mut budget = BucketDeleteDiagnosticBudget::with_limits(8, Duration::ZERO);

        let scan = scan_metadata_less_residue_with_budget(root.path(), &mut budget)
            .await
            .expect("an expired diagnostic budget should fail closed without an IO error");

        assert!(scan.diagnostic_truncated);
        assert_eq!(scan.entries_scanned, 0);
        assert_eq!(scan.diagnostic_bytes_read, 0);
    }

    #[tokio::test]
    async fn metadata_less_residue_scan_stops_at_diagnostic_budget() {
        let root = tempfile::tempdir().expect("temporary bucket root should be created");
        let bucket_path = root.path().join("bucket");
        tokio::fs::create_dir_all(&bucket_path)
            .await
            .expect("budget fixture directory should be created");
        for index in 0..(BUCKET_DELETE_DIAGNOSTIC_MAX_ENTRIES + 32) {
            std::fs::write(bucket_path.join(format!("orphan-{index:05}")), b"").expect("budget fixture file should be created");
        }

        let residue = scan_metadata_less_residue(&bucket_path)
            .await
            .expect("budgeted residue scan should fail closed without an IO error");
        assert!(residue.diagnostic_truncated);
        assert!(residue.has_residue_without_xlmeta());
        assert!(!residue.xlmeta_found);
        assert!(residue.entries_scanned <= BUCKET_DELETE_DIAGNOSTIC_MAX_ENTRIES);
        assert!(residue.files <= BUCKET_DELETE_DIAGNOSTIC_MAX_ENTRIES);
        assert_eq!(residue.diagnostic_bytes_read, 0);
    }

    #[tokio::test(start_paused = true)]
    async fn bucket_residue_scan_distinguishes_visible_and_tier_free_xlmeta() {
        // Classification keeps the production budget, but real filesystem
        // scheduling must not advance the clock before classification finishes.
        let (clock_guard_tx, clock_guard_rx) = std::sync::mpsc::channel::<()>();
        let clock_guard = tokio::task::spawn_blocking(move || clock_guard_rx.recv());
        let root = tempfile::tempdir().expect("temporary bucket root should be created");
        let bucket_path = root.path().join("bucket");
        let visible_path = bucket_path.join("visible").join(STORAGE_FORMAT_FILE);
        tokio::fs::create_dir_all(visible_path.parent().expect("visible xl.meta should have a parent"))
            .await
            .expect("visible object directory should be created");
        let mut visible = FileMeta::new();
        visible
            .add_version(FileInfo {
                version_id: Some(Uuid::new_v4()),
                mod_time: Some(OffsetDateTime::now_utc()),
                ..Default::default()
            })
            .expect("visible version should encode");
        tokio::fs::write(&visible_path, visible.marshal_msg().expect("visible xl.meta should marshal"))
            .await
            .expect("visible xl.meta should be written");

        let visible_scan = scan_metadata_less_residue(&bucket_path)
            .await
            .expect("visible xl.meta scan should succeed");
        assert!(!visible_scan.diagnostic_truncated, "{visible_scan:?}");
        assert_eq!(visible_scan.xlmeta_blocker, Some(BucketDeleteBlockerKind::VisibleVersion));

        tokio::fs::remove_dir_all(bucket_path.join("visible"))
            .await
            .expect("visible fixture should be removed");
        let free_path = bucket_path.join("free").join(STORAGE_FORMAT_FILE);
        tokio::fs::create_dir_all(free_path.parent().expect("free xl.meta should have a parent"))
            .await
            .expect("free-version object directory should be created");
        let source_version_id = Uuid::new_v4();
        let mut free = FileMeta::new();
        free.add_version(FileInfo {
            version_id: Some(source_version_id),
            transition_status: TRANSITION_COMPLETE.to_string(),
            transitioned_objname: "remote/object".to_string(),
            transition_version_id: Some(Uuid::new_v4()),
            transition_tier: "WARM".to_string(),
            mod_time: Some(OffsetDateTime::now_utc()),
            ..Default::default()
        })
        .expect("transitioned source should encode");
        let mut delete = FileInfo {
            version_id: Some(source_version_id),
            mod_time: Some(OffsetDateTime::now_utc()),
            ..Default::default()
        };
        delete.set_tier_free_version_id(&Uuid::new_v4().to_string());
        free.delete_version(&delete)
            .expect("transitioned source delete should create a free-version");
        tokio::fs::write(&free_path, free.marshal_msg().expect("free-version xl.meta should marshal"))
            .await
            .expect("free-version xl.meta should be written");

        let free_scan = scan_metadata_less_residue(&bucket_path)
            .await
            .expect("free-version xl.meta scan should succeed");
        assert!(!free_scan.diagnostic_truncated, "{free_scan:?}");
        assert_eq!(free_scan.xlmeta_blocker, Some(BucketDeleteBlockerKind::TierFreeVersion));

        tokio::fs::remove_dir_all(bucket_path.join("free"))
            .await
            .expect("free-version fixture should be removed");

        let exact_limit_path = bucket_path.join("exact-limit").join(STORAGE_FORMAT_FILE);
        tokio::fs::create_dir_all(exact_limit_path.parent().expect("exact-limit xl.meta should have a parent"))
            .await
            .expect("exact-limit object directory should be created");
        let exact_limit = tokio::fs::File::create(&exact_limit_path)
            .await
            .expect("exact-limit xl.meta should be created");
        exact_limit
            .set_len(BUCKET_DELETE_XLMETA_DIAGNOSTIC_MAX_BYTES)
            .await
            .expect("exact-limit xl.meta should be extended without allocating its contents");
        let exact_limit_scan = scan_metadata_less_residue(&bucket_path)
            .await
            .expect("exact-limit xl.meta scan should remain fail closed");
        assert!(!exact_limit_scan.diagnostic_truncated, "{exact_limit_scan:?}");
        assert_eq!(exact_limit_scan.xlmeta_blocker, Some(BucketDeleteBlockerKind::UnknownXlMeta));
        assert_eq!(exact_limit_scan.diagnostic_bytes_read, BUCKET_DELETE_XLMETA_DIAGNOSTIC_MAX_BYTES);
        tokio::fs::remove_dir_all(bucket_path.join("exact-limit"))
            .await
            .expect("exact-limit fixture should be removed");

        let oversized_path = bucket_path.join("oversized").join(STORAGE_FORMAT_FILE);
        tokio::fs::create_dir_all(oversized_path.parent().expect("oversized xl.meta should have a parent"))
            .await
            .expect("oversized object directory should be created");
        let oversized = tokio::fs::File::create(&oversized_path)
            .await
            .expect("oversized xl.meta should be created");
        oversized
            .set_len(BUCKET_DELETE_XLMETA_DIAGNOSTIC_MAX_BYTES + 1)
            .await
            .expect("oversized xl.meta should be extended without allocating its contents");
        let oversized_scan = scan_metadata_less_residue(&bucket_path)
            .await
            .expect("oversized xl.meta scan should remain fail closed");
        assert!(!oversized_scan.diagnostic_truncated, "{oversized_scan:?}");
        assert_eq!(oversized_scan.xlmeta_blocker, Some(BucketDeleteBlockerKind::UnknownXlMeta));
        assert_eq!(oversized_scan.diagnostic_bytes_read, 0);
        assert!(oversized_scan.diagnostic_bytes_read <= BUCKET_DELETE_XLMETA_DIAGNOSTIC_MAX_BYTES);

        drop(clock_guard_tx);
        let _ = clock_guard
            .await
            .expect("classification clock guard should exit after its sender is dropped");

        // An exhausted diagnostic budget may return before observing xl.meta.
        // It must report truncation rather than inventing a classification.
        let first_io_started = Arc::new(AtomicBool::new(false));
        let mut budget = BucketDeleteDiagnosticBudget::new().with_first_io_delay(
            BUCKET_DELETE_DIAGNOSTIC_MAX_ELAPSED + Duration::from_millis(100),
            first_io_started.clone(),
        );
        let truncated = scan_metadata_less_residue_with_budget(&bucket_path, &mut budget)
            .await
            .expect("a diagnostic timeout should return a partial scan");
        assert!(first_io_started.load(Ordering::SeqCst));
        assert!(truncated.diagnostic_truncated, "{truncated:?}");
        assert_eq!(truncated.xlmeta_blocker, None);
        assert_eq!(truncated.entries_scanned, 0);
        assert_eq!(truncated.diagnostic_bytes_read, 0);
    }

    #[tokio::test]
    #[serial]
    async fn bucket_namespace_reads_keep_pre_expansion_bucket_visible() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        let bucket = format!("pre-expansion-{}", Uuid::new_v4().simple());
        let object = "existing-object";
        ecstore.pools[0].disk_set[0]
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created in the original pool only");
        let mut reader = PutObjReader::from_vec(b"object written before pool expansion".to_vec());
        ecstore.pools[0]
            .put_object(&bucket, object, &mut reader, &ObjectOptions::default())
            .await
            .expect("object should be written in the original pool only");

        let buckets = ecstore
            .list_bucket(&BucketOptions::default())
            .await
            .expect("S3 bucket listing should remain available after adding an empty pool");
        assert!(buckets.iter().any(|entry| entry.name == bucket));

        let info = ecstore
            .get_bucket_info(&bucket, &BucketOptions::default())
            .await
            .expect("bucket validation should accept a bucket present in the original pool");
        assert_eq!(info.name, bucket);

        metadata_sys::init_bucket_metadata_sys(ecstore.clone(), vec![bucket.clone()]).await;
        let listed = ecstore
            .clone()
            .list_objects_v2(&bucket, "", None, None, 1000, false, None, false)
            .await
            .expect("ListObjectsV2 should remain available after adding an empty pool");
        assert!(listed.objects.iter().any(|entry| entry.name == object));
        ecstore.pools[1].disk_set[0]
            .get_bucket_info(&bucket, &BucketOptions::default())
            .await
            .expect("metadata initialization should heal the bucket volume into the expansion pool");

        let object_info = ecstore
            .get_object_info(&bucket, object, &ObjectOptions::default())
            .await
            .expect("an object in the original pool should remain readable after expansion");
        assert_eq!(object_info.name, object);
    }

    #[tokio::test]
    #[serial]
    async fn bucket_creation_accepts_exact_quorum_in_every_set() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        let bucket = format!("exact-set-quorum-{}", Uuid::new_v4().simple());
        let original_set = Arc::clone(&ecstore.pools[0].disk_set[0]);
        let offline = take_set_disks_offline(&ecstore, &original_set, &[0]).await;

        ecstore
            .make_bucket_on_sets(&bucket, &MakeBucketOptions::default())
            .await
            .expect("three of four disks should satisfy the original set quorum");

        restore_set_disks(&ecstore, &original_set, offline).await;
        for pool in &ecstore.pools {
            pool.disk_set[0]
                .get_bucket_info(&bucket, &BucketOptions::default())
                .await
                .expect("every set should expose a bucket created at its exact quorum");
        }
    }

    #[tokio::test]
    #[serial]
    async fn bucket_creation_rejects_cross_pool_quorum_subsidy() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        metadata_sys::init_bucket_metadata_sys(ecstore.clone(), Vec::new()).await;
        let bucket = format!("create-set-quorum-minus-one-{}", Uuid::new_v4().simple());
        let original_set = Arc::clone(&ecstore.pools[0].disk_set[0]);
        let offline = take_set_disks_offline(&ecstore, &original_set, &[0, 1]).await;

        let err = ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect_err("a healthy expansion pool must not subsidize a failed original set");
        restore_set_disks(&ecstore, &original_set, offline).await;
        assert!(
            matches!(&err, StorageError::InsufficientWriteQuorum(name, object) if name == &bucket && object.is_empty()),
            "unexpected error: {err}"
        );
        for pool in &ecstore.pools {
            assert_eq!(
                pool.disk_set[0]
                    .get_bucket_info(&bucket, &BucketOptions::default())
                    .await
                    .expect_err("failed creation must not leave an authoritative bucket"),
                StorageError::VolumeNotFound
            );
        }
    }

    #[tokio::test]
    #[serial]
    async fn bucket_creation_prioritizes_hard_set_error_over_existing_bucket() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        metadata_sys::init_bucket_metadata_sys(ecstore.clone(), Vec::new()).await;
        let bucket = format!("create-hard-error-{}", Uuid::new_v4().simple());
        ecstore.pools[0].disk_set[0]
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("the original pool should contain the bucket");
        let expansion_set = Arc::clone(&ecstore.pools[1].disk_set[0]);
        let offline = take_set_disks_offline(&ecstore, &expansion_set, &[0, 1]).await;

        let err = ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect_err("a hard set failure must outrank BucketExists from another set");
        restore_set_disks(&ecstore, &expansion_set, offline).await;
        assert!(
            matches!(&err, StorageError::InsufficientWriteQuorum(name, object) if name == &bucket && object.is_empty()),
            "unexpected error: {err}"
        );
        ecstore.pools[0].disk_set[0]
            .get_bucket_info(&bucket, &BucketOptions::default())
            .await
            .expect("failure in another set must not roll back a pre-existing bucket");
    }

    #[tokio::test]
    #[serial]
    async fn bucket_delete_rejects_cross_pool_quorum_subsidy_before_mutation() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        metadata_sys::init_bucket_metadata_sys(ecstore.clone(), Vec::new()).await;
        let bucket = format!("delete-set-quorum-minus-one-{}", Uuid::new_v4().simple());
        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created in every set before deletion");
        let original_set = Arc::clone(&ecstore.pools[0].disk_set[0]);
        let offline = take_set_disks_offline(&ecstore, &original_set, &[0, 1, 2]).await;

        let err = ecstore
            .delete_bucket(
                &bucket,
                &DeleteBucketOptions {
                    force: true,
                    ..Default::default()
                },
            )
            .await
            .expect_err("a healthy expansion pool must not subsidize a failed original set deletion");
        restore_set_disks(&ecstore, &original_set, offline).await;
        assert!(
            matches!(&err, StorageError::InsufficientWriteQuorum(name, object) if name == &bucket && object.is_empty()),
            "unexpected error: {err}"
        );
        for pool in &ecstore.pools {
            pool.disk_set[0]
                .get_bucket_info(&bucket, &BucketOptions::default())
                .await
                .expect("failed deletion preflight must preserve every bucket volume");
        }
        let (_, persisted) = metadata_sys::get_config_from_disk_with_presence_in(&ecstore.ctx, &bucket)
            .await
            .expect("failed deletion must preserve bucket metadata");
        assert!(persisted);
    }

    #[tokio::test]
    #[serial]
    async fn bucket_delete_rolls_back_sets_after_partial_failure() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        let bucket = format!("delete-set-rollback-{}", Uuid::new_v4().simple());
        ecstore
            .make_bucket_on_sets(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created in every set before deletion");
        let original_set = Arc::clone(&ecstore.pools[0].disk_set[0]);
        let offline = take_set_disks_offline(&ecstore, &original_set, &[0, 1]).await;

        let err = ecstore
            .delete_bucket_on_sets(&bucket, &DeleteBucketOptions::default())
            .await
            .expect_err("a failed set must fail the complete bucket deletion");
        restore_set_disks(&ecstore, &original_set, offline).await;
        assert_eq!(err, StorageError::ErasureWriteQuorum);
        for pool in &ecstore.pools {
            pool.disk_set[0]
                .get_bucket_info(&bucket, &BucketOptions::default())
                .await
                .expect("delete rollback should restore the bucket namespace in every set");
        }
    }

    #[tokio::test]
    #[serial]
    async fn bucket_config_update_accepts_pre_expansion_bucket() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        metadata_sys::init_bucket_metadata_sys(ecstore.clone(), Vec::new()).await;
        let bucket = format!("config-pre-expansion-{}", Uuid::new_v4().simple());
        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created with authoritative metadata");
        ecstore.pools[1].disk_set[0]
            .delete_bucket(&bucket, &DeleteBucketOptions::default())
            .await
            .expect("the expansion pool bucket volume should be removed for the regression state");

        let policy = br#"{"Version":"2012-10-17","Statement":[]}"#.to_vec();
        ecstore
            .update_bucket_metadata_config(&bucket, crate::bucket::metadata::BUCKET_POLICY_CONFIG, policy.clone())
            .await
            .expect("config mutation should validate existence per erasure set");
        let (stored, _) = ecstore
            .get_bucket_policy_raw(&bucket)
            .await
            .expect("updated bucket policy should remain readable");
        assert_eq!(stored.as_bytes(), policy);
    }

    #[tokio::test]
    #[serial]
    async fn bucket_metadata_init_recreates_pre_expansion_bucket_in_new_pool() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        let bucket = format!("metadata-expansion-{}", Uuid::new_v4().simple());
        ecstore.pools[0].disk_set[0]
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created in the original pool only");
        assert_eq!(
            ecstore.pools[1].disk_set[0]
                .get_bucket_info(&bucket, &BucketOptions::default())
                .await
                .expect_err("new pool should initially have no bucket volume"),
            StorageError::VolumeNotFound
        );

        metadata_sys::init_bucket_metadata_sys(ecstore.clone(), vec![bucket.clone()]).await;

        ecstore.pools[1].disk_set[0]
            .get_bucket_info(&bucket, &BucketOptions::default())
            .await
            .expect("metadata initialization should recreate the bucket volume in the new pool");
    }

    #[tokio::test]
    #[serial]
    async fn bucket_metadata_init_repairs_half_created_expansion_pool() {
        // Check both pool orders: an incomplete set must not hide a later
        // complete set, and a complete set must not weaken request validation.
        for complete_pool in 0..2 {
            let (temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
            let bucket = format!("partial-expansion-{}", Uuid::new_v4().simple());
            for pool_index in 0..2 {
                let present_disks = if pool_index == complete_pool { 4 } else { 2 };
                for disk_index in 0..present_disks {
                    tokio::fs::create_dir(
                        temp_dir
                            .path()
                            .join(format!("pool{pool_index}-disk{disk_index}"))
                            .join(&bucket),
                    )
                    .await
                    .expect("fixture bucket volume should be created");
                }
            }
            assert_eq!(
                ecstore
                    .get_bucket_info_from_sets(&bucket, &BucketOptions::default())
                    .await
                    .expect_err("request validation must reject a half-created expansion set"),
                StorageError::ErasureWriteQuorum
            );

            metadata_sys::init_bucket_metadata_sys(ecstore.clone(), vec![bucket.clone()]).await;

            for pool_index in 0..2 {
                for disk_index in 0..4 {
                    assert!(
                        temp_dir
                            .path()
                            .join(format!("pool{pool_index}-disk{disk_index}"))
                            .join(&bucket)
                            .is_dir(),
                        "metadata initialization must heal every missing expansion volume"
                    );
                }
            }
            ecstore
                .get_bucket_info_from_sets(&bucket, &BucketOptions::default())
                .await
                .expect("strict request validation should succeed after volume repair");
        }
    }

    #[tokio::test]
    #[serial]
    async fn bucket_metadata_init_does_not_combine_partial_set_evidence() {
        let (temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        let bucket = format!("no-quorum-expansion-{}", Uuid::new_v4().simple());
        for pool_index in 0..2 {
            for disk_index in 0..2 {
                tokio::fs::create_dir(
                    temp_dir
                        .path()
                        .join(format!("pool{pool_index}-disk{disk_index}"))
                        .join(&bucket),
                )
                .await
                .expect("fixture bucket volume should be created");
            }
        }
        assert_eq!(
            ecstore
                .bucket_exists_for_heal(&bucket)
                .await
                .expect_err("repair must require a complete quorum within one set"),
            StorageError::ErasureWriteQuorum
        );

        metadata_sys::init_bucket_metadata_sys(ecstore.clone(), vec![bucket.clone()]).await;

        for pool_index in 0..2 {
            for disk_index in 0..4 {
                assert_eq!(
                    temp_dir
                        .path()
                        .join(format!("pool{pool_index}-disk{disk_index}"))
                        .join(&bucket)
                        .is_dir(),
                    disk_index < 2,
                    "unproven bucket generations must not recreate missing volumes"
                );
            }
        }
    }

    #[tokio::test]
    #[serial]
    async fn bucket_metadata_init_does_not_recreate_stale_bucket_name() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        let bucket = format!("stale-expansion-{}", Uuid::new_v4().simple());

        metadata_sys::init_bucket_metadata_sys(ecstore.clone(), vec![bucket.clone()]).await;

        for pool in &ecstore.pools {
            let err = pool.disk_set[0]
                .get_bucket_info(&bucket, &BucketOptions::default())
                .await
                .expect_err("a stale startup listing must not recreate a deleted bucket");
            assert_eq!(err, StorageError::VolumeNotFound);
        }
    }

    #[tokio::test]
    #[serial]
    async fn bucket_info_read_quorum_tracks_erasure_layout() {
        for (drive_count, parity) in [(2, 1), (3, 1), (4, 2), (5, 2), (6, 3), (8, 4), (6, 2), (12, 6)] {
            let (_temp_dir, store) = setup_bucket_quorum_test_env(&[drive_count], Some(parity)).await;
            metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
            let bucket = format!("read-quorum-{drive_count}-{parity}");
            let object = "uncached-object";
            let body = b"erasure read quorum must follow the persisted layout".repeat(32_768);
            store
                .make_bucket(&bucket, &MakeBucketOptions::default())
                .await
                .expect("healthy namespace should accept bucket creation");
            store
                .put_object(&bucket, object, &mut PutObjReader::from_vec(body.clone()), &ObjectOptions::default())
                .await
                .expect("healthy erasure set should accept the seed object");
            let set = &store.pools[0].disk_set[0];
            let lock = set
                .new_ns_lock(&bucket, object)
                .await
                .expect("seed namespace lock should resolve");
            drop(
                lock.get_write_lock(Duration::from_secs(30))
                    .await
                    .expect("seed physical fanout must finish before taking disks offline"),
            );
            if (drive_count, parity) == (6, 3) {
                let mut kvs = rustfs_config::server_config::KVS::new();
                kvs.insert(crate::config::storageclass::CLASS_STANDARD.to_string(), "EC:2".to_string());
                set.set_test_storage_class_config(
                    crate::config::storageclass::lookup_config_for_pools_without_env(&kvs, &[drive_count])
                        .expect("a later storage-class change must not raise old objects' read quorum"),
                );
            }

            let offline_indexes = (0..parity).collect::<Vec<_>>();
            let offline = take_set_disks_offline(&store, set, &offline_indexes).await;
            let info = store
                .get_bucket_info(&bucket, &BucketOptions::default())
                .await
                .expect("bucket validation must admit the object's exact read quorum");
            assert_eq!(info.name, bucket);

            let mut reader = store
                .get_object_reader(&bucket, object, None, Default::default(), &ObjectOptions::default())
                .await
                .expect("the persisted layout should remain readable at its exact data-shard quorum");
            let mut restored = Vec::new();
            reader
                .stream
                .read_to_end(&mut restored)
                .await
                .expect("quorum read should reconstruct the body");
            assert_eq!(restored, body, "layout {drive_count}/{parity} must retain exact object contents");
            drop(reader);

            if drive_count - parity == drive_count / 2 {
                let error = store
                    .get_bucket_info_from_sets(&bucket, &BucketOptions::default())
                    .await
                    .expect_err("bucket mutations must retain their majority namespace check");
                assert_eq!(error, StorageError::ErasureWriteQuorum);
            }

            let below_quorum = take_set_disks_offline(&store, set, &[parity]).await;
            let read = store
                .get_object_reader(&bucket, object, None, Default::default(), &ObjectOptions::default())
                .await;
            match read {
                Ok(mut reader) => assert!(
                    reader.stream.read_to_end(&mut Vec::new()).await.is_err(),
                    "layout {drive_count}/{parity} must reject fewer than its data-shard quorum"
                ),
                Err(error) => assert!(
                    matches!(error, StorageError::ErasureReadQuorum | StorageError::InsufficientReadQuorum(_, _)),
                    "a missing shard must report read quorum loss, got {error}"
                ),
            }
            restore_set_disks(&store, set, below_quorum).await;
            restore_set_disks(&store, set, offline).await;
        }
    }

    #[tokio::test]
    #[serial]
    async fn lazy_metadata_load_uses_read_quorum_for_non_lock_bucket() {
        let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4], Some(2)).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = format!("lazy-read-quorum-{}", Uuid::new_v4().simple());
        store
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("healthy namespace should accept bucket creation");

        // Replace the instance metadata system with a cold cache while keeping the
        // persisted metadata and physical bucket in place.
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let set = &store.pools[0].disk_set[0];
        let offline = take_set_disks_offline(&store, set, &[0, 1]).await;

        metadata_sys::get_on_demand_migration_config_in(&store, &bucket)
            .await
            .expect("cold metadata load should admit the exact namespace read quorum");
        assert!(
            metadata_sys::get_in(&store.ctx, &bucket).await.is_ok(),
            "the lazy load should publish authoritative metadata"
        );
        assert_eq!(
            store
                .get_bucket_info_from_sets(&bucket, &BucketOptions::default())
                .await
                .expect_err("bucket mutations must retain their majority namespace check"),
            StorageError::ErasureWriteQuorum
        );

        restore_set_disks(&store, set, offline).await;
    }

    /// EC 2+2 with two disks offline still has object read quorum and has lost
    /// write quorum. A node that has not cached the bucket must still serve
    /// HEAD bucket, GET, HEAD object, and List; a write must fail.
    #[tokio::test]
    #[serial]
    async fn cold_bucket_reads_succeed_below_write_quorum() {
        let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4], Some(2)).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = format!("cold-read-quorum-{}", Uuid::new_v4().simple());
        let object = "seed-object";
        let body = b"cold bucket reads must survive lost write quorum".to_vec();
        store
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("healthy namespace should accept bucket creation");
        store
            .put_object(&bucket, object, &mut PutObjReader::from_vec(body.clone()), &ObjectOptions::default())
            .await
            .expect("healthy erasure set should accept the seed object");

        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        assert!(
            metadata_sys::get_in(&store.ctx, &bucket).await.is_err(),
            "the replacement metadata system must start without this bucket"
        );
        let set = &store.pools[0].disk_set[0];
        let lock = set
            .new_ns_lock(&bucket, object)
            .await
            .expect("seed namespace lock should resolve");
        drop(
            lock.get_write_lock(Duration::from_secs(30))
                .await
                .expect("seed physical fanout must finish before taking disks offline"),
        );
        let offline = take_set_disks_offline(&store, set, &[0, 1]).await;

        let info = store
            .get_bucket_info(&bucket, &BucketOptions::default())
            .await
            .expect("HEAD bucket must succeed at read quorum before the bucket is cached");
        assert_eq!(info.name, bucket);

        let mut reader = store
            .get_object_reader(&bucket, object, None, Default::default(), &ObjectOptions::default())
            .await
            .expect("GET must succeed for a bucket that was not loaded at startup");
        let mut restored = Vec::new();
        reader
            .stream
            .read_to_end(&mut restored)
            .await
            .expect("quorum read should reconstruct the body");
        assert_eq!(restored, body);
        drop(reader);

        let object_info = store
            .get_object_info(&bucket, object, &ObjectOptions::default())
            .await
            .expect("HEAD object must succeed at read quorum");
        assert_eq!(object_info.name, object);
        assert_eq!(object_info.size, i64::try_from(body.len()).expect("body length fits i64"));

        let listed = store
            .clone()
            .list_objects_v2(&bucket, "", None, None, 100, false, None, false)
            .await
            .expect("List must succeed at read quorum");
        assert!(
            listed.objects.iter().any(|item| item.name == object),
            "list should include the seed object, got {:?}",
            listed.objects.iter().map(|item| item.name.clone()).collect::<Vec<_>>()
        );
        assert!(
            metadata_sys::get_in(&store.ctx, &bucket).await.is_ok(),
            "the cold load should publish authoritative metadata"
        );

        let write_error = store
            .put_object(&bucket, "rejected-object", &mut PutObjReader::from_vec(body), &ObjectOptions::default())
            .await
            .expect_err("writes must still fail without write quorum");
        assert!(
            matches!(
                write_error,
                StorageError::InsufficientWriteQuorum(_, _) | StorageError::ErasureWriteQuorum
            ),
            "writes must fail with a write-quorum error, got {write_error}"
        );

        restore_set_disks(&store, set, offline).await;
    }

    #[tokio::test]
    #[serial]
    async fn lazy_metadata_load_rejects_below_read_quorum() {
        let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4], Some(2)).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = format!("lazy-below-quorum-{}", Uuid::new_v4().simple());
        store
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("healthy namespace should accept bucket creation");

        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let set = &store.pools[0].disk_set[0];
        let offline = take_set_disks_offline(&store, set, &[0, 1, 2]).await;

        let error = metadata_sys::get_on_demand_migration_config_in(&store, &bucket)
            .await
            .expect_err("cold metadata load must reject read quorum minus one");
        assert!(
            matches!(error, StorageError::ErasureReadQuorum | StorageError::InsufficientReadQuorum(_, _)),
            "below read quorum must fail with a read-quorum error, got {error}"
        );
        assert!(
            metadata_sys::get_in(&store.ctx, &bucket).await.is_err(),
            "a below-quorum load must not publish metadata"
        );

        restore_set_disks(&store, set, offline).await;
    }

    #[tokio::test]
    #[serial]
    async fn lazy_metadata_load_keeps_lock_creation_intent_fail_closed() {
        let (temp_dir, store) = setup_bucket_quorum_test_env(&[4], Some(2)).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = format!("lazy-lock-intent-{}", Uuid::new_v4().simple());
        let mut intent = BucketMetadata::new(&bucket);
        intent.lock_enabled = true;
        metadata_sys::set_new_bucket_metadata_intent_in(&store.ctx, intent)
            .await
            .expect("Object Lock creation intent should persist metadata and its incarnation sidecar");

        let set = &store.pools[0].disk_set[0];
        let offline = take_set_disks_offline(&store, set, &[0, 1]).await;
        assert_eq!(
            store
                .make_bucket_on_sets(&bucket, &MakeBucketOptions::default())
                .await
                .expect_err("physical creation must not commit below write quorum"),
            StorageError::ErasureWriteQuorum
        );
        assert!(!temp_dir.path().join("pool0-disk0").join(&bucket).exists());
        assert!(!temp_dir.path().join("pool0-disk1").join(&bucket).exists());
        assert!(temp_dir.path().join("pool0-disk2").join(&bucket).is_dir());
        assert!(temp_dir.path().join("pool0-disk3").join(&bucket).is_dir());

        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let error = metadata_sys::get_on_demand_migration_config_in(&store, &bucket)
            .await
            .expect_err("Object Lock intent must not become visible at read quorum");
        assert_eq!(error, StorageError::ErasureWriteQuorum);
        assert!(
            metadata_sys::get_in(&store.ctx, &bucket).await.is_err(),
            "the rejected Object Lock intent must not enter the metadata cache"
        );
        assert_eq!(
            store
                .get_bucket_info(&bucket, &BucketOptions::default())
                .await
                .expect_err("bucket-info read fallback must also reject a pending Object Lock intent"),
            StorageError::ErasureWriteQuorum
        );

        restore_set_disks(&store, set, offline).await;
    }

    #[tokio::test]
    #[serial]
    async fn lazy_metadata_load_admits_committed_lock_bucket_at_read_quorum() {
        let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4], Some(2)).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = format!("lazy-committed-lock-{}", Uuid::new_v4().simple());
        store
            .make_bucket(
                &bucket,
                &MakeBucketOptions {
                    lock_enabled: true,
                    ..Default::default()
                },
            )
            .await
            .expect("healthy namespace should accept Object Lock bucket creation");
        let (metadata, persisted) = metadata_sys::get_config_from_disk_with_presence_in(&store.ctx, &bucket)
            .await
            .expect("read committed Object Lock metadata");
        assert!(persisted && metadata.bucket_creation_committed);

        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let set = &store.pools[0].disk_set[0];
        let offline = take_set_disks_offline(&store, set, &[0, 1]).await;

        metadata_sys::get_on_demand_migration_config_in(&store, &bucket)
            .await
            .expect("committed Object Lock metadata should load at read quorum");
        assert!(
            metadata_sys::get_in(&store.ctx, &bucket).await.is_ok(),
            "committed Object Lock metadata should be published"
        );

        restore_set_disks(&store, set, offline).await;
    }

    #[tokio::test]
    #[serial]
    async fn creation_commit_backfill_runs_under_the_config_write_fence() {
        let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4], Some(2)).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = format!("lazy-commit-backfill-{}", Uuid::new_v4().simple());
        let mut intent = BucketMetadata::new(&bucket);
        intent.lock_enabled = true;
        metadata_sys::set_new_bucket_metadata_intent_in(&store.ctx, intent)
            .await
            .expect("persist pending Object Lock intent");
        store
            .make_bucket_on_sets(&bucket, &MakeBucketOptions::default())
            .await
            .expect("physical creation should commit at full write quorum");

        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        metadata_sys::get_object_lock_config_state_in(&store.ctx, &bucket)
            .await
            .expect("the fenced authority path should backfill the creation commit");
        let (metadata, persisted) = metadata_sys::get_config_from_disk_with_presence_in(&store.ctx, &bucket)
            .await
            .expect("read backfilled creation commit");
        assert!(persisted && metadata.bucket_creation_committed);

        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let set = &store.pools[0].disk_set[0];
        let offline = take_set_disks_offline(&store, set, &[0, 1]).await;
        metadata_sys::get_on_demand_migration_config_in(&store, &bucket)
            .await
            .expect("backfilled Object Lock bucket should load at read quorum");
        restore_set_disks(&store, set, offline).await;
    }

    #[tokio::test]
    #[serial]
    async fn enabling_object_lock_on_existing_bucket_establishes_commitment() {
        let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4], Some(2)).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = format!("enable-lock-legacy-{}", Uuid::new_v4().simple());
        store
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("create a versioned non-lock bucket");

        // Simulate a bucket persisted by a build before the commit field existed.
        let mut legacy = metadata_sys::get_in(&store.ctx, &bucket)
            .await
            .expect("cached metadata")
            .as_ref()
            .clone();
        legacy.lock_enabled = false;
        legacy.object_lock_config_xml = Vec::new();
        legacy.object_lock_config = None;
        legacy.bucket_creation_committed = false;
        metadata_sys::set_new_bucket_metadata_intent_in(&store.ctx, legacy)
            .await
            .expect("persist legacy-shaped metadata");

        let lock_xml = crate::bucket::utils::serialize(&*crate::store::ENABLED_OBJECT_LOCK_CONFIG)
            .expect("serialize Object Lock configuration");
        metadata_sys::update_in(&store.ctx, &bucket, "object-lock.xml", lock_xml)
            .await
            .expect("enabling Object Lock on an existing bucket must succeed");

        assert!(
            matches!(
                metadata_sys::get_object_lock_config_state_in(&store.ctx, &bucket)
                    .await
                    .expect("read the freshly enabled Object Lock state"),
                metadata_sys::ObjectLockConfigState::Configured { .. }
            ),
            "a healthy cluster must not report ErasureWriteQuorum after enabling Object Lock"
        );
        let (metadata, persisted) = metadata_sys::get_config_from_disk_with_presence_in(&store.ctx, &bucket)
            .await
            .expect("read persisted metadata");
        assert!(persisted && metadata.bucket_creation_committed);
    }

    #[tokio::test]
    #[serial]
    async fn creation_commit_backfill_preserves_concurrent_config_update() {
        let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4], Some(2)).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = format!("commit-backfill-pack-{}", Uuid::new_v4().simple());
        store
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("create bucket");

        // The cached snapshot on the migrating node still predates another
        // node's acknowledged configuration update.
        let mut stale = metadata_sys::get_in(&store.ctx, &bucket)
            .await
            .expect("cached metadata")
            .as_ref()
            .clone();
        stale.bucket_creation_committed = false;
        metadata_sys::set_new_bucket_metadata_intent_in(&store.ctx, stale.clone())
            .await
            .expect("persist the stale pre-upgrade snapshot");

        // A newer, still-uncommitted copy carrying an acknowledged Object Lock
        // update lands on disk after the migrating node read its snapshot.
        let mut newer = stale.clone();
        newer.lock_enabled = true;
        newer.object_lock_config_xml = crate::bucket::utils::serialize(&*crate::store::ENABLED_OBJECT_LOCK_CONFIG)
            .expect("serialize Object Lock configuration");
        newer
            .save_with_store(store.clone())
            .await
            .expect("persist the newer configuration");

        // A cold node must re-read the newest authoritative metadata under the
        // fence instead of rewriting its stale snapshot.
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        metadata_sys::get_object_lock_config_state_in(&store.ctx, &bucket)
            .await
            .expect("fenced commit migration");

        let (metadata, persisted) = metadata_sys::get_config_from_disk_with_presence_in(&store.ctx, &bucket)
            .await
            .expect("read persisted metadata");
        assert!(persisted && metadata.bucket_creation_committed);
        assert!(
            metadata.lock_enabled && !metadata.object_lock_config_xml.is_empty(),
            "the acknowledged Object Lock update must survive the backfill"
        );
    }

    #[tokio::test]
    #[serial]
    async fn creation_commit_backfill_keeps_recreated_generation() {
        let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4], Some(2)).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = format!("commit-backfill-recreate-{}", Uuid::new_v4().simple());
        store
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("create bucket");
        let old_incarnation = store.bucket_incarnation_id_from_disk(&bucket).await.expect("old incarnation");

        // A node that still holds the pre-recreation snapshot must never let it
        // outlive the replacement generation.
        let mut stale = metadata_sys::get_in(&store.ctx, &bucket)
            .await
            .expect("cached metadata")
            .as_ref()
            .clone();
        stale.lock_enabled = true;
        stale.object_lock_config_xml = crate::bucket::utils::serialize(&*crate::store::ENABLED_OBJECT_LOCK_CONFIG)
            .expect("serialize Object Lock configuration");
        stale.bucket_creation_committed = false;
        stale.save_with_store(store.clone()).await.expect("persist stale metadata");

        store
            .delete_bucket(&bucket, &crate::storage_api_contracts::bucket::DeleteBucketOptions::default())
            .await
            .expect("delete bucket");
        store
            .make_bucket(
                &bucket,
                &MakeBucketOptions {
                    lock_enabled: true,
                    ..Default::default()
                },
            )
            .await
            .expect("recreate bucket with Object Lock");
        let new_incarnation = store.bucket_incarnation_id_from_disk(&bucket).await.expect("new incarnation");
        assert_ne!(old_incarnation, new_incarnation);

        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        metadata_sys::get_object_lock_config_state_in(&store.ctx, &bucket)
            .await
            .expect("fenced backfill on the replacement generation");

        let (metadata, persisted) = metadata_sys::get_config_from_disk_with_presence_in(&store.ctx, &bucket)
            .await
            .expect("read persisted metadata");
        assert!(persisted);
        assert_eq!(
            metadata.bucket_incarnation_id, new_incarnation,
            "the replacement generation must stay authoritative"
        );
        assert!(metadata.bucket_creation_committed);
    }

    #[tokio::test]
    #[serial]
    async fn bucket_info_read_quorum_is_scoped_to_each_erasure_set() {
        let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4, 6], None).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = "read-quorum-mixed-pools";
        store
            .make_bucket(bucket, &MakeBucketOptions::default())
            .await
            .expect("healthy pools should accept bucket creation");

        let first_set = &store.pools[0].disk_set[0];
        let second_set = &store.pools[1].disk_set[0];
        let first_offline = take_set_disks_offline(&store, first_set, &[0, 1]).await;
        let second_offline = take_set_disks_offline(&store, second_set, &[0, 1, 2]).await;
        store
            .get_bucket_info(bucket, &BucketOptions::default())
            .await
            .expect("each set independently satisfies its namespace read quorum");

        for (set, extra_disk) in [(first_set, 2), (second_set, 3)] {
            let extra_offline = take_set_disks_offline(&store, set, &[extra_disk]).await;
            assert_eq!(
                store
                    .get_bucket_info(bucket, &BucketOptions::default())
                    .await
                    .expect_err("another pool must not subsidize a set below its read quorum"),
                StorageError::ErasureReadQuorum
            );
            restore_set_disks(&store, set, extra_offline).await;
        }
        restore_set_disks(&store, first_set, first_offline).await;
        restore_set_disks(&store, second_set, second_offline).await;
    }

    #[tokio::test]
    #[serial]
    async fn bucket_info_read_quorum_requires_authoritative_metadata() {
        for state in ["missing", "corrupt", "foreign", "incarnation"] {
            let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4], None).await;
            metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
            let bucket = format!("read-quorum-{state}-metadata");
            let mut metadata = if state == "missing" {
                store
                    .make_bucket_on_sets(&bucket, &MakeBucketOptions::default())
                    .await
                    .expect("simulate directories left before bucket metadata is published");
                BucketMetadata::new(&bucket)
            } else {
                store
                    .make_bucket(&bucket, &MakeBucketOptions::default())
                    .await
                    .expect("healthy bucket should publish metadata");
                metadata_sys::get_in(&store.ctx, &bucket)
                    .await
                    .expect("seed metadata should be cached")
                    .as_ref()
                    .clone()
            };
            let path = metadata.save_file_path();
            match state {
                "corrupt" => crate::config::com::save_config(store.clone(), &path, b"corrupt".to_vec())
                    .await
                    .expect("persist corrupt metadata while the cached copy remains valid"),
                "foreign" => {
                    metadata.name = "different-bucket".to_string();
                    let mut encoded = vec![1, 0, 1, 0];
                    encoded.extend(metadata.marshal_msg().expect("foreign metadata should encode"));
                    crate::config::com::save_config(store.clone(), &path, encoded)
                        .await
                        .expect("persist metadata for a different bucket at the requested path");
                }
                "incarnation" => crate::bucket::metadata::save_bucket_incarnation(store.clone(), &bucket, Uuid::new_v4())
                    .await
                    .expect("persist a different bucket generation"),
                _ => {}
            }

            let set = &store.pools[0].disk_set[0];
            let offline = take_set_disks_offline(&store, set, &[0, 1]).await;
            let error = store
                .get_bucket_info(&bucket, &BucketOptions::default())
                .await
                .expect_err("read admission must not trust residual directories or cached metadata");
            match state {
                "missing" => assert_eq!(error, StorageError::ErasureReadQuorum),
                "foreign" => assert_eq!(error, StorageError::FileCorrupt),
                "incarnation" => assert!(error.to_string().contains("sidecar does not match bucket metadata")),
                "corrupt" => assert!(error.to_string().contains("format invalid"), "unexpected corruption error: {error}"),
                _ => unreachable!(),
            }
            restore_set_disks(&store, set, offline).await;
        }
    }

    #[tokio::test]
    #[serial]
    async fn bucket_info_read_quorum_accepts_persisted_legacy_metadata() {
        let (_temp_dir, store) = setup_bucket_quorum_test_env(&[4], None).await;
        metadata_sys::init_bucket_metadata_sys(store.clone(), Vec::new()).await;
        let bucket = "interop";
        store
            .make_bucket_on_sets(bucket, &MakeBucketOptions::default())
            .await
            .expect("legacy bucket directories should exist");
        let hex = include_str!("../../tests/fixtures/minio/bucket_metadata.blob.hex")
            .split_whitespace()
            .collect::<String>();
        let body = (0..hex.len())
            .step_by(2)
            .map(|index| u8::from_str_radix(&hex[index..index + 2], 16).expect("pinned MinIO metadata fixture"))
            .collect();
        crate::config::com::save_config(store.clone(), &BucketMetadata::new(bucket).save_file_path(), body)
            .await
            .expect("legacy metadata should be persisted without an incarnation sidecar");

        let set = &store.pools[0].disk_set[0];
        let offline = take_set_disks_offline(&store, set, &[0, 1]).await;
        let info = store
            .get_bucket_info(bucket, &BucketOptions::default())
            .await
            .expect("persisted MinIO metadata should authorize reads at the namespace read quorum");
        assert_eq!(info.name, bucket);
        assert!(info.versioning);
        assert!(info.object_locking);
        restore_set_disks(&store, set, offline).await;
    }

    #[tokio::test]
    #[serial]
    async fn bucket_namespace_reads_report_missing_when_every_set_is_absent() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        let bucket = format!("absent-{}", Uuid::new_v4().simple());

        let buckets = ecstore
            .list_bucket(&BucketOptions::default())
            .await
            .expect("an empty healthy namespace should remain listable");
        assert!(buckets.iter().all(|entry| entry.name != bucket));

        let err = ecstore
            .get_bucket_info(&bucket, &BucketOptions::default())
            .await
            .expect_err("a bucket absent from every set must remain missing");
        assert_eq!(err, StorageError::VolumeNotFound);
    }

    #[tokio::test]
    #[serial]
    async fn bucket_namespace_reads_fail_closed_when_any_set_loses_quorum() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        metadata_sys::init_bucket_metadata_sys(ecstore.clone(), Vec::new()).await;
        let bucket = format!("degraded-expansion-{}", Uuid::new_v4().simple());
        ecstore.pools[0].disk_set[0]
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created in the original pool only");
        ecstore.pools[1].disk_set[0].disks.write().await[0] = None;
        ecstore.pools[1].disk_set[0].disks.write().await[1] = None;
        ecstore.pools[1].disk_set[0].disks.write().await[2] = None;

        let list_err = ecstore
            .list_bucket(&BucketOptions::default())
            .await
            .expect_err("listing must not hide an unavailable expansion pool");
        assert_eq!(list_err, StorageError::ErasureWriteQuorum);

        let info_err = ecstore
            .get_bucket_info(&bucket, &BucketOptions::default())
            .await
            .expect_err("bucket validation must fail when an expansion pool is unavailable");
        assert_eq!(info_err, StorageError::ErasureReadQuorum);
    }

    #[tokio::test]
    #[serial]
    async fn scanner_bucket_listing_unions_every_erasure_set() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        let bucket = format!("second-pool-only-{}", Uuid::new_v4().simple());
        ecstore.pools[1].disk_set[0]
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created in the second pool only");

        let listing = ecstore
            .list_bucket_for_scanner(&crate::storage_api_contracts::bucket::BucketOptions {
                no_metadata: true,
                ..Default::default()
            })
            .await
            .expect("scanner should enumerate every pool and set");

        assert!(listing.topology_complete);
        assert!(listing.buckets.iter().any(|entry| entry.name == bucket));
        assert_eq!(listing.set_buckets.len(), 2);
        assert!(
            listing
                .set_buckets
                .iter()
                .find(|scope| scope.pool_index == 0 && scope.set_index == 0)
                .is_some_and(|scope| scope.buckets.is_empty())
        );
        assert!(
            listing
                .set_buckets
                .iter()
                .find(|scope| scope.pool_index == 1 && scope.set_index == 0)
                .is_some_and(|scope| scope.buckets.iter().any(|entry| entry.name == bucket))
        );
    }

    #[tokio::test]
    async fn scanner_bucket_listing_marks_degraded_set_incomplete() {
        let (_temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        let bucket = format!("degraded-set-{}", Uuid::new_v4().simple());
        let set = &ecstore.pools[0].disk_set[0];
        set.make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created before a disk is removed");
        set.disks.write().await[0] = None;

        let listing = ecstore
            .list_bucket_for_scanner(&crate::storage_api_contracts::bucket::BucketOptions {
                no_metadata: true,
                ..Default::default()
            })
            .await
            .expect("a degraded set with quorum should still return candidate buckets");

        assert!(listing.buckets.iter().any(|entry| entry.name == bucket));
        assert!(!listing.topology_complete);
    }

    #[tokio::test]
    async fn scanner_bucket_listing_marks_divergent_disk_views_incomplete() {
        let (temp_dir, ecstore) = setup_multi_pool_bucket_test_env().await;
        let bucket = format!("divergent-set-{}", Uuid::new_v4().simple());
        ecstore.pools[0].disk_set[0]
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created before disk views diverge");
        for disk_index in 0..2 {
            tokio::fs::remove_dir_all(temp_dir.path().join(format!("pool0-disk{disk_index}")).join(&bucket))
                .await
                .expect("test bucket directory should be removed from a minority disk view");
        }

        let listing = ecstore
            .list_bucket_for_scanner(&crate::storage_api_contracts::bucket::BucketOptions {
                no_metadata: true,
                ..Default::default()
            })
            .await
            .expect("responsive disks should still produce a scanner candidate listing");

        assert!(listing.buckets.iter().all(|entry| entry.name != bucket));
        assert!(!listing.topology_complete);
    }

    // These tests share one isolated instance and mutate its bucket metadata;
    // serialize them so their assertions cannot observe each other's operations.
    #[tokio::test]
    #[serial]
    async fn bucket_delete_mark_delete_removes_empty_bucket_and_keeps_deleted_marker() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-mark-delete-{}", Uuid::new_v4().simple());

        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created");
        for disk_path in &disk_paths {
            tokio::fs::create_dir_all(disk_path.join(&bucket).join("empty-directory/nested/leaf"))
                .await
                .expect("empty directory remnant should be created");
        }
        assert!(metadata_sys::get_in(&ecstore.ctx, &bucket).await.is_ok());

        let generation_before_delete = ecstore.scanner_namespace_mutation_generation();
        ecstore
            .delete_bucket(
                &bucket,
                &DeleteBucketOptions {
                    srdelete_op: SRBucketDeleteOp::MarkDelete,
                    ..Default::default()
                },
            )
            .await
            .expect("MarkDelete should remove an empty bucket");
        assert_eq!(
            ecstore.scanner_namespace_mutation_generation(),
            generation_before_delete.saturating_add(2),
            "successful bucket deletion should fence scanner namespace activity before and after deletion"
        );

        assert!(
            !any_disk_path_exists(&disk_paths, &bucket).await,
            "MarkDelete should remove the bucket volume"
        );
        assert!(
            any_disk_path_exists(&disk_paths, bucket_deleted_marker_volume(&bucket)).await,
            "MarkDelete should persist the deleted-bucket marker"
        );
        assert!(
            metadata_sys::get_in(&ecstore.ctx, &bucket).await.is_err(),
            "deleted bucket metadata must be removed from the local cache"
        );
        let buckets = ecstore
            .list_bucket(&BucketOptions::default())
            .await
            .expect("bucket listing should succeed after MarkDelete");
        assert!(!buckets.iter().any(|info| info.name == bucket));
        ecstore
            .delete_all(RUSTFS_META_BUCKET, &bucket_deleted_marker_prefix(&bucket))
            .await
            .expect("deleted-bucket marker should be removed to simulate a partial failure");
        assert!(!any_disk_path_exists(&disk_paths, bucket_deleted_marker_volume(&bucket)).await);
        ecstore
            .delete_bucket(
                &bucket,
                &DeleteBucketOptions {
                    srdelete_op: SRBucketDeleteOp::MarkDelete,
                    ..Default::default()
                },
            )
            .await
            .expect("retried MarkDelete should recreate a missing tombstone");
        assert!(any_disk_path_exists(&disk_paths, bucket_deleted_marker_volume(&bucket)).await);
    }

    #[tokio::test]
    #[serial]
    async fn bucket_delete_mark_delete_rejects_non_empty_bucket_without_force() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-mark-delete-non-empty-{}", Uuid::new_v4().simple());

        create_bucket_with_object(&ecstore, &bucket, "object.txt").await;

        let err = ecstore
            .delete_bucket(
                &bucket,
                &DeleteBucketOptions {
                    srdelete_op: SRBucketDeleteOp::MarkDelete,
                    ..Default::default()
                },
            )
            .await
            .expect_err("MarkDelete should reject a non-empty bucket without force");

        assert!(matches!(err, StorageError::BucketNotEmpty(name) if name == bucket));
        assert!(any_disk_has_object_metadata(&disk_paths, &bucket).await);
        assert!(!any_disk_path_exists(&disk_paths, bucket_deleted_marker_volume(&bucket)).await);
    }

    #[tokio::test]
    #[serial]
    async fn bucket_delete_mark_delete_rejects_hidden_object_paths_without_force() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-mark-delete-hidden-{}", Uuid::new_v4().simple());

        create_bucket_with_object(&ecstore, &bucket, ".well-known/acme-challenge").await;
        let mut reader = PutObjReader::from_vec(b"delete bucket semantics".to_vec());
        ecstore
            .put_object(&bucket, ".rustfs.sys/object", &mut reader, &ObjectOptions::default())
            .await
            .expect("second hidden object should be written");

        let err = ecstore
            .delete_bucket(
                &bucket,
                &DeleteBucketOptions {
                    srdelete_op: SRBucketDeleteOp::MarkDelete,
                    ..Default::default()
                },
            )
            .await
            .expect_err("MarkDelete should reject a hidden object path without force");

        assert!(matches!(err, StorageError::BucketNotEmpty(name) if name == bucket));
        assert!(any_disk_has_object_metadata(&disk_paths, &bucket).await);
    }

    #[tokio::test]
    #[serial]
    async fn bucket_delete_purge_removes_bucket_data_and_internal_metadata() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-purge-{}", Uuid::new_v4().simple());
        let object = "object.txt";
        let metadata_prefix = format!("{RUSTFS_META_BUCKET}/{BUCKET_META_PREFIX}/{bucket}");

        create_bucket_with_object(&ecstore, &bucket, object).await;
        write_bucket_metadata_marker(&disk_paths, &metadata_prefix).await;
        ecstore
            .mark_bucket_deleted(&bucket)
            .await
            .expect("deleted-bucket marker should be created");
        assert!(any_disk_path_exists(&disk_paths, &metadata_prefix).await);
        assert!(any_disk_path_exists(&disk_paths, bucket_deleted_marker_volume(&bucket)).await);

        let generation_before_delete = ecstore.scanner_namespace_mutation_generation();
        ecstore
            .delete_bucket(
                &bucket,
                &DeleteBucketOptions {
                    force: true,
                    srdelete_op: SRBucketDeleteOp::Purge,
                    ..Default::default()
                },
            )
            .await
            .expect("Purge should force-delete bucket data");
        assert_eq!(
            ecstore.scanner_namespace_mutation_generation(),
            generation_before_delete.saturating_add(2),
            "successful bucket purge should fence scanner namespace activity before and after deletion"
        );

        assert!(!any_disk_path_exists(&disk_paths, &bucket).await, "Purge should remove the bucket volume");
        assert!(
            !any_disk_path_exists(&disk_paths, &metadata_prefix).await,
            "Purge should remove bucket metadata prefix"
        );
        assert!(
            !any_disk_path_exists(&disk_paths, bucket_deleted_marker_volume(&bucket)).await,
            "Purge should remove the deleted-bucket marker"
        );
        assert!(
            metadata_sys::get_in(&ecstore.ctx, &bucket).await.is_err(),
            "purged bucket metadata must be removed from the local cache"
        );
        write_bucket_metadata_marker(&disk_paths, &metadata_prefix).await;
        ecstore
            .mark_bucket_deleted(&bucket)
            .await
            .expect("stale deleted-bucket marker should be recreated");
        ecstore
            .delete_bucket(
                &bucket,
                &DeleteBucketOptions {
                    force: true,
                    srdelete_op: SRBucketDeleteOp::Purge,
                    ..Default::default()
                },
            )
            .await
            .expect("retried Purge should remove stale metadata without a bucket volume");
        assert!(!any_disk_path_exists(&disk_paths, &metadata_prefix).await);
        assert!(!any_disk_path_exists(&disk_paths, bucket_deleted_marker_volume(&bucket)).await);
    }

    /// Site replication replays make-with-versioning carrying the source's
    /// `lockEnabled` against a destination bucket that already exists. Gating the
    /// lock enable on `confirmed_missing` returned success while leaving the
    /// replica unlocked, so replicated versions could be deleted without the
    /// retention the source enforces.
    #[tokio::test(flavor = "multi_thread")]
    #[serial]
    async fn force_create_enables_object_lock_on_an_existing_bucket() {
        let (_disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-force-lock-{}", Uuid::new_v4().simple());

        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("plain bucket should be created");
        assert!(
            !metadata_sys::get_in(&ecstore.ctx, &bucket)
                .await
                .expect("metadata should load")
                .lock_enabled,
            "test setup: the bucket must start unlocked"
        );

        ecstore
            .make_bucket(
                &bucket,
                &MakeBucketOptions {
                    force_create: true,
                    lock_enabled: true,
                    ..Default::default()
                },
            )
            .await
            .expect("force create with lock_enabled should succeed on an existing bucket");

        let meta = metadata_sys::get_in(&ecstore.ctx, &bucket)
            .await
            .expect("metadata should load after the lock enable");
        assert!(meta.lock_enabled, "Object Lock must be enabled on the existing bucket");
        assert!(
            !meta.object_lock_config_xml.is_empty(),
            "the Object Lock configuration must be persisted, not just the flag"
        );
        assert!(
            !meta.versioning_config_xml.is_empty(),
            "Object Lock requires versioning, so that must be persisted too"
        );
    }

    /// rustfs/rustfs#8003: a force-create (site replication replay, admin
    /// import) against a bucket whose legacy migration stopped after the
    /// sidecar write must persist `.metadata.bin` under the stored incarnation
    /// rather than fail closed or replace the published identity.
    #[tokio::test(flavor = "multi_thread")]
    #[serial]
    async fn issue_8003_force_create_adopts_existing_incarnation_sidecar() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-8003-sidecar-{}", Uuid::new_v4().simple());

        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created");
        let stored = metadata_sys::get_in(&ecstore.ctx, &bucket)
            .await
            .expect("metadata should load")
            .bucket_incarnation_id;
        assert!(!stored.is_nil());

        let metadata_path = format!("{BUCKET_META_PREFIX}/{bucket}/{}", crate::bucket::metadata::BUCKET_METADATA_FILE);
        crate::config::com::delete_config(ecstore.clone(), &metadata_path)
            .await
            .expect("simulate the interrupted migration by removing only .metadata.bin");
        assert!(!any_disk_path_exists(&disk_paths, format!("{RUSTFS_META_BUCKET}/{metadata_path}")).await);
        metadata_sys::remove_bucket_metadata_in(&ecstore.ctx, &bucket)
            .await
            .expect("drop the cached copy so the next read hits disk");

        ecstore
            .make_bucket(
                &bucket,
                &MakeBucketOptions {
                    force_create: true,
                    ..Default::default()
                },
            )
            .await
            .expect("force create must repair the sidecar-only bucket");

        let (repaired, persisted) = metadata_sys::get_config_from_disk_with_presence_in(&ecstore.ctx, &bucket)
            .await
            .expect("repaired metadata should load");
        assert!(persisted, "force create must persist .metadata.bin again");
        assert!(repaired.bucket_incarnation_sidecar);
        assert_eq!(
            repaired.bucket_incarnation_id, stored,
            "the repair must keep the incarnation the sidecar already published"
        );
    }

    #[tokio::test(flavor = "multi_thread")]
    #[serial]
    async fn make_bucket_inherits_default_durability() {
        temp_env::async_with_vars([(crate::bucket::durability::ENV_NEW_BUCKET_DURABILITY_MODE, None::<&str>)], async {
            let (_disk_paths, ecstore) = setup_bucket_delete_test_env().await;
            let bucket = format!("bucket-default-durability-{}", Uuid::new_v4().simple());

            ecstore
                .make_bucket(&bucket, &MakeBucketOptions::default())
                .await
                .expect("new bucket should be created");

            let metadata = metadata_sys::get_in(&ecstore.ctx, &bucket)
                .await
                .expect("metadata should load for the new bucket");
            assert!(metadata.durability_config().is_none());
        })
        .await;
    }

    #[tokio::test(flavor = "multi_thread")]
    #[serial]
    async fn force_create_existing_bucket_keeps_durability_override() {
        let (_disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-force-durability-{}", Uuid::new_v4().simple());

        temp_env::async_with_vars([(crate::bucket::durability::ENV_NEW_BUCKET_DURABILITY_MODE, Some("inherit"))], async {
            ecstore
                .make_bucket(&bucket, &MakeBucketOptions::default())
                .await
                .expect("plain bucket should be created without a durability override");
        })
        .await;
        assert!(
            metadata_sys::get_in(&ecstore.ctx, &bucket)
                .await
                .expect("metadata should load after initial create")
                .durability_config()
                .is_none(),
            "test setup: the existing bucket must start without an override"
        );

        temp_env::async_with_vars([(crate::bucket::durability::ENV_NEW_BUCKET_DURABILITY_MODE, None::<&str>)], async {
            ecstore
                .make_bucket(
                    &bucket,
                    &MakeBucketOptions {
                        force_create: true,
                        lock_enabled: true,
                        ..Default::default()
                    },
                )
                .await
                .expect("force create should update existing bucket metadata");
        })
        .await;

        let metadata = metadata_sys::get_in(&ecstore.ctx, &bucket)
            .await
            .expect("metadata should load after force create");
        assert!(metadata.lock_enabled, "force create sanity check: Object Lock should be enabled");
        assert!(
            metadata.durability_config().is_none(),
            "force create must not apply the new-bucket default to existing bucket metadata"
        );
    }

    /// `DeleteBucket`'s emptiness check is a raw disk scan (`has_xlmeta_files`),
    /// not an S3-level listing, so "the client drained the bucket" and "the
    /// bucket is deletable" are two different contracts. Nothing pinned the
    /// second one, which is how the s3-tests lane ended up failing 219 cases on
    /// `nuke_prefixed_buckets` while every test body passed.
    #[tokio::test(flavor = "multi_thread")]
    #[serial]
    async fn bucket_delete_succeeds_after_the_last_object_version_is_deleted() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-delete-after-drain-{}", Uuid::new_v4().simple());
        let object = "object.txt";

        create_bucket_with_object(&ecstore, &bucket, object).await;

        ecstore
            .delete_object(&bucket, object, ObjectOptions::default())
            .await
            .expect("client delete of the only object should succeed");

        assert!(
            !any_disk_has_object_metadata(&disk_paths, &bucket).await,
            "deleting the last version must not leave xl.meta on disk: DeleteBucket scans the raw \
             bucket directory, so residue here is reported to clients as BucketNotEmpty"
        );

        ecstore
            .delete_bucket(&bucket, &DeleteBucketOptions::default())
            .await
            .expect("DeleteBucket must succeed once the client has drained the bucket");
    }

    #[tokio::test(flavor = "multi_thread")]
    #[serial]
    async fn bucket_delete_defers_zero_evidence_diagnostic_timeout_to_physical_empty_check() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-del-diag-{}", Uuid::new_v4().simple());

        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("empty bucket should be created");
        let first_io_started = Arc::new(AtomicBool::new(false));
        let diagnostic_budget = BucketDeleteDiagnosticBudget::new().with_first_io_delay(
            BUCKET_DELETE_DIAGNOSTIC_MAX_ELAPSED + Duration::from_millis(100),
            first_io_started.clone(),
        );

        ecstore
            .handle_delete_bucket_with_diagnostic_budget(&bucket, &DeleteBucketOptions::default(), diagnostic_budget)
            .await
            .expect("a zero-evidence diagnostic timeout must defer to the physical empty check");

        assert!(
            first_io_started.load(Ordering::SeqCst),
            "the regression must delay the first diagnostic read_dir"
        );
        assert!(
            !any_disk_path_exists(&disk_paths, &bucket).await,
            "the physical empty check should allow deletion of the actually empty bucket"
        );
    }

    #[tokio::test(flavor = "multi_thread")]
    #[serial]
    async fn bucket_delete_preserves_unobserved_residue_after_diagnostic_timeout() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-del-residue-{}", Uuid::new_v4().simple());
        let object = "object.txt";

        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created");
        write_metadata_less_part_on_all_disks(&disk_paths, &bucket, object).await;
        let first_io_started = Arc::new(AtomicBool::new(false));
        let diagnostic_budget = BucketDeleteDiagnosticBudget::new().with_first_io_delay(
            BUCKET_DELETE_DIAGNOSTIC_MAX_ELAPSED + Duration::from_millis(100),
            first_io_started.clone(),
        );

        let err = ecstore
            .handle_delete_bucket_with_diagnostic_budget(&bucket, &DeleteBucketOptions::default(), diagnostic_budget)
            .await
            .expect_err("physical empty-bucket enforcement must reject unobserved residue");
        assert!(
            first_io_started.load(Ordering::SeqCst),
            "the regression must time out before observing the residue"
        );
        match &err {
            StorageError::BucketNotEmpty(err_bucket) | StorageError::BucketNotEmptyWithDetails { bucket: err_bucket, .. } => {
                assert_eq!(err_bucket, &bucket)
            }
            other => panic!("expected an S3-compatible BucketNotEmpty error, got {other:?}"),
        }
        assert!(
            any_disk_path_exists(&disk_paths, Path::new(&bucket).join(object)).await,
            "physical empty-bucket enforcement must preserve unobserved residue"
        );

        ecstore
            .delete_bucket(
                &bucket,
                &DeleteBucketOptions {
                    force: true,
                    ..Default::default()
                },
            )
            .await
            .expect("explicit force should clean up the test-owned residue");
    }

    #[tokio::test(flavor = "multi_thread")]
    #[serial]
    async fn bucket_delete_succeeds_after_listing_and_deleting_an_unversioned_overwrite() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-delete-after-overwrite-{}", Uuid::new_v4().simple());
        let object = "object.txt";

        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("unversioned bucket should be created");

        let mut first_reader = PutObjReader::from_vec(b"version A".to_vec());
        let first = ecstore
            .put_object(&bucket, object, &mut first_reader, &ObjectOptions::default())
            .await
            .expect("version A should be written");
        let mut second_reader = PutObjReader::from_vec(b"version B".to_vec());
        let second = ecstore
            .put_object(&bucket, object, &mut second_reader, &ObjectOptions::default())
            .await
            .expect("version B should overwrite version A");
        assert_ne!(first.data_dir, second.data_dir, "the overwrite must publish a new body generation");

        let listing = ecstore
            .clone()
            .list_object_versions(&bucket, "", None, None, None, 1000)
            .await
            .expect("the overwritten object should remain listable for teardown");
        assert_eq!(listing.objects.len(), 1, "an unversioned overwrite should expose one current version");
        let current = &listing.objects[0];
        assert_eq!(current.name, object);
        assert!(current.is_latest, "the listed null version must be current");
        assert_eq!(current.version_id, None, "an unversioned object must be exposed as the null version");
        assert_eq!(
            current.data_dir, second.data_dir,
            "listing must expose version B, not the overwritten body"
        );

        for version in listing.objects {
            let version_id = version.version_id.map(|version_id| version_id.to_string());
            ecstore
                .delete_object(
                    &bucket,
                    &version.name,
                    ObjectOptions {
                        version_id,
                        ..Default::default()
                    },
                )
                .await
                .expect("each version returned by teardown listing should be deletable");
        }

        assert!(
            !any_disk_has_object_metadata(&disk_paths, &bucket).await,
            "deleting the listed null version must remove every xl.meta"
        );
        ecstore
            .delete_bucket(&bucket, &DeleteBucketOptions::default())
            .await
            .expect("DeleteBucket should succeed after the listed overwrite is deleted");
    }

    #[tokio::test]
    #[serial]
    async fn bucket_delete_default_s3_delete_still_rejects_non_empty_bucket() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-s3-delete-{}", Uuid::new_v4().simple());
        let object = "object.txt";

        create_bucket_with_object(&ecstore, &bucket, object).await;

        let generation_before_delete = ecstore.scanner_namespace_mutation_generation();
        let err = ecstore
            .delete_bucket(&bucket, &DeleteBucketOptions::default())
            .await
            .expect_err("default S3 DeleteBucket should reject non-empty buckets");

        assert!(matches!(err, StorageError::BucketNotEmpty(name) if name == bucket));
        assert_eq!(
            ecstore.scanner_namespace_mutation_generation(),
            generation_before_delete,
            "failed bucket deletion must not advance scanner namespace activity"
        );
        assert!(
            any_disk_has_object_metadata(&disk_paths, &bucket).await,
            "failed default S3 DeleteBucket must keep object data"
        );
        assert!(
            metadata_sys::get_in(&ecstore.ctx, &bucket).await.is_ok(),
            "failed default S3 DeleteBucket must keep metadata cache"
        );
    }

    #[tokio::test(flavor = "multi_thread")]
    #[serial]
    async fn bucket_delete_reports_metadata_less_residue_without_removing_it() {
        let (disk_paths, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-delete-orphan-datadir-{}", Uuid::new_v4().simple());
        let object = "object.txt";

        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created");
        write_metadata_less_part_on_all_disks(&disk_paths, &bucket, object).await;
        assert!(
            !any_disk_has_object_metadata(&disk_paths, &bucket).await,
            "test setup must reproduce a bucket with data dirs but no xl.meta"
        );

        let err = ecstore
            .delete_bucket(&bucket, &DeleteBucketOptions::default())
            .await
            .expect_err("metadata-less data dirs must block ordinary DeleteBucket");

        match err {
            StorageError::BucketNotEmptyWithDetails {
                bucket: err_bucket,
                details,
            } => {
                assert_eq!(err_bucket, bucket);
                assert!(
                    details.contains("metadata-less on-disk residue"),
                    "diagnostic should identify metadata-less residue, got: {details}"
                );
                assert!(details.contains("files="), "diagnostic should include a bounded count");
                assert!(details.contains("uuid_data_dirs="), "diagnostic should include data-dir count");
            }
            other => panic!("expected detailed BucketNotEmpty, got {other:?}"),
        }
        assert!(
            any_disk_path_exists(&disk_paths, Path::new(&bucket).join(object)).await,
            "ordinary DeleteBucket must preserve metadata-less data until an explicit operator action"
        );
    }

    #[tokio::test]
    #[serial]
    async fn bucket_delete_fences_put_started_after_empty_scan() {
        let (_, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-delete-empty-scan-race-{}", Uuid::new_v4().simple());
        let object = "committed-after-empty-scan";
        let payload = b"object started after DeleteBucket empty scan".to_vec();

        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created");

        let barrier = install_delete_bucket_empty_scan_barrier();
        let delete_store = ecstore.clone();
        let delete_bucket = bucket.clone();
        let delete = tokio::spawn(async move {
            delete_store
                .delete_bucket(&delete_bucket, &DeleteBucketOptions::default())
                .await
        });
        barrier.wait_until_paused().await;

        let put_store = ecstore.clone();
        let put_bucket = bucket.clone();
        let mut put = tokio::spawn(async move {
            let mut put_reader = PutObjReader::from_vec(payload);
            put_store
                .put_object(&put_bucket, object, &mut put_reader, &ObjectOptions::default())
                .await
        });
        assert!(
            tokio::time::timeout(Duration::from_millis(100), &mut put).await.is_err(),
            "PUT must wait behind the DeleteBucket lifecycle fence"
        );

        barrier.release();
        delete
            .await
            .expect("DeleteBucket task should join")
            .expect("DeleteBucket should commit while holding the lifecycle fence");
        let err = put
            .await
            .expect("PUT task should join")
            .expect_err("PUT must not recreate an object in the deleted bucket");
        assert!(matches!(err, StorageError::BucketNotFound(name) if name == bucket));
    }

    #[tokio::test]
    #[serial]
    async fn bucket_recreation_does_not_publish_unverified_usage() {
        let (_, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-usage-generation-{}", Uuid::new_v4().simple());
        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created");

        let mut snapshot = DataUsageInfo {
            last_update: Some(SystemTime::now()),
            buckets_count: 1,
            ..Default::default()
        };
        snapshot.buckets_usage.insert(
            bucket.clone(),
            BucketUsageInfo {
                size: 42,
                objects_count: 1,
                versions_count: 1,
                ..Default::default()
            },
        );
        snapshot.usage_snapshot_complete = true;
        snapshot.bucket_sizes.insert(bucket.clone(), 42);
        snapshot.calculate_totals();
        crate::data_usage::store_data_usage_in_backend(snapshot, ecstore.clone())
            .await
            .expect("usage snapshot should be stored");

        ecstore
            .delete_bucket(&bucket, &DeleteBucketOptions::default())
            .await
            .expect("empty bucket should be deleted");
        let deleted = crate::data_usage::load_data_usage_from_backend(ecstore.clone())
            .await
            .expect("usage snapshot should remain readable");
        assert!(!deleted.buckets_usage.contains_key(&bucket));

        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("same bucket name should be recreated after delete returns");
        crate::data_usage::record_bucket_object_write_memory(&bucket, None, 84).await;

        let mut recreated = crate::data_usage::load_data_usage_from_backend(ecstore.clone())
            .await
            .expect("recreated bucket usage base should load");
        crate::data_usage::apply_bucket_usage_memory_overlay(&mut recreated).await;
        assert!(
            !recreated.buckets_usage.contains_key(&bucket),
            "a request-path delta without an authoritative baseline must remain unavailable"
        );
        assert_eq!(crate::data_usage::get_bucket_usage_memory(&bucket).await, None);
    }

    #[tokio::test]
    #[serial]
    async fn bucket_create_removes_stale_usage_before_physical_creation() {
        let (_, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-create-stale-usage-{}", Uuid::new_v4().simple());
        let mut snapshot = DataUsageInfo {
            last_update: Some(SystemTime::now()),
            buckets_count: 1,
            ..Default::default()
        };
        snapshot.buckets_usage.insert(
            bucket.clone(),
            BucketUsageInfo {
                size: 42,
                objects_count: 1,
                versions_count: 1,
                ..Default::default()
            },
        );
        snapshot.usage_snapshot_complete = true;
        snapshot.bucket_sizes.insert(bucket.clone(), 42);
        snapshot.calculate_totals();
        crate::data_usage::store_data_usage_in_backend(snapshot, ecstore.clone())
            .await
            .expect("stale usage fixture should be stored");

        ecstore
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("CreateBucket should succeed");

        let persisted = crate::data_usage::load_data_usage_from_backend(ecstore.clone())
            .await
            .expect("usage snapshot should remain readable");
        assert!(
            !persisted.buckets_usage.contains_key(&bucket),
            "a newly created bucket must not inherit the predecessor generation's usage"
        );
        assert!(
            ecstore.get_bucket_info(&bucket, &BucketOptions::default()).await.is_ok(),
            "physical creation should happen after the usage fence succeeds"
        );
    }

    #[tokio::test]
    #[serial]
    async fn failed_create_rollback_does_not_run_unfenced_usage_cleanup() {
        let (_, ecstore) = setup_bucket_delete_test_env().await;
        let bucket = format!("bucket-create-rollback-{}", Uuid::new_v4().simple());
        ecstore
            .peer_sys
            .make_bucket(&bucket, &MakeBucketOptions::default())
            .await
            .expect("the partial-create fixture should expose a physical bucket");

        let mut snapshot = DataUsageInfo {
            last_update: Some(SystemTime::now()),
            buckets_count: 1,
            ..Default::default()
        };
        snapshot.buckets_usage.insert(
            bucket.clone(),
            BucketUsageInfo {
                size: 42,
                objects_count: 1,
                versions_count: 1,
                ..Default::default()
            },
        );
        snapshot.usage_snapshot_complete = true;
        snapshot.bucket_sizes.insert(bucket.clone(), 42);
        snapshot.calculate_totals();
        crate::data_usage::store_data_usage_in_backend(snapshot, ecstore.clone())
            .await
            .expect("the usage fixture should be stored");
        crate::bucket::metadata::save_bucket_incarnation(ecstore.clone(), &bucket, Uuid::new_v4())
            .await
            .expect("partial create should have an incarnation sidecar");

        ecstore.rollback_failed_bucket_creation(&bucket, None, None).await;

        assert!(
            ecstore
                .peer_sys
                .get_bucket_info(&bucket, &BucketOptions::default())
                .await
                .is_err(),
            "failed-create rollback should remove the partial physical bucket"
        );
        assert!(
            crate::bucket::metadata::load_bucket_incarnation(ecstore.clone(), &bucket)
                .await
                .expect("rollback sidecar lookup should succeed")
                .is_none(),
            "failed-create rollback must remove the orphan incarnation sidecar"
        );
        let persisted = crate::data_usage::load_data_usage_from_backend(ecstore.clone())
            .await
            .expect("the usage snapshot should remain readable");
        assert!(
            persisted.buckets_usage.contains_key(&bucket),
            "failed-create rollback must not start an unfenced usage cleanup"
        );

        crate::data_usage::store_data_usage_in_backend(DataUsageInfo::default(), ecstore.clone())
            .await
            .expect("the rollback usage fixture should be cleared");
    }

    #[tokio::test]
    #[serial]
    async fn bucket_create_fails_closed_when_usage_snapshot_cannot_be_fenced() {
        let (_, ecstore) = setup_bucket_delete_test_env().await;
        let deleted_bucket = format!("bucket-delete-corrupt-usage-{}", Uuid::new_v4().simple());
        ecstore
            .make_bucket(&deleted_bucket, &MakeBucketOptions::default())
            .await
            .expect("bucket should be created before corrupting usage");

        let usage_path = format!("{BUCKET_META_PREFIX}/{DATA_USAGE_OBJECT_NAME}");
        crate::config::com::save_config(ecstore.clone(), &usage_path, b"{".to_vec())
            .await
            .expect("corrupt usage fixture should be stored");

        ecstore
            .delete_bucket(&deleted_bucket, &DeleteBucketOptions::default())
            .await
            .expect("usage snapshot corruption must not block DeleteBucket");
        assert!(
            ecstore
                .get_bucket_info(&deleted_bucket, &crate::storage_api_contracts::bucket::BucketOptions::default())
                .await
                .is_err(),
            "successful DeleteBucket must remove the physical bucket"
        );

        let new_bucket = format!("bucket-create-corrupt-usage-{}", Uuid::new_v4().simple());
        let create_err = ecstore
            .make_bucket(&new_bucket, &MakeBucketOptions::default())
            .await
            .expect_err("MakeBucket must fail when stale usage cannot be fenced");
        assert!(!create_err.to_string().is_empty(), "the usage snapshot failure should be surfaced");
        assert!(
            ecstore
                .get_bucket_info(&new_bucket, &crate::storage_api_contracts::bucket::BucketOptions::default())
                .await
                .is_err(),
            "failed MakeBucket must not expose a new physical bucket"
        );

        let restored = serde_json::to_vec(&DataUsageInfo {
            last_update: Some(SystemTime::now()),
            ..Default::default()
        })
        .expect("restored usage fixture should encode");
        crate::config::com::save_config(ecstore, &usage_path, restored)
            .await
            .expect("usage fixture should be restored after the failure-path test");
    }

    /// Capture this module's log the way a stock deployment filters it.
    fn bucket_logs_at_default_level(emit: impl FnOnce()) -> String {
        use std::sync::{Arc, Mutex};
        use tracing_subscriber::EnvFilter;
        use tracing_subscriber::fmt::MakeWriter;
        use tracing_subscriber::layer::SubscriberExt;

        #[derive(Clone, Default)]
        struct CapturedLogs {
            buffer: Arc<Mutex<Vec<u8>>>,
        }
        struct CapturedLogWriter {
            buffer: Arc<Mutex<Vec<u8>>>,
        }
        impl std::io::Write for CapturedLogWriter {
            fn write(&mut self, buf: &[u8]) -> std::io::Result<usize> {
                self.buffer
                    .lock()
                    .expect("captured logs mutex should not be poisoned")
                    .extend_from_slice(buf);
                Ok(buf.len())
            }
            fn flush(&mut self) -> std::io::Result<()> {
                Ok(())
            }
        }
        impl<'a> MakeWriter<'a> for CapturedLogs {
            type Writer = CapturedLogWriter;
            fn make_writer(&'a self) -> Self::Writer {
                CapturedLogWriter {
                    buffer: Arc::clone(&self.buffer),
                }
            }
        }

        let logs = CapturedLogs::default();
        let subscriber = tracing_subscriber::registry()
            .with(EnvFilter::new(rustfs_config::DEFAULT_LOG_LEVEL))
            .with(
                tracing_subscriber::fmt::layer()
                    .with_writer(logs.clone())
                    .with_ansi(false)
                    .without_time(),
            );
        let _guard = tracing::subscriber::set_default(subscriber);
        let _callsite_pin = crate::test_tracing::pin_callsite_interest_for_test();

        emit();

        let buffer = logs
            .buffer
            .lock()
            .expect("captured logs mutex should not be poisoned")
            .clone();
        String::from_utf8(buffer).expect("captured logs should be valid UTF-8")
    }

    fn residue_sample(sample: &str) -> BucketMetadataLessResidue {
        BucketMetadataLessResidue {
            files: 1,
            entries_scanned: 3,
            sample: Some(sample.to_string()),
            ..Default::default()
        }
    }

    /// `DeleteBucket` answers from a raw per-disk residue scan, so it can refuse
    /// for a reason no S3 request can observe. When that happens the server's
    /// own record of which residue blocked it must survive the default log
    /// filter — otherwise a client that drained the bucket sees `BucketNotEmpty`
    /// and the server log holds no trace of the refusal at all.
    #[test]
    fn unreachable_residue_is_reported_at_the_default_log_level() {
        for kind in [
            BucketDeleteBlockerKind::UnknownXlMeta,
            BucketDeleteBlockerKind::OrphanDirectory,
            BucketDeleteBlockerKind::DiagnosticBudgetExceeded,
        ] {
            let logs = bucket_logs_at_default_level(|| {
                record_bucket_delete_blocker("drained-bucket", kind, &residue_sample("obj/8f2c/xl.meta"));
            });

            assert!(logs.contains("drained-bucket"), "the bucket must be named for {kind:?}: {logs}");
            assert!(logs.contains(kind.as_str()), "the blocker kind must be named for {kind:?}: {logs}");
            assert!(
                logs.contains("obj/8f2c/xl.meta"),
                "the residue sample is the only pointer to the leftover state for {kind:?}: {logs}"
            );
        }
    }

    /// A bucket that genuinely still holds a version is an ordinary 409: the
    /// client can list and delete what is left, so this must not be reported as
    /// a server-side fault.
    #[test]
    fn client_visible_blockers_stay_below_the_default_log_level() {
        for kind in [
            BucketDeleteBlockerKind::VisibleVersion,
            BucketDeleteBlockerKind::TierFreeVersion,
        ] {
            let logs = bucket_logs_at_default_level(|| {
                record_bucket_delete_blocker("still-full-bucket", kind, &residue_sample("obj/xl.meta"));
            });

            assert!(logs.is_empty(), "an ordinary non-empty bucket must not log an error for {kind:?}: {logs}");
        }
    }

    #[test]
    fn blocker_kinds_split_by_whether_the_client_can_reach_the_residue() {
        assert!(BucketDeleteBlockerKind::VisibleVersion.is_client_visible());
        assert!(BucketDeleteBlockerKind::TierFreeVersion.is_client_visible());
        assert!(!BucketDeleteBlockerKind::UnknownXlMeta.is_client_visible());
        assert!(!BucketDeleteBlockerKind::OrphanDirectory.is_client_visible());
        assert!(!BucketDeleteBlockerKind::DiagnosticBudgetExceeded.is_client_visible());
    }
}