@nxtedition/rocksdb 8.2.8 → 9.0.1

package/deps/rocksdb/rocksdb/db/db_properties_test.cc

@@ -188,40 +188,6 @@ TEST_F(DBPropertiesTest, GetAggregatedIntPropertyTest) {
 }
 
 namespace {
-void ResetTableProperties(TableProperties* tp) {
-  tp->data_size = 0;
-  tp->index_size = 0;
-  tp->filter_size = 0;
-  tp->raw_key_size = 0;
-  tp->raw_value_size = 0;
-  tp->num_data_blocks = 0;
-  tp->num_entries = 0;
-  tp->num_deletions = 0;
-  tp->num_merge_operands = 0;
-  tp->num_range_deletions = 0;
-}
-
-void ParseTablePropertiesString(std::string tp_string, TableProperties* tp) {
-  double dummy_double;
-  std::replace(tp_string.begin(), tp_string.end(), ';', ' ');
-  std::replace(tp_string.begin(), tp_string.end(), '=', ' ');
-  ResetTableProperties(tp);
-  sscanf(tp_string.c_str(),
-         "# data blocks %" SCNu64 " # entries %" SCNu64 " # deletions %" SCNu64
-         " # merge operands %" SCNu64 " # range deletions %" SCNu64
-         " raw key size %" SCNu64
-         " raw average key size %lf "
-         " raw value size %" SCNu64
-         " raw average value size %lf "
-         " data block size %" SCNu64 " index block size (user-key? %" SCNu64
-         ", delta-value? %" SCNu64 ") %" SCNu64 " filter block size %" SCNu64,
-         &tp->num_data_blocks, &tp->num_entries, &tp->num_deletions,
-         &tp->num_merge_operands, &tp->num_range_deletions, &tp->raw_key_size,
-         &dummy_double, &tp->raw_value_size, &dummy_double, &tp->data_size,
-         &tp->index_key_is_user_key, &tp->index_value_is_delta_encoded,
-         &tp->index_size, &tp->filter_size);
-}
-
 void VerifySimilar(uint64_t a, uint64_t b, double bias) {
   ASSERT_EQ(a == 0U, b == 0U);
   if (a == 0) {
@@ -1749,14 +1715,35 @@ TEST_F(DBPropertiesTest, SstFilesSize) {
     ASSERT_OK(Delete("key" + std::to_string(i)));
   }
   ASSERT_OK(Flush());
+
   uint64_t sst_size;
-  bool ok = db_->GetIntProperty(DB::Properties::kTotalSstFilesSize, &sst_size);
-  ASSERT_TRUE(ok);
+  ASSERT_TRUE(
+      db_->GetIntProperty(DB::Properties::kTotalSstFilesSize, &sst_size));
   ASSERT_GT(sst_size, 0);
   listener->size_before_compaction = sst_size;
+
+  uint64_t obsolete_sst_size;
+  ASSERT_TRUE(db_->GetIntProperty(DB::Properties::kObsoleteSstFilesSize,
+                                  &obsolete_sst_size));
+  ASSERT_EQ(obsolete_sst_size, 0);
+
+  // Hold files from being deleted so we can test property for size of obsolete
+  // SST files.
+  ASSERT_OK(db_->DisableFileDeletions());
+
   // Compact to clean all keys and trigger listener.
   ASSERT_OK(db_->CompactRange(CompactRangeOptions(), nullptr, nullptr));
   ASSERT_TRUE(listener->callback_triggered);
+
+  ASSERT_TRUE(db_->GetIntProperty(DB::Properties::kObsoleteSstFilesSize,
+                                  &obsolete_sst_size));
+  ASSERT_EQ(obsolete_sst_size, sst_size);
+
+  // Let the obsolete files be deleted.
+  ASSERT_OK(db_->EnableFileDeletions());
+  ASSERT_TRUE(db_->GetIntProperty(DB::Properties::kObsoleteSstFilesSize,
+                                  &obsolete_sst_size));
+  ASSERT_EQ(obsolete_sst_size, 0);
 }
 
 TEST_F(DBPropertiesTest, MinObsoleteSstNumberToKeep) {
@@ -2361,6 +2348,9 @@ TEST_F(DBPropertiesTest, TableMetaIndexKeys) {
         EXPECT_EQ("rocksdb.hashindex.prefixes",
                   PopMetaIndexKey(meta_iter.get()));
       }
+      if (bbto->format_version >= 6) {
+        EXPECT_EQ("rocksdb.index", PopMetaIndexKey(meta_iter.get()));
+      }
     }
     EXPECT_EQ("rocksdb.properties", PopMetaIndexKey(meta_iter.get()));
     EXPECT_EQ("NOT_FOUND", PopMetaIndexKey(meta_iter.get()));

package/deps/rocksdb/rocksdb/db/db_range_del_test.cc

@@ -682,6 +682,7 @@ TEST_F(DBRangeDelTest, TableEvictedDuringScan) {
     // soon as its refcount drops to zero.
     bbto.block_cache->EraseUnRefEntries();
   }
+  ASSERT_OK(iter->status());
   ASSERT_EQ(kNum, expected);
   delete iter;
   db_->ReleaseSnapshot(snapshot);
@@ -840,6 +841,7 @@ TEST_F(DBRangeDelTest, IteratorRemovesCoveredKeys) {
       ++expected;
     }
   }
+  ASSERT_OK(iter->status());
   ASSERT_EQ(kNum, expected);
   delete iter;
 }
@@ -908,6 +910,7 @@ TEST_F(DBRangeDelTest, IteratorIgnoresRangeDeletions) {
     std::string key;
     ASSERT_EQ(expected[i], iter->key());
   }
+  ASSERT_OK(iter->status());
   ASSERT_EQ(3, i);
   delete iter;
   db_->ReleaseSnapshot(snapshot);
@@ -1382,6 +1385,7 @@ TEST_F(DBRangeDelTest, UntruncatedTombstoneDoesNotDeleteNewerKey) {
     for (; iter->Valid(); iter->Next()) {
       ++keys_found;
     }
+    EXPECT_OK(iter->status());
     delete iter;
     return keys_found;
   };
@@ -1485,6 +1489,7 @@ TEST_F(DBRangeDelTest, DeletedMergeOperandReappearsIterPrev) {
   for (; iter->Valid(); iter->Prev()) {
     ++keys_found;
   }
+  ASSERT_OK(iter->status());
   delete iter;
   ASSERT_EQ(kNumKeys, keys_found);
 
@@ -1519,6 +1524,7 @@ TEST_F(DBRangeDelTest, SnapshotPreventsDroppedKeys) {
 
   iter->Next();
   ASSERT_FALSE(iter->Valid());
+  ASSERT_OK(iter->status());
 
   delete iter;
   db_->ReleaseSnapshot(snapshot);
@@ -1564,6 +1570,7 @@ TEST_F(DBRangeDelTest, SnapshotPreventsDroppedKeysInImmMemTables) {
 
   iter->Next();
   ASSERT_FALSE(iter->Valid());
+  ASSERT_OK(iter->status());
 }
 
 TEST_F(DBRangeDelTest, RangeTombstoneWrittenToMinimalSsts) {
@@ -1690,6 +1697,8 @@ TEST_F(DBRangeDelTest, LevelCompactOutputCutAtRangeTombstoneForTtlFiles) {
   ASSERT_EQ("0,1,0,1", FilesPerLevel());
 
   env_->MockSleepForSeconds(20 * 60 * 60);
+  // Prevent range tombstone from being dropped during compaction.
+  const Snapshot* snapshot = db_->GetSnapshot();
   ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(),
                              Key(11), Key(12)));
   ASSERT_OK(Put(Key(0), rnd.RandomString(1 << 10)));
@@ -1703,6 +1712,7 @@ TEST_F(DBRangeDelTest, LevelCompactOutputCutAtRangeTombstoneForTtlFiles) {
   // File 1: (qualified for TTL): Key(5) - Key(10)
   // File 1: DeleteRange [11, 12)
   ASSERT_EQ("0,3,0,1", FilesPerLevel());
+  db_->ReleaseSnapshot(snapshot);
 }
 
 // Test SST partitioner cut after every single key
@@ -1975,6 +1985,7 @@ TEST_F(DBRangeDelTest, IteratorRefresh) {
     ASSERT_EQ("key1", iter->key());
     iter->Next();
     ASSERT_FALSE(iter->Valid());
+    ASSERT_OK(iter->status());
 
     delete iter;
   }
@@ -3405,6 +3416,359 @@ TEST_F(DBRangeDelTest, AddRangeDelsSingleUserKeyTombstoneOnlyFile) {
   db_->ReleaseSnapshot(snapshot1);
 }
 
+TEST_F(DBRangeDelTest, NonBottommostCompactionDropRangetombstone) {
+  // L0: file 1: [DeleteRange[4, 5)], file 2: [3, 6, DeleteRange[8, 9)]
+  // L6 file 1: [2, 3], file 2: [7, 8]
+  // When compacting the two L0 files to L1, the compaction is non-bottommost
+  // since the compaction key range overlaps with L6 file 1. The range tombstone
+  // [4, 5) should be dropped since it does not overlap with any file in lower
+  // levels. The range tombstone [8, 9) should not be dropped.
+  Options opts = CurrentOptions();
+  opts.level_compaction_dynamic_level_bytes = false;
+  opts.num_levels = 7;
+  opts.level0_file_num_compaction_trigger = 3;
+  DestroyAndReopen(opts);
+
+  Random rnd(301);
+  // L6 file 1
+  ASSERT_OK(Put(Key(2), rnd.RandomString(100)));
+  ASSERT_OK(Put(Key(3), rnd.RandomString(100)));
+  ASSERT_OK(Flush());
+  // L6 file 2
+  ASSERT_OK(Put(Key(7), rnd.RandomString(100)));
+  ASSERT_OK(Put(Key(8), rnd.RandomString(100)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(6);
+  ASSERT_EQ(NumTableFilesAtLevel(6), 2);
+  // L0 file 1
+  ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(), Key(4),
+                             Key(5)));
+  ASSERT_OK(Flush());
+  // L0 file 2
+  ASSERT_OK(Put(Key(3), rnd.RandomString(100)));
+  ASSERT_OK(Put(Key(6), rnd.RandomString(100)));
+  ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(), Key(8),
+                             Key(9)));
+  ASSERT_OK(Flush());
+  // nothing is dropped during flush
+  std::string property;
+  db_->GetProperty(DB::Properties::kAggregatedTableProperties, &property);
+  TableProperties output_tp;
+  ParseTablePropertiesString(property, &output_tp);
+  ASSERT_EQ(output_tp.num_range_deletions, 2);
+  // Add one more L0 file to trigger L0->L1 compaction
+  ASSERT_OK(Put(Key(1), rnd.RandomString(100)));
+  ASSERT_OK(Put(Key(9), rnd.RandomString(100)));
+  ASSERT_OK(Flush());
+  ASSERT_OK(dbfull()->TEST_WaitForCompact());
+  ASSERT_EQ(NumTableFilesAtLevel(1), 1);
+  db_->GetProperty(DB::Properties::kAggregatedTableProperties, &property);
+  ParseTablePropertiesString(property, &output_tp);
+  ASSERT_EQ(output_tp.num_range_deletions, 1);
+
+  // Now create a snapshot protected range tombstone [4, 5), it should not
+  // be dropped.
+  ASSERT_OK(Put(Key(4), rnd.RandomString(100)));
+  const Snapshot* snapshot = db_->GetSnapshot();
+  ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(), Key(4),
+                             Key(5)));
+  CompactRangeOptions cro;
+  cro.bottommost_level_compaction = BottommostLevelCompaction::kForceOptimized;
+  ASSERT_OK(db_->CompactRange(cro, nullptr, nullptr));
+  // All compacted to L6
+  ASSERT_EQ("0,0,0,0,0,0,1", FilesPerLevel(0));
+  db_->GetProperty(DB::Properties::kAggregatedTableProperties, &property);
+  ParseTablePropertiesString(property, &output_tp);
+  ASSERT_EQ(output_tp.num_range_deletions, 1);
+  db_->ReleaseSnapshot(snapshot);
+}
+
+TEST_F(DBRangeDelTest, MemtableMaxRangeDeletions) {
+  // Tests option `memtable_max_range_deletions`.
+  Options options = CurrentOptions();
+  options.level_compaction_dynamic_file_size = false;
+  options.memtable_max_range_deletions = 50;
+  options.level0_file_num_compaction_trigger = 5;
+  DestroyAndReopen(options);
+
+  for (int i = 0; i < 50; ++i) {
+    // Intentionally delete overlapping ranges to see if the option
+    // checks number of range tombstone fragments instead.
+    ASSERT_OK(Put(Key(i), "val1"));
+    ASSERT_OK(Put(Key(i + 1), "val2"));
+    ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(),
+                               Key(i), Key(i + 2)));
+    ASSERT_OK(dbfull()->TEST_WaitForFlushMemTable());
+    ASSERT_EQ(0, NumTableFilesAtLevel(0));
+  }
+  // One more write to trigger flush.
+  ASSERT_OK(Put(Key(50), "val"));
+  ASSERT_OK(dbfull()->TEST_WaitForFlushMemTable());
+  ASSERT_EQ(1, NumTableFilesAtLevel(0));
+
+  // This should take effect for the next new memtable.
+  ASSERT_OK(db_->SetOptions({{"memtable_max_range_deletions", "1"}}));
+  ASSERT_OK(Flush());
+  ASSERT_EQ(2, NumTableFilesAtLevel(0));
+  ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(),
+                             Key(50), Key(100)));
+  ASSERT_OK(dbfull()->TEST_WaitForFlushMemTable());
+  ASSERT_EQ(2, NumTableFilesAtLevel(0));
+  // One more write to trigger flush.
+  ASSERT_OK(Put(Key(50), "new val"));
+  ASSERT_OK(dbfull()->TEST_WaitForFlushMemTable());
+  ASSERT_EQ(3, NumTableFilesAtLevel(0));
+}
+
+TEST_F(DBRangeDelTest, RangeDelReseekAfterFileReadError) {
+  // This is to test a bug that is fixed in
+  // https://github.com/facebook/rocksdb/pull/11786.
+  Options opts = CurrentOptions();
+  opts.num_levels = 7;
+
+  // Set up LSM
+  //
+  // L4: F1: [key1] F2: [key2]
+  // L5:   F3:[DeleteRange(key3, key6)]
+  // L6:               F4:[key3, key6]
+  // Will inject error when reading from F2.
+  // SeekToFirst() should land on key1.
+  // Next() should encounter error when reading from F2,
+  // and range del reseek should not reset this status.
+  Random rnd(301);
+  // L6
+  ASSERT_OK(Put(Key(3), rnd.RandomString(100)));
+  ASSERT_OK(Put(Key(6), rnd.RandomString(100)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(6);
+  // L5
+  ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(), Key(3),
+                             Key(6)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(5);
+  // L4
+  ASSERT_OK(Put(Key(2), rnd.RandomString(100)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(4);
+  std::string fname;
+  std::vector<LiveFileMetaData> live_files;
+  db_->GetLiveFilesMetaData(&live_files);
+  for (auto& meta : live_files) {
+    if (meta.level == 4) {
+      fname = meta.name;
+      break;
+    }
+  }
+  ASSERT_TRUE(!fname.empty());
+  ASSERT_OK(Put(Key(1), rnd.RandomString(100)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(4);
+
+  SyncPoint::GetInstance()->SetCallBack(
+      "RandomAccessFileReader::Read::BeforeReturn", [&fname](void* pair_ptr) {
+        auto p =
+            reinterpret_cast<std::pair<std::string*, IOStatus*>*>(pair_ptr);
+        if (p->first->find(fname) != std::string::npos) {
+          *p->second = IOStatus::IOError();
+          p->second->SetRetryable(true);
+        }
+      });
+  SyncPoint::GetInstance()->EnableProcessing();
+  std::unique_ptr<Iterator> iter{db_->NewIterator(ReadOptions())};
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_OK(iter->status());
+  ASSERT_EQ(iter->key(), Key(1));
+  iter->Next();
+  ASSERT_FALSE(iter->Valid());
+  ASSERT_NOK(iter->status());
+  ASSERT_TRUE(iter->status().IsIOError());
+  iter.reset();
+  SyncPoint::GetInstance()->ClearAllCallBacks();
+  SyncPoint::GetInstance()->DisableProcessing();
+
+  // Reverse scan
+  // LSM setup
+  // L4: F1: [key2]  F2: [key7, key8]
+  // L5:        F3:[[key3, key6)]
+  // L6:        F4:[key1, key5]
+  // Ingest error when read from F1.
+  // SeekToLast() should land on key8.
+  // During Prev(), MergingIterator will encounter error when reading from F1
+  // and do a range del reseek (it sees key5 covered by a range tombstone).
+  DestroyAndReopen(opts);
+  // L6
+  ASSERT_OK(Put(Key(1), rnd.RandomString(100)));
+  ASSERT_OK(Put(Key(5), rnd.RandomString(100)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(6);
+  // L5
+  ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(), Key(3),
+                             Key(6)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(5);
+  // L4
+  ASSERT_OK(Put(Key(2), rnd.RandomString(100)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(4);
+  live_files.clear();
+  db_->GetLiveFilesMetaData(&live_files);
+  for (auto& meta : live_files) {
+    if (meta.level == 4) {
+      fname = meta.name;
+      break;
+    }
+  }
+  ASSERT_TRUE(!fname.empty());
+  ASSERT_OK(Put(Key(7), rnd.RandomString(100)));
+  ASSERT_OK(Put(Key(8), rnd.RandomString(100)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(4);
+
+  SyncPoint::GetInstance()->SetCallBack(
+      "RandomAccessFileReader::Read::AnyOffset", [&fname](void* pair_ptr) {
+        auto p =
+            reinterpret_cast<std::pair<std::string*, IOStatus*>*>(pair_ptr);
+        if (p->first->find(fname) != std::string::npos) {
+          *p->second = IOStatus::IOError();
+          p->second->SetRetryable(true);
+        }
+      });
+  SyncPoint::GetInstance()->EnableProcessing();
+  iter.reset(db_->NewIterator(ReadOptions()));
+  iter->SeekToLast();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_OK(iter->status());
+  ASSERT_EQ(iter->key(), Key(8));
+  // Note that for reverse scan, DBIter will need to ensure
+  // the key it returns is the one with the highest sequence number.
+  // To return key7, it internally calls MergingIterator::Prev()
+  // until it reaches a previous user key.
+  iter->Prev();
+  ASSERT_FALSE(iter->Valid());
+  ASSERT_NOK(iter->status());
+  ASSERT_TRUE(iter->status().IsIOError());
+
+  iter.reset();
+}
+
+TEST_F(DBRangeDelTest, ReleaseSnapshotAfterIteratorCreation) {
+  // Test that range tombstone code path in LevelIterator
+  // does access ReadOptions::snapshot after Iterator creation.
+  //
+  // Put some data in L2 so that range tombstone in L1 will not be dropped.
+  ASSERT_OK(Put(Key(0), "v"));
+  ASSERT_OK(Put(Key(100), "v"));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(2);
+
+  // two L1 file with range del
+  ASSERT_OK(Put(Key(1), "v"));
+  ASSERT_OK(Put(Key(2), "v"));
+  ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(), Key(3),
+                             Key(4)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(1);
+
+  ASSERT_OK(Put(Key(5), "v"));
+  ASSERT_OK(Put(Key(6), "v"));
+  ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(), Key(5),
+                             Key(6)));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(1);
+
+  ASSERT_EQ(2, NumTableFilesAtLevel(1));
+  ASSERT_EQ(1, NumTableFilesAtLevel(2));
+
+  const Snapshot* snapshot = db_->GetSnapshot();
+  ReadOptions ro;
+  ro.snapshot = snapshot;
+
+  Iterator* iter = db_->NewIterator(ro);
+  db_->ReleaseSnapshot(snapshot);
+
+  iter->Seek(Key(1));
+  std::vector<int> expected_keys{1, 2, 6, 100};
+  for (int i : expected_keys) {
+    ASSERT_OK(iter->status());
+    ASSERT_TRUE(iter->Valid());
+    ASSERT_EQ(iter->key(), Key(i));
+    iter->Next();
+  }
+  ASSERT_TRUE(!iter->Valid() && iter->status().ok());
+
+  delete iter;
+}
+
+TEST_F(DBRangeDelTest, RefreshWithSnapshot) {
+  ASSERT_OK(Put(Key(4), "4"));
+  ASSERT_OK(Put(Key(6), "6"));
+  const Snapshot* snapshot = db_->GetSnapshot();
+  ASSERT_OK(db_->DeleteRange(WriteOptions(), db_->DefaultColumnFamily(), Key(3),
+                             Key(5)));
+
+  std::unique_ptr<Iterator> iter{db_->NewIterator(ReadOptions())};
+  // Live Memtable
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_EQ(iter->key(), Key(6));
+  ASSERT_OK(iter->Refresh(snapshot));
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_EQ(iter->key(), Key(4));
+  // Immutable Memtable
+  ASSERT_OK(dbfull()->TEST_SwitchMemtable());
+  ASSERT_OK(iter->Refresh(nullptr));
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_EQ(iter->key(), Key(6));
+  ASSERT_OK(iter->Refresh(snapshot));
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_EQ(iter->key(), Key(4));
+  // L0
+  ASSERT_OK(Flush());
+  ASSERT_EQ(1, NumTableFilesAtLevel(0));
+  ASSERT_OK(iter->Refresh(nullptr));
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_EQ(iter->key(), Key(6));
+  ASSERT_OK(iter->Refresh(snapshot));
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_EQ(iter->key(), Key(4));
+  // L1
+  MoveFilesToLevel(1);
+  ASSERT_EQ(1, NumTableFilesAtLevel(1));
+  ASSERT_OK(iter->Refresh(nullptr));
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_EQ(iter->key(), Key(6));
+  ASSERT_OK(iter->Refresh(snapshot));
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_EQ(iter->key(), Key(4));
+  // L1 with two file.
+  // Test that when LevelIterator enters a new file,
+  // it remembers which snapshot sequence number to use.
+  ASSERT_OK(Put(Key(2), "2"));
+  ASSERT_OK(Flush());
+  MoveFilesToLevel(1);
+  ASSERT_EQ(2, NumTableFilesAtLevel(1));
+  ASSERT_OK(iter->Refresh(nullptr));
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  // LevelIterator is at the first file
+  ASSERT_EQ(iter->key(), Key(2));
+  ASSERT_OK(iter->Refresh(snapshot));
+  // Will enter the second file, and create a new range tombstone iterator.
+  // It should use the snapshot sequence number.
+  iter->SeekToFirst();
+  ASSERT_TRUE(iter->Valid());
+  ASSERT_EQ(iter->key(), Key(4));
+  iter.reset();
+  db_->ReleaseSnapshot(snapshot);
+}
 }  // namespace ROCKSDB_NAMESPACE
 
 int main(int argc, char** argv) {

package/deps/rocksdb/rocksdb/db/db_rate_limiter_test.cc

@@ -220,6 +220,7 @@ TEST_P(DBRateLimiterOnReadTest, Iterator) {
       ++expected;
     }
   }
+  ASSERT_OK(iter->status());
   // Reverse scan does not read evenly (one block per iteration) due to
   // descending seqno ordering, so wait until after the loop to check total.
   ASSERT_EQ(expected, options_.rate_limiter->GetTotalRequests(Env::IO_USER));
@@ -235,9 +236,18 @@ TEST_P(DBRateLimiterOnReadTest, VerifyChecksum) {
   ASSERT_EQ(0, options_.rate_limiter->GetTotalRequests(Env::IO_USER));
 
   ASSERT_OK(db_->VerifyChecksum(GetReadOptions()));
-  // There are 3 reads per file: ReadMetaIndexBlock,
-  // VerifyChecksumInMetaBlocks, VerifyChecksumInBlocks
-  int expected = kNumFiles * 3;
+  // In BufferedIO,
+  // there are 7 reads per file, each of which will be rate-limited.
+  // During open:  read footer, meta index block, properties block, index block.
+  // During actual checksum verification: read meta index block, verify checksum
+  // in meta blocks and verify checksum in file blocks.
+  //
+  // In DirectIO, where we support tail prefetching, during table open, we only
+  // do 1 read instead of 4 as described above. Actual checksum verification
+  // reads stay the same.
+  int num_read_per_file = (!use_direct_io_) ? 7 : 4;
+  int expected = kNumFiles * num_read_per_file;
+
   ASSERT_EQ(expected, options_.rate_limiter->GetTotalRequests(Env::IO_USER));
 }
 

package/deps/rocksdb/rocksdb/db/db_readonly_with_timestamp_test.cc

@@ -240,6 +240,7 @@ TEST_F(DBReadOnlyTestWithTimestamp, IteratorAndGet) {
       get_value_and_check(db_, read_opts, it->key(), it->value(),
                           write_timestamps[i]);
     }
+    ASSERT_OK(it->status());
     size_t expected_count = kMaxKey - start_keys[i] + 1;
     ASSERT_EQ(expected_count, count);
 
@@ -252,6 +253,7 @@ TEST_F(DBReadOnlyTestWithTimestamp, IteratorAndGet) {
       get_value_and_check(db_, read_opts, it->key(), it->value(),
                           write_timestamps[i]);
     }
+    ASSERT_OK(it->status());
     ASSERT_EQ(static_cast<size_t>(kMaxKey) - start_keys[i] + 1, count);
 
     // SeekToFirst()/SeekToLast() with lower/upper bounds.
@@ -273,6 +275,7 @@ TEST_F(DBReadOnlyTestWithTimestamp, IteratorAndGet) {
         get_value_and_check(db_, read_opts, it->key(), it->value(),
                             write_timestamps[i]);
       }
+      ASSERT_OK(it->status());
       ASSERT_EQ(r - std::max(l, start_keys[i]), count);
 
       for (it->SeekToLast(), key = std::min(r, kMaxKey + 1), count = 0;
@@ -282,6 +285,7 @@ TEST_F(DBReadOnlyTestWithTimestamp, IteratorAndGet) {
         get_value_and_check(db_, read_opts, it->key(), it->value(),
                             write_timestamps[i]);
       }
+      ASSERT_OK(it->status());
       l += (kMaxKey / 100);
       r -= (kMaxKey / 100);
     }
@@ -328,6 +332,7 @@ TEST_F(DBReadOnlyTestWithTimestamp, Iterators) {
     CheckIterUserEntry(iters[0], Key1(key), kTypeValue,
                        "value" + std::to_string(key), write_timestamp);
   }
+  ASSERT_OK(iters[0]->status());
 
   size_t expected_count = kMaxKey - 0 + 1;
   ASSERT_EQ(expected_count, count);
@@ -336,6 +341,53 @@ TEST_F(DBReadOnlyTestWithTimestamp, Iterators) {
   Close();
 }
 
+TEST_F(DBReadOnlyTestWithTimestamp, FullHistoryTsLowSanityCheckFail) {
+  Options options = CurrentOptions();
+  options.env = env_;
+  options.comparator = test::BytewiseComparatorWithU64TsWrapper();
+  // Use UDT in memtable only feature for this test, so we can control that
+  // newly set `full_history_ts_low` collapse history when Flush happens.
+  options.persist_user_defined_timestamps = false;
+  options.allow_concurrent_memtable_write = false;
+  DestroyAndReopen(options);
+
+  std::string write_ts;
+  PutFixed64(&write_ts, 1);
+  ASSERT_OK(db_->Put(WriteOptions(), "foo", write_ts, "val1"));
+
+  std::string full_history_ts_low;
+  PutFixed64(&full_history_ts_low, 3);
+  ASSERT_OK(db_->IncreaseFullHistoryTsLow(db_->DefaultColumnFamily(),
+                                          full_history_ts_low));
+  ASSERT_OK(Flush(0));
+
+  // Reopen the database in read only mode to test its timestamp support.
+  Close();
+  ASSERT_OK(ReadOnlyReopen(options));
+
+  // Reading below full_history_ts_low fails a sanity check.
+  std::string read_ts;
+  PutFixed64(&read_ts, 2);
+  Slice read_ts_slice = read_ts;
+  ReadOptions read_opts;
+  read_opts.timestamp = &read_ts_slice;
+
+  // Get()
+  std::string value;
+  ASSERT_TRUE(db_->Get(read_opts, "foo", &value).IsInvalidArgument());
+  // NewIterator()
+  std::unique_ptr<Iterator> iter(
+      db_->NewIterator(read_opts, db_->DefaultColumnFamily()));
+  ASSERT_TRUE(iter->status().IsInvalidArgument());
+
+  // NewIterators()
+  std::vector<ColumnFamilyHandle*> cfhs = {db_->DefaultColumnFamily()};
+  std::vector<Iterator*> iterators;
+  ASSERT_TRUE(
+      db_->NewIterators(read_opts, cfhs, &iterators).IsInvalidArgument());
+  Close();
+}
+
 TEST_F(DBReadOnlyTestWithTimestamp, IteratorsReadTimestampSizeMismatch) {
   const int kNumKeysPerFile = 128;
   const uint64_t kMaxKey = 1024;