rocksdb-native 2.2.0 → 2.3.0

package/vendor/librocksdb/vendor/rocksdb/db/compaction/compaction_picker_universal.cc

@@ -227,6 +227,7 @@ class UniversalCompactionBuilder {
   const InternalKeyComparator* icmp_;
   double score_;
   std::vector<SortedRun> sorted_runs_;
+  uint64_t max_run_size_;
   const std::string& cf_name_;
   const MutableCFOptions& mutable_cf_options_;
   const MutableDBOptions& mutable_db_options_;
@@ -235,7 +236,8 @@ class UniversalCompactionBuilder {
   LogBuffer* log_buffer_;
 
   static std::vector<UniversalCompactionBuilder::SortedRun> CalculateSortedRuns(
-      const VersionStorageInfo& vstorage, int last_level);
+      const VersionStorageInfo& vstorage, int last_level,
+      uint64_t* max_run_size);
 
   // Pick a path ID to place a newly generated file, with its estimated file
   // size.
@@ -440,11 +442,15 @@ void UniversalCompactionBuilder::SortedRun::DumpSizeInfo(
 
 std::vector<UniversalCompactionBuilder::SortedRun>
 UniversalCompactionBuilder::CalculateSortedRuns(
-    const VersionStorageInfo& vstorage, int last_level) {
+    const VersionStorageInfo& vstorage, int last_level,
+    uint64_t* max_run_size) {
+  assert(max_run_size);
+  *max_run_size = 0;
   std::vector<UniversalCompactionBuilder::SortedRun> ret;
   for (FileMetaData* f : vstorage.LevelFiles(0)) {
     ret.emplace_back(0, f, f->fd.GetFileSize(), f->compensated_file_size,
                      f->being_compacted);
+    *max_run_size = std::max(*max_run_size, f->fd.GetFileSize());
   }
   for (int level = 1; level <= last_level; level++) {
     uint64_t total_compensated_size = 0U;
@@ -466,6 +472,7 @@ UniversalCompactionBuilder::CalculateSortedRuns(
       ret.emplace_back(level, nullptr, total_size, total_compensated_size,
                        being_compacted);
     }
+    *max_run_size = std::max(*max_run_size, total_size);
   }
   return ret;
 }
@@ -477,13 +484,16 @@ Compaction* UniversalCompactionBuilder::PickCompaction() {
   score_ = vstorage_->CompactionScore(kLevel0);
   int max_output_level =
       vstorage_->MaxOutputLevel(ioptions_.allow_ingest_behind);
-  sorted_runs_ = CalculateSortedRuns(*vstorage_, max_output_level);
+  max_run_size_ = 0;
+  sorted_runs_ =
+      CalculateSortedRuns(*vstorage_, max_output_level, &max_run_size_);
+  int file_num_compaction_trigger =
+      mutable_cf_options_.level0_file_num_compaction_trigger;
 
   if (sorted_runs_.size() == 0 ||
       (vstorage_->FilesMarkedForPeriodicCompaction().empty() &&
        vstorage_->FilesMarkedForCompaction().empty() &&
-       sorted_runs_.size() < (unsigned int)mutable_cf_options_
-                                 .level0_file_num_compaction_trigger)) {
+       sorted_runs_.size() < (unsigned int)file_num_compaction_trigger)) {
     ROCKS_LOG_BUFFER(log_buffer_, "[%s] Universal: nothing to do\n",
                      cf_name_.c_str());
     TEST_SYNC_POINT_CALLBACK(
@@ -505,11 +515,9 @@ Compaction* UniversalCompactionBuilder::PickCompaction() {
     TEST_SYNC_POINT_CALLBACK("PostPickPeriodicCompaction", c);
   }
 
-  // Check for size amplification.
   if (c == nullptr &&
-      sorted_runs_.size() >=
-          static_cast<size_t>(
-              mutable_cf_options_.level0_file_num_compaction_trigger)) {
+      sorted_runs_.size() >= static_cast<size_t>(file_num_compaction_trigger)) {
+    // Check for size amplification.
     if ((c = PickCompactionToReduceSizeAmp()) != nullptr) {
       TEST_SYNC_POINT("PickCompactionToReduceSizeAmpReturnNonnullptr");
       ROCKS_LOG_BUFFER(log_buffer_, "[%s] Universal: compacting for size amp\n",
@@ -527,13 +535,48 @@ Compaction* UniversalCompactionBuilder::PickCompaction() {
                          cf_name_.c_str());
       } else {
         // Size amplification and file size ratios are within configured limits.
-        // If max read amplification is exceeding configured limits, then force
-        // compaction without looking at filesize ratios and try to reduce
-        // the number of files to fewer than level0_file_num_compaction_trigger.
+        // If max read amplification exceeds configured limits, then force
+        // compaction to reduce the number sorted runs without looking at file
+        // size ratios.
+
         // This is guaranteed by NeedsCompaction()
         assert(sorted_runs_.size() >=
-               static_cast<size_t>(
-                   mutable_cf_options_.level0_file_num_compaction_trigger));
+               static_cast<size_t>(file_num_compaction_trigger));
+        int max_num_runs =
+            mutable_cf_options_.compaction_options_universal.max_read_amp;
+        if (max_num_runs < 0) {
+          // any value < -1 is not valid
+          assert(max_num_runs == -1);
+          // By default, fall back to `level0_file_num_compaction_trigger`
+          max_num_runs = file_num_compaction_trigger;
+        } else if (max_num_runs == 0) {
+          if (mutable_cf_options_.compaction_options_universal.stop_style ==
+              kCompactionStopStyleTotalSize) {
+            // 0 means auto-tuning by RocksDB. We estimate max num run based on
+            // max_run_size, size_ratio and write buffer size:
+            // Assume the size of the lowest level size is equal to
+            // write_buffer_size. Each subsequent level is the max size without
+            // triggering size_ratio compaction. `max_num_runs` is the minimum
+            // number of levels required such that the target size of the
+            // largest level is at least `max_run_size_`.
+            max_num_runs = 1;
+            double cur_level_max_size =
+                static_cast<double>(mutable_cf_options_.write_buffer_size);
+            double total_run_size = 0;
+            while (cur_level_max_size < static_cast<double>(max_run_size_)) {
+              // This loop should not take too many iterations since
+              // cur_level_max_size at least doubles each iteration.
+              total_run_size += cur_level_max_size;
+              cur_level_max_size = (100.0 + ratio) / 100.0 * total_run_size;
+              ++max_num_runs;
+            }
+          } else {
+            // TODO: implement the auto-tune logic for this stop style
+            max_num_runs = file_num_compaction_trigger;
+          }
+        } else {
+          // max_num_runs > 0, it's the limit on the number of sorted run
+        }
         // Get the total number of sorted runs that are not being compacted
         int num_sr_not_compacted = 0;
         for (size_t i = 0; i < sorted_runs_.size(); i++) {
@@ -544,17 +587,25 @@ Compaction* UniversalCompactionBuilder::PickCompaction() {
 
         // The number of sorted runs that are not being compacted is greater
         // than the maximum allowed number of sorted runs
-        if (num_sr_not_compacted >
-            mutable_cf_options_.level0_file_num_compaction_trigger) {
-          unsigned int num_files =
-              num_sr_not_compacted -
-              mutable_cf_options_.level0_file_num_compaction_trigger + 1;
+        if (num_sr_not_compacted > max_num_runs) {
+          unsigned int num_files = num_sr_not_compacted - max_num_runs + 1;
           if ((c = PickCompactionToReduceSortedRuns(UINT_MAX, num_files)) !=
               nullptr) {
             ROCKS_LOG_BUFFER(log_buffer_,
-                             "[%s] Universal: compacting for file num -- %u\n",
-                             cf_name_.c_str(), num_files);
+                             "[%s] Universal: compacting for file num, to "
+                             "compact file num -- %u, max num runs allowed"
+                             "-- %d, max_run_size -- %" PRIu64 "\n",
+                             cf_name_.c_str(), num_files, max_num_runs,
+                             max_run_size_);
           }
+        } else {
+          ROCKS_LOG_BUFFER(
+              log_buffer_,
+              "[%s] Universal: skipping compaction for file num, num runs not "
+              "being compacted -- %u, max num runs allowed -- %d, max_run_size "
+              "-- %" PRIu64 "\n",
+              cf_name_.c_str(), num_sr_not_compacted, max_num_runs,
+              max_run_size_);
         }
       }
     }
@@ -1139,7 +1190,8 @@ Compaction* UniversalCompactionBuilder::PickIncrementalForReduceSizeAmp(
   // from bottom_start_idx and bottom_end_idx, but for now, we use
   // SetupOtherInputs() for simplicity.
   int parent_index = -1;  // Create and use bottom_start_idx?
-  if (!picker_->SetupOtherInputs(cf_name_, vstorage_, &second_last_level_inputs,
+  if (!picker_->SetupOtherInputs(cf_name_, mutable_cf_options_, vstorage_,
+                                 &second_last_level_inputs,
                                  &bottom_level_inputs, &parent_index,
                                  /*base_index=*/-1)) {
     return nullptr;
@@ -1310,8 +1362,9 @@ Compaction* UniversalCompactionBuilder::PickDeleteTriggeredCompaction() {
       int parent_index = -1;
 
       output_level_inputs.level = output_level;
-      if (!picker_->SetupOtherInputs(cf_name_, vstorage_, &start_level_inputs,
-                                     &output_level_inputs, &parent_index, -1)) {
+      if (!picker_->SetupOtherInputs(cf_name_, mutable_cf_options_, vstorage_,
+                                     &start_level_inputs, &output_level_inputs,
+                                     &parent_index, -1)) {
         return nullptr;
       }
       inputs.push_back(start_level_inputs);

package/vendor/librocksdb/vendor/rocksdb/db/compaction/compaction_service_job.cc

@@ -275,6 +275,8 @@ Status CompactionServiceCompactionJob::Run() {
   log_buffer_->FlushBufferToLog();
   LogCompaction();
   const uint64_t start_micros = db_options_.clock->NowMicros();
+  c->GetOrInitInputTableProperties();
+
   // Pick the only sub-compaction we should have
   assert(compact_->sub_compact_states.size() == 1);
   SubcompactionState* sub_compact = compact_->sub_compact_states.data();

package/vendor/librocksdb/vendor/rocksdb/db/convenience.cc

@@ -71,6 +71,9 @@ Status VerifySstFileChecksumInternal(const Options& options,
   } else {
     return s;
   }
+  if (!s.ok()) {
+    return s;
+  }
   std::unique_ptr<TableReader> table_reader;
   std::unique_ptr<RandomAccessFileReader> file_reader(
       new RandomAccessFileReader(

package/vendor/librocksdb/vendor/rocksdb/db/db_filesnapshot.cc

@@ -21,6 +21,7 @@
 #include "rocksdb/db.h"
 #include "rocksdb/env.h"
 #include "rocksdb/metadata.h"
+#include "rocksdb/transaction_log.h"
 #include "rocksdb/types.h"
 #include "test_util/sync_point.h"
 #include "util/file_checksum_helper.h"
@@ -91,7 +92,12 @@ Status DBImpl::GetLiveFiles(std::vector<std::string>& ret,
   return Status::OK();
 }
 
-Status DBImpl::GetSortedWalFiles(VectorLogPtr& files) {
+Status DBImpl::GetSortedWalFiles(VectorWalPtr& files) {
+  return GetSortedWalFilesImpl(files,
+                               /*need_seqnos*/ true);
+}
+
+Status DBImpl::GetSortedWalFilesImpl(VectorWalPtr& files, bool need_seqnos) {
   // Record tracked WALs as a (minimum) cross-check for directory scan
   std::vector<uint64_t> required_by_manifest;
 
@@ -117,7 +123,10 @@ Status DBImpl::GetSortedWalFiles(VectorLogPtr& files) {
     }
   }
 
-  Status s = wal_manager_.GetSortedWalFiles(files);
+  // NOTE: need to include archived WALs because needed WALs might have been
+  // archived since getting required_by_manifest set
+  Status s = wal_manager_.GetSortedWalFiles(files, need_seqnos,
+                                            /*include_archived*/ true);
 
   // DisableFileDeletions / EnableFileDeletions not supported in read-only DB
   if (deletions_disabled.ok()) {
@@ -151,10 +160,33 @@ Status DBImpl::GetSortedWalFiles(VectorLogPtr& files) {
     }
   }
 
+  if (s.ok()) {
+    size_t wal_count = files.size();
+    ROCKS_LOG_INFO(immutable_db_options_.info_log,
+                   "Number of WAL files %" ROCKSDB_PRIszt " (%" ROCKSDB_PRIszt
+                   " required by manifest)",
+                   wal_count, required_by_manifest.size());
+#ifndef NDEBUG
+    std::ostringstream wal_names;
+    for (const auto& wal : files) {
+      wal_names << wal->PathName() << " ";
+    }
+
+    std::ostringstream wal_required_by_manifest_names;
+    for (const auto& wal : required_by_manifest) {
+      wal_required_by_manifest_names << wal << ".log ";
+    }
+
+    ROCKS_LOG_INFO(immutable_db_options_.info_log,
+                   "Log files : %s .Log files required by manifest: %s.",
+                   wal_names.str().c_str(),
+                   wal_required_by_manifest_names.str().c_str());
+#endif  // NDEBUG
+  }
   return s;
 }
 
-Status DBImpl::GetCurrentWalFile(std::unique_ptr<LogFile>* current_log_file) {
+Status DBImpl::GetCurrentWalFile(std::unique_ptr<WalFile>* current_log_file) {
   uint64_t current_logfile_number;
   {
     InstrumentedMutexLock l(&mutex_);
@@ -175,14 +207,20 @@ Status DBImpl::GetLiveFilesStorageInfo(
   // NOTE: This implementation was largely migrated from Checkpoint.
 
   Status s;
-  VectorLogPtr live_wal_files;
+  VectorWalPtr live_wal_files;
   bool flush_memtable = true;
   if (!immutable_db_options_.allow_2pc) {
     if (opts.wal_size_for_flush == std::numeric_limits<uint64_t>::max()) {
       flush_memtable = false;
     } else if (opts.wal_size_for_flush > 0) {
-      // If the outstanding log files are small, we skip the flush.
-      s = GetSortedWalFiles(live_wal_files);
+      // FIXME: avoid querying the filesystem for current WAL state
+      // If the outstanding WAL files are small, we skip the flush.
+      // Don't take archived log size into account when calculating wal
+      // size for flush, and don't need to verify consistency with manifest
+      // here & now.
+      s = wal_manager_.GetSortedWalFiles(live_wal_files,
+                                         /* need_seqnos */ false,
+                                         /*include_archived*/ false);
 
       if (!s.ok()) {
         return s;
@@ -193,6 +231,7 @@ Status DBImpl::GetLiveFilesStorageInfo(
       // We may be able to cover 2PC case too.
       uint64_t total_wal_size = 0;
       for (auto& wal : live_wal_files) {
+        assert(wal->Type() == kAliveLogFile);
         total_wal_size += wal->SizeFileBytes();
       }
       if (total_wal_size < opts.wal_size_for_flush) {
@@ -206,12 +245,18 @@ Status DBImpl::GetLiveFilesStorageInfo(
   // metadata.
   mutex_.Lock();
   if (flush_memtable) {
-    Status status = FlushForGetLiveFiles();
-    if (!status.ok()) {
-      mutex_.Unlock();
-      ROCKS_LOG_ERROR(immutable_db_options_.info_log, "Cannot Flush data %s\n",
-                      status.ToString().c_str());
-      return status;
+    bool wal_locked = lock_wal_count_ > 0;
+    if (wal_locked) {
+      ROCKS_LOG_INFO(immutable_db_options_.info_log,
+                     "Can't FlushForGetLiveFiles while WAL is locked");
+    } else {
+      Status status = FlushForGetLiveFiles();
+      if (!status.ok()) {
+        mutex_.Unlock();
+        ROCKS_LOG_ERROR(immutable_db_options_.info_log,
+                        "Cannot Flush data %s\n", status.ToString().c_str());
+        return status;
+      }
     }
   }
 
@@ -287,6 +332,8 @@ Status DBImpl::GetLiveFilesStorageInfo(
   const uint64_t options_number = versions_->options_file_number();
   const uint64_t options_size = versions_->options_file_size_;
   const uint64_t min_log_num = MinLogNumberToKeep();
+  // Ensure consistency with manifest for track_and_verify_wals_in_manifest
+  const uint64_t max_log_num = logfile_number_;
 
   mutex_.Unlock();
 
@@ -350,10 +397,13 @@ Status DBImpl::GetLiveFilesStorageInfo(
   TEST_SYNC_POINT("CheckpointImpl::CreateCheckpoint:SavedLiveFiles2");
 
   if (s.ok()) {
-    // To maximize the effectiveness of track_and_verify_wals_in_manifest,
-    // sync WAL when it is enabled.
-    s = FlushWAL(
-        immutable_db_options_.track_and_verify_wals_in_manifest /* sync */);
+    // FlushWAL is required to ensure we can physically copy everything
+    // logically written to the WAL. (Sync not strictly required for
+    // active WAL to be copied rather than hard linked, even when
+    // Checkpoint guarantees that the copied-to file is sync-ed. Plus we can't
+    // help track_and_verify_wals_in_manifest after manifest_size is
+    // already determined.)
+    s = FlushWAL(/*sync=*/false);
     if (s.IsNotSupported()) {  // read-only DB or similar
       s = Status::OK();
     }
@@ -362,25 +412,52 @@ Status DBImpl::GetLiveFilesStorageInfo(
   TEST_SYNC_POINT("CheckpointImpl::CreateCustomCheckpoint:AfterGetLive1");
   TEST_SYNC_POINT("CheckpointImpl::CreateCustomCheckpoint:AfterGetLive2");
 
-  // If we have more than one column family, we also need to get WAL files.
+  // Even after WAL flush, there could be multiple WALs that are not
+  // fully synced. Although the output DB of a Checkpoint or Backup needs
+  // to be fully synced on return, we don't strictly need to sync this
+  // DB (the input DB). If we allow Checkpoint to hard link an inactive
+  // WAL that isn't fully synced, that could result in an insufficiently
+  // sync-ed Checkpoint. Here we get the set of WALs that are potentially
+  // unsynced or still being written to, to prevent them from being hard
+  // linked. Enforcing max_log_num from above ensures any new WALs after
+  // GetOpenWalSizes() and before GetSortedWalFiles() are not included in
+  // the results.
+  // NOTE: we might still hard link a file that is open for writing, even
+  // if we don't do any more writes to it.
+  //
+  // In a step toward reducing unnecessary file metadata queries, we also
+  // get and use our known flushed sizes for those WALs.
+  // FIXME: eventually we should not be using filesystem queries at all for
+  // the required set of WAL files.
+  //
+  // However for recycled log files, we just copy the whole file,
+  // for better or worse.
+  //
+  std::map<uint64_t, uint64_t> open_wal_number_to_size;
+  bool recycling_log_files = immutable_db_options_.recycle_log_file_num > 0;
+  if (s.ok() && !recycling_log_files) {
+    s = GetOpenWalSizes(open_wal_number_to_size);
+  }
+
+  // [old comment] If we have more than one column family, we also need to get
+  // WAL files.
   if (s.ok()) {
-    s = GetSortedWalFiles(live_wal_files);
+    // FIXME: avoid querying the filesystem for current WAL state
+    s = GetSortedWalFilesImpl(live_wal_files,
+                              /* need_seqnos */ false);
   }
   if (!s.ok()) {
     return s;
   }
 
-  size_t wal_size = live_wal_files.size();
-
-  ROCKS_LOG_INFO(immutable_db_options_.info_log,
-                 "Number of log files %" ROCKSDB_PRIszt, live_wal_files.size());
-
+  size_t wal_count = live_wal_files.size();
   // Link WAL files. Copy exact size of last one because it is the only one
   // that has changes after the last flush.
   auto wal_dir = immutable_db_options_.GetWalDir();
-  for (size_t i = 0; s.ok() && i < wal_size; ++i) {
+  for (size_t i = 0; s.ok() && i < wal_count; ++i) {
     if ((live_wal_files[i]->Type() == kAliveLogFile) &&
-        (!flush_memtable || live_wal_files[i]->LogNumber() >= min_log_num)) {
+        (!flush_memtable || live_wal_files[i]->LogNumber() >= min_log_num) &&
+        live_wal_files[i]->LogNumber() <= max_log_num) {
       results.emplace_back();
       LiveFileStorageInfo& info = results.back();
       auto f = live_wal_files[i]->PathName();
@@ -389,12 +466,29 @@ Status DBImpl::GetLiveFilesStorageInfo(
       info.directory = wal_dir;
       info.file_number = live_wal_files[i]->LogNumber();
       info.file_type = kWalFile;
-      info.size = live_wal_files[i]->SizeFileBytes();
-      // Trim the log either if its the last one, or log file recycling is
-      // enabled. In the latter case, a hard link doesn't prevent the file
-      // from being renamed and recycled. So we need to copy it instead.
-      info.trim_to_size = (i + 1 == wal_size) ||
-                          (immutable_db_options_.recycle_log_file_num > 0);
+      if (recycling_log_files) {
+        info.size = live_wal_files[i]->SizeFileBytes();
+        // Recyclable WAL files must be copied instead of hard linked
+        info.trim_to_size = true;
+      } else {
+        auto it = open_wal_number_to_size.find(info.file_number);
+        if (it == open_wal_number_to_size.end()) {
+          // Known fully synced and no future writes (in part from
+          // max_log_num check). Ok to hard link
+          info.size = live_wal_files[i]->SizeFileBytes();
+          assert(!info.trim_to_size);
+        } else {
+          // Marked as (possibly) still open -> use our known flushed size
+          // and force file copy instead of hard link
+          info.size = it->second;
+          info.trim_to_size = true;
+          // FIXME: this is needed as long as db_stress uses
+          // SetReadUnsyncedData(false), because it will only be able to
+          // copy the synced portion of the WAL, which under
+          // SetReadUnsyncedData(false) is given by the reported file size.
+          info.size = std::min(info.size, live_wal_files[i]->SizeFileBytes());
+        }
+      }
       if (opts.include_checksum_info) {
         info.file_checksum_func_name = kUnknownFileChecksumFuncName;
         info.file_checksum = kUnknownFileChecksum;