Fix AddFile() conflict with compaction output [WaitForAddFile()]

Summary:
Since AddFile unlock/lock the mutex inside LogAndApply() we need to ensure that during this period other compactions cannot run since such compactions are not aware of the file we are ingesting and could create a compaction that overlap wit this file

this diff add
- WaitForAddFile() call that will ensure that no AddFile() calls are being processed right now
- Call `WaitForAddFile()` in 3 locations
-- When doing manual Compaction
-- When starting automatic Compaction
-- When  doing CompactFiles()

Test Plan: unit test

Reviewers: lightmark, yiwu, andrewkr, sdong

Reviewed By: sdong

Subscribers: andrewkr, yoshinorim, jkedgar, dhruba

Differential Revision: https://reviews.facebook.net/D64383

db/db_impl.cc

@@ -341,6 +341,8 @@ DBImpl::DBImpl(const DBOptions& options, const std::string& dbname)
       next_job_id_(1),
       has_unpersisted_data_(false),
       env_options_(db_options_),
+      num_running_addfile_(0),
+      addfile_cv_(&mutex_),
 #ifndef ROCKSDB_LITE
       wal_manager_(db_options_, env_options_),
 #endif  // ROCKSDB_LITE
@@ -1994,6 +1996,7 @@ Status DBImpl::CompactRange(const CompactRangeOptions& options,
   int max_level_with_files = 0;
   {
     InstrumentedMutexLock l(&mutex_);
+    WaitForAddFile();
     Version* base = cfd->current();
     for (int level = 1; level < base->storage_info()->num_non_empty_levels();
          level++) {
@@ -2109,6 +2112,10 @@ Status DBImpl::CompactFiles(
   {
     InstrumentedMutexLock l(&mutex_);
 
+    // This call will unlock/lock the mutex to wait for current running
+    // AddFile() calls to finish.
+    WaitForAddFile();
+
     s = CompactFilesImpl(compact_options, cfd, sv->current,
                          input_file_names, output_level,
                          output_path_id, &job_context, &log_buffer);
@@ -3190,6 +3197,11 @@ void DBImpl::BackgroundCallCompaction(void* arg) {
   LogBuffer log_buffer(InfoLogLevel::INFO_LEVEL, db_options_.info_log.get());
   {
     InstrumentedMutexLock l(&mutex_);
+
+    // This call will unlock/lock the mutex to wait for current running
+    // AddFile() calls to finish.
+    WaitForAddFile();
+
     num_running_compactions_++;
 
     auto pending_outputs_inserted_elem =

db/db_impl.h

@@ -650,6 +650,10 @@ class DBImpl : public DB {
   int PickLevelForIngestedFile(ColumnFamilyData* cfd,
                                const ExternalSstFileInfo& file_info);
 
+  // Wait for current AddFile() calls to finish.
+  // REQUIRES: mutex_ held
+  void WaitForAddFile();
+
   Status CompactFilesImpl(
       const CompactionOptions& compact_options, ColumnFamilyData* cfd,
       Version* version, const std::vector<std::string>& input_file_names,
@@ -659,6 +663,10 @@ class DBImpl : public DB {
                                  const std::string& file_path,
                                  ExternalSstFileInfo* file_info);
 
+#else
+  // AddFile is not supported in ROCKSDB_LITE so this function
+  // will be no-op
+  void WaitForAddFile() {}
 #endif  // ROCKSDB_LITE
 
   ColumnFamilyData* GetColumnFamilyDataByName(const std::string& cf_name);
@@ -973,6 +981,14 @@ class DBImpl : public DB {
   // REQUIRES: mutex held
   std::unordered_set<Compaction*> running_compactions_;
 
+  // Number of running AddFile() calls.
+  // REQUIRES: mutex held
+  int num_running_addfile_;
+
+  // A condition variable that will be signaled whenever
+  // num_running_addfile_ goes to 0.
+  InstrumentedCondVar addfile_cv_;
+
 #ifndef ROCKSDB_LITE
   WalManager wal_manager_;
 #endif  // ROCKSDB_LITE

db/db_impl_add_file.cc

@@ -237,12 +237,16 @@ Status DBImpl::AddFile(ColumnFamilyHandle* column_family,
 
   {
     InstrumentedMutexLock l(&mutex_);
+    TEST_SYNC_POINT("DBImpl::AddFile:MutexLock");
+
     const MutableCFOptions mutable_cf_options =
         *cfd->GetLatestMutableCFOptions();
 
     WriteThread::Writer w;
     write_thread_.EnterUnbatched(&w, &mutex_);
 
+    num_running_addfile_++;
+
     if (!skip_snapshot_check && !snapshots_.empty()) {
       // Check that no snapshots are being held
       status =
@@ -316,7 +320,13 @@ Status DBImpl::AddFile(ColumnFamilyHandle* column_family,
       ReleaseFileNumberFromPendingOutputs(
           pending_outputs_inserted_elem_list[i]);
     }
-  }
+
+    num_running_addfile_--;
+    if (num_running_addfile_ == 0) {
+      addfile_cv_.SignalAll();
+    }
+    TEST_SYNC_POINT("DBImpl::AddFile:MutexUnlock");
+  }  // mutex_ is unlocked here;
 
   if (!status.ok()) {
     // We failed to add the files to the database
@@ -395,6 +405,13 @@ int DBImpl::PickLevelForIngestedFile(ColumnFamilyData* cfd,
 
   return target_level;
 }
+
+void DBImpl::WaitForAddFile() {
+  mutex_.AssertHeld();
+  while (num_running_addfile_ > 0) {
+    addfile_cv_.Wait();
+  }
+}
 #endif  // ROCKSDB_LITE
 
 }  // namespace rocksdb

db/external_sst_file_test.cc

@@ -984,6 +984,87 @@ TEST_F(ExternalSSTFileTest, PickedLevel) {
   rocksdb::SyncPoint::GetInstance()->DisableProcessing();
 }
 
+TEST_F(ExternalSSTFileTest, PickedLevelBug) {
+  Options options = CurrentOptions();
+  options.disable_auto_compactions = false;
+  options.level0_file_num_compaction_trigger = 3;
+  options.num_levels = 2;
+  options.env = env_;
+  DestroyAndReopen(options);
+
+  std::vector<int> file_keys;
+
+  // file #1 in L0
+  file_keys = {0, 5, 7};
+  for (int k : file_keys) {
+    ASSERT_OK(Put(Key(k), Key(k)));
+  }
+  ASSERT_OK(Flush());
+
+  // file #2 in L0
+  file_keys = {4, 6, 8, 9};
+  for (int k : file_keys) {
+    ASSERT_OK(Put(Key(k), Key(k)));
+  }
+  ASSERT_OK(Flush());
+
+  // We have 2 overlapping files in L0
+  EXPECT_EQ(FilesPerLevel(), "2");
+
+  rocksdb::SyncPoint::GetInstance()->LoadDependency({
+      {"DBImpl::AddFile:MutexLock", "ExternalSSTFileTest::PickedLevelBug:0"},
+      {"ExternalSSTFileTest::PickedLevelBug:1", "DBImpl::AddFile:MutexUnlock"},
+  });
+
+  std::atomic<bool> bg_compact_started(false);
+  rocksdb::SyncPoint::GetInstance()->SetCallBack(
+      "DBImpl::BackgroundCompaction:Start",
+      [&](void* arg) { bg_compact_started.store(true); });
+
+  rocksdb::SyncPoint::GetInstance()->EnableProcessing();
+
+  // Start a thread that will ingest a new file
+  std::thread bg_addfile([&]() {
+    file_keys = {1, 2, 3};
+    ASSERT_OK(GenerateAndAddExternalFile(options, file_keys, 1));
+  });
+
+  // Wait for AddFile to start picking levels and writing MANIFEST
+  TEST_SYNC_POINT("ExternalSSTFileTest::PickedLevelBug:0");
+
+  // While writing the MANIFEST start a thread that will ask for compaction
+  std::thread bg_compact([&]() {
+    ASSERT_OK(db_->CompactRange(CompactRangeOptions(), nullptr, nullptr));
+  });
+
+  // We need to verify that no compactions can run while AddFile is
+  // ingesting the files into the levels it find suitable. So we will
+  // wait for 2 seconds to give a chance for compactions to run during
+  // this period, and then make sure that no compactions where able to run
+  env_->SleepForMicroseconds(1000000 * 2);
+  ASSERT_FALSE(bg_compact_started.load());
+
+  // Hold AddFile from finishing writing the MANIFEST
+  TEST_SYNC_POINT("ExternalSSTFileTest::PickedLevelBug:1");
+
+  bg_addfile.join();
+  bg_compact.join();
+
+  dbfull()->TEST_WaitForCompact();
+
+  int total_keys = 0;
+  Iterator* iter = db_->NewIterator(ReadOptions());
+  for (iter->SeekToFirst(); iter->Valid(); iter->Next()) {
+    ASSERT_OK(iter->status());
+    total_keys++;
+  }
+  ASSERT_EQ(total_keys, 10);
+
+  delete iter;
+
+  rocksdb::SyncPoint::GetInstance()->DisableProcessing();
+}
+
 TEST_F(ExternalSSTFileTest, PickedLevelDynamic) {
   Options options = CurrentOptions();
   options.disable_auto_compactions = false;