Consistency Check Function

Summary: Added a function/command to check the consistency of live files' meta data

Test Plan:
Manual test (size mismatch, file not exist).
Command test script.

Reviewers: haobo

Reviewed By: haobo

CC: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D16935

HISTORY.md

@@ -13,6 +13,8 @@
 * Chagned Options.prefix_extractor from raw pointer to shared_ptr (take ownership)
   Changed HashSkipListRepFactory and HashLinkListRepFactory constructor to not take SliceTransform object (use Options.prefix_extractor implicitly)
 * Added Env::GetThreadPoolQueueLen(), which returns the waiting queue length of thread pools
+* Added DB::CheckConsistency(), which checks the consistency of live files' metadata
+  Added a corresponding command "checkconsistency" in ldb tool
 
 ### New Features
 * If we find one truncated record at the end of the MANIFEST or WAL files,

db/db_impl_readonly.cc

@@ -99,4 +99,46 @@ Status DB::OpenForReadOnly(const Options& options, const std::string& dbname,
   return s;
 }
 
+Status DB::CheckConsistency(const Options& options,
+    const std::string& name) {
+  DB *db = nullptr;
+  Status st;
+
+  st = DB::OpenForReadOnly(options, name, &db);
+  if (!st.ok()) {
+    return st;
+  }
+
+  std::vector<LiveFileMetaData> metadata;
+  db->GetLiveFilesMetaData(&metadata);
+
+  for (const auto& md : metadata) {
+    std::string file_path = name + md.name;
+
+    if (!db->GetEnv()->FileExists(file_path)) {
+      st = Status::Corruption("sst file " + md.name + " doesn't exist");
+      break;
+    }
+
+    uint64_t fsize = 0;
+    st = db->GetEnv()->GetFileSize(file_path, &fsize);
+    if (!st.ok()) {
+      st = Status::Corruption(
+          "Failed to determine the actual size of file " + md.name +
+          ": " + st.ToString());
+      break;
+    }
+    if (fsize != md.size) {
+      st = Status::Corruption(
+          "sst file size mismatch: " + md.name +
+          ". Size recorded in manifest " + std::to_string(md.size) +
+          ", actual size " + std::to_string(fsize));
+      break;
+    }
+  }
+
+  delete db;
+  return st;
 }
+
+}   // namespace rocksdb

include/rocksdb/db.h

@@ -91,6 +91,14 @@ class DB {
       const std::string& name, DB** dbptr,
       bool error_if_log_file_exist = false);
 
+  // Check the consistency of live files' metadata.
+  // It will return Corruption Status when a file in manifest
+  // doesn't actually exist or doesn't match the actual file size.
+  // Note: This call should be invoked only when the database is
+  // not already open and serving data.
+  static Status CheckConsistency(const Options& options,
+      const std::string& name);
+
   DB() { }
   virtual ~DB();
 

tools/ldb_test.py

@@ -129,6 +129,8 @@ class LDBTestCase(unittest.TestCase):
         # It is weird that GET and SCAN raise exception for
         # non-existent key, while delete does not
 
+        self.assertRunOK("checkconsistency", "OK")
+
     def dumpDb(self, params, dumpFile):
         return 0 == run_err_null("./ldb dump %s > %s" % (params, dumpFile))
 
@@ -201,6 +203,7 @@ class LDBTestCase(unittest.TestCase):
         self.assertRunOK("scan", "a1 : b1\na2 : b2\na3 : b3\na4 : b4")
         self.assertRunOK("delete --hex 0x6133", "OK")
         self.assertRunOK("scan", "a1 : b1\na2 : b2\na4 : b4")
+        self.assertRunOK("checkconsistency", "OK")
 
     def testTtlPutGet(self):
         print "Running testTtlPutGet..."
@@ -215,6 +218,7 @@ class LDBTestCase(unittest.TestCase):
         self.assertRunOK("put a3 b3 --create_if_missing", "OK")
         # fails because timstamp's length is greater than value's
         self.assertRunFAIL("get --ttl a3")
+        self.assertRunOK("checkconsistency", "OK")
 
     def testInvalidCmdLines(self):
         print "Running testInvalidCmdLines..."
@@ -354,5 +358,26 @@ class LDBTestCase(unittest.TestCase):
                 origDbPath, os.path.join(origDbPath, "LOG"))))
         self.assertRunOK("scan", "x1 : y1\nx2 : y2\nx3 : y3\nx4 : y4")
 
+    def testCheckConsistency(self):
+        print "Running testCheckConsistency..."
+
+        dbPath = os.path.join(self.TMP_DIR, self.DB_NAME)
+        self.assertRunOK("put x1 y1 --create_if_missing", "OK")
+        self.assertRunOK("put x2 y2", "OK")
+        self.assertRunOK("get x1", "y1")
+        self.assertRunOK("checkconsistency", "OK")
+
+        sstFilePath = my_check_output("ls %s" % os.path.join(dbPath, "*.sst"),
+                                      shell=True)
+
+        # Modify the file
+        my_check_output("echo 'evil' > %s" % sstFilePath, shell=True)
+        self.assertRunFAIL("checkconsistency")
+
+        # Delete the file
+        my_check_output("rm -f %s" % sstFilePath, shell=True)
+        self.assertRunFAIL("checkconsistency")
+
+
 if __name__ == "__main__":
     unittest.main()

util/ldb_cmd.cc

@@ -154,6 +154,8 @@ LDBCommand* LDBCommand::SelectCommand(
     return new ManifestDumpCommand(cmdParams, option_map, flags);
   } else if (cmd == InternalDumpCommand::Name()) {
     return new InternalDumpCommand(cmdParams, option_map, flags);
+  } else if (cmd == CheckConsistencyCommand::Name()) {
+    return new CheckConsistencyCommand(cmdParams, option_map, flags);
   }
   return nullptr;
 }
@@ -1749,5 +1751,29 @@ void DBQuerierCommand::DoCommand() {
   }
 }
 
-
+CheckConsistencyCommand::CheckConsistencyCommand(const vector<string>& params,
+    const map<string, string>& options, const vector<string>& flags) :
+  LDBCommand(options, flags, false,
+             BuildCmdLineOptions({})) {
 }
+
+void CheckConsistencyCommand::Help(string& ret) {
+  ret.append("  ");
+  ret.append(CheckConsistencyCommand::Name());
+  ret.append("\n");
+}
+
+void CheckConsistencyCommand::DoCommand() {
+  Options opt = PrepareOptionsForOpenDB();
+  if (!exec_state_.IsNotStarted()) {
+    return;
+  }
+  Status st = DB::CheckConsistency(opt, db_path_);
+  if (st.ok()) {
+    fprintf(stdout, "OK\n");
+  } else {
+    exec_state_ = LDBCommandExecuteResult::FAILED(st.ToString());
+  }
+}
+
+}   // namespace rocksdb

util/ldb_cmd.h

@@ -686,4 +686,20 @@ private:
   static const char* DELETE_CMD;
 };
 
+class CheckConsistencyCommand : public LDBCommand {
+public:
+  static string Name() { return "checkconsistency"; }
+
+  CheckConsistencyCommand(const vector<string>& params,
+      const map<string, string>& options, const vector<string>& flags);
+
+  virtual void DoCommand();
+
+  virtual bool NoDBOpen() {
+    return true;
+  }
+
+  static void Help(string& ret);
+};
+
 } // namespace rocksdb

util/ldb_tool.cc

@@ -53,6 +53,7 @@ public:
     DeleteCommand::Help(ret);
     DBQuerierCommand::Help(ret);
     ApproxSizeCommand::Help(ret);
+    CheckConsistencyCommand::Help(ret);
 
     ret.append("\n\n");
     ret.append("Admin Commands:\n");