rocksdb/db/version_edit.cc

408 lines
11 KiB
C++
Raw Normal View History

// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
//
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
#include "db/version_edit.h"
#include "db/version_set.h"
#include "util/coding.h"
#include "rocksdb/slice.h"
namespace rocksdb {
// Tag numbers for serialized VersionEdit. These numbers are written to
// disk and should not be changed.
enum Tag {
kComparator = 1,
kLogNumber = 2,
kNextFileNumber = 3,
kLastSequence = 4,
kCompactPointer = 5,
kDeletedFile = 6,
kNewFile = 7,
// 8 was used for large value refs
kPrevLogNumber = 9,
// these are new formats divergent from open source leveldb
kNewFile2 = 100,
kNewFile3 = 102,
kColumnFamily = 200, // specify column family for version edit
kColumnFamilyAdd = 201,
kColumnFamilyDrop = 202,
kMaxColumnFamily = 203,
};
uint64_t PackFileNumberAndPathId(uint64_t number, uint64_t path_id) {
assert(number <= kFileNumberMask);
return number | (path_id * (kFileNumberMask + 1));
}
void VersionEdit::Clear() {
comparator_.clear();
max_level_ = 0;
log_number_ = 0;
prev_log_number_ = 0;
last_sequence_ = 0;
next_file_number_ = 0;
max_column_family_ = 0;
has_comparator_ = false;
has_log_number_ = false;
has_prev_log_number_ = false;
has_next_file_number_ = false;
has_last_sequence_ = false;
has_max_column_family_ = false;
deleted_files_.clear();
new_files_.clear();
column_family_ = 0;
is_column_family_add_ = 0;
is_column_family_drop_ = 0;
column_family_name_.clear();
}
bool VersionEdit::EncodeTo(std::string* dst) const {
if (has_comparator_) {
PutVarint32(dst, kComparator);
PutLengthPrefixedSlice(dst, comparator_);
}
if (has_log_number_) {
PutVarint32(dst, kLogNumber);
PutVarint64(dst, log_number_);
}
if (has_prev_log_number_) {
PutVarint32(dst, kPrevLogNumber);
PutVarint64(dst, prev_log_number_);
}
if (has_next_file_number_) {
PutVarint32(dst, kNextFileNumber);
PutVarint64(dst, next_file_number_);
}
if (has_last_sequence_) {
PutVarint32(dst, kLastSequence);
PutVarint64(dst, last_sequence_);
}
if (has_max_column_family_) {
PutVarint32(dst, kMaxColumnFamily);
PutVarint32(dst, max_column_family_);
}
for (const auto& deleted : deleted_files_) {
PutVarint32(dst, kDeletedFile);
PutVarint32(dst, deleted.first /* level */);
PutVarint64(dst, deleted.second /* file number */);
}
for (size_t i = 0; i < new_files_.size(); i++) {
const FileMetaData& f = new_files_[i].second;
if (!f.smallest.Valid() || !f.largest.Valid()) {
return false;
}
if (f.fd.GetPathId() == 0) {
// Use older format to make sure user can roll back the build if they
// don't config multiple DB paths.
PutVarint32(dst, kNewFile2);
} else {
PutVarint32(dst, kNewFile3);
}
PutVarint32(dst, new_files_[i].first); // level
PutVarint64(dst, f.fd.GetNumber());
if (f.fd.GetPathId() != 0) {
PutVarint32(dst, f.fd.GetPathId());
}
PutVarint64(dst, f.fd.GetFileSize());
PutLengthPrefixedSlice(dst, f.smallest.Encode());
PutLengthPrefixedSlice(dst, f.largest.Encode());
PutVarint64(dst, f.smallest_seqno);
PutVarint64(dst, f.largest_seqno);
}
// 0 is default and does not need to be explicitly written
if (column_family_ != 0) {
PutVarint32(dst, kColumnFamily);
PutVarint32(dst, column_family_);
}
if (is_column_family_add_) {
PutVarint32(dst, kColumnFamilyAdd);
PutLengthPrefixedSlice(dst, Slice(column_family_name_));
}
if (is_column_family_drop_) {
PutVarint32(dst, kColumnFamilyDrop);
}
return true;
}
static bool GetInternalKey(Slice* input, InternalKey* dst) {
Slice str;
if (GetLengthPrefixedSlice(input, &str)) {
dst->DecodeFrom(str);
[fix] SIGSEGV when VersionEdit in MANIFEST is corrupted Summary: This was reported by our customers in task #4295529. Cause: * MANIFEST file contains a VersionEdit, which contains file entries whose 'smallest' and 'largest' internal keys are empty. String with zero characters. Root cause of corruption was not investigated. We should report corruption when this happens. However, we currently SIGSEGV. Here's what happens: * VersionEdit encodes zero-strings happily and stores them in smallest and largest InternalKeys. InternalKey::Encode() does assert when `rep_.empty()`, but we don't assert in production environemnts. Also, we should never assert as a result of DB corruption. * As part of our ConsistencyCheck, we call GetLiveFilesMetaData() * GetLiveFilesMetadata() calls `file->largest.user_key().ToString()` * user_key() function does: 1. assert(size > 8) (ooops, no assert), 2. returns `Slice(internal_key.data(), internal_key.size() - 8)` * since `internal_key.size()` is unsigned int, this call translates to `Slice(whatever, 1298471928561892576182756)`. Bazinga. Fix: * VersionEdit checks if InternalKey is valid in `VersionEdit::GetInternalKey()`. If it's invalid, returns corruption. Lessons learned: * Always keep in mind that even if you `assert()`, production code will continue execution even if assert fails. * Never `assert` based on DB corruption. Assert only if the code should guarantee that assert can't fail. Test Plan: dumped offending manifest. Before: assert. Now: corruption Reviewers: dhruba, haobo, sdong Reviewed By: dhruba CC: leveldb Differential Revision: https://reviews.facebook.net/D18507
2014-05-08 01:52:12 +02:00
return dst->Valid();
} else {
return false;
}
}
bool VersionEdit::GetLevel(Slice* input, int* level, const char** msg) {
uint32_t v;
if (GetVarint32(input, &v)) {
*level = v;
if (max_level_ < *level) {
max_level_ = *level;
}
return true;
} else {
return false;
}
}
Status VersionEdit::DecodeFrom(const Slice& src) {
Clear();
Slice input = src;
const char* msg = nullptr;
uint32_t tag;
// Temporary storage for parsing
int level;
FileMetaData f;
Slice str;
InternalKey key;
while (msg == nullptr && GetVarint32(&input, &tag)) {
switch (tag) {
case kComparator:
if (GetLengthPrefixedSlice(&input, &str)) {
comparator_ = str.ToString();
has_comparator_ = true;
} else {
msg = "comparator name";
}
break;
case kLogNumber:
if (GetVarint64(&input, &log_number_)) {
has_log_number_ = true;
} else {
msg = "log number";
}
break;
case kPrevLogNumber:
if (GetVarint64(&input, &prev_log_number_)) {
has_prev_log_number_ = true;
} else {
msg = "previous log number";
}
break;
case kNextFileNumber:
if (GetVarint64(&input, &next_file_number_)) {
has_next_file_number_ = true;
} else {
msg = "next file number";
}
break;
case kLastSequence:
if (GetVarint64(&input, &last_sequence_)) {
has_last_sequence_ = true;
} else {
msg = "last sequence number";
}
break;
case kMaxColumnFamily:
if (GetVarint32(&input, &max_column_family_)) {
has_max_column_family_ = true;
} else {
msg = "max column family";
}
break;
case kCompactPointer:
if (GetLevel(&input, &level, &msg) &&
GetInternalKey(&input, &key)) {
// we don't use compact pointers anymore,
// but we should not fail if they are still
// in manifest
} else {
if (!msg) {
msg = "compaction pointer";
}
}
break;
case kDeletedFile: {
uint64_t number;
if (GetLevel(&input, &level, &msg) && GetVarint64(&input, &number)) {
deleted_files_.insert(std::make_pair(level, number));
} else {
if (!msg) {
msg = "deleted file";
}
}
break;
}
case kNewFile: {
uint64_t number;
uint64_t file_size;
if (GetLevel(&input, &level, &msg) && GetVarint64(&input, &number) &&
GetVarint64(&input, &file_size) &&
GetInternalKey(&input, &f.smallest) &&
GetInternalKey(&input, &f.largest)) {
f.fd = FileDescriptor(number, 0, file_size);
new_files_.push_back(std::make_pair(level, f));
} else {
if (!msg) {
msg = "new-file entry";
}
}
break;
}
case kNewFile2: {
uint64_t number;
uint64_t file_size;
if (GetLevel(&input, &level, &msg) && GetVarint64(&input, &number) &&
GetVarint64(&input, &file_size) &&
GetInternalKey(&input, &f.smallest) &&
GetInternalKey(&input, &f.largest) &&
GetVarint64(&input, &f.smallest_seqno) &&
GetVarint64(&input, &f.largest_seqno)) {
f.fd = FileDescriptor(number, 0, file_size);
new_files_.push_back(std::make_pair(level, f));
} else {
if (!msg) {
msg = "new-file2 entry";
}
}
break;
}
case kNewFile3: {
uint64_t number;
uint32_t path_id;
uint64_t file_size;
if (GetLevel(&input, &level, &msg) && GetVarint64(&input, &number) &&
GetVarint32(&input, &path_id) && GetVarint64(&input, &file_size) &&
GetInternalKey(&input, &f.smallest) &&
GetInternalKey(&input, &f.largest) &&
GetVarint64(&input, &f.smallest_seqno) &&
GetVarint64(&input, &f.largest_seqno)) {
f.fd = FileDescriptor(number, path_id, file_size);
new_files_.push_back(std::make_pair(level, f));
} else {
if (!msg) {
msg = "new-file3 entry";
}
}
break;
}
case kColumnFamily:
if (!GetVarint32(&input, &column_family_)) {
if (!msg) {
msg = "set column family id";
}
}
break;
case kColumnFamilyAdd:
if (GetLengthPrefixedSlice(&input, &str)) {
is_column_family_add_ = true;
column_family_name_ = str.ToString();
} else {
if (!msg) {
msg = "column family add";
}
}
break;
case kColumnFamilyDrop:
is_column_family_drop_ = true;
break;
default:
msg = "unknown tag";
break;
}
}
if (msg == nullptr && !input.empty()) {
msg = "invalid tag";
}
Status result;
if (msg != nullptr) {
result = Status::Corruption("VersionEdit", msg);
}
return result;
}
std::string VersionEdit::DebugString(bool hex_key) const {
std::string r;
r.append("VersionEdit {");
if (has_comparator_) {
r.append("\n Comparator: ");
r.append(comparator_);
}
if (has_log_number_) {
r.append("\n LogNumber: ");
AppendNumberTo(&r, log_number_);
}
if (has_prev_log_number_) {
r.append("\n PrevLogNumber: ");
AppendNumberTo(&r, prev_log_number_);
}
if (has_next_file_number_) {
r.append("\n NextFile: ");
AppendNumberTo(&r, next_file_number_);
}
if (has_last_sequence_) {
r.append("\n LastSeq: ");
AppendNumberTo(&r, last_sequence_);
}
for (DeletedFileSet::const_iterator iter = deleted_files_.begin();
iter != deleted_files_.end();
++iter) {
r.append("\n DeleteFile: ");
AppendNumberTo(&r, iter->first);
r.append(" ");
AppendNumberTo(&r, iter->second);
}
for (size_t i = 0; i < new_files_.size(); i++) {
const FileMetaData& f = new_files_[i].second;
r.append("\n AddFile: ");
AppendNumberTo(&r, new_files_[i].first);
r.append(" ");
AppendNumberTo(&r, f.fd.GetNumber());
r.append(" ");
AppendNumberTo(&r, f.fd.GetFileSize());
r.append(" ");
r.append(f.smallest.DebugString(hex_key));
r.append(" .. ");
r.append(f.largest.DebugString(hex_key));
}
r.append("\n ColumnFamily: ");
AppendNumberTo(&r, column_family_);
if (is_column_family_add_) {
r.append("\n ColumnFamilyAdd: ");
r.append(column_family_name_);
}
if (is_column_family_drop_) {
r.append("\n ColumnFamilyDrop");
}
if (has_max_column_family_) {
r.append("\n MaxColumnFamily: ");
2014-03-11 20:14:09 +01:00
AppendNumberTo(&r, max_column_family_);
}
r.append("\n}\n");
return r;
}
} // namespace rocksdb