|
|
|
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
|
|
|
|
// This source code is licensed under both the GPLv2 (found in the
|
|
|
|
// COPYING file in the root directory) and Apache 2.0 License
|
|
|
|
// (found in the LICENSE.Apache file in the root directory).
|
|
|
|
|
|
|
|
#include "table/mock_table.h"
|
|
|
|
|
|
|
|
#include "db/dbformat.h"
|
Introduce a new storage specific Env API (#5761)
Summary:
The current Env API encompasses both storage/file operations, as well as OS related operations. Most of the APIs return a Status, which does not have enough metadata about an error, such as whether its retry-able or not, scope (i.e fault domain) of the error etc., that may be required in order to properly handle a storage error. The file APIs also do not provide enough control over the IO SLA, such as timeout, prioritization, hinting about placement and redundancy etc.
This PR separates out the file/storage APIs from Env into a new FileSystem class. The APIs are updated to return an IOStatus with metadata about the error, as well as to take an IOOptions structure as input in order to allow more control over the IO.
The user can set both ```options.env``` and ```options.file_system``` to specify that RocksDB should use the former for OS related operations and the latter for storage operations. Internally, a ```CompositeEnvWrapper``` has been introduced that inherits from ```Env``` and redirects individual methods to either an ```Env``` implementation or the ```FileSystem``` as appropriate. When options are sanitized during ```DB::Open```, ```options.env``` is replaced with a newly allocated ```CompositeEnvWrapper``` instance if both env and file_system have been specified. This way, the rest of the RocksDB code can continue to function as before.
This PR also ports PosixEnv to the new API by splitting it into two - PosixEnv and PosixFileSystem. PosixEnv is defined as a sub-class of CompositeEnvWrapper, and threading/time functions are overridden with Posix specific implementations in order to avoid an extra level of indirection.
The ```CompositeEnvWrapper``` translates ```IOStatus``` return code to ```Status```, and sets the severity to ```kSoftError``` if the io_status is retryable. The error handling code in RocksDB can then recover the DB automatically.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5761
Differential Revision: D18868376
Pulled By: anand1976
fbshipit-source-id: 39efe18a162ea746fabac6360ff529baba48486f
5 years ago
|
|
|
#include "env/composite_env_wrapper.h"
|
|
|
|
#include "file/random_access_file_reader.h"
|
|
|
|
#include "port/port.h"
|
|
|
|
#include "rocksdb/table_properties.h"
|
|
|
|
#include "table/get_context.h"
|
|
|
|
#include "util/coding.h"
|
|
|
|
|
|
|
|
namespace rocksdb {
|
|
|
|
namespace mock {
|
|
|
|
|
|
|
|
namespace {
|
|
|
|
|
|
|
|
const InternalKeyComparator icmp_(BytewiseComparator());
|
|
|
|
|
|
|
|
} // namespace
|
|
|
|
|
|
|
|
stl_wrappers::KVMap MakeMockFile(
|
|
|
|
std::initializer_list<std::pair<const std::string, std::string>> l) {
|
|
|
|
return stl_wrappers::KVMap(l, stl_wrappers::LessOfComparator(&icmp_));
|
|
|
|
}
|
|
|
|
|
|
|
|
InternalIterator* MockTableReader::NewIterator(
|
|
|
|
const ReadOptions&, const SliceTransform* /* prefix_extractor */,
|
|
|
|
Arena* /*arena*/, bool /*skip_filters*/, TableReaderCaller /*caller*/,
|
|
|
|
size_t /*compaction_readahead_size*/) {
|
|
|
|
return new MockTableIterator(table_);
|
|
|
|
}
|
|
|
|
|
|
|
|
Status MockTableReader::Get(const ReadOptions&, const Slice& key,
|
|
|
|
GetContext* get_context,
|
|
|
|
const SliceTransform* /*prefix_extractor*/,
|
|
|
|
bool /*skip_filters*/) {
|
|
|
|
std::unique_ptr<MockTableIterator> iter(new MockTableIterator(table_));
|
|
|
|
for (iter->Seek(key); iter->Valid(); iter->Next()) {
|
|
|
|
ParsedInternalKey parsed_key;
|
|
|
|
if (!ParseInternalKey(iter->key(), &parsed_key)) {
|
|
|
|
return Status::Corruption(Slice());
|
|
|
|
}
|
|
|
|
|
|
|
|
bool dont_care __attribute__((__unused__));
|
|
|
|
if (!get_context->SaveValue(parsed_key, iter->value(), &dont_care)) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
|
|
|
std::shared_ptr<const TableProperties> MockTableReader::GetTableProperties()
|
|
|
|
const {
|
|
|
|
return std::shared_ptr<const TableProperties>(new TableProperties());
|
|
|
|
}
|
|
|
|
|
|
|
|
MockTableFactory::MockTableFactory() : next_id_(1) {}
|
|
|
|
|
|
|
|
Status MockTableFactory::NewTableReader(
|
|
|
|
const TableReaderOptions& /*table_reader_options*/,
|
|
|
|
std::unique_ptr<RandomAccessFileReader>&& file, uint64_t /*file_size*/,
|
|
|
|
std::unique_ptr<TableReader>* table_reader,
|
|
|
|
bool /*prefetch_index_and_filter_in_cache*/) const {
|
|
|
|
uint32_t id = GetIDFromFile(file.get());
|
|
|
|
|
|
|
|
MutexLock lock_guard(&file_system_.mutex);
|
|
|
|
|
|
|
|
auto it = file_system_.files.find(id);
|
|
|
|
if (it == file_system_.files.end()) {
|
|
|
|
return Status::IOError("Mock file not found");
|
|
|
|
}
|
|
|
|
|
|
|
|
table_reader->reset(new MockTableReader(it->second));
|
|
|
|
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
|
|
|
TableBuilder* MockTableFactory::NewTableBuilder(
|
|
|
|
const TableBuilderOptions& /*table_builder_options*/,
|
|
|
|
uint32_t /*column_family_id*/, WritableFileWriter* file) const {
|
|
|
|
uint32_t id = GetAndWriteNextID(file);
|
|
|
|
|
|
|
|
return new MockTableBuilder(id, &file_system_);
|
|
|
|
}
|
|
|
|
|
|
|
|
Status MockTableFactory::CreateMockTable(Env* env, const std::string& fname,
|
|
|
|
stl_wrappers::KVMap file_contents) {
|
|
|
|
std::unique_ptr<WritableFile> file;
|
|
|
|
auto s = env->NewWritableFile(fname, &file, EnvOptions());
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
Introduce a new storage specific Env API (#5761)
Summary:
The current Env API encompasses both storage/file operations, as well as OS related operations. Most of the APIs return a Status, which does not have enough metadata about an error, such as whether its retry-able or not, scope (i.e fault domain) of the error etc., that may be required in order to properly handle a storage error. The file APIs also do not provide enough control over the IO SLA, such as timeout, prioritization, hinting about placement and redundancy etc.
This PR separates out the file/storage APIs from Env into a new FileSystem class. The APIs are updated to return an IOStatus with metadata about the error, as well as to take an IOOptions structure as input in order to allow more control over the IO.
The user can set both ```options.env``` and ```options.file_system``` to specify that RocksDB should use the former for OS related operations and the latter for storage operations. Internally, a ```CompositeEnvWrapper``` has been introduced that inherits from ```Env``` and redirects individual methods to either an ```Env``` implementation or the ```FileSystem``` as appropriate. When options are sanitized during ```DB::Open```, ```options.env``` is replaced with a newly allocated ```CompositeEnvWrapper``` instance if both env and file_system have been specified. This way, the rest of the RocksDB code can continue to function as before.
This PR also ports PosixEnv to the new API by splitting it into two - PosixEnv and PosixFileSystem. PosixEnv is defined as a sub-class of CompositeEnvWrapper, and threading/time functions are overridden with Posix specific implementations in order to avoid an extra level of indirection.
The ```CompositeEnvWrapper``` translates ```IOStatus``` return code to ```Status```, and sets the severity to ```kSoftError``` if the io_status is retryable. The error handling code in RocksDB can then recover the DB automatically.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5761
Differential Revision: D18868376
Pulled By: anand1976
fbshipit-source-id: 39efe18a162ea746fabac6360ff529baba48486f
5 years ago
|
|
|
WritableFileWriter file_writer(NewLegacyWritableFileWrapper(std::move(file)),
|
|
|
|
fname, EnvOptions());
|
|
|
|
|
|
|
|
uint32_t id = GetAndWriteNextID(&file_writer);
|
|
|
|
file_system_.files.insert({id, std::move(file_contents)});
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
|
|
|
uint32_t MockTableFactory::GetAndWriteNextID(WritableFileWriter* file) const {
|
|
|
|
uint32_t next_id = next_id_.fetch_add(1);
|
|
|
|
char buf[4];
|
|
|
|
EncodeFixed32(buf, next_id);
|
|
|
|
file->Append(Slice(buf, 4));
|
|
|
|
return next_id;
|
|
|
|
}
|
|
|
|
|
Move rate_limiter, write buffering, most perf context instrumentation and most random kill out of Env
Summary: We want to keep Env a think layer for better portability. Less platform dependent codes should be moved out of Env. In this patch, I create a wrapper of file readers and writers, and put rate limiting, write buffering, as well as most perf context instrumentation and random kill out of Env. It will make it easier to maintain multiple Env in the future.
Test Plan: Run all existing unit tests.
Reviewers: anthony, kradhakrishnan, IslamAbdelRahman, yhchiang, igor
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D42321
10 years ago
|
|
|
uint32_t MockTableFactory::GetIDFromFile(RandomAccessFileReader* file) const {
|
|
|
|
char buf[4];
|
|
|
|
Slice result;
|
|
|
|
file->Read(0, 4, &result, buf);
|
|
|
|
assert(result.size() == 4);
|
|
|
|
return DecodeFixed32(buf);
|
|
|
|
}
|
|
|
|
|
|
|
|
void MockTableFactory::AssertSingleFile(
|
|
|
|
const stl_wrappers::KVMap& file_contents) {
|
|
|
|
ASSERT_EQ(file_system_.files.size(), 1U);
|
|
|
|
ASSERT_EQ(file_contents, file_system_.files.begin()->second);
|
|
|
|
}
|
|
|
|
|
|
|
|
void MockTableFactory::AssertLatestFile(
|
|
|
|
const stl_wrappers::KVMap& file_contents) {
|
|
|
|
ASSERT_GE(file_system_.files.size(), 1U);
|
|
|
|
auto latest = file_system_.files.end();
|
|
|
|
--latest;
|
|
|
|
|
|
|
|
if (file_contents != latest->second) {
|
|
|
|
std::cout << "Wrong content! Content of latest file:" << std::endl;
|
|
|
|
for (const auto& kv : latest->second) {
|
|
|
|
ParsedInternalKey ikey;
|
|
|
|
std::string key, value;
|
|
|
|
std::tie(key, value) = kv;
|
|
|
|
ParseInternalKey(Slice(key), &ikey);
|
|
|
|
std::cout << ikey.DebugString(false) << " -> " << value << std::endl;
|
|
|
|
}
|
|
|
|
FAIL();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
} // namespace mock
|
|
|
|
} // namespace rocksdb
|