rocksdb/tools/sst_dump_tool.cc

889 lines
32 KiB
C++
Raw Normal View History

// 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).
//
#ifndef ROCKSDB_LITE
#include "tools/sst_dump_tool_imp.h"
#include <cinttypes>
#include <chrono>
#include <iostream>
#include <map>
#include <memory>
#include <sstream>
#include <vector>
#include "db/blob/blob_index.h"
#include "db/memtable.h"
#include "db/write_batch_internal.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
2019-12-13 23:47:08 +01:00
#include "env/composite_env_wrapper.h"
#include "options/cf_options.h"
#include "rocksdb/db.h"
#include "rocksdb/env.h"
#include "rocksdb/iterator.h"
#include "rocksdb/slice_transform.h"
#include "rocksdb/status.h"
#include "rocksdb/table_properties.h"
#include "rocksdb/utilities/ldb_cmd.h"
#include "table/block_based/block.h"
#include "table/block_based/block_based_table_builder.h"
#include "table/block_based/block_based_table_factory.h"
#include "table/block_based/block_builder.h"
#include "table/format.h"
#include "table/meta_blocks.h"
#include "table/plain/plain_table_factory.h"
#include "table/table_reader.h"
#include "util/compression.h"
#include "util/random.h"
#include "port/port.h"
namespace ROCKSDB_NAMESPACE {
SstFileDumper::SstFileDumper(const Options& options,
const std::string& file_path,
size_t readahead_size, bool verify_checksum,
bool output_hex, bool decode_blob_index)
: file_name_(file_path),
read_num_(0),
output_hex_(output_hex),
decode_blob_index_(decode_blob_index),
options_(options),
ioptions_(options_),
moptions_(ColumnFamilyOptions(options_)),
read_options_(verify_checksum, false),
internal_comparator_(BytewiseComparator()) {
read_options_.readahead_size = readahead_size;
fprintf(stdout, "Process %s\n", file_path.c_str());
Disable pre-fetching of index and filter blocks for sst_dump_tool. Summary: BlockBasedTable pre-fetches the filter and index blocks on Open call. This is an optimistic optimization targeted for runtime scenario. The optimization is unnecessary for sst_dump_tool - Added a provision to disable pre-fetching of index and filter blocks in BlockBasedTable - Disabled pre-fetching for the sst_dump tool Stack for reference : #01 0x00000000005ed944 in snappy::InternalUncompress<snappy::SnappyArrayWriter> () from /home/engshare/third-party2/snappy/1.0.3/src/snappy-1.0.3/snappy.cc:148 #02 0x00000000005edeee in snappy::RawUncompress () from /home/engshare/third-party2/snappy/1.0.3/src/snappy-1.0.3/snappy.cc:947 #03 0x00000000004e0b4d in rocksdb::UncompressBlockContents () from /data/users/paultuckfield/rocksdb/./util/compression.h:69 #04 0x00000000004e145c in rocksdb::ReadBlockContents () from /data/users/paultuckfield/rocksdb/table/format.cc:334 #05 0x00000000004ca424 in rocksdb::(anonymous namespace)::ReadBlockFromFile () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:70 #06 0x00000000004cccad in rocksdb::BlockBasedTable::CreateIndexReader () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:173 #07 0x00000000004d17e5 in rocksdb::BlockBasedTable::Open () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:553 #08 0x00000000004c8184 in rocksdb::BlockBasedTableFactory::NewTableReader () from /data/users/paultuckfield/rocksdb/table/block_based_table_factory.cc:51 #09 0x0000000000598463 in rocksdb::SstFileReader::NewTableReader () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:69 #10 0x00000000005986c2 in rocksdb::SstFileReader::SstFileReader () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:26 #11 0x0000000000599047 in rocksdb::SSTDumpTool::Run () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:332 #12 0x0000000000409b06 in main () from /data/users/paultuckfield/rocksdb/tools/sst_dump.cc:12 Test Plan: - Added a unit test to trigger the code. - Also did some manual verification. - Passed all unit tests task #6296048 Reviewers: igor, rven, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D34041
2015-02-26 01:34:26 +01:00
init_result_ = GetTableReader(file_name_);
}
2014-11-09 19:01:50 +01:00
extern const uint64_t kBlockBasedTableMagicNumber;
extern const uint64_t kLegacyBlockBasedTableMagicNumber;
extern const uint64_t kPlainTableMagicNumber;
extern const uint64_t kLegacyPlainTableMagicNumber;
const char* testFileName = "test_file_name";
static const std::vector<std::pair<CompressionType, const char*>>
kCompressions = {
{CompressionType::kNoCompression, "kNoCompression"},
{CompressionType::kSnappyCompression, "kSnappyCompression"},
{CompressionType::kZlibCompression, "kZlibCompression"},
{CompressionType::kBZip2Compression, "kBZip2Compression"},
{CompressionType::kLZ4Compression, "kLZ4Compression"},
{CompressionType::kLZ4HCCompression, "kLZ4HCCompression"},
{CompressionType::kXpressCompression, "kXpressCompression"},
{CompressionType::kZSTD, "kZSTD"}};
Status SstFileDumper::GetTableReader(const std::string& file_path) {
// Warning about 'magic_number' being uninitialized shows up only in UBsan
// builds. Though access is guarded by 's.ok()' checks, fix the issue to
// avoid any warnings.
uint64_t magic_number = Footer::kInvalidTableMagicNumber;
// read table magic number
Footer footer;
std::unique_ptr<RandomAccessFile> file;
uint64_t file_size = 0;
Status s = options_.env->NewRandomAccessFile(file_path, &file, soptions_);
if (s.ok()) {
s = options_.env->GetFileSize(file_path, &file_size);
}
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
2019-12-13 23:47:08 +01:00
file_.reset(new RandomAccessFileReader(NewLegacyRandomAccessFileWrapper(file),
file_path));
FilePrefetchBuffer prefetch_buffer(nullptr, 0, 0, true /* enable */,
false /* track_min_offset */);
const uint64_t kSstDumpTailPrefetchSize = 512 * 1024;
uint64_t prefetch_size = (file_size > kSstDumpTailPrefetchSize)
? kSstDumpTailPrefetchSize
: file_size;
uint64_t prefetch_off = file_size - prefetch_size;
prefetch_buffer.Prefetch(file_.get(), prefetch_off,
static_cast<size_t>(prefetch_size));
if (s.ok()) {
s = ReadFooterFromFile(file_.get(), &prefetch_buffer, file_size, &footer);
}
if (s.ok()) {
magic_number = footer.table_magic_number();
}
if (s.ok()) {
if (magic_number == kPlainTableMagicNumber ||
magic_number == kLegacyPlainTableMagicNumber) {
soptions_.use_mmap_reads = true;
options_.env->NewRandomAccessFile(file_path, &file, soptions_);
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
2019-12-13 23:47:08 +01:00
file_.reset(new RandomAccessFileReader(
NewLegacyRandomAccessFileWrapper(file), file_path));
}
options_.comparator = &internal_comparator_;
// For old sst format, ReadTableProperties might fail but file can be read
if (ReadTableProperties(magic_number, file_.get(), file_size,
(magic_number == kBlockBasedTableMagicNumber)
? &prefetch_buffer
: nullptr)
.ok()) {
SetTableOptionsByMagicNumber(magic_number);
} else {
SetOldTableOptions();
}
}
if (s.ok()) {
s = NewTableReader(ioptions_, soptions_, internal_comparator_, file_size,
&table_reader_);
}
return s;
}
Status SstFileDumper::NewTableReader(
const ImmutableCFOptions& /*ioptions*/, const EnvOptions& /*soptions*/,
const InternalKeyComparator& /*internal_comparator*/, uint64_t file_size,
std::unique_ptr<TableReader>* /*table_reader*/) {
auto t_opt =
TableReaderOptions(ioptions_, moptions_.prefix_extractor.get(), soptions_,
internal_comparator_, false /* skip_filters */,
false /* imortal */, true /* force_direct_prefetch */);
// Allow open file with global sequence number for backward compatibility.
t_opt.largest_seqno = kMaxSequenceNumber;
Disable pre-fetching of index and filter blocks for sst_dump_tool. Summary: BlockBasedTable pre-fetches the filter and index blocks on Open call. This is an optimistic optimization targeted for runtime scenario. The optimization is unnecessary for sst_dump_tool - Added a provision to disable pre-fetching of index and filter blocks in BlockBasedTable - Disabled pre-fetching for the sst_dump tool Stack for reference : #01 0x00000000005ed944 in snappy::InternalUncompress<snappy::SnappyArrayWriter> () from /home/engshare/third-party2/snappy/1.0.3/src/snappy-1.0.3/snappy.cc:148 #02 0x00000000005edeee in snappy::RawUncompress () from /home/engshare/third-party2/snappy/1.0.3/src/snappy-1.0.3/snappy.cc:947 #03 0x00000000004e0b4d in rocksdb::UncompressBlockContents () from /data/users/paultuckfield/rocksdb/./util/compression.h:69 #04 0x00000000004e145c in rocksdb::ReadBlockContents () from /data/users/paultuckfield/rocksdb/table/format.cc:334 #05 0x00000000004ca424 in rocksdb::(anonymous namespace)::ReadBlockFromFile () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:70 #06 0x00000000004cccad in rocksdb::BlockBasedTable::CreateIndexReader () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:173 #07 0x00000000004d17e5 in rocksdb::BlockBasedTable::Open () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:553 #08 0x00000000004c8184 in rocksdb::BlockBasedTableFactory::NewTableReader () from /data/users/paultuckfield/rocksdb/table/block_based_table_factory.cc:51 #09 0x0000000000598463 in rocksdb::SstFileReader::NewTableReader () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:69 #10 0x00000000005986c2 in rocksdb::SstFileReader::SstFileReader () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:26 #11 0x0000000000599047 in rocksdb::SSTDumpTool::Run () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:332 #12 0x0000000000409b06 in main () from /data/users/paultuckfield/rocksdb/tools/sst_dump.cc:12 Test Plan: - Added a unit test to trigger the code. - Also did some manual verification. - Passed all unit tests task #6296048 Reviewers: igor, rven, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D34041
2015-02-26 01:34:26 +01:00
// We need to turn off pre-fetching of index and filter nodes for
// BlockBasedTable
if (BlockBasedTableFactory::kName == options_.table_factory->Name()) {
return options_.table_factory->NewTableReader(t_opt, std::move(file_),
file_size, &table_reader_,
/*enable_prefetch=*/false);
Disable pre-fetching of index and filter blocks for sst_dump_tool. Summary: BlockBasedTable pre-fetches the filter and index blocks on Open call. This is an optimistic optimization targeted for runtime scenario. The optimization is unnecessary for sst_dump_tool - Added a provision to disable pre-fetching of index and filter blocks in BlockBasedTable - Disabled pre-fetching for the sst_dump tool Stack for reference : #01 0x00000000005ed944 in snappy::InternalUncompress<snappy::SnappyArrayWriter> () from /home/engshare/third-party2/snappy/1.0.3/src/snappy-1.0.3/snappy.cc:148 #02 0x00000000005edeee in snappy::RawUncompress () from /home/engshare/third-party2/snappy/1.0.3/src/snappy-1.0.3/snappy.cc:947 #03 0x00000000004e0b4d in rocksdb::UncompressBlockContents () from /data/users/paultuckfield/rocksdb/./util/compression.h:69 #04 0x00000000004e145c in rocksdb::ReadBlockContents () from /data/users/paultuckfield/rocksdb/table/format.cc:334 #05 0x00000000004ca424 in rocksdb::(anonymous namespace)::ReadBlockFromFile () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:70 #06 0x00000000004cccad in rocksdb::BlockBasedTable::CreateIndexReader () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:173 #07 0x00000000004d17e5 in rocksdb::BlockBasedTable::Open () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:553 #08 0x00000000004c8184 in rocksdb::BlockBasedTableFactory::NewTableReader () from /data/users/paultuckfield/rocksdb/table/block_based_table_factory.cc:51 #09 0x0000000000598463 in rocksdb::SstFileReader::NewTableReader () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:69 #10 0x00000000005986c2 in rocksdb::SstFileReader::SstFileReader () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:26 #11 0x0000000000599047 in rocksdb::SSTDumpTool::Run () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:332 #12 0x0000000000409b06 in main () from /data/users/paultuckfield/rocksdb/tools/sst_dump.cc:12 Test Plan: - Added a unit test to trigger the code. - Also did some manual verification. - Passed all unit tests task #6296048 Reviewers: igor, rven, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D34041
2015-02-26 01:34:26 +01:00
}
// For all other factory implementation
return options_.table_factory->NewTableReader(t_opt, std::move(file_),
file_size, &table_reader_);
Disable pre-fetching of index and filter blocks for sst_dump_tool. Summary: BlockBasedTable pre-fetches the filter and index blocks on Open call. This is an optimistic optimization targeted for runtime scenario. The optimization is unnecessary for sst_dump_tool - Added a provision to disable pre-fetching of index and filter blocks in BlockBasedTable - Disabled pre-fetching for the sst_dump tool Stack for reference : #01 0x00000000005ed944 in snappy::InternalUncompress<snappy::SnappyArrayWriter> () from /home/engshare/third-party2/snappy/1.0.3/src/snappy-1.0.3/snappy.cc:148 #02 0x00000000005edeee in snappy::RawUncompress () from /home/engshare/third-party2/snappy/1.0.3/src/snappy-1.0.3/snappy.cc:947 #03 0x00000000004e0b4d in rocksdb::UncompressBlockContents () from /data/users/paultuckfield/rocksdb/./util/compression.h:69 #04 0x00000000004e145c in rocksdb::ReadBlockContents () from /data/users/paultuckfield/rocksdb/table/format.cc:334 #05 0x00000000004ca424 in rocksdb::(anonymous namespace)::ReadBlockFromFile () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:70 #06 0x00000000004cccad in rocksdb::BlockBasedTable::CreateIndexReader () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:173 #07 0x00000000004d17e5 in rocksdb::BlockBasedTable::Open () from /data/users/paultuckfield/rocksdb/table/block_based_table_reader.cc:553 #08 0x00000000004c8184 in rocksdb::BlockBasedTableFactory::NewTableReader () from /data/users/paultuckfield/rocksdb/table/block_based_table_factory.cc:51 #09 0x0000000000598463 in rocksdb::SstFileReader::NewTableReader () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:69 #10 0x00000000005986c2 in rocksdb::SstFileReader::SstFileReader () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:26 #11 0x0000000000599047 in rocksdb::SSTDumpTool::Run () from /data/users/paultuckfield/rocksdb/util/sst_dump_tool.cc:332 #12 0x0000000000409b06 in main () from /data/users/paultuckfield/rocksdb/tools/sst_dump.cc:12 Test Plan: - Added a unit test to trigger the code. - Also did some manual verification. - Passed all unit tests task #6296048 Reviewers: igor, rven, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D34041
2015-02-26 01:34:26 +01:00
}
Status SstFileDumper::VerifyChecksum() {
// We could pass specific readahead setting into read options if needed.
return table_reader_->VerifyChecksum(read_options_,
TableReaderCaller::kSSTDumpTool);
}
Status SstFileDumper::DumpTable(const std::string& out_filename) {
std::unique_ptr<WritableFile> out_file;
Env* env = options_.env;
env->NewWritableFile(out_filename, &out_file, soptions_);
Status s = table_reader_->DumpTable(out_file.get());
out_file->Close();
return s;
}
uint64_t SstFileDumper::CalculateCompressedTableSize(
sst_dump recompress show #blocks compressed and not compressed (#5791) Summary: Closes https://github.com/facebook/rocksdb/issues/1474 Helps show when the 12.5% threshold for GoodCompressionRatio (originally from ldb) is hit. Example output: ``` > ./sst_dump --file=/tmp/test.sst --command=recompress from [] to [] Process /tmp/test.sst Sst file format: block-based Block Size: 16384 Compression: kNoCompression Size: 122579836 Blocks: 2300 Compressed: 0 ( 0.0%) Not compressed (ratio): 2300 (100.0%) Not compressed (abort): 0 ( 0.0%) Compression: kSnappyCompression Size: 46289962 Blocks: 2300 Compressed: 2119 ( 92.1%) Not compressed (ratio): 181 ( 7.9%) Not compressed (abort): 0 ( 0.0%) Compression: kZlibCompression Size: 29689825 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) Unsupported compression type: kBZip2Compression. Compression: kLZ4Compression Size: 44785490 Blocks: 2300 Compressed: 1950 ( 84.8%) Not compressed (ratio): 350 ( 15.2%) Not compressed (abort): 0 ( 0.0%) Compression: kLZ4HCCompression Size: 37498895 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) Unsupported compression type: kXpressCompression. Compression: kZSTD Size: 32208707 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/5791 Differential Revision: D17347870 fbshipit-source-id: af10849c010b46b20e54162b70123c2805ffe526
2019-09-14 01:29:16 +02:00
const TableBuilderOptions& tb_options, size_t block_size,
uint64_t* num_data_blocks) {
std::unique_ptr<WritableFile> out_file;
std::unique_ptr<Env> env(NewMemEnv(options_.env));
env->NewWritableFile(testFileName, &out_file, soptions_);
std::unique_ptr<WritableFileWriter> dest_writer;
dest_writer.reset(
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
2019-12-13 23:47:08 +01:00
new WritableFileWriter(NewLegacyWritableFileWrapper(std::move(out_file)),
testFileName, soptions_));
BlockBasedTableOptions table_options;
table_options.block_size = block_size;
BlockBasedTableFactory block_based_tf(table_options);
std::unique_ptr<TableBuilder> table_builder;
table_builder.reset(block_based_tf.NewTableBuilder(
tb_options,
TablePropertiesCollectorFactory::Context::kUnknownColumnFamily,
dest_writer.get()));
std::unique_ptr<InternalIterator> iter(table_reader_->NewIterator(
read_options_, moptions_.prefix_extractor.get(), /*arena=*/nullptr,
/*skip_filters=*/false, TableReaderCaller::kSSTDumpTool));
for (iter->SeekToFirst(); iter->Valid(); iter->Next()) {
table_builder->Add(iter->key(), iter->value());
}
Properly report IO errors when IndexType::kBinarySearchWithFirstKey is used (#6621) Summary: Context: Index type `kBinarySearchWithFirstKey` added the ability for sst file iterator to sometimes report a key from index without reading the corresponding data block. This is useful when sst blocks are cut at some meaningful boundaries (e.g. one block per key prefix), and many seeks land between blocks (e.g. for each prefix, the ranges of keys in different sst files are nearly disjoint, so a typical seek needs to read a data block from only one file even if all files have the prefix). But this added a new error condition, which rocksdb code was really not equipped to deal with: `InternalIterator::value()` may fail with an IO error or Status::Incomplete, but it's just a method returning a Slice, with no way to report error instead. Before this PR, this type of error wasn't handled at all (an empty slice was returned), and kBinarySearchWithFirstKey implementation was considered a prototype. Now that we (LogDevice) have experimented with kBinarySearchWithFirstKey for a while and confirmed that it's really useful, this PR is adding the missing error handling. It's a pretty inconvenient situation implementation-wise. The error needs to be reported from InternalIterator when trying to access value. But there are ~700 call sites of `InternalIterator::value()`, most of which either can't hit the error condition (because the iterator is reading from memtable or from index or something) or wouldn't benefit from the deferred loading of the value (e.g. compaction iterator that reads all values anyway). Adding error handling to all these call sites would needlessly bloat the code. So instead I made the deferred value loading optional: only the call sites that may use deferred loading have to call the new method `PrepareValue()` before calling `value()`. The feature is enabled with a new bool argument `allow_unprepared_value` to a bunch of methods that create iterators (it wouldn't make sense to put it in ReadOptions because it's completely internal to iterators, with virtually no user-visible effect). Lmk if you have better ideas. Note that the deferred value loading only happens for *internal* iterators. The user-visible iterator (DBIter) always prepares the value before returning from Seek/Next/etc. We could go further and add an API to defer that value loading too, but that's most likely not useful for LogDevice, so it doesn't seem worth the complexity for now. Pull Request resolved: https://github.com/facebook/rocksdb/pull/6621 Test Plan: make -j5 check . Will also deploy to some logdevice test clusters and look at stats. Reviewed By: siying Differential Revision: D20786930 Pulled By: al13n321 fbshipit-source-id: 6da77d918bad3780522e918f17f4d5513d3e99ee
2020-04-16 02:37:23 +02:00
if (!iter->status().ok()) {
fputs(iter->status().ToString().c_str(), stderr);
exit(1);
}
Status s = table_builder->Finish();
if (!s.ok()) {
fputs(s.ToString().c_str(), stderr);
exit(1);
}
uint64_t size = table_builder->FileSize();
sst_dump recompress show #blocks compressed and not compressed (#5791) Summary: Closes https://github.com/facebook/rocksdb/issues/1474 Helps show when the 12.5% threshold for GoodCompressionRatio (originally from ldb) is hit. Example output: ``` > ./sst_dump --file=/tmp/test.sst --command=recompress from [] to [] Process /tmp/test.sst Sst file format: block-based Block Size: 16384 Compression: kNoCompression Size: 122579836 Blocks: 2300 Compressed: 0 ( 0.0%) Not compressed (ratio): 2300 (100.0%) Not compressed (abort): 0 ( 0.0%) Compression: kSnappyCompression Size: 46289962 Blocks: 2300 Compressed: 2119 ( 92.1%) Not compressed (ratio): 181 ( 7.9%) Not compressed (abort): 0 ( 0.0%) Compression: kZlibCompression Size: 29689825 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) Unsupported compression type: kBZip2Compression. Compression: kLZ4Compression Size: 44785490 Blocks: 2300 Compressed: 1950 ( 84.8%) Not compressed (ratio): 350 ( 15.2%) Not compressed (abort): 0 ( 0.0%) Compression: kLZ4HCCompression Size: 37498895 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) Unsupported compression type: kXpressCompression. Compression: kZSTD Size: 32208707 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/5791 Differential Revision: D17347870 fbshipit-source-id: af10849c010b46b20e54162b70123c2805ffe526
2019-09-14 01:29:16 +02:00
assert(num_data_blocks != nullptr);
*num_data_blocks = table_builder->GetTableProperties().num_data_blocks;
env->DeleteFile(testFileName);
return size;
}
int SstFileDumper::ShowAllCompressionSizes(
size_t block_size,
const std::vector<std::pair<CompressionType, const char*>>&
compression_types,
int32_t compress_level_from,
int32_t compress_level_to) {
fprintf(stdout, "Block Size: %" ROCKSDB_PRIszt "\n", block_size);
for (auto& i : compression_types) {
if (CompressionTypeSupported(i.first)) {
fprintf(stdout, "Compression: %-24s\n", i.second);
CompressionOptions compress_opt;
for(int32_t j = compress_level_from; j <= compress_level_to;
j++) {
fprintf(stdout, "Compression level: %d", j);
compress_opt.level = j;
ShowCompressionSize(block_size, i.first, compress_opt);
}
} else {
fprintf(stdout, "Unsupported compression type: %s.\n", i.second);
}
}
return 0;
}
int SstFileDumper::ShowCompressionSize(
size_t block_size,
CompressionType compress_type,
const CompressionOptions& compress_opt) {
Options opts;
opts.statistics = ROCKSDB_NAMESPACE::CreateDBStatistics();
sst_dump recompress show #blocks compressed and not compressed (#5791) Summary: Closes https://github.com/facebook/rocksdb/issues/1474 Helps show when the 12.5% threshold for GoodCompressionRatio (originally from ldb) is hit. Example output: ``` > ./sst_dump --file=/tmp/test.sst --command=recompress from [] to [] Process /tmp/test.sst Sst file format: block-based Block Size: 16384 Compression: kNoCompression Size: 122579836 Blocks: 2300 Compressed: 0 ( 0.0%) Not compressed (ratio): 2300 (100.0%) Not compressed (abort): 0 ( 0.0%) Compression: kSnappyCompression Size: 46289962 Blocks: 2300 Compressed: 2119 ( 92.1%) Not compressed (ratio): 181 ( 7.9%) Not compressed (abort): 0 ( 0.0%) Compression: kZlibCompression Size: 29689825 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) Unsupported compression type: kBZip2Compression. Compression: kLZ4Compression Size: 44785490 Blocks: 2300 Compressed: 1950 ( 84.8%) Not compressed (ratio): 350 ( 15.2%) Not compressed (abort): 0 ( 0.0%) Compression: kLZ4HCCompression Size: 37498895 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) Unsupported compression type: kXpressCompression. Compression: kZSTD Size: 32208707 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/5791 Differential Revision: D17347870 fbshipit-source-id: af10849c010b46b20e54162b70123c2805ffe526
2019-09-14 01:29:16 +02:00
opts.statistics->set_stats_level(StatsLevel::kAll);
const ImmutableCFOptions imoptions(opts);
const ColumnFamilyOptions cfo(opts);
const MutableCFOptions moptions(cfo);
ROCKSDB_NAMESPACE::InternalKeyComparator ikc(opts.comparator);
std::vector<std::unique_ptr<IntTblPropCollectorFactory> >
block_based_table_factories;
std::string column_family_name;
int unknown_level = -1;
TableBuilderOptions tb_opts(
imoptions, moptions, ikc, &block_based_table_factories, compress_type,
0 /* sample_for_compression */, compress_opt,
false /* skip_filters */, column_family_name, unknown_level);
uint64_t num_data_blocks = 0;
std::chrono::steady_clock::time_point start = std::chrono::steady_clock::now();
uint64_t file_size =
CalculateCompressedTableSize(tb_opts, block_size, &num_data_blocks);
std::chrono::steady_clock::time_point end = std::chrono::steady_clock::now();
fprintf(stdout, " Size: %10" PRIu64, file_size);
fprintf(stdout, " Blocks: %6" PRIu64, num_data_blocks);
fprintf(stdout, " Time Taken: %10s microsecs",
std::to_string(std::chrono::duration_cast<std::chrono::microseconds>
(end-start).count()).c_str());
const uint64_t compressed_blocks =
opts.statistics->getAndResetTickerCount(NUMBER_BLOCK_COMPRESSED);
const uint64_t not_compressed_blocks =
opts.statistics->getAndResetTickerCount(NUMBER_BLOCK_NOT_COMPRESSED);
// When the option enable_index_compression is true,
// NUMBER_BLOCK_COMPRESSED is incremented for index block(s).
if ((compressed_blocks + not_compressed_blocks) > num_data_blocks) {
num_data_blocks = compressed_blocks + not_compressed_blocks;
}
const uint64_t ratio_not_compressed_blocks =
(num_data_blocks - compressed_blocks) - not_compressed_blocks;
const double compressed_pcnt =
(0 == num_data_blocks) ? 0.0
: ((static_cast<double>(compressed_blocks) /
static_cast<double>(num_data_blocks)) *
100.0);
const double ratio_not_compressed_pcnt =
(0 == num_data_blocks)
? 0.0
: ((static_cast<double>(ratio_not_compressed_blocks) /
static_cast<double>(num_data_blocks)) *
100.0);
const double not_compressed_pcnt =
(0 == num_data_blocks)
? 0.0
: ((static_cast<double>(not_compressed_blocks) /
static_cast<double>(num_data_blocks)) *
100.0);
fprintf(stdout, " Compressed: %6" PRIu64 " (%5.1f%%)", compressed_blocks,
compressed_pcnt);
fprintf(stdout, " Not compressed (ratio): %6" PRIu64 " (%5.1f%%)",
ratio_not_compressed_blocks, ratio_not_compressed_pcnt);
fprintf(stdout, " Not compressed (abort): %6" PRIu64 " (%5.1f%%)\n",
not_compressed_blocks, not_compressed_pcnt);
return 0;
}
Status SstFileDumper::ReadTableProperties(uint64_t table_magic_number,
RandomAccessFileReader* file,
uint64_t file_size,
FilePrefetchBuffer* prefetch_buffer) {
TableProperties* table_properties = nullptr;
Status s = ROCKSDB_NAMESPACE::ReadTableProperties(
file, file_size, table_magic_number, ioptions_, &table_properties,
/* compression_type_missing= */ false,
/* memory_allocator= */ nullptr, prefetch_buffer);
if (s.ok()) {
table_properties_.reset(table_properties);
} else {
fprintf(stdout, "Not able to read table properties\n");
}
return s;
}
Status SstFileDumper::SetTableOptionsByMagicNumber(
uint64_t table_magic_number) {
assert(table_properties_);
if (table_magic_number == kBlockBasedTableMagicNumber ||
table_magic_number == kLegacyBlockBasedTableMagicNumber) {
BlockBasedTableFactory* bbtf = new BlockBasedTableFactory();
// To force tail prefetching, we fake reporting two useful reads of 512KB
// from the tail.
// It needs at least two data points to warm up the stats.
bbtf->tail_prefetch_stats()->RecordEffectiveSize(512 * 1024);
bbtf->tail_prefetch_stats()->RecordEffectiveSize(512 * 1024);
options_.table_factory.reset(bbtf);
fprintf(stdout, "Sst file format: block-based\n");
auto& props = table_properties_->user_collected_properties;
auto pos = props.find(BlockBasedTablePropertyNames::kIndexType);
if (pos != props.end()) {
auto index_type_on_file = static_cast<BlockBasedTableOptions::IndexType>(
DecodeFixed32(pos->second.c_str()));
if (index_type_on_file ==
BlockBasedTableOptions::IndexType::kHashSearch) {
options_.prefix_extractor.reset(NewNoopTransform());
}
}
} else if (table_magic_number == kPlainTableMagicNumber ||
table_magic_number == kLegacyPlainTableMagicNumber) {
options_.allow_mmap_reads = true;
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = kPlainTableVariableLength;
plain_table_options.bloom_bits_per_key = 0;
plain_table_options.hash_table_ratio = 0;
plain_table_options.index_sparseness = 1;
plain_table_options.huge_page_tlb_size = 0;
plain_table_options.encoding_type = kPlain;
plain_table_options.full_scan_mode = true;
options_.table_factory.reset(NewPlainTableFactory(plain_table_options));
fprintf(stdout, "Sst file format: plain table\n");
} else {
char error_msg_buffer[80];
snprintf(error_msg_buffer, sizeof(error_msg_buffer) - 1,
"Unsupported table magic number --- %lx",
(long)table_magic_number);
return Status::InvalidArgument(error_msg_buffer);
}
return Status::OK();
}
Status SstFileDumper::SetOldTableOptions() {
assert(table_properties_ == nullptr);
options_.table_factory = std::make_shared<BlockBasedTableFactory>();
fprintf(stdout, "Sst file format: block-based(old version)\n");
return Status::OK();
}
Status SstFileDumper::ReadSequential(bool print_kv, uint64_t read_num,
bool has_from, const std::string& from_key,
bool has_to, const std::string& to_key,
bool use_from_as_prefix) {
if (!table_reader_) {
return init_result_;
}
InternalIterator* iter = table_reader_->NewIterator(
read_options_, moptions_.prefix_extractor.get(),
/*arena=*/nullptr, /*skip_filters=*/false,
TableReaderCaller::kSSTDumpTool);
uint64_t i = 0;
if (has_from) {
InternalKey ikey;
ikey.SetMinPossibleForUserKey(from_key);
iter->Seek(ikey.Encode());
} else {
iter->SeekToFirst();
}
for (; iter->Valid(); iter->Next()) {
Slice key = iter->key();
Slice value = iter->value();
++i;
if (read_num > 0 && i > read_num)
break;
ParsedInternalKey ikey;
if (!ParseInternalKey(key, &ikey)) {
std::cerr << "Internal Key ["
<< key.ToString(true /* in hex*/)
<< "] parse error!\n";
continue;
}
// the key returned is not prefixed with out 'from' key
if (use_from_as_prefix && !ikey.user_key.starts_with(from_key)) {
break;
}
// If end marker was specified, we stop before it
if (has_to && BytewiseComparator()->Compare(ikey.user_key, to_key) >= 0) {
break;
}
if (print_kv) {
if (!decode_blob_index_ || ikey.type != kTypeBlobIndex) {
fprintf(stdout, "%s => %s\n", ikey.DebugString(output_hex_).c_str(),
value.ToString(output_hex_).c_str());
} else {
BlobIndex blob_index;
const Status s = blob_index.DecodeFrom(value);
if (!s.ok()) {
fprintf(stderr, "%s => error decoding blob index\n",
ikey.DebugString(output_hex_).c_str());
continue;
}
fprintf(stdout, "%s => %s\n", ikey.DebugString(output_hex_).c_str(),
blob_index.DebugString(output_hex_).c_str());
}
}
}
read_num_ += i;
Status ret = iter->status();
delete iter;
return ret;
}
Status SstFileDumper::ReadTableProperties(
std::shared_ptr<const TableProperties>* table_properties) {
if (!table_reader_) {
return init_result_;
}
*table_properties = table_reader_->GetTableProperties();
return init_result_;
}
namespace {
void print_help() {
fprintf(
stderr,
R"(sst_dump --file=<data_dir_OR_sst_file> [--command=check|scan|raw|recompress]
Improve sst_dump help message Summary: Current Message ``` sst_dump [--command=check|scan|none|raw] [--verify_checksum] --file=data_dir_OR_sst_file [--output_hex] [--input_key_hex] [--from=<user_key>] [--to=<user_key>] [--read_num=NUM] [--show_properties] [--show_compression_sizes] [--show_compression_sizes [--set_block_size=<block_size>]] ``` New message ``` sst_dump --file=<data_dir_OR_sst_file> [--command=check|scan|raw] --file=<data_dir_OR_sst_file> Path to SST file or directory containing SST files --command=check|scan|raw check: Iterate over entries in files but dont print anything except if an error is encounterd (default command) scan: Iterate over entries in files and print them to screen raw: Dump all the table contents to <file_name>_dump.txt --output_hex Can be combined with scan command to print the keys and values in Hex --from=<user_key> Key to start reading from when executing check|scan --to=<user_key> Key to stop reading at when executing check|scan --read_num=<num> Maximum number of entries to read when executing check|scan --verify_checksum Verify file checksum when executing check|scan --input_key_hex Can be combined with --from and --to to indicate that these values are encoded in Hex --show_properties Print table properties after iterating over the file --show_compression_sizes Independent command that will recreate the SST file using 16K block size with different compressions and report the size of the file using such compression --set_block_size=<block_size> Can be combined with --show_compression_sizes to set the block size that will be used when trying different compression algorithms ``` Test Plan: none Reviewers: yhchiang, andrewkr, kradhakrishnan, yiwu, sdong Reviewed By: sdong Subscribers: andrewkr, dhruba Differential Revision: https://reviews.facebook.net/D56325
2016-04-08 21:05:02 +02:00
--file=<data_dir_OR_sst_file>
Path to SST file or directory containing SST files
--env_uri=<uri of underlying Env>
URI of underlying Env
--command=check|scan|raw|verify
sst_dump recompress show #blocks compressed and not compressed (#5791) Summary: Closes https://github.com/facebook/rocksdb/issues/1474 Helps show when the 12.5% threshold for GoodCompressionRatio (originally from ldb) is hit. Example output: ``` > ./sst_dump --file=/tmp/test.sst --command=recompress from [] to [] Process /tmp/test.sst Sst file format: block-based Block Size: 16384 Compression: kNoCompression Size: 122579836 Blocks: 2300 Compressed: 0 ( 0.0%) Not compressed (ratio): 2300 (100.0%) Not compressed (abort): 0 ( 0.0%) Compression: kSnappyCompression Size: 46289962 Blocks: 2300 Compressed: 2119 ( 92.1%) Not compressed (ratio): 181 ( 7.9%) Not compressed (abort): 0 ( 0.0%) Compression: kZlibCompression Size: 29689825 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) Unsupported compression type: kBZip2Compression. Compression: kLZ4Compression Size: 44785490 Blocks: 2300 Compressed: 1950 ( 84.8%) Not compressed (ratio): 350 ( 15.2%) Not compressed (abort): 0 ( 0.0%) Compression: kLZ4HCCompression Size: 37498895 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) Unsupported compression type: kXpressCompression. Compression: kZSTD Size: 32208707 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/5791 Differential Revision: D17347870 fbshipit-source-id: af10849c010b46b20e54162b70123c2805ffe526
2019-09-14 01:29:16 +02:00
check: Iterate over entries in files but don't print anything except if an error is encountered (default command)
Improve sst_dump help message Summary: Current Message ``` sst_dump [--command=check|scan|none|raw] [--verify_checksum] --file=data_dir_OR_sst_file [--output_hex] [--input_key_hex] [--from=<user_key>] [--to=<user_key>] [--read_num=NUM] [--show_properties] [--show_compression_sizes] [--show_compression_sizes [--set_block_size=<block_size>]] ``` New message ``` sst_dump --file=<data_dir_OR_sst_file> [--command=check|scan|raw] --file=<data_dir_OR_sst_file> Path to SST file or directory containing SST files --command=check|scan|raw check: Iterate over entries in files but dont print anything except if an error is encounterd (default command) scan: Iterate over entries in files and print them to screen raw: Dump all the table contents to <file_name>_dump.txt --output_hex Can be combined with scan command to print the keys and values in Hex --from=<user_key> Key to start reading from when executing check|scan --to=<user_key> Key to stop reading at when executing check|scan --read_num=<num> Maximum number of entries to read when executing check|scan --verify_checksum Verify file checksum when executing check|scan --input_key_hex Can be combined with --from and --to to indicate that these values are encoded in Hex --show_properties Print table properties after iterating over the file --show_compression_sizes Independent command that will recreate the SST file using 16K block size with different compressions and report the size of the file using such compression --set_block_size=<block_size> Can be combined with --show_compression_sizes to set the block size that will be used when trying different compression algorithms ``` Test Plan: none Reviewers: yhchiang, andrewkr, kradhakrishnan, yiwu, sdong Reviewed By: sdong Subscribers: andrewkr, dhruba Differential Revision: https://reviews.facebook.net/D56325
2016-04-08 21:05:02 +02:00
scan: Iterate over entries in files and print them to screen
raw: Dump all the table contents to <file_name>_dump.txt
sst_dump recompress show #blocks compressed and not compressed (#5791) Summary: Closes https://github.com/facebook/rocksdb/issues/1474 Helps show when the 12.5% threshold for GoodCompressionRatio (originally from ldb) is hit. Example output: ``` > ./sst_dump --file=/tmp/test.sst --command=recompress from [] to [] Process /tmp/test.sst Sst file format: block-based Block Size: 16384 Compression: kNoCompression Size: 122579836 Blocks: 2300 Compressed: 0 ( 0.0%) Not compressed (ratio): 2300 (100.0%) Not compressed (abort): 0 ( 0.0%) Compression: kSnappyCompression Size: 46289962 Blocks: 2300 Compressed: 2119 ( 92.1%) Not compressed (ratio): 181 ( 7.9%) Not compressed (abort): 0 ( 0.0%) Compression: kZlibCompression Size: 29689825 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) Unsupported compression type: kBZip2Compression. Compression: kLZ4Compression Size: 44785490 Blocks: 2300 Compressed: 1950 ( 84.8%) Not compressed (ratio): 350 ( 15.2%) Not compressed (abort): 0 ( 0.0%) Compression: kLZ4HCCompression Size: 37498895 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) Unsupported compression type: kXpressCompression. Compression: kZSTD Size: 32208707 Blocks: 2300 Compressed: 2301 (100.0%) Not compressed (ratio): 0 ( 0.0%) Not compressed (abort): 0 ( 0.0%) ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/5791 Differential Revision: D17347870 fbshipit-source-id: af10849c010b46b20e54162b70123c2805ffe526
2019-09-14 01:29:16 +02:00
verify: Iterate all the blocks in files verifying checksum to detect possible corruption but don't print anything except if a corruption is encountered
recompress: reports the SST file size if recompressed with different
compression types
Improve sst_dump help message Summary: Current Message ``` sst_dump [--command=check|scan|none|raw] [--verify_checksum] --file=data_dir_OR_sst_file [--output_hex] [--input_key_hex] [--from=<user_key>] [--to=<user_key>] [--read_num=NUM] [--show_properties] [--show_compression_sizes] [--show_compression_sizes [--set_block_size=<block_size>]] ``` New message ``` sst_dump --file=<data_dir_OR_sst_file> [--command=check|scan|raw] --file=<data_dir_OR_sst_file> Path to SST file or directory containing SST files --command=check|scan|raw check: Iterate over entries in files but dont print anything except if an error is encounterd (default command) scan: Iterate over entries in files and print them to screen raw: Dump all the table contents to <file_name>_dump.txt --output_hex Can be combined with scan command to print the keys and values in Hex --from=<user_key> Key to start reading from when executing check|scan --to=<user_key> Key to stop reading at when executing check|scan --read_num=<num> Maximum number of entries to read when executing check|scan --verify_checksum Verify file checksum when executing check|scan --input_key_hex Can be combined with --from and --to to indicate that these values are encoded in Hex --show_properties Print table properties after iterating over the file --show_compression_sizes Independent command that will recreate the SST file using 16K block size with different compressions and report the size of the file using such compression --set_block_size=<block_size> Can be combined with --show_compression_sizes to set the block size that will be used when trying different compression algorithms ``` Test Plan: none Reviewers: yhchiang, andrewkr, kradhakrishnan, yiwu, sdong Reviewed By: sdong Subscribers: andrewkr, dhruba Differential Revision: https://reviews.facebook.net/D56325
2016-04-08 21:05:02 +02:00
--output_hex
Can be combined with scan command to print the keys and values in Hex
--decode_blob_index
Decode blob indexes and print them in a human-readable format during scans.
Improve sst_dump help message Summary: Current Message ``` sst_dump [--command=check|scan|none|raw] [--verify_checksum] --file=data_dir_OR_sst_file [--output_hex] [--input_key_hex] [--from=<user_key>] [--to=<user_key>] [--read_num=NUM] [--show_properties] [--show_compression_sizes] [--show_compression_sizes [--set_block_size=<block_size>]] ``` New message ``` sst_dump --file=<data_dir_OR_sst_file> [--command=check|scan|raw] --file=<data_dir_OR_sst_file> Path to SST file or directory containing SST files --command=check|scan|raw check: Iterate over entries in files but dont print anything except if an error is encounterd (default command) scan: Iterate over entries in files and print them to screen raw: Dump all the table contents to <file_name>_dump.txt --output_hex Can be combined with scan command to print the keys and values in Hex --from=<user_key> Key to start reading from when executing check|scan --to=<user_key> Key to stop reading at when executing check|scan --read_num=<num> Maximum number of entries to read when executing check|scan --verify_checksum Verify file checksum when executing check|scan --input_key_hex Can be combined with --from and --to to indicate that these values are encoded in Hex --show_properties Print table properties after iterating over the file --show_compression_sizes Independent command that will recreate the SST file using 16K block size with different compressions and report the size of the file using such compression --set_block_size=<block_size> Can be combined with --show_compression_sizes to set the block size that will be used when trying different compression algorithms ``` Test Plan: none Reviewers: yhchiang, andrewkr, kradhakrishnan, yiwu, sdong Reviewed By: sdong Subscribers: andrewkr, dhruba Differential Revision: https://reviews.facebook.net/D56325
2016-04-08 21:05:02 +02:00
--from=<user_key>
Key to start reading from when executing check|scan
--to=<user_key>
Key to stop reading at when executing check|scan
--prefix=<user_key>
Returns all keys with this prefix when executing check|scan
Cannot be used in conjunction with --from
Improve sst_dump help message Summary: Current Message ``` sst_dump [--command=check|scan|none|raw] [--verify_checksum] --file=data_dir_OR_sst_file [--output_hex] [--input_key_hex] [--from=<user_key>] [--to=<user_key>] [--read_num=NUM] [--show_properties] [--show_compression_sizes] [--show_compression_sizes [--set_block_size=<block_size>]] ``` New message ``` sst_dump --file=<data_dir_OR_sst_file> [--command=check|scan|raw] --file=<data_dir_OR_sst_file> Path to SST file or directory containing SST files --command=check|scan|raw check: Iterate over entries in files but dont print anything except if an error is encounterd (default command) scan: Iterate over entries in files and print them to screen raw: Dump all the table contents to <file_name>_dump.txt --output_hex Can be combined with scan command to print the keys and values in Hex --from=<user_key> Key to start reading from when executing check|scan --to=<user_key> Key to stop reading at when executing check|scan --read_num=<num> Maximum number of entries to read when executing check|scan --verify_checksum Verify file checksum when executing check|scan --input_key_hex Can be combined with --from and --to to indicate that these values are encoded in Hex --show_properties Print table properties after iterating over the file --show_compression_sizes Independent command that will recreate the SST file using 16K block size with different compressions and report the size of the file using such compression --set_block_size=<block_size> Can be combined with --show_compression_sizes to set the block size that will be used when trying different compression algorithms ``` Test Plan: none Reviewers: yhchiang, andrewkr, kradhakrishnan, yiwu, sdong Reviewed By: sdong Subscribers: andrewkr, dhruba Differential Revision: https://reviews.facebook.net/D56325
2016-04-08 21:05:02 +02:00
--read_num=<num>
Maximum number of entries to read when executing check|scan
--verify_checksum
Verify file checksum when executing check|scan
--input_key_hex
Can be combined with --from and --to to indicate that these values are encoded in Hex
--show_properties
Print table properties after iterating over the file when executing
check|scan|raw
Improve sst_dump help message Summary: Current Message ``` sst_dump [--command=check|scan|none|raw] [--verify_checksum] --file=data_dir_OR_sst_file [--output_hex] [--input_key_hex] [--from=<user_key>] [--to=<user_key>] [--read_num=NUM] [--show_properties] [--show_compression_sizes] [--show_compression_sizes [--set_block_size=<block_size>]] ``` New message ``` sst_dump --file=<data_dir_OR_sst_file> [--command=check|scan|raw] --file=<data_dir_OR_sst_file> Path to SST file or directory containing SST files --command=check|scan|raw check: Iterate over entries in files but dont print anything except if an error is encounterd (default command) scan: Iterate over entries in files and print them to screen raw: Dump all the table contents to <file_name>_dump.txt --output_hex Can be combined with scan command to print the keys and values in Hex --from=<user_key> Key to start reading from when executing check|scan --to=<user_key> Key to stop reading at when executing check|scan --read_num=<num> Maximum number of entries to read when executing check|scan --verify_checksum Verify file checksum when executing check|scan --input_key_hex Can be combined with --from and --to to indicate that these values are encoded in Hex --show_properties Print table properties after iterating over the file --show_compression_sizes Independent command that will recreate the SST file using 16K block size with different compressions and report the size of the file using such compression --set_block_size=<block_size> Can be combined with --show_compression_sizes to set the block size that will be used when trying different compression algorithms ``` Test Plan: none Reviewers: yhchiang, andrewkr, kradhakrishnan, yiwu, sdong Reviewed By: sdong Subscribers: andrewkr, dhruba Differential Revision: https://reviews.facebook.net/D56325
2016-04-08 21:05:02 +02:00
--set_block_size=<block_size>
Can be combined with --command=recompress to set the block size that will
be used when trying different compression algorithms
--compression_types=<comma-separated list of CompressionType members, e.g.,
kSnappyCompression>
Can be combined with --command=recompress to run recompression for this
list of compression types
--parse_internal_key=<0xKEY>
Convenience option to parse an internal key on the command line. Dumps the
internal key in hex format {'key' @ SN: type}
--compression_level_from=<compression_level>
Compression level to start compressing when executing recompress. One compression type
and compression_level_to must also be specified
--compression_level_to=<compression_level>
Compression level to stop compressing when executing recompress. One compression type
and compression_level_from must also be specified
Improve sst_dump help message Summary: Current Message ``` sst_dump [--command=check|scan|none|raw] [--verify_checksum] --file=data_dir_OR_sst_file [--output_hex] [--input_key_hex] [--from=<user_key>] [--to=<user_key>] [--read_num=NUM] [--show_properties] [--show_compression_sizes] [--show_compression_sizes [--set_block_size=<block_size>]] ``` New message ``` sst_dump --file=<data_dir_OR_sst_file> [--command=check|scan|raw] --file=<data_dir_OR_sst_file> Path to SST file or directory containing SST files --command=check|scan|raw check: Iterate over entries in files but dont print anything except if an error is encounterd (default command) scan: Iterate over entries in files and print them to screen raw: Dump all the table contents to <file_name>_dump.txt --output_hex Can be combined with scan command to print the keys and values in Hex --from=<user_key> Key to start reading from when executing check|scan --to=<user_key> Key to stop reading at when executing check|scan --read_num=<num> Maximum number of entries to read when executing check|scan --verify_checksum Verify file checksum when executing check|scan --input_key_hex Can be combined with --from and --to to indicate that these values are encoded in Hex --show_properties Print table properties after iterating over the file --show_compression_sizes Independent command that will recreate the SST file using 16K block size with different compressions and report the size of the file using such compression --set_block_size=<block_size> Can be combined with --show_compression_sizes to set the block size that will be used when trying different compression algorithms ``` Test Plan: none Reviewers: yhchiang, andrewkr, kradhakrishnan, yiwu, sdong Reviewed By: sdong Subscribers: andrewkr, dhruba Differential Revision: https://reviews.facebook.net/D56325
2016-04-08 21:05:02 +02:00
)");
}
// arg_name would include all prefix, e.g. "--my_arg="
// arg_val is the parses value.
// True if there is a match. False otherwise.
// Woud exit after printing errmsg if cannot be parsed.
bool ParseIntArg(const char* arg, const std::string arg_name,
const std::string err_msg, int64_t* arg_val) {
if (strncmp(arg, arg_name.c_str(), arg_name.size()) == 0) {
std::string input_str = arg + arg_name.size();
std::istringstream iss(input_str);
iss >> *arg_val;
if (iss.fail()) {
fprintf(stderr, "%s\n", err_msg.c_str());
exit(1);
}
return true;
}
return false;
}
} // namespace
int SSTDumpTool::Run(int argc, char** argv, Options options) {
const char* env_uri = nullptr;
const char* dir_or_file = nullptr;
uint64_t read_num = std::numeric_limits<uint64_t>::max();
std::string command;
char junk;
uint64_t n;
bool verify_checksum = false;
bool output_hex = false;
bool decode_blob_index = false;
bool input_key_hex = false;
bool has_from = false;
bool has_to = false;
bool use_from_as_prefix = false;
bool show_properties = false;
bool show_summary = false;
bool set_block_size = false;
bool has_compression_level_from = false;
bool has_compression_level_to = false;
bool has_specified_compression_types = false;
std::string from_key;
std::string to_key;
std::string block_size_str;
std::string compression_level_from_str;
std::string compression_level_to_str;
size_t block_size = 0;
size_t readahead_size = 2 * 1024 * 1024;
std::vector<std::pair<CompressionType, const char*>> compression_types;
uint64_t total_num_files = 0;
uint64_t total_num_data_blocks = 0;
uint64_t total_data_block_size = 0;
uint64_t total_index_block_size = 0;
uint64_t total_filter_block_size = 0;
int32_t compress_level_from = CompressionOptions::kDefaultCompressionLevel;
int32_t compress_level_to = CompressionOptions::kDefaultCompressionLevel;
int64_t tmp_val;
for (int i = 1; i < argc; i++) {
if (strncmp(argv[i], "--env_uri=", 10) == 0) {
env_uri = argv[i] + 10;
} else if (strncmp(argv[i], "--file=", 7) == 0) {
dir_or_file = argv[i] + 7;
} else if (strcmp(argv[i], "--output_hex") == 0) {
output_hex = true;
} else if (strcmp(argv[i], "--decode_blob_index") == 0) {
decode_blob_index = true;
} else if (strcmp(argv[i], "--input_key_hex") == 0) {
input_key_hex = true;
} else if (sscanf(argv[i], "--read_num=%lu%c", (unsigned long*)&n, &junk) ==
1) {
read_num = n;
} else if (strcmp(argv[i], "--verify_checksum") == 0) {
verify_checksum = true;
} else if (strncmp(argv[i], "--command=", 10) == 0) {
command = argv[i] + 10;
} else if (strncmp(argv[i], "--from=", 7) == 0) {
from_key = argv[i] + 7;
has_from = true;
} else if (strncmp(argv[i], "--to=", 5) == 0) {
to_key = argv[i] + 5;
has_to = true;
} else if (strncmp(argv[i], "--prefix=", 9) == 0) {
from_key = argv[i] + 9;
use_from_as_prefix = true;
} else if (strcmp(argv[i], "--show_properties") == 0) {
show_properties = true;
} else if (strcmp(argv[i], "--show_summary") == 0) {
show_summary = true;
} else if (ParseIntArg(argv[i], "--set_block_size=",
"block size must be numeric", &tmp_val)) {
set_block_size = true;
block_size = static_cast<size_t>(tmp_val);
} else if (ParseIntArg(argv[i], "--readahead_size=",
"readahead_size must be numeric", &tmp_val)) {
readahead_size = static_cast<size_t>(tmp_val);
} else if (strncmp(argv[i], "--compression_types=", 20) == 0) {
std::string compression_types_csv = argv[i] + 20;
std::istringstream iss(compression_types_csv);
std::string compression_type;
has_specified_compression_types = true;
while (std::getline(iss, compression_type, ',')) {
auto iter = std::find_if(
kCompressions.begin(), kCompressions.end(),
[&compression_type](std::pair<CompressionType, const char*> curr) {
return curr.second == compression_type;
});
if (iter == kCompressions.end()) {
fprintf(stderr, "%s is not a valid CompressionType\n",
compression_type.c_str());
exit(1);
}
compression_types.emplace_back(*iter);
}
} else if (strncmp(argv[i], "--parse_internal_key=", 21) == 0) {
std::string in_key(argv[i] + 21);
try {
in_key = ROCKSDB_NAMESPACE::LDBCommand::HexToString(in_key);
} catch (...) {
std::cerr << "ERROR: Invalid key input '"
<< in_key
<< "' Use 0x{hex representation of internal rocksdb key}" << std::endl;
return -1;
}
Slice sl_key = ROCKSDB_NAMESPACE::Slice(in_key);
ParsedInternalKey ikey;
int retc = 0;
if (!ParseInternalKey(sl_key, &ikey)) {
std::cerr << "Internal Key [" << sl_key.ToString(true /* in hex*/)
<< "] parse error!\n";
retc = -1;
}
fprintf(stdout, "key=%s\n", ikey.DebugString(true).c_str());
return retc;
} else if (ParseIntArg(argv[i], "--compression_level_from=",
"compression_level_from must be numeric",
&tmp_val)) {
has_compression_level_from = true;
compress_level_from = static_cast<int>(tmp_val);
} else if (ParseIntArg(argv[i], "--compression_level_to=",
"compression_level_to must be numeric", &tmp_val)) {
has_compression_level_to = true;
compress_level_to = static_cast<int>(tmp_val);
} else {
fprintf(stderr, "Unrecognized argument '%s'\n\n", argv[i]);
print_help();
exit(1);
}
}
if(has_compression_level_from && has_compression_level_to) {
if(!has_specified_compression_types || compression_types.size() != 1) {
fprintf(stderr, "Specify one compression type.\n\n");
exit(1);
}
} else if(has_compression_level_from || has_compression_level_to) {
fprintf(stderr, "Specify both --compression_level_from and "
"--compression_level_to.\n\n");
exit(1);
}
if (use_from_as_prefix && has_from) {
fprintf(stderr, "Cannot specify --prefix and --from\n\n");
exit(1);
}
if (input_key_hex) {
if (has_from || use_from_as_prefix) {
from_key = ROCKSDB_NAMESPACE::LDBCommand::HexToString(from_key);
}
if (has_to) {
to_key = ROCKSDB_NAMESPACE::LDBCommand::HexToString(to_key);
}
}
if (dir_or_file == nullptr) {
fprintf(stderr, "file or directory must be specified.\n\n");
print_help();
exit(1);
}
std::shared_ptr<ROCKSDB_NAMESPACE::Env> env_guard;
// If caller of SSTDumpTool::Run(...) does not specify a different env other
// than Env::Default(), then try to load custom env based on dir_or_file.
// Otherwise, the caller is responsible for creating custom env.
if (!options.env || options.env == ROCKSDB_NAMESPACE::Env::Default()) {
Env* env = Env::Default();
Status s = Env::LoadEnv(env_uri ? env_uri : "", &env, &env_guard);
if (!s.ok() && !s.IsNotFound()) {
fprintf(stderr, "LoadEnv: %s\n", s.ToString().c_str());
exit(1);
}
options.env = env;
} else {
fprintf(stdout, "options.env is %p\n", options.env);
}
std::vector<std::string> filenames;
ROCKSDB_NAMESPACE::Env* env = options.env;
ROCKSDB_NAMESPACE::Status st = env->GetChildren(dir_or_file, &filenames);
bool dir = true;
if (!st.ok()) {
filenames.clear();
filenames.push_back(dir_or_file);
dir = false;
}
fprintf(stdout, "from [%s] to [%s]\n",
ROCKSDB_NAMESPACE::Slice(from_key).ToString(true).c_str(),
ROCKSDB_NAMESPACE::Slice(to_key).ToString(true).c_str());
uint64_t total_read = 0;
for (size_t i = 0; i < filenames.size(); i++) {
std::string filename = filenames.at(i);
if (filename.length() <= 4 ||
filename.rfind(".sst") != filename.length() - 4) {
// ignore
continue;
}
if (dir) {
filename = std::string(dir_or_file) + "/" + filename;
}
ROCKSDB_NAMESPACE::SstFileDumper dumper(options, filename, readahead_size,
verify_checksum, output_hex,
decode_blob_index);
if (!dumper.getStatus().ok()) {
fprintf(stderr, "%s: %s\n", filename.c_str(),
dumper.getStatus().ToString().c_str());
continue;
}
if (command == "recompress") {
dumper.ShowAllCompressionSizes(
set_block_size ? block_size : 16384,
compression_types.empty() ? kCompressions : compression_types,
compress_level_from, compress_level_to);
return 0;
}
if (command == "raw") {
std::string out_filename = filename.substr(0, filename.length() - 4);
out_filename.append("_dump.txt");
st = dumper.DumpTable(out_filename);
if (!st.ok()) {
fprintf(stderr, "%s: %s\n", filename.c_str(), st.ToString().c_str());
exit(1);
} else {
fprintf(stdout, "raw dump written to file %s\n", &out_filename[0]);
}
continue;
}
// scan all files in give file path.
if (command == "" || command == "scan" || command == "check") {
st = dumper.ReadSequential(
command == "scan", read_num > 0 ? (read_num - total_read) : read_num,
has_from || use_from_as_prefix, from_key, has_to, to_key,
use_from_as_prefix);
if (!st.ok()) {
fprintf(stderr, "%s: %s\n", filename.c_str(),
st.ToString().c_str());
}
total_read += dumper.GetReadNumber();
if (read_num > 0 && total_read > read_num) {
break;
}
}
if (command == "verify") {
st = dumper.VerifyChecksum();
if (!st.ok()) {
fprintf(stderr, "%s is corrupted: %s\n", filename.c_str(),
st.ToString().c_str());
} else {
fprintf(stdout, "The file is ok\n");
}
continue;
}
if (show_properties || show_summary) {
const ROCKSDB_NAMESPACE::TableProperties* table_properties;
std::shared_ptr<const ROCKSDB_NAMESPACE::TableProperties>
table_properties_from_reader;
st = dumper.ReadTableProperties(&table_properties_from_reader);
if (!st.ok()) {
fprintf(stderr, "%s: %s\n", filename.c_str(), st.ToString().c_str());
fprintf(stderr, "Try to use initial table properties\n");
table_properties = dumper.GetInitTableProperties();
} else {
table_properties = table_properties_from_reader.get();
}
if (table_properties != nullptr) {
if (show_properties) {
fprintf(stdout,
"Table Properties:\n"
"------------------------------\n"
" %s",
table_properties->ToString("\n ", ": ").c_str());
}
total_num_files += 1;
total_num_data_blocks += table_properties->num_data_blocks;
total_data_block_size += table_properties->data_size;
total_index_block_size += table_properties->index_size;
total_filter_block_size += table_properties->filter_size;
if (show_properties) {
fprintf(stdout,
"Raw user collected properties\n"
"------------------------------\n");
for (const auto& kv : table_properties->user_collected_properties) {
std::string prop_name = kv.first;
std::string prop_val = Slice(kv.second).ToString(true);
fprintf(stdout, " # %s: 0x%s\n", prop_name.c_str(),
prop_val.c_str());
}
}
} else {
fprintf(stderr, "Reader unexpectedly returned null properties\n");
}
}
}
if (show_summary) {
fprintf(stdout, "total number of files: %" PRIu64 "\n", total_num_files);
fprintf(stdout, "total number of data blocks: %" PRIu64 "\n",
total_num_data_blocks);
fprintf(stdout, "total data block size: %" PRIu64 "\n",
total_data_block_size);
fprintf(stdout, "total index block size: %" PRIu64 "\n",
total_index_block_size);
fprintf(stdout, "total filter block size: %" PRIu64 "\n",
total_filter_block_size);
}
return 0;
}
} // namespace ROCKSDB_NAMESPACE
#endif // ROCKSDB_LITE