You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
rocksdb/db_stress_tool/db_stress_common.h

573 lines
20 KiB

// 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).
//
// 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.
//
// The test uses an array to compare against values written to the database.
// Keys written to the array are in 1:1 correspondence to the actual values in
// the database according to the formula in the function GenerateValue.
// Space is reserved in the array from 0 to FLAGS_max_key and values are
// randomly written/deleted/read from those positions. During verification we
// compare all the positions in the array. To shorten/elongate the running
// time, you could change the settings: FLAGS_max_key, FLAGS_ops_per_thread,
// (sometimes also FLAGS_threads).
//
// NOTE that if FLAGS_test_batches_snapshots is set, the test will have
// different behavior. See comment of the flag for details.
#ifdef GFLAGS
#pragma once
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <sys/types.h>
#include <algorithm>
#include <array>
#include <chrono>
#include <cinttypes>
#include <exception>
#include <queue>
#include <thread>
#include "db/db_impl/db_impl.h"
#include "db/version_set.h"
#include "db_stress_tool/db_stress_env_wrapper.h"
#include "db_stress_tool/db_stress_listener.h"
#include "db_stress_tool/db_stress_shared_state.h"
#include "db_stress_tool/db_stress_test_base.h"
#include "hdfs/env_hdfs.h"
#include "logging/logging.h"
#include "monitoring/histogram.h"
#include "options/options_helper.h"
#include "port/port.h"
#include "rocksdb/cache.h"
#include "rocksdb/env.h"
#include "rocksdb/slice.h"
#include "rocksdb/slice_transform.h"
#include "rocksdb/statistics.h"
#include "rocksdb/utilities/backupable_db.h"
#include "rocksdb/utilities/checkpoint.h"
#include "rocksdb/utilities/db_ttl.h"
#include "rocksdb/utilities/debug.h"
#include "rocksdb/utilities/options_util.h"
#include "rocksdb/utilities/transaction.h"
#include "rocksdb/utilities/transaction_db.h"
#include "rocksdb/write_batch.h"
#include "test_util/testutil.h"
#include "util/coding.h"
#include "util/compression.h"
#include "util/crc32c.h"
#include "util/gflags_compat.h"
#include "util/mutexlock.h"
#include "util/random.h"
#include "util/string_util.h"
#include "utilities/blob_db/blob_db.h"
#include "utilities/merge_operators.h"
using GFLAGS_NAMESPACE::ParseCommandLineFlags;
using GFLAGS_NAMESPACE::RegisterFlagValidator;
using GFLAGS_NAMESPACE::SetUsageMessage;
DECLARE_uint64(seed);
DECLARE_bool(read_only);
DECLARE_int64(max_key);
DECLARE_double(hot_key_alpha);
DECLARE_int32(max_key_len);
DECLARE_string(key_len_percent_dist);
DECLARE_int32(key_window_scale_factor);
DECLARE_int32(column_families);
DECLARE_string(options_file);
DECLARE_int64(active_width);
DECLARE_bool(test_batches_snapshots);
DECLARE_bool(atomic_flush);
DECLARE_bool(test_cf_consistency);
DECLARE_int32(threads);
DECLARE_int32(ttl);
DECLARE_int32(value_size_mult);
DECLARE_int32(compaction_readahead_size);
DECLARE_bool(enable_pipelined_write);
DECLARE_bool(verify_before_write);
DECLARE_bool(histogram);
DECLARE_bool(destroy_db_initially);
DECLARE_bool(verbose);
DECLARE_bool(progress_reports);
DECLARE_uint64(db_write_buffer_size);
DECLARE_int32(write_buffer_size);
DECLARE_int32(max_write_buffer_number);
DECLARE_int32(min_write_buffer_number_to_merge);
DECLARE_int32(max_write_buffer_number_to_maintain);
DECLARE_int64(max_write_buffer_size_to_maintain);
DECLARE_double(memtable_prefix_bloom_size_ratio);
DECLARE_bool(memtable_whole_key_filtering);
DECLARE_int32(open_files);
DECLARE_int64(compressed_cache_size);
DECLARE_int32(compaction_style);
DECLARE_int32(num_levels);
DECLARE_int32(level0_file_num_compaction_trigger);
DECLARE_int32(level0_slowdown_writes_trigger);
DECLARE_int32(level0_stop_writes_trigger);
DECLARE_int32(block_size);
DECLARE_int32(format_version);
DECLARE_int32(index_block_restart_interval);
DECLARE_int32(max_background_compactions);
DECLARE_int32(num_bottom_pri_threads);
DECLARE_int32(compaction_thread_pool_adjust_interval);
DECLARE_int32(compaction_thread_pool_variations);
DECLARE_int32(max_background_flushes);
DECLARE_int32(universal_size_ratio);
DECLARE_int32(universal_min_merge_width);
DECLARE_int32(universal_max_merge_width);
DECLARE_int32(universal_max_size_amplification_percent);
DECLARE_int32(clear_column_family_one_in);
DECLARE_int32(get_live_files_one_in);
DECLARE_int32(get_sorted_wal_files_one_in);
DECLARE_int32(get_current_wal_file_one_in);
DECLARE_int32(set_options_one_in);
DECLARE_int32(set_in_place_one_in);
DECLARE_int64(cache_size);
DECLARE_bool(cache_index_and_filter_blocks);
DECLARE_int32(top_level_index_pinning);
DECLARE_int32(partition_pinning);
DECLARE_int32(unpartitioned_pinning);
DECLARE_bool(use_clock_cache);
DECLARE_uint64(subcompactions);
DECLARE_uint64(periodic_compaction_seconds);
DECLARE_uint64(compaction_ttl);
DECLARE_bool(allow_concurrent_memtable_write);
DECLARE_bool(enable_write_thread_adaptive_yield);
DECLARE_int32(reopen);
DECLARE_double(bloom_bits);
DECLARE_bool(use_block_based_filter);
Experimental (production candidate) SST schema for Ribbon filter (#7658) Summary: Added experimental public API for Ribbon filter: NewExperimentalRibbonFilterPolicy(). This experimental API will take a "Bloom equivalent" bits per key, and configure the Ribbon filter for the same FP rate as Bloom would have but ~30% space savings. (Note: optimize_filters_for_memory is not yet implemented for Ribbon filter. That can be added with no effect on schema.) Internally, the Ribbon filter is configured using a "one_in_fp_rate" value, which is 1 over desired FP rate. For example, use 100 for 1% FP rate. I'm expecting this will be used in the future for configuring Bloom-like filters, as I expect people to more commonly hold constant the filter accuracy and change the space vs. time trade-off, rather than hold constant the space (per key) and change the accuracy vs. time trade-off, though we might make that available. ### Benchmarking ``` $ ./filter_bench -impl=2 -quick -m_keys_total_max=200 -average_keys_per_filter=100000 -net_includes_hashing Building... Build avg ns/key: 34.1341 Number of filters: 1993 Total size (MB): 238.488 Reported total allocated memory (MB): 262.875 Reported internal fragmentation: 10.2255% Bits/key stored: 10.0029 ---------------------------- Mixed inside/outside queries... Single filter net ns/op: 18.7508 Random filter net ns/op: 258.246 Average FP rate %: 0.968672 ---------------------------- Done. (For more info, run with -legend or -help.) $ ./filter_bench -impl=3 -quick -m_keys_total_max=200 -average_keys_per_filter=100000 -net_includes_hashing Building... Build avg ns/key: 130.851 Number of filters: 1993 Total size (MB): 168.166 Reported total allocated memory (MB): 183.211 Reported internal fragmentation: 8.94626% Bits/key stored: 7.05341 ---------------------------- Mixed inside/outside queries... Single filter net ns/op: 58.4523 Random filter net ns/op: 363.717 Average FP rate %: 0.952978 ---------------------------- Done. (For more info, run with -legend or -help.) ``` 168.166 / 238.488 = 0.705 -> 29.5% space reduction 130.851 / 34.1341 = 3.83x construction time for this Ribbon filter vs. lastest Bloom filter (could make that as little as about 2.5x for less space reduction) ### Working around a hashing "flaw" bloom_test discovered a flaw in the simple hashing applied in StandardHasher when num_starts == 1 (num_slots == 128), showing an excessively high FP rate. The problem is that when many entries, on the order of number of hash bits or kCoeffBits, are associated with the same start location, the correlation between the CoeffRow and ResultRow (for efficiency) can lead to a solution that is "universal," or nearly so, for entries mapping to that start location. (Normally, variance in start location breaks the effective association between CoeffRow and ResultRow; the same value for CoeffRow is effectively different if start locations are different.) Without kUseSmash and with num_starts > 1 (thus num_starts ~= num_slots), this flaw should be completely irrelevant. Even with 10M slots, the chances of a single slot having just 16 (or more) entries map to it--not enough to cause an FP problem, which would be local to that slot if it happened--is 1 in millions. This spreadsheet formula shows that: =1/(10000000*(1 - POISSON(15, 1, TRUE))) As kUseSmash==false (the setting for Standard128RibbonBitsBuilder) is intended for CPU efficiency of filters with many more entries/slots than kCoeffBits, a very reasonable work-around is to disallow num_starts==1 when !kUseSmash, by making the minimum non-zero number of slots 2*kCoeffBits. This is the work-around I've applied. This also means that the new Ribbon filter schema (Standard128RibbonBitsBuilder) is not space-efficient for less than a few hundred entries. Because of this, I have made it fall back on constructing a Bloom filter, under existing schema, when that is more space efficient for small filters. (We can change this in the future if we want.) TODO: better unit tests for this case in ribbon_test, and probably update StandardHasher for kUseSmash case so that it can scale nicely to small filters. ### Other related changes * Add Ribbon filter to stress/crash test * Add Ribbon filter to filter_bench as -impl=3 * Add option string support, as in "filter_policy=experimental_ribbon:5.678;" where 5.678 is the Bloom equivalent bits per key. * Rename internal mode BloomFilterPolicy::kAuto to kAutoBloom * Add a general BuiltinFilterBitsBuilder::CalculateNumEntry based on binary searching CalculateSpace (inefficient), so that subclasses (especially experimental ones) don't have to provide an efficient implementation inverting CalculateSpace. * Minor refactor FastLocalBloomBitsBuilder for new base class XXH3pFilterBitsBuilder shared with new Standard128RibbonBitsBuilder, which allows the latter to fall back on Bloom construction in some extreme cases. * Mostly updated bloom_test for Ribbon filter, though a test like FullBloomTest::Schema is a next TODO to ensure schema stability (in case this becomes production-ready schema as it is). * Add some APIs to ribbon_impl.h for configuring Ribbon filters. Although these are reasonably covered by bloom_test, TODO more unit tests in ribbon_test * Added a "tool" FindOccupancyForSuccessRate to ribbon_test to get data for constructing the linear approximations in GetNumSlotsFor95PctSuccess. Pull Request resolved: https://github.com/facebook/rocksdb/pull/7658 Test Plan: Some unit tests updated but other testing is left TODO. This is considered experimental but laying down schema compatibility as early as possible in case it proves production-quality. Also tested in stress/crash test. Reviewed By: jay-zhuang Differential Revision: D24899349 Pulled By: pdillinger fbshipit-source-id: 9715f3e6371c959d923aea8077c9423c7a9f82b8
4 years ago
DECLARE_bool(use_ribbon_filter);
DECLARE_bool(partition_filters);
Minimize memory internal fragmentation for Bloom filters (#6427) Summary: New experimental option BBTO::optimize_filters_for_memory builds filters that maximize their use of "usable size" from malloc_usable_size, which is also used to compute block cache charges. Rather than always "rounding up," we track state in the BloomFilterPolicy object to mix essentially "rounding down" and "rounding up" so that the average FP rate of all generated filters is the same as without the option. (YMMV as heavily accessed filters might be unluckily lower accuracy.) Thus, the option near-minimizes what the block cache considers as "memory used" for a given target Bloom filter false positive rate and Bloom filter implementation. There are no forward or backward compatibility issues with this change, though it only works on the format_version=5 Bloom filter. With Jemalloc, we see about 10% reduction in memory footprint (and block cache charge) for Bloom filters, but 1-2% increase in storage footprint, due to encoding efficiency losses (FP rate is non-linear with bits/key). Why not weighted random round up/down rather than state tracking? By only requiring malloc_usable_size, we don't actually know what the next larger and next smaller usable sizes for the allocator are. We pick a requested size, accept and use whatever usable size it has, and use the difference to inform our next choice. This allows us to narrow in on the right balance without tracking/predicting usable sizes. Why not weight history of generated filter false positive rates by number of keys? This could lead to excess skew in small filters after generating a large filter. Results from filter_bench with jemalloc (irrelevant details omitted): (normal keys/filter, but high variance) $ ./filter_bench -quick -impl=2 -average_keys_per_filter=30000 -vary_key_count_ratio=0.9 Build avg ns/key: 29.6278 Number of filters: 5516 Total size (MB): 200.046 Reported total allocated memory (MB): 220.597 Reported internal fragmentation: 10.2732% Bits/key stored: 10.0097 Average FP rate %: 0.965228 $ ./filter_bench -quick -impl=2 -average_keys_per_filter=30000 -vary_key_count_ratio=0.9 -optimize_filters_for_memory Build avg ns/key: 30.5104 Number of filters: 5464 Total size (MB): 200.015 Reported total allocated memory (MB): 200.322 Reported internal fragmentation: 0.153709% Bits/key stored: 10.1011 Average FP rate %: 0.966313 (very few keys / filter, optimization not as effective due to ~59 byte internal fragmentation in blocked Bloom filter representation) $ ./filter_bench -quick -impl=2 -average_keys_per_filter=1000 -vary_key_count_ratio=0.9 Build avg ns/key: 29.5649 Number of filters: 162950 Total size (MB): 200.001 Reported total allocated memory (MB): 224.624 Reported internal fragmentation: 12.3117% Bits/key stored: 10.2951 Average FP rate %: 0.821534 $ ./filter_bench -quick -impl=2 -average_keys_per_filter=1000 -vary_key_count_ratio=0.9 -optimize_filters_for_memory Build avg ns/key: 31.8057 Number of filters: 159849 Total size (MB): 200 Reported total allocated memory (MB): 208.846 Reported internal fragmentation: 4.42297% Bits/key stored: 10.4948 Average FP rate %: 0.811006 (high keys/filter) $ ./filter_bench -quick -impl=2 -average_keys_per_filter=1000000 -vary_key_count_ratio=0.9 Build avg ns/key: 29.7017 Number of filters: 164 Total size (MB): 200.352 Reported total allocated memory (MB): 221.5 Reported internal fragmentation: 10.5552% Bits/key stored: 10.0003 Average FP rate %: 0.969358 $ ./filter_bench -quick -impl=2 -average_keys_per_filter=1000000 -vary_key_count_ratio=0.9 -optimize_filters_for_memory Build avg ns/key: 30.7131 Number of filters: 160 Total size (MB): 200.928 Reported total allocated memory (MB): 200.938 Reported internal fragmentation: 0.00448054% Bits/key stored: 10.1852 Average FP rate %: 0.963387 And from db_bench (block cache) with jemalloc: $ ./db_bench -db=/dev/shm/dbbench.no_optimize -benchmarks=fillrandom -format_version=5 -value_size=90 -bloom_bits=10 -num=2000000 -threads=8 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=false $ ./db_bench -db=/dev/shm/dbbench -benchmarks=fillrandom -format_version=5 -value_size=90 -bloom_bits=10 -num=2000000 -threads=8 -optimize_filters_for_memory -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=false $ (for FILE in /dev/shm/dbbench.no_optimize/*.sst; do ./sst_dump --file=$FILE --show_properties | grep 'filter block' ; done) | awk '{ t += $4; } END { print t; }' 17063835 $ (for FILE in /dev/shm/dbbench/*.sst; do ./sst_dump --file=$FILE --show_properties | grep 'filter block' ; done) | awk '{ t += $4; } END { print t; }' 17430747 $ #^ 2.1% additional filter storage $ ./db_bench -db=/dev/shm/dbbench.no_optimize -use_existing_db -benchmarks=readrandom,stats -statistics -bloom_bits=10 -num=2000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=false -duration=10 -cache_index_and_filter_blocks -cache_size=1000000000 rocksdb.block.cache.index.add COUNT : 33 rocksdb.block.cache.index.bytes.insert COUNT : 8440400 rocksdb.block.cache.filter.add COUNT : 33 rocksdb.block.cache.filter.bytes.insert COUNT : 21087528 rocksdb.bloom.filter.useful COUNT : 4963889 rocksdb.bloom.filter.full.positive COUNT : 1214081 rocksdb.bloom.filter.full.true.positive COUNT : 1161999 $ #^ 1.04 % observed FP rate $ ./db_bench -db=/dev/shm/dbbench -use_existing_db -benchmarks=readrandom,stats -statistics -bloom_bits=10 -num=2000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=false -optimize_filters_for_memory -duration=10 -cache_index_and_filter_blocks -cache_size=1000000000 rocksdb.block.cache.index.add COUNT : 33 rocksdb.block.cache.index.bytes.insert COUNT : 8448592 rocksdb.block.cache.filter.add COUNT : 33 rocksdb.block.cache.filter.bytes.insert COUNT : 18220328 rocksdb.bloom.filter.useful COUNT : 5360933 rocksdb.bloom.filter.full.positive COUNT : 1321315 rocksdb.bloom.filter.full.true.positive COUNT : 1262999 $ #^ 1.08 % observed FP rate, 13.6% less memory usage for filters (Due to specific key density, this example tends to generate filters that are "worse than average" for internal fragmentation. "Better than average" cases can show little or no improvement.) Pull Request resolved: https://github.com/facebook/rocksdb/pull/6427 Test Plan: unit test added, 'make check' with gcc, clang and valgrind Reviewed By: siying Differential Revision: D22124374 Pulled By: pdillinger fbshipit-source-id: f3e3aa152f9043ddf4fae25799e76341d0d8714e
4 years ago
DECLARE_bool(optimize_filters_for_memory);
DECLARE_int32(index_type);
DECLARE_string(db);
DECLARE_string(secondaries_base);
DECLARE_bool(test_secondary);
DECLARE_string(expected_values_path);
DECLARE_bool(verify_checksum);
DECLARE_bool(mmap_read);
DECLARE_bool(mmap_write);
DECLARE_bool(use_direct_reads);
DECLARE_bool(use_direct_io_for_flush_and_compaction);
DECLARE_bool(mock_direct_io);
DECLARE_bool(statistics);
DECLARE_bool(sync);
DECLARE_bool(use_fsync);
DECLARE_int32(kill_random_test);
DECLARE_string(kill_exclude_prefixes);
DECLARE_bool(disable_wal);
DECLARE_uint64(recycle_log_file_num);
DECLARE_int64(target_file_size_base);
DECLARE_int32(target_file_size_multiplier);
DECLARE_uint64(max_bytes_for_level_base);
DECLARE_double(max_bytes_for_level_multiplier);
DECLARE_int32(range_deletion_width);
DECLARE_uint64(rate_limiter_bytes_per_sec);
DECLARE_bool(rate_limit_bg_reads);
DECLARE_uint64(sst_file_manager_bytes_per_sec);
DECLARE_uint64(sst_file_manager_bytes_per_truncate);
DECLARE_bool(use_txn);
DECLARE_uint64(txn_write_policy);
DECLARE_bool(unordered_write);
DECLARE_int32(backup_one_in);
DECLARE_uint64(backup_max_size);
DECLARE_int32(checkpoint_one_in);
DECLARE_int32(ingest_external_file_one_in);
DECLARE_int32(ingest_external_file_width);
DECLARE_int32(compact_files_one_in);
DECLARE_int32(compact_range_one_in);
DECLARE_int32(mark_for_compaction_one_file_in);
DECLARE_int32(flush_one_in);
DECLARE_int32(pause_background_one_in);
DECLARE_int32(compact_range_width);
DECLARE_int32(acquire_snapshot_one_in);
DECLARE_bool(compare_full_db_state_snapshot);
DECLARE_uint64(snapshot_hold_ops);
DECLARE_bool(long_running_snapshots);
DECLARE_bool(use_multiget);
DECLARE_int32(readpercent);
DECLARE_int32(prefixpercent);
DECLARE_int32(writepercent);
DECLARE_int32(delpercent);
DECLARE_int32(delrangepercent);
DECLARE_int32(nooverwritepercent);
DECLARE_int32(iterpercent);
DECLARE_uint64(num_iterations);
DECLARE_string(compression_type);
DECLARE_string(bottommost_compression_type);
DECLARE_int32(compression_max_dict_bytes);
DECLARE_int32(compression_zstd_max_train_bytes);
DECLARE_int32(compression_parallel_threads);
Limit buffering for collecting samples for compression dictionary (#7970) Summary: For dictionary compression, we need to collect some representative samples of the data to be compressed, which we use to either generate or train (when `CompressionOptions::zstd_max_train_bytes > 0`) a dictionary. Previously, the strategy was to buffer all the data blocks during flush, and up to the target file size during compaction. That strategy allowed us to randomly pick samples from as wide a range as possible that'd be guaranteed to land in a single output file. However, some users try to make huge files in memory-constrained environments, where this strategy can cause OOM. This PR introduces an option, `CompressionOptions::max_dict_buffer_bytes`, that limits how much data blocks are buffered before we switch to unbuffered mode (which means creating the per-SST dictionary, writing out the buffered data, and compressing/writing new blocks as soon as they are built). It is not strict as we currently buffer more than just data blocks -- also keys are buffered. But it does make a step towards giving users predictable memory usage. Related changes include: - Changed sampling for dictionary compression to select unique data blocks when there is limited availability of data blocks - Made use of `BlockBuilder::SwapAndReset()` to save an allocation+memcpy when buffering data blocks for building a dictionary - Changed `ParseBoolean()` to accept an input containing characters after the boolean. This is necessary since, with this PR, a value for `CompressionOptions::enabled` is no longer necessarily the final component in the `CompressionOptions` string. Pull Request resolved: https://github.com/facebook/rocksdb/pull/7970 Test Plan: - updated `CompressionOptions` unit tests to verify limit is respected (to the extent expected in the current implementation) in various scenarios of flush/compaction to bottommost/non-bottommost level - looked at jemalloc heap profiles right before and after switching to unbuffered mode during flush/compaction. Verified memory usage in buffering is proportional to the limit set. Reviewed By: pdillinger Differential Revision: D26467994 Pulled By: ajkr fbshipit-source-id: 3da4ef9fba59974e4ef40e40c01611002c861465
3 years ago
DECLARE_uint64(compression_max_dict_buffer_bytes);
DECLARE_string(checksum_type);
DECLARE_string(hdfs);
DECLARE_string(env_uri);
DECLARE_string(fs_uri);
DECLARE_uint64(ops_per_thread);
DECLARE_uint64(log2_keys_per_lock);
DECLARE_uint64(max_manifest_file_size);
DECLARE_bool(in_place_update);
DECLARE_int32(secondary_catch_up_one_in);
DECLARE_string(memtablerep);
DECLARE_int32(prefix_size);
DECLARE_bool(use_merge);
DECLARE_bool(use_full_merge_v1);
DECLARE_int32(sync_wal_one_in);
DECLARE_bool(avoid_unnecessary_blocking_io);
DECLARE_bool(write_dbid_to_manifest);
DECLARE_bool(avoid_flush_during_recovery);
DECLARE_uint64(max_write_batch_group_size_bytes);
DECLARE_bool(level_compaction_dynamic_level_bytes);
DECLARE_int32(verify_checksum_one_in);
DECLARE_int32(verify_db_one_in);
DECLARE_int32(continuous_verification_interval);
DECLARE_int32(get_property_one_in);
DECLARE_string(file_checksum_impl);
#ifndef ROCKSDB_LITE
// Options for StackableDB-based BlobDB
DECLARE_bool(use_blob_db);
DECLARE_uint64(blob_db_min_blob_size);
DECLARE_uint64(blob_db_bytes_per_sync);
DECLARE_uint64(blob_db_file_size);
DECLARE_bool(blob_db_enable_gc);
DECLARE_double(blob_db_gc_cutoff);
#endif // !ROCKSDB_LITE
// Options for integrated BlobDB
DECLARE_bool(allow_setting_blob_options_dynamically);
DECLARE_bool(enable_blob_files);
DECLARE_uint64(min_blob_size);
DECLARE_uint64(blob_file_size);
DECLARE_string(blob_compression_type);
DECLARE_bool(enable_blob_garbage_collection);
DECLARE_double(blob_garbage_collection_age_cutoff);
DECLARE_int32(approximate_size_one_in);
DECLARE_bool(sync_fault_injection);
DECLARE_bool(best_efforts_recovery);
DECLARE_bool(skip_verifydb);
DECLARE_bool(enable_compaction_filter);
DECLARE_bool(paranoid_file_checks);
Integrity protection for live updates to WriteBatch (#7748) Summary: This PR adds the foundation classes for key-value integrity protection and the first use case: protecting live updates from the source buffers added to `WriteBatch` through the destination buffer in `MemTable`. The width of the protection info is not yet configurable -- only eight bytes per key is supported. This PR allows users to enable protection by constructing `WriteBatch` with `protection_bytes_per_key == 8`. It does not yet expose a way for users to get integrity protection via other write APIs (e.g., `Put()`, `Merge()`, `Delete()`, etc.). The foundation classes (`ProtectionInfo.*`) embed the coverage info in their type, and provide `Protect.*()` and `Strip.*()` functions to navigate between types with different coverage. For making bytes per key configurable (for powers of two up to eight) in the future, these classes are templated on the unsigned integer type used to store the protection info. That integer contains the XOR'd result of hashes with independent seeds for all covered fields. For integer fields, the hash is computed on the raw unadjusted bytes, so the result is endian-dependent. The most significant bytes are truncated when the hash value (8 bytes) is wider than the protection integer. When `WriteBatch` is constructed with `protection_bytes_per_key == 8`, we hold a `ProtectionInfoKVOTC` (i.e., one that covers key, value, optype aka `ValueType`, timestamp, and CF ID) for each entry added to the batch. The protection info is generated from the original buffers passed by the user, as well as the original metadata generated internally. When writing to memtable, each entry is transformed to a `ProtectionInfoKVOTS` (i.e., dropping coverage of CF ID and adding coverage of sequence number), since at that point we know the sequence number, and have already selected a memtable corresponding to a particular CF. This protection info is verified once the entry is encoded in the `MemTable` buffer. Pull Request resolved: https://github.com/facebook/rocksdb/pull/7748 Test Plan: - an integration test to verify a wide variety of single-byte changes to the encoded `MemTable` buffer are caught - add to stress/crash test to verify it works in variety of configs/operations without intentional corruption - [deferred] unit tests for `ProtectionInfo.*` classes for edge cases like KV swap, `SliceParts` and `Slice` APIs are interchangeable, etc. Reviewed By: pdillinger Differential Revision: D25754492 Pulled By: ajkr fbshipit-source-id: e481bac6c03c2ab268be41359730f1ceb9964866
3 years ago
DECLARE_uint64(batch_protection_bytes_per_key);
DECLARE_uint64(user_timestamp_size);
constexpr long KB = 1024;
constexpr int kRandomValueMaxFactor = 3;
constexpr int kValueMaxLen = 100;
// wrapped posix or hdfs environment
extern ROCKSDB_NAMESPACE::Env* db_stress_env;
#ifndef NDEBUG
namespace ROCKSDB_NAMESPACE {
class FaultInjectionTestFS;
} // namespace ROCKSDB_NAMESPACE
extern std::shared_ptr<ROCKSDB_NAMESPACE::FaultInjectionTestFS> fault_fs_guard;
#endif
extern enum ROCKSDB_NAMESPACE::CompressionType compression_type_e;
extern enum ROCKSDB_NAMESPACE::CompressionType bottommost_compression_type_e;
extern enum ROCKSDB_NAMESPACE::ChecksumType checksum_type_e;
enum RepFactory { kSkipList, kHashSkipList, kVectorRep };
inline enum RepFactory StringToRepFactory(const char* ctype) {
assert(ctype);
if (!strcasecmp(ctype, "skip_list"))
return kSkipList;
else if (!strcasecmp(ctype, "prefix_hash"))
return kHashSkipList;
else if (!strcasecmp(ctype, "vector"))
return kVectorRep;
fprintf(stdout, "Cannot parse memreptable %s\n", ctype);
return kSkipList;
}
extern enum RepFactory FLAGS_rep_factory;
namespace ROCKSDB_NAMESPACE {
inline enum ROCKSDB_NAMESPACE::CompressionType StringToCompressionType(
const char* ctype) {
assert(ctype);
ROCKSDB_NAMESPACE::CompressionType ret_compression_type;
if (!strcasecmp(ctype, "disable")) {
ret_compression_type = ROCKSDB_NAMESPACE::kDisableCompressionOption;
} else if (!strcasecmp(ctype, "none")) {
ret_compression_type = ROCKSDB_NAMESPACE::kNoCompression;
} else if (!strcasecmp(ctype, "snappy")) {
ret_compression_type = ROCKSDB_NAMESPACE::kSnappyCompression;
} else if (!strcasecmp(ctype, "zlib")) {
ret_compression_type = ROCKSDB_NAMESPACE::kZlibCompression;
} else if (!strcasecmp(ctype, "bzip2")) {
ret_compression_type = ROCKSDB_NAMESPACE::kBZip2Compression;
} else if (!strcasecmp(ctype, "lz4")) {
ret_compression_type = ROCKSDB_NAMESPACE::kLZ4Compression;
} else if (!strcasecmp(ctype, "lz4hc")) {
ret_compression_type = ROCKSDB_NAMESPACE::kLZ4HCCompression;
} else if (!strcasecmp(ctype, "xpress")) {
ret_compression_type = ROCKSDB_NAMESPACE::kXpressCompression;
} else if (!strcasecmp(ctype, "zstd")) {
ret_compression_type = ROCKSDB_NAMESPACE::kZSTD;
} else {
fprintf(stderr, "Cannot parse compression type '%s'\n", ctype);
ret_compression_type =
ROCKSDB_NAMESPACE::kSnappyCompression; // default value
}
if (ret_compression_type != ROCKSDB_NAMESPACE::kDisableCompressionOption &&
!CompressionTypeSupported(ret_compression_type)) {
// Use no compression will be more portable but considering this is
// only a stress test and snappy is widely available. Use snappy here.
ret_compression_type = ROCKSDB_NAMESPACE::kSnappyCompression;
}
return ret_compression_type;
}
inline enum ROCKSDB_NAMESPACE::ChecksumType StringToChecksumType(
const char* ctype) {
assert(ctype);
auto iter = ROCKSDB_NAMESPACE::checksum_type_string_map.find(ctype);
if (iter != ROCKSDB_NAMESPACE::checksum_type_string_map.end()) {
return iter->second;
}
fprintf(stderr, "Cannot parse checksum type '%s'\n", ctype);
return ROCKSDB_NAMESPACE::kCRC32c;
}
inline std::string ChecksumTypeToString(ROCKSDB_NAMESPACE::ChecksumType ctype) {
auto iter = std::find_if(
ROCKSDB_NAMESPACE::checksum_type_string_map.begin(),
ROCKSDB_NAMESPACE::checksum_type_string_map.end(),
[&](const std::pair<std::string, ROCKSDB_NAMESPACE::ChecksumType>&
name_and_enum_val) { return name_and_enum_val.second == ctype; });
assert(iter != ROCKSDB_NAMESPACE::checksum_type_string_map.end());
return iter->first;
}
inline std::vector<std::string> SplitString(std::string src) {
std::vector<std::string> ret;
if (src.empty()) {
return ret;
}
size_t pos = 0;
size_t pos_comma;
while ((pos_comma = src.find(',', pos)) != std::string::npos) {
ret.push_back(src.substr(pos, pos_comma - pos));
pos = pos_comma + 1;
}
ret.push_back(src.substr(pos, src.length()));
return ret;
}
#ifdef _MSC_VER
#pragma warning(push)
// truncation of constant value on static_cast
#pragma warning(disable : 4309)
#endif
inline bool GetNextPrefix(const ROCKSDB_NAMESPACE::Slice& src, std::string* v) {
std::string ret = src.ToString();
for (int i = static_cast<int>(ret.size()) - 1; i >= 0; i--) {
if (ret[i] != static_cast<char>(255)) {
ret[i] = ret[i] + 1;
break;
} else if (i != 0) {
ret[i] = 0;
} else {
// all FF. No next prefix
return false;
}
}
*v = ret;
return true;
}
#ifdef _MSC_VER
#pragma warning(pop)
#endif
// convert long to a big-endian slice key
extern inline std::string GetStringFromInt(int64_t val) {
std::string little_endian_key;
std::string big_endian_key;
PutFixed64(&little_endian_key, val);
assert(little_endian_key.size() == sizeof(val));
big_endian_key.resize(sizeof(val));
for (size_t i = 0; i < sizeof(val); ++i) {
big_endian_key[i] = little_endian_key[sizeof(val) - 1 - i];
}
return big_endian_key;
}
// A struct for maintaining the parameters for generating variable length keys
struct KeyGenContext {
// Number of adjacent keys in one cycle of key lengths
uint64_t window;
// Number of keys of each possible length in a given window
std::vector<uint64_t> weights;
};
extern KeyGenContext key_gen_ctx;
// Generate a variable length key string from the given int64 val. The
// order of the keys is preserved. The key could be anywhere from 8 to
// max_key_len * 8 bytes.
// The algorithm picks the length based on the
// offset of the val within a configured window and the distribution of the
// number of keys of various lengths in that window. For example, if x, y, x are
// the weights assigned to each possible key length, the keys generated would be
// - {0}...{x-1}
// {(x-1),0}..{(x-1),(y-1)},{(x-1),(y-1),0}..{(x-1),(y-1),(z-1)} and so on.
// Additionally, a trailer of 0-7 bytes could be appended.
extern inline std::string Key(int64_t val) {
uint64_t window = key_gen_ctx.window;
size_t levels = key_gen_ctx.weights.size();
std::string key;
for (size_t level = 0; level < levels; ++level) {
uint64_t weight = key_gen_ctx.weights[level];
uint64_t offset = static_cast<uint64_t>(val) % window;
uint64_t mult = static_cast<uint64_t>(val) / window;
uint64_t pfx = mult * weight + (offset >= weight ? weight - 1 : offset);
key.append(GetStringFromInt(pfx));
if (offset < weight) {
// Use the bottom 3 bits of offset as the number of trailing 'x's in the
// key. If the next key is going to be of the next level, then skip the
// trailer as it would break ordering. If the key length is already at max,
// skip the trailer.
if (offset < weight - 1 && level < levels - 1) {
size_t trailer_len = offset & 0x7;
key.append(trailer_len, 'x');
}
break;
}
val = offset - weight;
window -= weight;
}
return key;
}
// Given a string key, map it to an index into the expected values buffer
extern inline bool GetIntVal(std::string big_endian_key, uint64_t* key_p) {
size_t size_key = big_endian_key.size();
std::vector<uint64_t> prefixes;
assert(size_key <= key_gen_ctx.weights.size() * sizeof(uint64_t));
std::string little_endian_key;
little_endian_key.resize(size_key);
for (size_t start = 0; start + sizeof(uint64_t) <= size_key;
start += sizeof(uint64_t)) {
size_t end = start + sizeof(uint64_t);
for (size_t i = 0; i < sizeof(uint64_t); ++i) {
little_endian_key[start + i] = big_endian_key[end - 1 - i];
}
Slice little_endian_slice =
Slice(&little_endian_key[start], sizeof(uint64_t));
uint64_t pfx;
if (!GetFixed64(&little_endian_slice, &pfx)) {
return false;
}
prefixes.emplace_back(pfx);
}
uint64_t key = 0;
for (size_t i = 0; i < prefixes.size(); ++i) {
uint64_t pfx = prefixes[i];
key += (pfx / key_gen_ctx.weights[i]) * key_gen_ctx.window +
pfx % key_gen_ctx.weights[i];
if (i < prefixes.size() - 1) {
// The encoding writes a `key_gen_ctx.weights[i] - 1` that counts for
// `key_gen_ctx.weights[i]` when there are more prefixes to come. So we
// need to add back the one here as we're at a non-last prefix.
++key;
}
}
*key_p = key;
return true;
}
// Given a string prefix, map it to the first corresponding index in the
// expected values buffer.
inline bool GetFirstIntValInPrefix(std::string big_endian_prefix,
uint64_t* key_p) {
size_t size_key = big_endian_prefix.size();
// Pad with zeros to make it a multiple of 8. This function may be called
// with a prefix, in which case we return the first index that falls
// inside or outside that prefix, dependeing on whether the prefix is
// the start of upper bound of a scan
unsigned int pad = sizeof(uint64_t) - (size_key % sizeof(uint64_t));
if (pad < sizeof(uint64_t)) {
big_endian_prefix.append(pad, '\0');
}
return GetIntVal(std::move(big_endian_prefix), key_p);
}
extern inline uint64_t GetPrefixKeyCount(const std::string& prefix,
const std::string& ub) {
uint64_t start = 0;
uint64_t end = 0;
if (!GetFirstIntValInPrefix(prefix, &start) ||
!GetFirstIntValInPrefix(ub, &end)) {
return 0;
}
return end - start;
}
extern inline std::string StringToHex(const std::string& str) {
std::string result = "0x";
result.append(Slice(str).ToString(true));
return result;
}
// Unified output format for double parameters
extern inline std::string FormatDoubleParam(double param) {
return std::to_string(param);
}
// Make sure that double parameter is a value we can reproduce by
// re-inputting the value printed.
extern inline void SanitizeDoubleParam(double* param) {
*param = std::atof(FormatDoubleParam(*param).c_str());
}
extern void PoolSizeChangeThread(void* v);
extern void DbVerificationThread(void* v);
extern void PrintKeyValue(int cf, uint64_t key, const char* value, size_t sz);
extern int64_t GenerateOneKey(ThreadState* thread, uint64_t iteration);
extern std::vector<int64_t> GenerateNKeys(ThreadState* thread, int num_keys,
uint64_t iteration);
extern size_t GenerateValue(uint32_t rand, char* v, size_t max_sz);
extern StressTest* CreateCfConsistencyStressTest();
extern StressTest* CreateBatchedOpsStressTest();
extern StressTest* CreateNonBatchedOpsStressTest();
extern void InitializeHotKeyGenerator(double alpha);
extern int64_t GetOneHotKeyID(double rand_seed, int64_t max_key);
extern std::string GenerateTimestampForRead();
extern std::string NowNanosStr();
std::shared_ptr<FileChecksumGenFactory> GetFileChecksumImpl(
const std::string& name);
} // namespace ROCKSDB_NAMESPACE
#endif // GFLAGS