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/cf_consistency_stress.cc

641 lines
21 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.
#ifdef GFLAGS
#include "db_stress_tool/db_stress_common.h"
#include "file/file_util.h"
namespace ROCKSDB_NAMESPACE {
class CfConsistencyStressTest : public StressTest {
public:
CfConsistencyStressTest() : batch_id_(0) {}
~CfConsistencyStressTest() override {}
bool IsStateTracked() const override { return false; }
Status TestPut(ThreadState* thread, WriteOptions& write_opts,
const ReadOptions& /* read_opts */,
const std::vector<int>& rand_column_families,
const std::vector<int64_t>& rand_keys,
char (&value)[100]) override {
Add the PutEntity API to the stress/crash tests (#10760) Summary: The patch adds the `PutEntity` API to the non-batched, batched, and CF consistency stress tests. Namely, when the new `db_stress` command line parameter `use_put_entity_one_in` is greater than zero, one in N writes on average is performed using `PutEntity` rather than `Put`. The wide-column entity written has the generated value in its default column; in addition, it contains up to three additional columns where the original generated value is divided up between the column name and the column value (with the column name containing the first k characters of the generated value, and the column value containing the rest). Whether `PutEntity` is used (and if so, how many columns the entity has) is completely determined by the "value base" used to generate the value (that is, there is no randomness involved). Assuming the same `use_put_entity_one_in` setting is used across `db_stress` invocations, this enables us to reconstruct and validate the entity during subsequent `db_stress` runs. Note that `PutEntity` is currently incompatible with `Merge`, transactions, and user-defined timestamps; these combinations are currently disabled/disallowed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10760 Test Plan: Ran some batched, non-batched, and CF consistency stress tests using the script. Reviewed By: riversand963 Differential Revision: D39939032 Pulled By: ltamasi fbshipit-source-id: eafdf124e95993fb7d73158e3b006d11819f7fa9
2 years ago
assert(!rand_column_families.empty());
assert(!rand_keys.empty());
const std::string k = Key(rand_keys[0]);
const uint32_t value_base = batch_id_.fetch_add(1);
const size_t sz = GenerateValue(value_base, value, sizeof(value));
const Slice v(value, sz);
WriteBatch batch;
Add the PutEntity API to the stress/crash tests (#10760) Summary: The patch adds the `PutEntity` API to the non-batched, batched, and CF consistency stress tests. Namely, when the new `db_stress` command line parameter `use_put_entity_one_in` is greater than zero, one in N writes on average is performed using `PutEntity` rather than `Put`. The wide-column entity written has the generated value in its default column; in addition, it contains up to three additional columns where the original generated value is divided up between the column name and the column value (with the column name containing the first k characters of the generated value, and the column value containing the rest). Whether `PutEntity` is used (and if so, how many columns the entity has) is completely determined by the "value base" used to generate the value (that is, there is no randomness involved). Assuming the same `use_put_entity_one_in` setting is used across `db_stress` invocations, this enables us to reconstruct and validate the entity during subsequent `db_stress` runs. Note that `PutEntity` is currently incompatible with `Merge`, transactions, and user-defined timestamps; these combinations are currently disabled/disallowed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10760 Test Plan: Ran some batched, non-batched, and CF consistency stress tests using the script. Reviewed By: riversand963 Differential Revision: D39939032 Pulled By: ltamasi fbshipit-source-id: eafdf124e95993fb7d73158e3b006d11819f7fa9
2 years ago
const bool use_put_entity = !FLAGS_use_merge &&
FLAGS_use_put_entity_one_in > 0 &&
(value_base % FLAGS_use_put_entity_one_in) == 0;
for (auto cf : rand_column_families) {
Add the PutEntity API to the stress/crash tests (#10760) Summary: The patch adds the `PutEntity` API to the non-batched, batched, and CF consistency stress tests. Namely, when the new `db_stress` command line parameter `use_put_entity_one_in` is greater than zero, one in N writes on average is performed using `PutEntity` rather than `Put`. The wide-column entity written has the generated value in its default column; in addition, it contains up to three additional columns where the original generated value is divided up between the column name and the column value (with the column name containing the first k characters of the generated value, and the column value containing the rest). Whether `PutEntity` is used (and if so, how many columns the entity has) is completely determined by the "value base" used to generate the value (that is, there is no randomness involved). Assuming the same `use_put_entity_one_in` setting is used across `db_stress` invocations, this enables us to reconstruct and validate the entity during subsequent `db_stress` runs. Note that `PutEntity` is currently incompatible with `Merge`, transactions, and user-defined timestamps; these combinations are currently disabled/disallowed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10760 Test Plan: Ran some batched, non-batched, and CF consistency stress tests using the script. Reviewed By: riversand963 Differential Revision: D39939032 Pulled By: ltamasi fbshipit-source-id: eafdf124e95993fb7d73158e3b006d11819f7fa9
2 years ago
ColumnFamilyHandle* const cfh = column_families_[cf];
assert(cfh);
if (FLAGS_use_merge) {
Add the PutEntity API to the stress/crash tests (#10760) Summary: The patch adds the `PutEntity` API to the non-batched, batched, and CF consistency stress tests. Namely, when the new `db_stress` command line parameter `use_put_entity_one_in` is greater than zero, one in N writes on average is performed using `PutEntity` rather than `Put`. The wide-column entity written has the generated value in its default column; in addition, it contains up to three additional columns where the original generated value is divided up between the column name and the column value (with the column name containing the first k characters of the generated value, and the column value containing the rest). Whether `PutEntity` is used (and if so, how many columns the entity has) is completely determined by the "value base" used to generate the value (that is, there is no randomness involved). Assuming the same `use_put_entity_one_in` setting is used across `db_stress` invocations, this enables us to reconstruct and validate the entity during subsequent `db_stress` runs. Note that `PutEntity` is currently incompatible with `Merge`, transactions, and user-defined timestamps; these combinations are currently disabled/disallowed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10760 Test Plan: Ran some batched, non-batched, and CF consistency stress tests using the script. Reviewed By: riversand963 Differential Revision: D39939032 Pulled By: ltamasi fbshipit-source-id: eafdf124e95993fb7d73158e3b006d11819f7fa9
2 years ago
batch.Merge(cfh, k, v);
} else if (use_put_entity) {
batch.PutEntity(cfh, k, GenerateWideColumns(value_base, v));
} else {
batch.Put(cfh, k, v);
}
}
Add the PutEntity API to the stress/crash tests (#10760) Summary: The patch adds the `PutEntity` API to the non-batched, batched, and CF consistency stress tests. Namely, when the new `db_stress` command line parameter `use_put_entity_one_in` is greater than zero, one in N writes on average is performed using `PutEntity` rather than `Put`. The wide-column entity written has the generated value in its default column; in addition, it contains up to three additional columns where the original generated value is divided up between the column name and the column value (with the column name containing the first k characters of the generated value, and the column value containing the rest). Whether `PutEntity` is used (and if so, how many columns the entity has) is completely determined by the "value base" used to generate the value (that is, there is no randomness involved). Assuming the same `use_put_entity_one_in` setting is used across `db_stress` invocations, this enables us to reconstruct and validate the entity during subsequent `db_stress` runs. Note that `PutEntity` is currently incompatible with `Merge`, transactions, and user-defined timestamps; these combinations are currently disabled/disallowed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10760 Test Plan: Ran some batched, non-batched, and CF consistency stress tests using the script. Reviewed By: riversand963 Differential Revision: D39939032 Pulled By: ltamasi fbshipit-source-id: eafdf124e95993fb7d73158e3b006d11819f7fa9
2 years ago
Status s = db_->Write(write_opts, &batch);
Add the PutEntity API to the stress/crash tests (#10760) Summary: The patch adds the `PutEntity` API to the non-batched, batched, and CF consistency stress tests. Namely, when the new `db_stress` command line parameter `use_put_entity_one_in` is greater than zero, one in N writes on average is performed using `PutEntity` rather than `Put`. The wide-column entity written has the generated value in its default column; in addition, it contains up to three additional columns where the original generated value is divided up between the column name and the column value (with the column name containing the first k characters of the generated value, and the column value containing the rest). Whether `PutEntity` is used (and if so, how many columns the entity has) is completely determined by the "value base" used to generate the value (that is, there is no randomness involved). Assuming the same `use_put_entity_one_in` setting is used across `db_stress` invocations, this enables us to reconstruct and validate the entity during subsequent `db_stress` runs. Note that `PutEntity` is currently incompatible with `Merge`, transactions, and user-defined timestamps; these combinations are currently disabled/disallowed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10760 Test Plan: Ran some batched, non-batched, and CF consistency stress tests using the script. Reviewed By: riversand963 Differential Revision: D39939032 Pulled By: ltamasi fbshipit-source-id: eafdf124e95993fb7d73158e3b006d11819f7fa9
2 years ago
if (!s.ok()) {
fprintf(stderr, "multi put or merge error: %s\n", s.ToString().c_str());
thread->stats.AddErrors(1);
} else {
auto num = static_cast<long>(rand_column_families.size());
thread->stats.AddBytesForWrites(num, (sz + 1) * num);
}
return s;
}
Status TestDelete(ThreadState* thread, WriteOptions& write_opts,
const std::vector<int>& rand_column_families,
const std::vector<int64_t>& rand_keys) override {
std::string key_str = Key(rand_keys[0]);
Slice key = key_str;
WriteBatch batch;
for (auto cf : rand_column_families) {
ColumnFamilyHandle* cfh = column_families_[cf];
batch.Delete(cfh, key);
}
Status s = db_->Write(write_opts, &batch);
if (!s.ok()) {
fprintf(stderr, "multidel error: %s\n", s.ToString().c_str());
thread->stats.AddErrors(1);
} else {
thread->stats.AddDeletes(static_cast<long>(rand_column_families.size()));
}
return s;
}
Status TestDeleteRange(ThreadState* thread, WriteOptions& write_opts,
const std::vector<int>& rand_column_families,
const std::vector<int64_t>& rand_keys) override {
int64_t rand_key = rand_keys[0];
auto shared = thread->shared;
int64_t max_key = shared->GetMaxKey();
if (rand_key > max_key - FLAGS_range_deletion_width) {
rand_key =
thread->rand.Next() % (max_key - FLAGS_range_deletion_width + 1);
}
std::string key_str = Key(rand_key);
Slice key = key_str;
std::string end_key_str = Key(rand_key + FLAGS_range_deletion_width);
Slice end_key = end_key_str;
WriteBatch batch;
for (auto cf : rand_column_families) {
ColumnFamilyHandle* cfh = column_families_[rand_column_families[cf]];
batch.DeleteRange(cfh, key, end_key);
}
Status s = db_->Write(write_opts, &batch);
if (!s.ok()) {
fprintf(stderr, "multi del range error: %s\n", s.ToString().c_str());
thread->stats.AddErrors(1);
} else {
thread->stats.AddRangeDeletions(
static_cast<long>(rand_column_families.size()));
}
return s;
}
void TestIngestExternalFile(
ThreadState* /* thread */,
const std::vector<int>& /* rand_column_families */,
const std::vector<int64_t>& /* rand_keys */) override {
assert(false);
fprintf(stderr,
"CfConsistencyStressTest does not support TestIngestExternalFile "
"because it's not possible to verify the result\n");
std::terminate();
}
Status TestGet(ThreadState* thread, const ReadOptions& readoptions,
const std::vector<int>& rand_column_families,
const std::vector<int64_t>& rand_keys) override {
std::string key_str = Key(rand_keys[0]);
Slice key = key_str;
Status s;
bool is_consistent = true;
if (thread->rand.OneIn(2)) {
// 1/2 chance, does a random read from random CF
auto cfh =
column_families_[rand_column_families[thread->rand.Next() %
rand_column_families.size()]];
std::string from_db;
s = db_->Get(readoptions, cfh, key, &from_db);
} else {
// 1/2 chance, comparing one key is the same across all CFs
const Snapshot* snapshot = db_->GetSnapshot();
ReadOptions readoptionscopy = readoptions;
readoptionscopy.snapshot = snapshot;
std::string value0;
s = db_->Get(readoptionscopy, column_families_[rand_column_families[0]],
key, &value0);
if (s.ok() || s.IsNotFound()) {
bool found = s.ok();
for (size_t i = 1; i < rand_column_families.size(); i++) {
std::string value1;
s = db_->Get(readoptionscopy,
column_families_[rand_column_families[i]], key, &value1);
if (!s.ok() && !s.IsNotFound()) {
break;
}
if (!found && s.ok()) {
fprintf(stderr, "Get() return different results with key %s\n",
Slice(key_str).ToString(true).c_str());
fprintf(stderr, "CF %s is not found\n",
column_family_names_[0].c_str());
fprintf(stderr, "CF %s returns value %s\n",
column_family_names_[i].c_str(),
Slice(value1).ToString(true).c_str());
is_consistent = false;
} else if (found && s.IsNotFound()) {
fprintf(stderr, "Get() return different results with key %s\n",
Slice(key_str).ToString(true).c_str());
fprintf(stderr, "CF %s returns value %s\n",
column_family_names_[0].c_str(),
Slice(value0).ToString(true).c_str());
fprintf(stderr, "CF %s is not found\n",
column_family_names_[i].c_str());
is_consistent = false;
} else if (s.ok() && value0 != value1) {
fprintf(stderr, "Get() return different results with key %s\n",
Slice(key_str).ToString(true).c_str());
fprintf(stderr, "CF %s returns value %s\n",
column_family_names_[0].c_str(),
Slice(value0).ToString(true).c_str());
fprintf(stderr, "CF %s returns value %s\n",
column_family_names_[i].c_str(),
Slice(value1).ToString(true).c_str());
is_consistent = false;
}
if (!is_consistent) {
break;
}
}
}
db_->ReleaseSnapshot(snapshot);
}
if (!is_consistent) {
fprintf(stderr, "TestGet error: is_consistent is false\n");
thread->stats.AddErrors(1);
// Fail fast to preserve the DB state.
thread->shared->SetVerificationFailure();
} else if (s.ok()) {
thread->stats.AddGets(1, 1);
} else if (s.IsNotFound()) {
thread->stats.AddGets(1, 0);
} else {
fprintf(stderr, "TestGet error: %s\n", s.ToString().c_str());
thread->stats.AddErrors(1);
}
return s;
}
std::vector<Status> TestMultiGet(
ThreadState* thread, const ReadOptions& read_opts,
const std::vector<int>& rand_column_families,
const std::vector<int64_t>& rand_keys) override {
size_t num_keys = rand_keys.size();
std::vector<std::string> key_str;
std::vector<Slice> keys;
keys.reserve(num_keys);
key_str.reserve(num_keys);
std::vector<PinnableSlice> values(num_keys);
std::vector<Status> statuses(num_keys);
ColumnFamilyHandle* cfh = column_families_[rand_column_families[0]];
Add rate-limiting support to batched MultiGet() (#10159) Summary: **Context/Summary:** https://github.com/facebook/rocksdb/pull/9424 added rate-limiting support for user reads, which does not include batched `MultiGet()`s that call `RandomAccessFileReader::MultiRead()`. The reason is that it's harder (compared with RandomAccessFileReader::Read()) to implement the ideal rate-limiting where we first call `RateLimiter::RequestToken()` for allowed bytes to multi-read and then consume those bytes by satisfying as many requests in `MultiRead()` as possible. For example, it can be tricky to decide whether we want partially fulfilled requests within one `MultiRead()` or not. However, due to a recent urgent user request, we decide to pursue an elementary (but a conditionally ineffective) solution where we accumulate enough rate limiter requests toward the total bytes needed by one `MultiRead()` before doing that `MultiRead()`. This is not ideal when the total bytes are huge as we will actually consume a huge bandwidth from rate-limiter causing a burst on disk. This is not what we ultimately want with rate limiter. Therefore a follow-up work is noted through TODO comments. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10159 Test Plan: - Modified existing unit test `DBRateLimiterOnReadTest/DBRateLimiterOnReadTest.NewMultiGet` - Traced the underlying system calls `io_uring_enter` and verified they are 10 seconds apart from each other correctly under the setting of `strace -ftt -e trace=io_uring_enter ./db_bench -benchmarks=multireadrandom -db=/dev/shm/testdb2 -readonly -num=50 -threads=1 -multiread_batched=1 -batch_size=100 -duration=10 -rate_limiter_bytes_per_sec=200 -rate_limiter_refill_period_us=1000000 -rate_limit_bg_reads=1 -disable_auto_compactions=1 -rate_limit_user_ops=1` where each `MultiRead()` read about 2000 bytes (inspected by debugger) and the rate limiter grants 200 bytes per seconds. - Stress test: - Verified `./db_stress (-test_cf_consistency=1/test_batches_snapshots=1) -use_multiget=1 -cache_size=1048576 -rate_limiter_bytes_per_sec=10241024 -rate_limit_bg_reads=1 -rate_limit_user_ops=1` work Reviewed By: ajkr, anand1976 Differential Revision: D37135172 Pulled By: hx235 fbshipit-source-id: 73b8e8f14761e5d4b77235dfe5d41f4eea968bcd
3 years ago
ReadOptions readoptionscopy = read_opts;
readoptionscopy.rate_limiter_priority =
FLAGS_rate_limit_user_ops ? Env::IO_USER : Env::IO_TOTAL;
for (size_t i = 0; i < num_keys; ++i) {
key_str.emplace_back(Key(rand_keys[i]));
keys.emplace_back(key_str.back());
}
Add rate-limiting support to batched MultiGet() (#10159) Summary: **Context/Summary:** https://github.com/facebook/rocksdb/pull/9424 added rate-limiting support for user reads, which does not include batched `MultiGet()`s that call `RandomAccessFileReader::MultiRead()`. The reason is that it's harder (compared with RandomAccessFileReader::Read()) to implement the ideal rate-limiting where we first call `RateLimiter::RequestToken()` for allowed bytes to multi-read and then consume those bytes by satisfying as many requests in `MultiRead()` as possible. For example, it can be tricky to decide whether we want partially fulfilled requests within one `MultiRead()` or not. However, due to a recent urgent user request, we decide to pursue an elementary (but a conditionally ineffective) solution where we accumulate enough rate limiter requests toward the total bytes needed by one `MultiRead()` before doing that `MultiRead()`. This is not ideal when the total bytes are huge as we will actually consume a huge bandwidth from rate-limiter causing a burst on disk. This is not what we ultimately want with rate limiter. Therefore a follow-up work is noted through TODO comments. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10159 Test Plan: - Modified existing unit test `DBRateLimiterOnReadTest/DBRateLimiterOnReadTest.NewMultiGet` - Traced the underlying system calls `io_uring_enter` and verified they are 10 seconds apart from each other correctly under the setting of `strace -ftt -e trace=io_uring_enter ./db_bench -benchmarks=multireadrandom -db=/dev/shm/testdb2 -readonly -num=50 -threads=1 -multiread_batched=1 -batch_size=100 -duration=10 -rate_limiter_bytes_per_sec=200 -rate_limiter_refill_period_us=1000000 -rate_limit_bg_reads=1 -disable_auto_compactions=1 -rate_limit_user_ops=1` where each `MultiRead()` read about 2000 bytes (inspected by debugger) and the rate limiter grants 200 bytes per seconds. - Stress test: - Verified `./db_stress (-test_cf_consistency=1/test_batches_snapshots=1) -use_multiget=1 -cache_size=1048576 -rate_limiter_bytes_per_sec=10241024 -rate_limit_bg_reads=1 -rate_limit_user_ops=1` work Reviewed By: ajkr, anand1976 Differential Revision: D37135172 Pulled By: hx235 fbshipit-source-id: 73b8e8f14761e5d4b77235dfe5d41f4eea968bcd
3 years ago
db_->MultiGet(readoptionscopy, cfh, num_keys, keys.data(), values.data(),
statuses.data());
for (auto s : statuses) {
if (s.ok()) {
// found case
thread->stats.AddGets(1, 1);
} else if (s.IsNotFound()) {
// not found case
thread->stats.AddGets(1, 0);
} else {
// errors case
fprintf(stderr, "MultiGet error: %s\n", s.ToString().c_str());
thread->stats.AddErrors(1);
}
}
return statuses;
}
Status TestPrefixScan(ThreadState* thread, const ReadOptions& readoptions,
const std::vector<int>& rand_column_families,
const std::vector<int64_t>& rand_keys) override {
assert(!rand_column_families.empty());
assert(!rand_keys.empty());
const std::string key = Key(rand_keys[0]);
const size_t prefix_to_use =
(FLAGS_prefix_size < 0) ? 7 : static_cast<size_t>(FLAGS_prefix_size);
const Slice prefix(key.data(), prefix_to_use);
std::string upper_bound;
Slice ub_slice;
ReadOptions ro_copy = readoptions;
// Get the next prefix first and then see if we want to set upper bound.
// We'll use the next prefix in an assertion later on
if (GetNextPrefix(prefix, &upper_bound) && thread->rand.OneIn(2)) {
ub_slice = Slice(upper_bound);
ro_copy.iterate_upper_bound = &ub_slice;
}
ColumnFamilyHandle* const cfh =
column_families_[rand_column_families[thread->rand.Uniform(
static_cast<int>(rand_column_families.size()))]];
assert(cfh);
std::unique_ptr<Iterator> iter(db_->NewIterator(ro_copy, cfh));
uint64_t count = 0;
Status s;
for (iter->Seek(prefix); iter->Valid() && iter->key().starts_with(prefix);
iter->Next()) {
++count;
const WideColumns expected_columns = GenerateExpectedWideColumns(
GetValueBase(iter->value()), iter->value());
if (iter->columns() != expected_columns) {
s = Status::Corruption(
"Value and columns inconsistent",
DebugString(iter->value(), iter->columns(), expected_columns));
break;
}
}
assert(prefix_to_use == 0 ||
count <= GetPrefixKeyCount(prefix.ToString(), upper_bound));
if (s.ok()) {
s = iter->status();
}
if (!s.ok()) {
fprintf(stderr, "TestPrefixScan error: %s\n", s.ToString().c_str());
thread->stats.AddErrors(1);
return s;
}
thread->stats.AddPrefixes(1, count);
return Status::OK();
}
ColumnFamilyHandle* GetControlCfh(ThreadState* thread,
int /*column_family_id*/
) override {
// All column families should contain the same data. Randomly pick one.
return column_families_[thread->rand.Next() % column_families_.size()];
}
void VerifyDb(ThreadState* thread) const override {
Add rate limiter priority to ReadOptions (#9424) Summary: Users can set the priority for file reads associated with their operation by setting `ReadOptions::rate_limiter_priority` to something other than `Env::IO_TOTAL`. Rate limiting `VerifyChecksum()` and `VerifyFileChecksums()` is the motivation for this PR, so it also includes benchmarks and minor bug fixes to get that working. `RandomAccessFileReader::Read()` already had support for rate limiting compaction reads. I changed that rate limiting to be non-specific to compaction, but rather performed according to the passed in `Env::IOPriority`. Now the compaction read rate limiting is supported by setting `rate_limiter_priority = Env::IO_LOW` on its `ReadOptions`. There is no default value for the new `Env::IOPriority` parameter to `RandomAccessFileReader::Read()`. That means this PR goes through all callers (in some cases multiple layers up the call stack) to find a `ReadOptions` to provide the priority. There are TODOs for cases I believe it would be good to let user control the priority some day (e.g., file footer reads), and no TODO in cases I believe it doesn't matter (e.g., trace file reads). The API doc only lists the missing cases where a file read associated with a provided `ReadOptions` cannot be rate limited. For cases like file ingestion checksum calculation, there is no API to provide `ReadOptions` or `Env::IOPriority`, so I didn't count that as missing. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9424 Test Plan: - new unit tests - new benchmarks on ~50MB database with 1MB/s read rate limit and 100ms refill interval; verified with strace reads are chunked (at 0.1MB per chunk) and spaced roughly 100ms apart. - setup command: `./db_bench -benchmarks=fillrandom,compact -db=/tmp/testdb -target_file_size_base=1048576 -disable_auto_compactions=true -file_checksum=true` - benchmarks command: `strace -ttfe pread64 ./db_bench -benchmarks=verifychecksum,verifyfilechecksums -use_existing_db=true -db=/tmp/testdb -rate_limiter_bytes_per_sec=1048576 -rate_limit_bg_reads=1 -rate_limit_user_ops=true -file_checksum=true` - crash test using IO_USER priority on non-validation reads with https://github.com/facebook/rocksdb/issues/9567 reverted: `python3 tools/db_crashtest.py blackbox --max_key=1000000 --write_buffer_size=524288 --target_file_size_base=524288 --level_compaction_dynamic_level_bytes=true --duration=3600 --rate_limit_bg_reads=true --rate_limit_user_ops=true --rate_limiter_bytes_per_sec=10485760 --interval=10` Reviewed By: hx235 Differential Revision: D33747386 Pulled By: ajkr fbshipit-source-id: a2d985e97912fba8c54763798e04f006ccc56e0c
3 years ago
// This `ReadOptions` is for validation purposes. Ignore
// `FLAGS_rate_limit_user_ops` to avoid slowing any validation.
ReadOptions options(FLAGS_verify_checksum, true);
// We must set total_order_seek to true because we are doing a SeekToFirst
// on a column family whose memtables may support (by default) prefix-based
// iterator. In this case, NewIterator with options.total_order_seek being
// false returns a prefix-based iterator. Calling SeekToFirst using this
// iterator causes the iterator to become invalid. That means we cannot
// iterate the memtable using this iterator any more, although the memtable
// contains the most up-to-date key-values.
options.total_order_seek = true;
ManagedSnapshot snapshot_guard(db_);
options.snapshot = snapshot_guard.snapshot();
const size_t num = column_families_.size();
std::vector<std::unique_ptr<Iterator>> iters;
iters.reserve(num);
for (size_t i = 0; i < num; ++i) {
iters.emplace_back(db_->NewIterator(options, column_families_[i]));
iters.back()->SeekToFirst();
}
std::vector<Status> statuses(num, Status::OK());
assert(thread);
auto shared = thread->shared;
assert(shared);
do {
if (shared->HasVerificationFailedYet()) {
break;
}
size_t valid_cnt = 0;
for (size_t i = 0; i < num; ++i) {
const auto& iter = iters[i];
assert(iter);
if (iter->Valid()) {
const WideColumns expected_columns = GenerateExpectedWideColumns(
GetValueBase(iter->value()), iter->value());
if (iter->columns() != expected_columns) {
statuses[i] = Status::Corruption(
"Value and columns inconsistent",
DebugString(iter->value(), iter->columns(), expected_columns));
} else {
++valid_cnt;
}
} else {
statuses[i] = iter->status();
}
}
if (valid_cnt == 0) {
for (size_t i = 0; i < num; ++i) {
const auto& s = statuses[i];
if (!s.ok()) {
fprintf(stderr, "Iterator on cf %s has error: %s\n",
column_families_[i]->GetName().c_str(),
s.ToString().c_str());
shared->SetVerificationFailure();
}
}
break;
}
if (valid_cnt < num) {
shared->SetVerificationFailure();
for (size_t i = 0; i < num; ++i) {
assert(iters[i]);
if (!iters[i]->Valid()) {
if (statuses[i].ok()) {
fprintf(stderr, "Finished scanning cf %s\n",
column_families_[i]->GetName().c_str());
} else {
fprintf(stderr, "Iterator on cf %s has error: %s\n",
column_families_[i]->GetName().c_str(),
statuses[i].ToString().c_str());
}
} else {
fprintf(stderr, "cf %s has remaining data to scan\n",
column_families_[i]->GetName().c_str());
}
}
break;
}
if (shared->HasVerificationFailedYet()) {
break;
}
// If the program reaches here, then all column families' iterators are
// still valid.
assert(valid_cnt == num);
if (shared->PrintingVerificationResults()) {
continue;
}
assert(iters[0]);
const Slice key = iters[0]->key();
const Slice value = iters[0]->value();
int num_mismatched_cfs = 0;
for (size_t i = 1; i < num; ++i) {
assert(iters[i]);
const int cmp = key.compare(iters[i]->key());
if (cmp != 0) {
++num_mismatched_cfs;
if (1 == num_mismatched_cfs) {
fprintf(stderr, "Verification failed\n");
fprintf(stderr, "Latest Sequence Number: %" PRIu64 "\n",
db_->GetLatestSequenceNumber());
fprintf(stderr, "[%s] %s => %s\n",
column_families_[0]->GetName().c_str(),
key.ToString(true /* hex */).c_str(),
value.ToString(true /* hex */).c_str());
}
fprintf(stderr, "[%s] %s => %s\n",
column_families_[i]->GetName().c_str(),
iters[i]->key().ToString(true /* hex */).c_str(),
iters[i]->value().ToString(true /* hex */).c_str());
#ifndef ROCKSDB_LITE
Slice begin_key;
Slice end_key;
if (cmp < 0) {
begin_key = key;
end_key = iters[i]->key();
} else {
begin_key = iters[i]->key();
end_key = key;
}
const auto print_key_versions = [&](ColumnFamilyHandle* cfh) {
constexpr size_t kMaxNumIKeys = 8;
std::vector<KeyVersion> versions;
const Status s = GetAllKeyVersions(db_, cfh, begin_key, end_key,
kMaxNumIKeys, &versions);
if (!s.ok()) {
fprintf(stderr, "%s\n", s.ToString().c_str());
return;
}
assert(cfh);
fprintf(stderr,
"Internal keys in CF '%s', [%s, %s] (max %" ROCKSDB_PRIszt
")\n",
cfh->GetName().c_str(),
begin_key.ToString(true /* hex */).c_str(),
end_key.ToString(true /* hex */).c_str(), kMaxNumIKeys);
for (const KeyVersion& kv : versions) {
fprintf(stderr, " key %s seq %" PRIu64 " type %d\n",
Slice(kv.user_key).ToString(true).c_str(), kv.sequence,
kv.type);
}
};
if (1 == num_mismatched_cfs) {
print_key_versions(column_families_[0]);
}
print_key_versions(column_families_[i]);
#endif // ROCKSDB_LITE
shared->SetVerificationFailure();
}
}
shared->FinishPrintingVerificationResults();
for (auto& iter : iters) {
assert(iter);
iter->Next();
}
} while (true);
}
#ifndef ROCKSDB_LITE
void ContinuouslyVerifyDb(ThreadState* thread) const override {
assert(thread);
Status status;
DB* db_ptr = cmp_db_ ? cmp_db_ : db_;
const auto& cfhs = cmp_db_ ? cmp_cfhs_ : column_families_;
// Take a snapshot to preserve the state of primary db.
ManagedSnapshot snapshot_guard(db_);
SharedState* shared = thread->shared;
assert(shared);
if (cmp_db_) {
status = cmp_db_->TryCatchUpWithPrimary();
if (!status.ok()) {
fprintf(stderr, "TryCatchUpWithPrimary: %s\n",
status.ToString().c_str());
shared->SetShouldStopTest();
assert(false);
return;
}
}
const auto checksum_column_family = [](Iterator* iter,
uint32_t* checksum) -> Status {
assert(nullptr != checksum);
uint32_t ret = 0;
for (iter->SeekToFirst(); iter->Valid(); iter->Next()) {
ret = crc32c::Extend(ret, iter->key().data(), iter->key().size());
ret = crc32c::Extend(ret, iter->value().data(), iter->value().size());
for (const auto& column : iter->columns()) {
ret = crc32c::Extend(ret, column.name().data(), column.name().size());
ret =
crc32c::Extend(ret, column.value().data(), column.value().size());
}
}
*checksum = ret;
return iter->status();
};
Add rate limiter priority to ReadOptions (#9424) Summary: Users can set the priority for file reads associated with their operation by setting `ReadOptions::rate_limiter_priority` to something other than `Env::IO_TOTAL`. Rate limiting `VerifyChecksum()` and `VerifyFileChecksums()` is the motivation for this PR, so it also includes benchmarks and minor bug fixes to get that working. `RandomAccessFileReader::Read()` already had support for rate limiting compaction reads. I changed that rate limiting to be non-specific to compaction, but rather performed according to the passed in `Env::IOPriority`. Now the compaction read rate limiting is supported by setting `rate_limiter_priority = Env::IO_LOW` on its `ReadOptions`. There is no default value for the new `Env::IOPriority` parameter to `RandomAccessFileReader::Read()`. That means this PR goes through all callers (in some cases multiple layers up the call stack) to find a `ReadOptions` to provide the priority. There are TODOs for cases I believe it would be good to let user control the priority some day (e.g., file footer reads), and no TODO in cases I believe it doesn't matter (e.g., trace file reads). The API doc only lists the missing cases where a file read associated with a provided `ReadOptions` cannot be rate limited. For cases like file ingestion checksum calculation, there is no API to provide `ReadOptions` or `Env::IOPriority`, so I didn't count that as missing. Pull Request resolved: https://github.com/facebook/rocksdb/pull/9424 Test Plan: - new unit tests - new benchmarks on ~50MB database with 1MB/s read rate limit and 100ms refill interval; verified with strace reads are chunked (at 0.1MB per chunk) and spaced roughly 100ms apart. - setup command: `./db_bench -benchmarks=fillrandom,compact -db=/tmp/testdb -target_file_size_base=1048576 -disable_auto_compactions=true -file_checksum=true` - benchmarks command: `strace -ttfe pread64 ./db_bench -benchmarks=verifychecksum,verifyfilechecksums -use_existing_db=true -db=/tmp/testdb -rate_limiter_bytes_per_sec=1048576 -rate_limit_bg_reads=1 -rate_limit_user_ops=true -file_checksum=true` - crash test using IO_USER priority on non-validation reads with https://github.com/facebook/rocksdb/issues/9567 reverted: `python3 tools/db_crashtest.py blackbox --max_key=1000000 --write_buffer_size=524288 --target_file_size_base=524288 --level_compaction_dynamic_level_bytes=true --duration=3600 --rate_limit_bg_reads=true --rate_limit_user_ops=true --rate_limiter_bytes_per_sec=10485760 --interval=10` Reviewed By: hx235 Differential Revision: D33747386 Pulled By: ajkr fbshipit-source-id: a2d985e97912fba8c54763798e04f006ccc56e0c
3 years ago
// This `ReadOptions` is for validation purposes. Ignore
// `FLAGS_rate_limit_user_ops` to avoid slowing any validation.
ReadOptions ropts(FLAGS_verify_checksum, true);
ropts.total_order_seek = true;
if (nullptr == cmp_db_) {
ropts.snapshot = snapshot_guard.snapshot();
}
uint32_t crc = 0;
{
// Compute crc for all key-values of default column family.
std::unique_ptr<Iterator> it(db_ptr->NewIterator(ropts));
status = checksum_column_family(it.get(), &crc);
if (!status.ok()) {
fprintf(stderr, "Computing checksum of default cf: %s\n",
status.ToString().c_str());
assert(false);
}
}
// Since we currently intentionally disallow reading from the secondary
// instance with snapshot, we cannot achieve cross-cf consistency if WAL is
// enabled because there is no guarantee that secondary instance replays
// the primary's WAL to a consistent point where all cfs have the same
// data.
if (status.ok() && FLAGS_disable_wal) {
uint32_t tmp_crc = 0;
for (ColumnFamilyHandle* cfh : cfhs) {
if (cfh == db_ptr->DefaultColumnFamily()) {
continue;
}
std::unique_ptr<Iterator> it(db_ptr->NewIterator(ropts, cfh));
status = checksum_column_family(it.get(), &tmp_crc);
if (!status.ok() || tmp_crc != crc) {
break;
}
}
if (!status.ok()) {
fprintf(stderr, "status: %s\n", status.ToString().c_str());
shared->SetShouldStopTest();
assert(false);
} else if (tmp_crc != crc) {
fprintf(stderr, "tmp_crc=%" PRIu32 " crc=%" PRIu32 "\n", tmp_crc, crc);
shared->SetShouldStopTest();
assert(false);
}
}
}
#else // ROCKSDB_LITE
void ContinuouslyVerifyDb(ThreadState* /*thread*/) const override {}
#endif // !ROCKSDB_LITE
std::vector<int> GenerateColumnFamilies(
const int /* num_column_families */,
int /* rand_column_family */) const override {
std::vector<int> ret;
int num = static_cast<int>(column_families_.size());
int k = 0;
std::generate_n(back_inserter(ret), num, [&k]() -> int { return k++; });
return ret;
}
private:
Add the PutEntity API to the stress/crash tests (#10760) Summary: The patch adds the `PutEntity` API to the non-batched, batched, and CF consistency stress tests. Namely, when the new `db_stress` command line parameter `use_put_entity_one_in` is greater than zero, one in N writes on average is performed using `PutEntity` rather than `Put`. The wide-column entity written has the generated value in its default column; in addition, it contains up to three additional columns where the original generated value is divided up between the column name and the column value (with the column name containing the first k characters of the generated value, and the column value containing the rest). Whether `PutEntity` is used (and if so, how many columns the entity has) is completely determined by the "value base" used to generate the value (that is, there is no randomness involved). Assuming the same `use_put_entity_one_in` setting is used across `db_stress` invocations, this enables us to reconstruct and validate the entity during subsequent `db_stress` runs. Note that `PutEntity` is currently incompatible with `Merge`, transactions, and user-defined timestamps; these combinations are currently disabled/disallowed. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10760 Test Plan: Ran some batched, non-batched, and CF consistency stress tests using the script. Reviewed By: riversand963 Differential Revision: D39939032 Pulled By: ltamasi fbshipit-source-id: eafdf124e95993fb7d73158e3b006d11819f7fa9
2 years ago
std::atomic<uint32_t> batch_id_;
};
StressTest* CreateCfConsistencyStressTest() {
return new CfConsistencyStressTest();
}
} // namespace ROCKSDB_NAMESPACE
#endif // GFLAGS