2017-04-06 02:14:05 +02:00
|
|
|
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
|
2017-07-16 01:03:42 +02:00
|
|
|
// 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).
|
2017-04-06 02:14:05 +02:00
|
|
|
//
|
|
|
|
// 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.
|
2019-06-06 22:52:39 +02:00
|
|
|
#include <cinttypes>
|
2017-04-06 02:14:05 +02:00
|
|
|
|
|
|
|
#include "db/builder.h"
|
2020-08-15 05:11:35 +02:00
|
|
|
#include "db/db_impl/db_impl.h"
|
2018-06-28 21:23:57 +02:00
|
|
|
#include "db/error_handler.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"
|
2019-09-16 19:31:27 +02:00
|
|
|
#include "file/read_write_util.h"
|
2019-05-30 05:44:08 +02:00
|
|
|
#include "file/sst_file_manager_impl.h"
|
2019-09-16 19:31:27 +02:00
|
|
|
#include "file/writable_file_writer.h"
|
2019-06-18 00:17:43 +02:00
|
|
|
#include "monitoring/persistent_stats_history.h"
|
2020-08-15 05:11:35 +02:00
|
|
|
#include "monitoring/stats_dump_scheduler.h"
|
2017-04-06 04:02:00 +02:00
|
|
|
#include "options/options_helper.h"
|
2017-04-07 05:06:34 +02:00
|
|
|
#include "rocksdb/wal_filter.h"
|
2019-05-30 23:47:29 +02:00
|
|
|
#include "table/block_based/block_based_table_factory.h"
|
2019-05-30 20:21:38 +02:00
|
|
|
#include "test_util/sync_point.h"
|
2019-05-31 02:39:43 +02:00
|
|
|
#include "util/rate_limiter.h"
|
2017-04-06 02:14:05 +02:00
|
|
|
|
2020-02-20 21:07:53 +01:00
|
|
|
namespace ROCKSDB_NAMESPACE {
|
2018-12-17 22:12:22 +01:00
|
|
|
Options SanitizeOptions(const std::string& dbname, const Options& src) {
|
2017-04-06 02:14:05 +02:00
|
|
|
auto db_options = SanitizeOptions(dbname, DBOptions(src));
|
|
|
|
ImmutableDBOptions immutable_db_options(db_options);
|
|
|
|
auto cf_options =
|
|
|
|
SanitizeOptions(immutable_db_options, ColumnFamilyOptions(src));
|
|
|
|
return Options(db_options, cf_options);
|
|
|
|
}
|
|
|
|
|
|
|
|
DBOptions SanitizeOptions(const std::string& dbname, const DBOptions& src) {
|
|
|
|
DBOptions result(src);
|
|
|
|
|
2020-03-24 05:50:42 +01:00
|
|
|
if (result.env == nullptr) {
|
|
|
|
result.env = Env::Default();
|
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
|
|
|
}
|
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
// result.max_open_files means an "infinite" open files.
|
|
|
|
if (result.max_open_files != -1) {
|
|
|
|
int max_max_open_files = port::GetMaxOpenFiles();
|
|
|
|
if (max_max_open_files == -1) {
|
2017-05-04 05:46:17 +02:00
|
|
|
max_max_open_files = 0x400000;
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
ClipToRange(&result.max_open_files, 20, max_max_open_files);
|
2018-12-29 03:00:00 +01:00
|
|
|
TEST_SYNC_POINT_CALLBACK("SanitizeOptions::AfterChangeMaxOpenFiles",
|
|
|
|
&result.max_open_files);
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
if (result.info_log == nullptr) {
|
|
|
|
Status s = CreateLoggerFromOptions(dbname, result, &result.info_log);
|
|
|
|
if (!s.ok()) {
|
|
|
|
// No place suitable for logging
|
|
|
|
result.info_log = nullptr;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!result.write_buffer_manager) {
|
|
|
|
result.write_buffer_manager.reset(
|
|
|
|
new WriteBufferManager(result.db_write_buffer_size));
|
|
|
|
}
|
2018-12-17 22:12:22 +01:00
|
|
|
auto bg_job_limits = DBImpl::GetBGJobLimits(
|
|
|
|
result.max_background_flushes, result.max_background_compactions,
|
|
|
|
result.max_background_jobs, true /* parallelize_compactions */);
|
2017-05-24 20:25:38 +02:00
|
|
|
result.env->IncBackgroundThreadsIfNeeded(bg_job_limits.max_compactions,
|
2017-04-06 02:14:05 +02:00
|
|
|
Env::Priority::LOW);
|
2017-05-24 20:25:38 +02:00
|
|
|
result.env->IncBackgroundThreadsIfNeeded(bg_job_limits.max_flushes,
|
2017-04-06 02:14:05 +02:00
|
|
|
Env::Priority::HIGH);
|
|
|
|
|
|
|
|
if (result.rate_limiter.get() != nullptr) {
|
|
|
|
if (result.bytes_per_sync == 0) {
|
|
|
|
result.bytes_per_sync = 1024 * 1024;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2017-05-24 18:52:08 +02:00
|
|
|
if (result.delayed_write_rate == 0) {
|
|
|
|
if (result.rate_limiter.get() != nullptr) {
|
|
|
|
result.delayed_write_rate = result.rate_limiter->GetBytesPerSecond();
|
|
|
|
}
|
|
|
|
if (result.delayed_write_rate == 0) {
|
|
|
|
result.delayed_write_rate = 16 * 1024 * 1024;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
if (result.WAL_ttl_seconds > 0 || result.WAL_size_limit_MB > 0) {
|
|
|
|
result.recycle_log_file_num = false;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (result.recycle_log_file_num &&
|
2020-08-18 03:19:45 +02:00
|
|
|
(result.wal_recovery_mode ==
|
|
|
|
WALRecoveryMode::kTolerateCorruptedTailRecords ||
|
|
|
|
result.wal_recovery_mode == WALRecoveryMode::kPointInTimeRecovery ||
|
2017-04-06 02:14:05 +02:00
|
|
|
result.wal_recovery_mode == WALRecoveryMode::kAbsoluteConsistency)) {
|
2020-08-18 03:19:45 +02:00
|
|
|
// - kTolerateCorruptedTailRecords is inconsistent with recycle log file
|
|
|
|
// feature. WAL recycling expects recovery success upon encountering a
|
|
|
|
// corrupt record at the point where new data ends and recycled data
|
|
|
|
// remains at the tail. However, `kTolerateCorruptedTailRecords` must fail
|
|
|
|
// upon encountering any such corrupt record, as it cannot differentiate
|
|
|
|
// between this and a real corruption, which would cause committed updates
|
|
|
|
// to be truncated -- a violation of the recovery guarantee.
|
|
|
|
// - kPointInTimeRecovery and kAbsoluteConsistency are incompatible with
|
|
|
|
// recycle log file feature temporarily due to a bug found introducing a
|
|
|
|
// hole in the recovered data
|
|
|
|
// (https://github.com/facebook/rocksdb/pull/7252#issuecomment-673766236).
|
|
|
|
// Besides this bug, we believe the features are fundamentally compatible.
|
2020-01-31 16:26:46 +01:00
|
|
|
result.recycle_log_file_num = 0;
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
if (result.wal_dir.empty()) {
|
|
|
|
// Use dbname as default
|
|
|
|
result.wal_dir = dbname;
|
|
|
|
}
|
|
|
|
if (result.wal_dir.back() == '/') {
|
|
|
|
result.wal_dir = result.wal_dir.substr(0, result.wal_dir.size() - 1);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (result.db_paths.size() == 0) {
|
|
|
|
result.db_paths.emplace_back(dbname, std::numeric_limits<uint64_t>::max());
|
|
|
|
}
|
|
|
|
|
2018-12-17 22:12:22 +01:00
|
|
|
if (result.use_direct_reads && result.compaction_readahead_size == 0) {
|
2017-05-05 20:58:10 +02:00
|
|
|
TEST_SYNC_POINT_CALLBACK("SanitizeOptions:direct_io", nullptr);
|
2017-04-06 02:14:05 +02:00
|
|
|
result.compaction_readahead_size = 1024 * 1024 * 2;
|
|
|
|
}
|
|
|
|
|
2018-12-17 22:12:22 +01:00
|
|
|
if (result.compaction_readahead_size > 0 || result.use_direct_reads) {
|
2017-04-06 02:14:05 +02:00
|
|
|
result.new_table_reader_for_compaction_inputs = true;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Force flush on DB open if 2PC is enabled, since with 2PC we have no
|
|
|
|
// guarantee that consecutive log files have consecutive sequence id, which
|
|
|
|
// make recovery complicated.
|
|
|
|
if (result.allow_2pc) {
|
|
|
|
result.avoid_flush_during_recovery = false;
|
|
|
|
}
|
|
|
|
|
2017-10-27 22:25:54 +02:00
|
|
|
#ifndef ROCKSDB_LITE
|
2019-07-07 06:04:22 +02:00
|
|
|
ImmutableDBOptions immutable_db_options(result);
|
|
|
|
if (!IsWalDirSameAsDBPath(&immutable_db_options)) {
|
|
|
|
// Either the WAL dir and db_paths[0]/db_name are not the same, or we
|
|
|
|
// cannot tell for sure. In either case, assume they're different and
|
|
|
|
// explicitly cleanup the trash log files (bypass DeleteScheduler)
|
|
|
|
// Do this first so even if we end up calling
|
|
|
|
// DeleteScheduler::CleanupDirectory on the same dir later, it will be
|
|
|
|
// safe
|
|
|
|
std::vector<std::string> filenames;
|
2020-08-25 01:41:42 +02:00
|
|
|
result.env->GetChildren(result.wal_dir, &filenames).PermitUncheckedError();
|
2019-07-07 06:04:22 +02:00
|
|
|
for (std::string& filename : filenames) {
|
2019-09-20 21:00:55 +02:00
|
|
|
if (filename.find(".log.trash", filename.length() -
|
|
|
|
std::string(".log.trash").length()) !=
|
|
|
|
std::string::npos) {
|
2019-07-07 06:04:22 +02:00
|
|
|
std::string trash_file = result.wal_dir + "/" + filename;
|
2020-08-25 01:41:42 +02:00
|
|
|
result.env->DeleteFile(trash_file).PermitUncheckedError();
|
2019-07-07 06:04:22 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2017-10-27 22:25:54 +02:00
|
|
|
// When the DB is stopped, it's possible that there are some .trash files that
|
|
|
|
// were not deleted yet, when we open the DB we will find these .trash files
|
|
|
|
// and schedule them to be deleted (or delete immediately if SstFileManager
|
|
|
|
// was not used)
|
|
|
|
auto sfm = static_cast<SstFileManagerImpl*>(result.sst_file_manager.get());
|
|
|
|
for (size_t i = 0; i < result.db_paths.size(); i++) {
|
|
|
|
DeleteScheduler::CleanupDirectory(result.env, sfm, result.db_paths[i].path);
|
|
|
|
}
|
|
|
|
|
Auto recovery from out of space errors (#4164)
Summary:
This commit implements automatic recovery from a Status::NoSpace() error
during background operations such as write callback, flush and
compaction. The broad design is as follows -
1. Compaction errors are treated as soft errors and don't put the
database in read-only mode. A compaction is delayed until enough free
disk space is available to accomodate the compaction outputs, which is
estimated based on the input size. This means that users can continue to
write, and we rely on the WriteController to delay or stop writes if the
compaction debt becomes too high due to persistent low disk space
condition
2. Errors during write callback and flush are treated as hard errors,
i.e the database is put in read-only mode and goes back to read-write
only fater certain recovery actions are taken.
3. Both types of recovery rely on the SstFileManagerImpl to poll for
sufficient disk space. We assume that there is a 1-1 mapping between an
SFM and the underlying OS storage container. For cases where multiple
DBs are hosted on a single storage container, the user is expected to
allocate a single SFM instance and use the same one for all the DBs. If
no SFM is specified by the user, DBImpl::Open() will allocate one, but
this will be one per DB and each DB will recover independently. The
recovery implemented by SFM is as follows -
a) On the first occurance of an out of space error during compaction,
subsequent
compactions will be delayed until the disk free space check indicates
enough available space. The required space is computed as the sum of
input sizes.
b) The free space check requirement will be removed once the amount of
free space is greater than the size reserved by in progress
compactions when the first error occured
c) If the out of space error is a hard error, a background thread in
SFM will poll for sufficient headroom before triggering the recovery
of the database and putting it in write-only mode. The headroom is
calculated as the sum of the write_buffer_size of all the DB instances
associated with the SFM
4. EventListener callbacks will be called at the start and completion of
automatic recovery. Users can disable the auto recov ery in the start
callback, and later initiate it manually by calling DB::Resume()
Todo:
1. More extensive testing
2. Add disk full condition to db_stress (follow-on PR)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4164
Differential Revision: D9846378
Pulled By: anand1976
fbshipit-source-id: 80ea875dbd7f00205e19c82215ff6e37da10da4a
2018-09-15 22:36:19 +02:00
|
|
|
// Create a default SstFileManager for purposes of tracking compaction size
|
|
|
|
// and facilitating recovery from out of space errors.
|
|
|
|
if (result.sst_file_manager.get() == nullptr) {
|
|
|
|
std::shared_ptr<SstFileManager> sst_file_manager(
|
|
|
|
NewSstFileManager(result.env, result.info_log));
|
|
|
|
result.sst_file_manager = sst_file_manager;
|
|
|
|
}
|
|
|
|
#endif
|
Add an option to prevent DB::Open() from querying sizes of all sst files (#6353)
Summary:
When paranoid_checks is on, DBImpl::CheckConsistency() iterates over all sst files and calls Env::GetFileSize() for each of them. As far as I could understand, this is pretty arbitrary and doesn't affect correctness - if filesystem doesn't corrupt fsynced files, the file sizes will always match; if it does, it may as well corrupt contents as well as sizes, and rocksdb doesn't check contents on open.
If there are thousands of sst files, getting all their sizes takes a while. If, on top of that, Env is overridden to use some remote storage instead of local filesystem, it can be *really* slow and overload the remote storage service. This PR adds an option to not do GetFileSize(); instead it does GetChildren() for parent directory to check that all the expected sst files are at least present, but doesn't check their sizes.
We can't just disable paranoid_checks instead because paranoid_checks do a few other important things: make the DB read-only on write errors, print error messages on read errors, etc.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6353
Test Plan: ran the added sanity check unit test. Will try it out in a LogDevice test cluster where the GetFileSize() calls are causing a lot of trouble.
Differential Revision: D19656425
Pulled By: al13n321
fbshipit-source-id: c2c421b367633033760d1f56747bad206d1fbf82
2020-02-04 10:24:29 +01:00
|
|
|
|
|
|
|
if (!result.paranoid_checks) {
|
|
|
|
result.skip_checking_sst_file_sizes_on_db_open = true;
|
|
|
|
ROCKS_LOG_INFO(result.info_log,
|
|
|
|
"file size check will be skipped during open.");
|
|
|
|
}
|
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
|
|
|
namespace {
|
|
|
|
Status SanitizeOptionsByTable(
|
|
|
|
const DBOptions& db_opts,
|
|
|
|
const std::vector<ColumnFamilyDescriptor>& column_families) {
|
|
|
|
Status s;
|
|
|
|
for (auto cf : column_families) {
|
|
|
|
s = cf.options.table_factory->SanitizeOptions(db_opts, cf.options);
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return Status::OK();
|
|
|
|
}
|
2019-06-04 04:47:02 +02:00
|
|
|
} // namespace
|
2017-04-06 02:14:05 +02:00
|
|
|
|
2019-06-04 04:47:02 +02:00
|
|
|
Status DBImpl::ValidateOptions(
|
2017-04-06 02:14:05 +02:00
|
|
|
const DBOptions& db_options,
|
|
|
|
const std::vector<ColumnFamilyDescriptor>& column_families) {
|
|
|
|
Status s;
|
|
|
|
for (auto& cfd : column_families) {
|
2019-06-04 04:47:02 +02:00
|
|
|
s = ColumnFamilyData::ValidateOptions(db_options, cfd.options);
|
2017-04-06 02:14:05 +02:00
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
}
|
2019-06-04 04:47:02 +02:00
|
|
|
s = ValidateOptions(db_options);
|
|
|
|
return s;
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
|
2019-06-04 04:47:02 +02:00
|
|
|
Status DBImpl::ValidateOptions(const DBOptions& db_options) {
|
2017-04-06 02:14:05 +02:00
|
|
|
if (db_options.db_paths.size() > 4) {
|
|
|
|
return Status::NotSupported(
|
|
|
|
"More than four DB paths are not supported yet. ");
|
|
|
|
}
|
|
|
|
|
|
|
|
if (db_options.allow_mmap_reads && db_options.use_direct_reads) {
|
|
|
|
// Protect against assert in PosixMMapReadableFile constructor
|
|
|
|
return Status::NotSupported(
|
|
|
|
"If memory mapped reads (allow_mmap_reads) are enabled "
|
|
|
|
"then direct I/O reads (use_direct_reads) must be disabled. ");
|
|
|
|
}
|
|
|
|
|
2017-04-13 22:07:33 +02:00
|
|
|
if (db_options.allow_mmap_writes &&
|
|
|
|
db_options.use_direct_io_for_flush_and_compaction) {
|
2017-04-06 02:14:05 +02:00
|
|
|
return Status::NotSupported(
|
|
|
|
"If memory mapped writes (allow_mmap_writes) are enabled "
|
2017-04-13 22:07:33 +02:00
|
|
|
"then direct I/O writes (use_direct_io_for_flush_and_compaction) must "
|
|
|
|
"be disabled. ");
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
if (db_options.keep_log_file_num == 0) {
|
|
|
|
return Status::InvalidArgument("keep_log_file_num must be greater than 0");
|
|
|
|
}
|
|
|
|
|
2019-05-14 02:43:47 +02:00
|
|
|
if (db_options.unordered_write &&
|
|
|
|
!db_options.allow_concurrent_memtable_write) {
|
|
|
|
return Status::InvalidArgument(
|
|
|
|
"unordered_write is incompatible with !allow_concurrent_memtable_write");
|
|
|
|
}
|
|
|
|
|
|
|
|
if (db_options.unordered_write && db_options.enable_pipelined_write) {
|
|
|
|
return Status::InvalidArgument(
|
|
|
|
"unordered_write is incompatible with enable_pipelined_write");
|
|
|
|
}
|
|
|
|
|
2019-09-28 02:15:48 +02:00
|
|
|
if (db_options.atomic_flush && db_options.enable_pipelined_write) {
|
|
|
|
return Status::InvalidArgument(
|
|
|
|
"atomic_flush is incompatible with enable_pipelined_write");
|
|
|
|
}
|
|
|
|
|
2020-03-21 03:17:54 +01:00
|
|
|
// TODO remove this restriction
|
|
|
|
if (db_options.atomic_flush && db_options.best_efforts_recovery) {
|
|
|
|
return Status::InvalidArgument(
|
|
|
|
"atomic_flush is currently incompatible with best-efforts recovery");
|
|
|
|
}
|
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
return Status::OK();
|
|
|
|
}
|
2019-06-04 04:47:02 +02:00
|
|
|
|
2020-07-10 22:39:47 +02:00
|
|
|
Status DBImpl::NewDB(std::vector<std::string>* new_filenames) {
|
2017-04-06 02:14:05 +02:00
|
|
|
VersionEdit new_db;
|
2019-09-03 17:50:47 +02:00
|
|
|
Status s = SetIdentityFile(env_, dbname_);
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
if (immutable_db_options_.write_dbid_to_manifest) {
|
|
|
|
std::string temp_db_id;
|
|
|
|
GetDbIdentityFromIdentityFile(&temp_db_id);
|
|
|
|
new_db.SetDBId(temp_db_id);
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
new_db.SetLogNumber(0);
|
|
|
|
new_db.SetNextFile(2);
|
|
|
|
new_db.SetLastSequence(0);
|
|
|
|
|
|
|
|
ROCKS_LOG_INFO(immutable_db_options_.info_log, "Creating manifest 1 \n");
|
|
|
|
const std::string manifest = DescriptorFileName(dbname_, 1);
|
|
|
|
{
|
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
|
|
|
std::unique_ptr<FSWritableFile> file;
|
|
|
|
FileOptions file_options = fs_->OptimizeForManifestWrite(file_options_);
|
|
|
|
s = NewWritableFile(fs_.get(), manifest, &file, file_options);
|
2017-04-06 02:14:05 +02:00
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
file->SetPreallocationBlockSize(
|
|
|
|
immutable_db_options_.manifest_preallocation_size);
|
2018-11-09 20:17:34 +01:00
|
|
|
std::unique_ptr<WritableFileWriter> file_writer(new WritableFileWriter(
|
2020-09-08 19:49:01 +02:00
|
|
|
std::move(file), manifest, file_options, env_, io_tracer_,
|
|
|
|
nullptr /* stats */, immutable_db_options_.listeners));
|
2017-04-06 02:14:05 +02:00
|
|
|
log::Writer log(std::move(file_writer), 0, false);
|
|
|
|
std::string record;
|
|
|
|
new_db.EncodeTo(&record);
|
|
|
|
s = log.AddRecord(record);
|
|
|
|
if (s.ok()) {
|
|
|
|
s = SyncManifest(env_, &immutable_db_options_, log.file());
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (s.ok()) {
|
|
|
|
// Make "CURRENT" file that points to the new manifest file.
|
Pass IOStatus to write path and set retryable IO Error as hard error in BG jobs (#6487)
Summary:
In the current code base, we use Status to get and store the returned status from the call. Specifically, for IO related functions, the current Status cannot reflect the IO Error details such as error scope, error retryable attribute, and others. With the implementation of https://github.com/facebook/rocksdb/issues/5761, we have the new Wrapper for IO, which returns IOStatus instead of Status. However, the IOStatus is purged at the lower level of write path and transferred to Status.
The first job of this PR is to pass the IOStatus to the write path (flush, WAL write, and Compaction). The second job is to identify the Retryable IO Error as HardError, and set the bg_error_ as HardError. In this case, the DB Instance becomes read only. User is informed of the Status and need to take actions to deal with it (e.g., call db->Resume()).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6487
Test Plan: Added the testing case to error_handler_fs_test. Pass make asan_check
Reviewed By: anand1976
Differential Revision: D20685017
Pulled By: zhichao-cao
fbshipit-source-id: ff85f042896243abcd6ef37877834e26f36b6eb0
2020-03-28 00:03:05 +01:00
|
|
|
s = SetCurrentFile(fs_.get(), dbname_, 1, directories_.GetDbDir());
|
2020-07-10 22:39:47 +02:00
|
|
|
if (new_filenames) {
|
|
|
|
new_filenames->emplace_back(
|
|
|
|
manifest.substr(manifest.find_last_of("/\\") + 1));
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
} else {
|
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
|
|
|
fs_->DeleteFile(manifest, IOOptions(), nullptr);
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
2020-03-03 01:14:00 +01:00
|
|
|
IOStatus DBImpl::CreateAndNewDirectory(
|
|
|
|
FileSystem* fs, const std::string& dirname,
|
|
|
|
std::unique_ptr<FSDirectory>* directory) {
|
2017-04-06 02:14:05 +02:00
|
|
|
// We call CreateDirIfMissing() as the directory may already exist (if we
|
|
|
|
// are reopening a DB), when this happens we don't want creating the
|
|
|
|
// directory to cause an error. However, we need to check if creating the
|
|
|
|
// directory fails or else we may get an obscure message about the lock
|
|
|
|
// file not existing. One real-world example of this occurring is if
|
|
|
|
// env->CreateDirIfMissing() doesn't create intermediate directories, e.g.
|
|
|
|
// when dbname_ is "dir/db" but when "dir" doesn't exist.
|
2020-03-03 01:14:00 +01:00
|
|
|
IOStatus io_s = fs->CreateDirIfMissing(dirname, IOOptions(), nullptr);
|
|
|
|
if (!io_s.ok()) {
|
|
|
|
return io_s;
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
2020-03-03 01:14:00 +01:00
|
|
|
return fs->NewDirectory(dirname, IOOptions(), directory, nullptr);
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
|
2020-03-03 01:14:00 +01:00
|
|
|
IOStatus Directories::SetDirectories(FileSystem* fs, const std::string& dbname,
|
|
|
|
const std::string& wal_dir,
|
|
|
|
const std::vector<DbPath>& data_paths) {
|
|
|
|
IOStatus io_s = DBImpl::CreateAndNewDirectory(fs, dbname, &db_dir_);
|
|
|
|
if (!io_s.ok()) {
|
|
|
|
return io_s;
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
if (!wal_dir.empty() && dbname != wal_dir) {
|
2020-03-03 01:14:00 +01:00
|
|
|
io_s = DBImpl::CreateAndNewDirectory(fs, wal_dir, &wal_dir_);
|
|
|
|
if (!io_s.ok()) {
|
|
|
|
return io_s;
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
data_dirs_.clear();
|
|
|
|
for (auto& p : data_paths) {
|
|
|
|
const std::string db_path = p.path;
|
|
|
|
if (db_path == dbname) {
|
|
|
|
data_dirs_.emplace_back(nullptr);
|
|
|
|
} else {
|
2020-03-03 01:14:00 +01:00
|
|
|
std::unique_ptr<FSDirectory> path_directory;
|
|
|
|
io_s = DBImpl::CreateAndNewDirectory(fs, db_path, &path_directory);
|
|
|
|
if (!io_s.ok()) {
|
|
|
|
return io_s;
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
data_dirs_.emplace_back(path_directory.release());
|
|
|
|
}
|
|
|
|
}
|
|
|
|
assert(data_dirs_.size() == data_paths.size());
|
2020-03-03 01:14:00 +01:00
|
|
|
return IOStatus::OK();
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
Status DBImpl::Recover(
|
|
|
|
const std::vector<ColumnFamilyDescriptor>& column_families, bool read_only,
|
2020-01-29 20:39:11 +01:00
|
|
|
bool error_if_log_file_exist, bool error_if_data_exists_in_logs,
|
|
|
|
uint64_t* recovered_seq) {
|
2017-04-06 02:14:05 +02:00
|
|
|
mutex_.AssertHeld();
|
|
|
|
|
|
|
|
bool is_new_db = false;
|
|
|
|
assert(db_lock_ == nullptr);
|
2020-07-10 22:39:47 +02:00
|
|
|
std::vector<std::string> files_in_dbname;
|
2017-04-06 02:14:05 +02:00
|
|
|
if (!read_only) {
|
2020-03-03 01:14:00 +01:00
|
|
|
Status s = directories_.SetDirectories(fs_.get(), dbname_,
|
2017-04-06 02:14:05 +02:00
|
|
|
immutable_db_options_.wal_dir,
|
|
|
|
immutable_db_options_.db_paths);
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
|
|
|
s = env_->LockFile(LockFileName(dbname_), &db_lock_);
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
2020-01-07 21:30:40 +01:00
|
|
|
std::string current_fname = CurrentFileName(dbname_);
|
2020-06-15 23:09:53 +02:00
|
|
|
// Path to any MANIFEST file in the db dir. It does not matter which one.
|
|
|
|
// Since best-efforts recovery ignores CURRENT file, existence of a
|
|
|
|
// MANIFEST indicates the recovery to recover existing db. If no MANIFEST
|
|
|
|
// can be found, a new db will be created.
|
|
|
|
std::string manifest_path;
|
|
|
|
if (!immutable_db_options_.best_efforts_recovery) {
|
|
|
|
s = env_->FileExists(current_fname);
|
|
|
|
} else {
|
|
|
|
s = Status::NotFound();
|
2020-07-10 22:39:47 +02:00
|
|
|
Status io_s = env_->GetChildren(dbname_, &files_in_dbname);
|
|
|
|
if (!io_s.ok()) {
|
|
|
|
s = io_s;
|
|
|
|
files_in_dbname.clear();
|
|
|
|
}
|
|
|
|
for (const std::string& file : files_in_dbname) {
|
2020-06-15 23:09:53 +02:00
|
|
|
uint64_t number = 0;
|
|
|
|
FileType type = kLogFile; // initialize
|
|
|
|
if (ParseFileName(file, &number, &type) && type == kDescriptorFile) {
|
|
|
|
// Found MANIFEST (descriptor log), thus best-efforts recovery does
|
|
|
|
// not have to treat the db as empty.
|
|
|
|
s = Status::OK();
|
|
|
|
manifest_path = dbname_ + "/" + file;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
if (s.IsNotFound()) {
|
|
|
|
if (immutable_db_options_.create_if_missing) {
|
2020-07-10 22:39:47 +02:00
|
|
|
s = NewDB(&files_in_dbname);
|
2017-04-06 02:14:05 +02:00
|
|
|
is_new_db = true;
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
return Status::InvalidArgument(
|
2020-01-07 21:30:40 +01:00
|
|
|
current_fname, "does not exist (create_if_missing is false)");
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
} else if (s.ok()) {
|
|
|
|
if (immutable_db_options_.error_if_exists) {
|
2018-12-17 22:12:22 +01:00
|
|
|
return Status::InvalidArgument(dbname_,
|
|
|
|
"exists (error_if_exists is true)");
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
} else {
|
|
|
|
// Unexpected error reading file
|
|
|
|
assert(s.IsIOError());
|
|
|
|
return s;
|
|
|
|
}
|
Introduce a new storage specific Env API (#5761)
Summary:
The current Env API encompasses both storage/file operations, as well as OS related operations. Most of the APIs return a Status, which does not have enough metadata about an error, such as whether its retry-able or not, scope (i.e fault domain) of the error etc., that may be required in order to properly handle a storage error. The file APIs also do not provide enough control over the IO SLA, such as timeout, prioritization, hinting about placement and redundancy etc.
This PR separates out the file/storage APIs from Env into a new FileSystem class. The APIs are updated to return an IOStatus with metadata about the error, as well as to take an IOOptions structure as input in order to allow more control over the IO.
The user can set both ```options.env``` and ```options.file_system``` to specify that RocksDB should use the former for OS related operations and the latter for storage operations. Internally, a ```CompositeEnvWrapper``` has been introduced that inherits from ```Env``` and redirects individual methods to either an ```Env``` implementation or the ```FileSystem``` as appropriate. When options are sanitized during ```DB::Open```, ```options.env``` is replaced with a newly allocated ```CompositeEnvWrapper``` instance if both env and file_system have been specified. This way, the rest of the RocksDB code can continue to function as before.
This PR also ports PosixEnv to the new API by splitting it into two - PosixEnv and PosixFileSystem. PosixEnv is defined as a sub-class of CompositeEnvWrapper, and threading/time functions are overridden with Posix specific implementations in order to avoid an extra level of indirection.
The ```CompositeEnvWrapper``` translates ```IOStatus``` return code to ```Status```, and sets the severity to ```kSoftError``` if the io_status is retryable. The error handling code in RocksDB can then recover the DB automatically.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5761
Differential Revision: D18868376
Pulled By: anand1976
fbshipit-source-id: 39efe18a162ea746fabac6360ff529baba48486f
2019-12-13 23:47:08 +01:00
|
|
|
// Verify compatibility of file_options_ and filesystem
|
2018-02-26 23:46:12 +01:00
|
|
|
{
|
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
|
|
|
std::unique_ptr<FSRandomAccessFile> idfile;
|
|
|
|
FileOptions customized_fs(file_options_);
|
|
|
|
customized_fs.use_direct_reads |=
|
2018-02-26 23:46:12 +01:00
|
|
|
immutable_db_options_.use_direct_io_for_flush_and_compaction;
|
2020-06-15 23:09:53 +02:00
|
|
|
const std::string& fname =
|
|
|
|
manifest_path.empty() ? current_fname : manifest_path;
|
|
|
|
s = fs_->NewRandomAccessFile(fname, customized_fs, &idfile, nullptr);
|
2018-02-26 23:46:12 +01:00
|
|
|
if (!s.ok()) {
|
2018-08-09 19:58:34 +02:00
|
|
|
std::string error_str = s.ToString();
|
2018-02-26 23:46:12 +01:00
|
|
|
// Check if unsupported Direct I/O is the root cause
|
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
|
|
|
customized_fs.use_direct_reads = false;
|
2020-06-15 23:09:53 +02:00
|
|
|
s = fs_->NewRandomAccessFile(fname, customized_fs, &idfile, nullptr);
|
2018-02-26 23:46:12 +01:00
|
|
|
if (s.ok()) {
|
|
|
|
return Status::InvalidArgument(
|
|
|
|
"Direct I/O is not supported by the specified DB.");
|
|
|
|
} else {
|
|
|
|
return Status::InvalidArgument(
|
2018-08-09 19:58:34 +02:00
|
|
|
"Found options incompatible with filesystem", error_str.c_str());
|
2018-02-26 23:46:12 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2020-07-10 22:39:47 +02:00
|
|
|
} else if (immutable_db_options_.best_efforts_recovery) {
|
|
|
|
assert(files_in_dbname.empty());
|
|
|
|
Status s = env_->GetChildren(dbname_, &files_in_dbname);
|
|
|
|
if (s.IsNotFound()) {
|
|
|
|
return Status::InvalidArgument(dbname_,
|
|
|
|
"does not exist (open for read only)");
|
|
|
|
} else if (s.IsIOError()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
assert(s.ok());
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
2019-09-03 17:50:47 +02:00
|
|
|
assert(db_id_.empty());
|
2020-03-21 03:17:54 +01:00
|
|
|
Status s;
|
|
|
|
bool missing_table_file = false;
|
|
|
|
if (!immutable_db_options_.best_efforts_recovery) {
|
|
|
|
s = versions_->Recover(column_families, read_only, &db_id_);
|
|
|
|
} else {
|
2020-07-10 22:39:47 +02:00
|
|
|
assert(!files_in_dbname.empty());
|
|
|
|
s = versions_->TryRecover(column_families, read_only, files_in_dbname,
|
|
|
|
&db_id_, &missing_table_file);
|
2020-03-21 03:17:54 +01:00
|
|
|
if (s.ok()) {
|
2020-05-08 21:59:02 +02:00
|
|
|
// TryRecover may delete previous column_family_set_.
|
|
|
|
column_family_memtables_.reset(
|
|
|
|
new ColumnFamilyMemTablesImpl(versions_->GetColumnFamilySet()));
|
2020-04-24 01:18:28 +02:00
|
|
|
s = FinishBestEffortsRecovery();
|
2020-03-21 03:17:54 +01:00
|
|
|
}
|
|
|
|
}
|
2019-09-03 17:50:47 +02:00
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
// Happens when immutable_db_options_.write_dbid_to_manifest is set to true
|
|
|
|
// the very first time.
|
|
|
|
if (db_id_.empty()) {
|
|
|
|
// Check for the IDENTITY file and create it if not there.
|
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
|
|
|
s = fs_->FileExists(IdentityFileName(dbname_), IOOptions(), nullptr);
|
2019-09-03 17:50:47 +02:00
|
|
|
// Typically Identity file is created in NewDB() and for some reason if
|
|
|
|
// it is no longer available then at this point DB ID is not in Identity
|
|
|
|
// file or Manifest.
|
|
|
|
if (s.IsNotFound()) {
|
|
|
|
s = SetIdentityFile(env_, dbname_);
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
} else if (!s.ok()) {
|
|
|
|
assert(s.IsIOError());
|
|
|
|
return s;
|
|
|
|
}
|
2020-02-07 21:29:41 +01:00
|
|
|
s = GetDbIdentityFromIdentityFile(&db_id_);
|
|
|
|
if (immutable_db_options_.write_dbid_to_manifest && s.ok()) {
|
2019-09-03 17:50:47 +02:00
|
|
|
VersionEdit edit;
|
|
|
|
edit.SetDBId(db_id_);
|
|
|
|
Options options;
|
|
|
|
MutableCFOptions mutable_cf_options(options);
|
|
|
|
versions_->db_id_ = db_id_;
|
2020-02-07 21:29:41 +01:00
|
|
|
s = versions_->LogAndApply(versions_->GetColumnFamilySet()->GetDefault(),
|
2019-09-03 17:50:47 +02:00
|
|
|
mutable_cf_options, &edit, &mutex_, nullptr,
|
|
|
|
false);
|
|
|
|
}
|
|
|
|
} else {
|
2020-02-07 21:29:41 +01:00
|
|
|
s = SetIdentityFile(env_, dbname_, db_id_);
|
2019-09-03 17:50:47 +02:00
|
|
|
}
|
2019-06-18 00:17:43 +02:00
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
if (immutable_db_options_.paranoid_checks && s.ok()) {
|
|
|
|
s = CheckConsistency();
|
|
|
|
}
|
2018-04-06 04:49:06 +02:00
|
|
|
if (s.ok() && !read_only) {
|
2020-03-03 01:14:00 +01:00
|
|
|
std::map<std::string, std::shared_ptr<FSDirectory>> created_dirs;
|
2018-04-06 04:49:06 +02:00
|
|
|
for (auto cfd : *versions_->GetColumnFamilySet()) {
|
2020-02-03 23:11:50 +01:00
|
|
|
s = cfd->AddDirectories(&created_dirs);
|
2018-04-06 04:49:06 +02:00
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2019-06-18 00:17:43 +02:00
|
|
|
// DB mutex is already held
|
|
|
|
if (s.ok() && immutable_db_options_.persist_stats_to_disk) {
|
|
|
|
s = InitPersistStatsColumnFamily();
|
|
|
|
}
|
2018-09-26 19:34:56 +02:00
|
|
|
|
2020-07-10 22:39:47 +02:00
|
|
|
std::vector<std::string> files_in_wal_dir;
|
2017-04-06 02:14:05 +02:00
|
|
|
if (s.ok()) {
|
2020-02-07 21:29:41 +01:00
|
|
|
// Initial max_total_in_memory_state_ before recovery logs. Log recovery
|
|
|
|
// may check this value to decide whether to flush.
|
|
|
|
max_total_in_memory_state_ = 0;
|
|
|
|
for (auto cfd : *versions_->GetColumnFamilySet()) {
|
|
|
|
auto* mutable_cf_options = cfd->GetLatestMutableCFOptions();
|
|
|
|
max_total_in_memory_state_ += mutable_cf_options->write_buffer_size *
|
|
|
|
mutable_cf_options->max_write_buffer_number;
|
|
|
|
}
|
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
SequenceNumber next_sequence(kMaxSequenceNumber);
|
|
|
|
default_cf_handle_ = new ColumnFamilyHandleImpl(
|
|
|
|
versions_->GetColumnFamilySet()->GetDefault(), this, &mutex_);
|
|
|
|
default_cf_internal_stats_ = default_cf_handle_->cfd()->internal_stats();
|
2019-06-18 00:17:43 +02:00
|
|
|
// TODO(Zhongyi): handle single_column_family_mode_ when
|
|
|
|
// persistent_stats is enabled
|
2017-04-06 02:14:05 +02:00
|
|
|
single_column_family_mode_ =
|
|
|
|
versions_->GetColumnFamilySet()->NumberOfColumnFamilies() == 1;
|
|
|
|
|
|
|
|
// Recover from all newer log files than the ones named in the
|
|
|
|
// descriptor (new log files may have been added by the previous
|
|
|
|
// incarnation without registering them in the descriptor).
|
|
|
|
//
|
|
|
|
// Note that prev_log_number() is no longer used, but we pay
|
|
|
|
// attention to it in case we are recovering a database
|
|
|
|
// produced by an older version of rocksdb.
|
2020-03-21 03:17:54 +01:00
|
|
|
if (!immutable_db_options_.best_efforts_recovery) {
|
2020-07-10 22:39:47 +02:00
|
|
|
s = env_->GetChildren(immutable_db_options_.wal_dir, &files_in_wal_dir);
|
2020-03-21 03:17:54 +01:00
|
|
|
}
|
2019-01-16 01:41:50 +01:00
|
|
|
if (s.IsNotFound()) {
|
|
|
|
return Status::InvalidArgument("wal_dir not found",
|
|
|
|
immutable_db_options_.wal_dir);
|
|
|
|
} else if (!s.ok()) {
|
2017-04-06 02:14:05 +02:00
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
|
|
|
std::vector<uint64_t> logs;
|
2020-07-10 22:39:47 +02:00
|
|
|
for (const auto& file : files_in_wal_dir) {
|
2017-04-06 02:14:05 +02:00
|
|
|
uint64_t number;
|
|
|
|
FileType type;
|
2020-07-10 22:39:47 +02:00
|
|
|
if (ParseFileName(file, &number, &type) && type == kLogFile) {
|
2017-04-06 02:14:05 +02:00
|
|
|
if (is_new_db) {
|
|
|
|
return Status::Corruption(
|
|
|
|
"While creating a new Db, wal_dir contains "
|
|
|
|
"existing log file: ",
|
2020-07-10 22:39:47 +02:00
|
|
|
file);
|
2017-04-06 02:14:05 +02:00
|
|
|
} else {
|
|
|
|
logs.push_back(number);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (logs.size() > 0) {
|
|
|
|
if (error_if_log_file_exist) {
|
|
|
|
return Status::Corruption(
|
|
|
|
"The db was opened in readonly mode with error_if_log_file_exist"
|
|
|
|
"flag but a log file already exists");
|
|
|
|
} else if (error_if_data_exists_in_logs) {
|
|
|
|
for (auto& log : logs) {
|
|
|
|
std::string fname = LogFileName(immutable_db_options_.wal_dir, log);
|
|
|
|
uint64_t bytes;
|
|
|
|
s = env_->GetFileSize(fname, &bytes);
|
|
|
|
if (s.ok()) {
|
|
|
|
if (bytes > 0) {
|
|
|
|
return Status::Corruption(
|
|
|
|
"error_if_data_exists_in_logs is set but there are data "
|
|
|
|
" in log files.");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!logs.empty()) {
|
|
|
|
// Recover in the order in which the logs were generated
|
|
|
|
std::sort(logs.begin(), logs.end());
|
2020-01-29 20:39:11 +01:00
|
|
|
bool corrupted_log_found = false;
|
|
|
|
s = RecoverLogFiles(logs, &next_sequence, read_only,
|
|
|
|
&corrupted_log_found);
|
|
|
|
if (corrupted_log_found && recovered_seq != nullptr) {
|
|
|
|
*recovered_seq = next_sequence;
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
if (!s.ok()) {
|
|
|
|
// Clear memtables if recovery failed
|
|
|
|
for (auto cfd : *versions_->GetColumnFamilySet()) {
|
|
|
|
cfd->CreateNewMemtable(*cfd->GetLatestMutableCFOptions(),
|
|
|
|
kMaxSequenceNumber);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-12-08 02:03:49 +01:00
|
|
|
if (read_only) {
|
|
|
|
// If we are opening as read-only, we need to update options_file_number_
|
|
|
|
// to reflect the most recent OPTIONS file. It does not matter for regular
|
|
|
|
// read-write db instance because options_file_number_ will later be
|
|
|
|
// updated to versions_->NewFileNumber() in RenameTempFileToOptionsFile.
|
2020-07-10 22:39:47 +02:00
|
|
|
std::vector<std::string> filenames;
|
2018-12-08 02:03:49 +01:00
|
|
|
if (s.ok()) {
|
2020-07-10 22:39:47 +02:00
|
|
|
const std::string normalized_dbname = NormalizePath(dbname_);
|
|
|
|
const std::string normalized_wal_dir =
|
|
|
|
NormalizePath(immutable_db_options_.wal_dir);
|
|
|
|
if (immutable_db_options_.best_efforts_recovery) {
|
|
|
|
filenames = std::move(files_in_dbname);
|
|
|
|
} else if (normalized_dbname == normalized_wal_dir) {
|
|
|
|
filenames = std::move(files_in_wal_dir);
|
|
|
|
} else {
|
|
|
|
s = env_->GetChildren(GetName(), &filenames);
|
|
|
|
}
|
2018-12-08 02:03:49 +01:00
|
|
|
}
|
|
|
|
if (s.ok()) {
|
|
|
|
uint64_t number = 0;
|
|
|
|
uint64_t options_file_number = 0;
|
|
|
|
FileType type;
|
2020-07-10 22:39:47 +02:00
|
|
|
for (const auto& fname : filenames) {
|
2018-12-08 02:03:49 +01:00
|
|
|
if (ParseFileName(fname, &number, &type) && type == kOptionsFile) {
|
|
|
|
options_file_number = std::max(number, options_file_number);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
versions_->options_file_number_ = options_file_number;
|
|
|
|
}
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
2019-06-18 00:17:43 +02:00
|
|
|
Status DBImpl::PersistentStatsProcessFormatVersion() {
|
|
|
|
mutex_.AssertHeld();
|
|
|
|
Status s;
|
|
|
|
// persist version when stats CF doesn't exist
|
|
|
|
bool should_persist_format_version = !persistent_stats_cfd_exists_;
|
|
|
|
mutex_.Unlock();
|
|
|
|
if (persistent_stats_cfd_exists_) {
|
|
|
|
// Check persistent stats format version compatibility. Drop and recreate
|
|
|
|
// persistent stats CF if format version is incompatible
|
|
|
|
uint64_t format_version_recovered = 0;
|
|
|
|
Status s_format = DecodePersistentStatsVersionNumber(
|
|
|
|
this, StatsVersionKeyType::kFormatVersion, &format_version_recovered);
|
|
|
|
uint64_t compatible_version_recovered = 0;
|
|
|
|
Status s_compatible = DecodePersistentStatsVersionNumber(
|
|
|
|
this, StatsVersionKeyType::kCompatibleVersion,
|
|
|
|
&compatible_version_recovered);
|
|
|
|
// abort reading from existing stats CF if any of following is true:
|
|
|
|
// 1. failed to read format version or compatible version from disk
|
|
|
|
// 2. sst's format version is greater than current format version, meaning
|
|
|
|
// this sst is encoded with a newer RocksDB release, and current compatible
|
|
|
|
// version is below the sst's compatible version
|
|
|
|
if (!s_format.ok() || !s_compatible.ok() ||
|
|
|
|
(kStatsCFCurrentFormatVersion < format_version_recovered &&
|
|
|
|
kStatsCFCompatibleFormatVersion < compatible_version_recovered)) {
|
|
|
|
if (!s_format.ok() || !s_compatible.ok()) {
|
|
|
|
ROCKS_LOG_INFO(
|
|
|
|
immutable_db_options_.info_log,
|
|
|
|
"Reading persistent stats version key failed. Format key: %s, "
|
|
|
|
"compatible key: %s",
|
|
|
|
s_format.ToString().c_str(), s_compatible.ToString().c_str());
|
|
|
|
} else {
|
|
|
|
ROCKS_LOG_INFO(
|
|
|
|
immutable_db_options_.info_log,
|
|
|
|
"Disable persistent stats due to corrupted or incompatible format "
|
|
|
|
"version\n");
|
|
|
|
}
|
|
|
|
DropColumnFamily(persist_stats_cf_handle_);
|
|
|
|
DestroyColumnFamilyHandle(persist_stats_cf_handle_);
|
|
|
|
ColumnFamilyHandle* handle = nullptr;
|
|
|
|
ColumnFamilyOptions cfo;
|
|
|
|
OptimizeForPersistentStats(&cfo);
|
|
|
|
s = CreateColumnFamily(cfo, kPersistentStatsColumnFamilyName, &handle);
|
|
|
|
persist_stats_cf_handle_ = static_cast<ColumnFamilyHandleImpl*>(handle);
|
|
|
|
// should also persist version here because old stats CF is discarded
|
|
|
|
should_persist_format_version = true;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (s.ok() && should_persist_format_version) {
|
|
|
|
// Persistent stats CF being created for the first time, need to write
|
|
|
|
// format version key
|
|
|
|
WriteBatch batch;
|
|
|
|
batch.Put(persist_stats_cf_handle_, kFormatVersionKeyString,
|
|
|
|
ToString(kStatsCFCurrentFormatVersion));
|
|
|
|
batch.Put(persist_stats_cf_handle_, kCompatibleVersionKeyString,
|
|
|
|
ToString(kStatsCFCompatibleFormatVersion));
|
|
|
|
WriteOptions wo;
|
|
|
|
wo.low_pri = true;
|
|
|
|
wo.no_slowdown = true;
|
|
|
|
wo.sync = false;
|
|
|
|
s = Write(wo, &batch);
|
|
|
|
}
|
|
|
|
mutex_.Lock();
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
|
|
|
Status DBImpl::InitPersistStatsColumnFamily() {
|
|
|
|
mutex_.AssertHeld();
|
|
|
|
assert(!persist_stats_cf_handle_);
|
|
|
|
ColumnFamilyData* persistent_stats_cfd =
|
|
|
|
versions_->GetColumnFamilySet()->GetColumnFamily(
|
|
|
|
kPersistentStatsColumnFamilyName);
|
|
|
|
persistent_stats_cfd_exists_ = persistent_stats_cfd != nullptr;
|
|
|
|
|
|
|
|
Status s;
|
|
|
|
if (persistent_stats_cfd != nullptr) {
|
|
|
|
// We are recovering from a DB which already contains persistent stats CF,
|
|
|
|
// the CF is already created in VersionSet::ApplyOneVersionEdit, but
|
|
|
|
// column family handle was not. Need to explicitly create handle here.
|
|
|
|
persist_stats_cf_handle_ =
|
|
|
|
new ColumnFamilyHandleImpl(persistent_stats_cfd, this, &mutex_);
|
|
|
|
} else {
|
|
|
|
mutex_.Unlock();
|
|
|
|
ColumnFamilyHandle* handle = nullptr;
|
|
|
|
ColumnFamilyOptions cfo;
|
|
|
|
OptimizeForPersistentStats(&cfo);
|
|
|
|
s = CreateColumnFamily(cfo, kPersistentStatsColumnFamilyName, &handle);
|
|
|
|
persist_stats_cf_handle_ = static_cast<ColumnFamilyHandleImpl*>(handle);
|
|
|
|
mutex_.Lock();
|
|
|
|
}
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
// REQUIRES: log_numbers are sorted in ascending order
|
|
|
|
Status DBImpl::RecoverLogFiles(const std::vector<uint64_t>& log_numbers,
|
2020-01-29 20:39:11 +01:00
|
|
|
SequenceNumber* next_sequence, bool read_only,
|
|
|
|
bool* corrupted_log_found) {
|
2017-04-06 02:14:05 +02:00
|
|
|
struct LogReporter : public log::Reader::Reporter {
|
|
|
|
Env* env;
|
|
|
|
Logger* info_log;
|
|
|
|
const char* fname;
|
|
|
|
Status* status; // nullptr if immutable_db_options_.paranoid_checks==false
|
2019-02-14 22:52:47 +01:00
|
|
|
void Corruption(size_t bytes, const Status& s) override {
|
2017-04-06 02:14:05 +02:00
|
|
|
ROCKS_LOG_WARN(info_log, "%s%s: dropping %d bytes; %s",
|
2020-06-12 03:39:21 +02:00
|
|
|
(status == nullptr ? "(ignoring error) " : ""), fname,
|
|
|
|
static_cast<int>(bytes), s.ToString().c_str());
|
|
|
|
if (status != nullptr && status->ok()) {
|
|
|
|
*status = s;
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
mutex_.AssertHeld();
|
|
|
|
Status status;
|
|
|
|
std::unordered_map<int, VersionEdit> version_edits;
|
|
|
|
// no need to refcount because iteration is under mutex
|
|
|
|
for (auto cfd : *versions_->GetColumnFamilySet()) {
|
|
|
|
VersionEdit edit;
|
|
|
|
edit.SetColumnFamily(cfd->GetID());
|
|
|
|
version_edits.insert({cfd->GetID(), edit});
|
|
|
|
}
|
|
|
|
int job_id = next_job_id_.fetch_add(1);
|
|
|
|
{
|
|
|
|
auto stream = event_logger_.Log();
|
|
|
|
stream << "job" << job_id << "event"
|
|
|
|
<< "recovery_started";
|
|
|
|
stream << "log_files";
|
|
|
|
stream.StartArray();
|
|
|
|
for (auto log_number : log_numbers) {
|
|
|
|
stream << log_number;
|
|
|
|
}
|
|
|
|
stream.EndArray();
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifndef ROCKSDB_LITE
|
|
|
|
if (immutable_db_options_.wal_filter != nullptr) {
|
|
|
|
std::map<std::string, uint32_t> cf_name_id_map;
|
|
|
|
std::map<uint32_t, uint64_t> cf_lognumber_map;
|
|
|
|
for (auto cfd : *versions_->GetColumnFamilySet()) {
|
2018-12-17 22:12:22 +01:00
|
|
|
cf_name_id_map.insert(std::make_pair(cfd->GetName(), cfd->GetID()));
|
2017-04-06 02:14:05 +02:00
|
|
|
cf_lognumber_map.insert(
|
2018-12-17 22:12:22 +01:00
|
|
|
std::make_pair(cfd->GetID(), cfd->GetLogNumber()));
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
immutable_db_options_.wal_filter->ColumnFamilyLogNumberMap(cf_lognumber_map,
|
|
|
|
cf_name_id_map);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
bool stop_replay_by_wal_filter = false;
|
|
|
|
bool stop_replay_for_corruption = false;
|
|
|
|
bool flushed = false;
|
2017-09-23 02:24:14 +02:00
|
|
|
uint64_t corrupted_log_number = kMaxSequenceNumber;
|
2019-06-08 00:31:40 +02:00
|
|
|
uint64_t min_log_number = MinLogNumberToKeep();
|
2017-04-06 02:14:05 +02:00
|
|
|
for (auto log_number : log_numbers) {
|
2019-06-08 00:31:40 +02:00
|
|
|
if (log_number < min_log_number) {
|
Skip deleted WALs during recovery
Summary:
This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction)
This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2.
Closes https://github.com/facebook/rocksdb/pull/3765
Differential Revision: D7747618
Pulled By: siying
fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
2018-05-04 00:35:11 +02:00
|
|
|
ROCKS_LOG_INFO(immutable_db_options_.info_log,
|
|
|
|
"Skipping log #%" PRIu64
|
|
|
|
" since it is older than min log to keep #%" PRIu64,
|
2019-06-08 00:31:40 +02:00
|
|
|
log_number, min_log_number);
|
Skip deleted WALs during recovery
Summary:
This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction)
This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2.
Closes https://github.com/facebook/rocksdb/pull/3765
Differential Revision: D7747618
Pulled By: siying
fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
2018-05-04 00:35:11 +02:00
|
|
|
continue;
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
// The previous incarnation may not have written any MANIFEST
|
|
|
|
// records after allocating this log number. So we manually
|
|
|
|
// update the file number allocation counter in VersionSet.
|
2017-10-10 22:07:00 +02:00
|
|
|
versions_->MarkFileNumberUsed(log_number);
|
2017-04-06 02:14:05 +02:00
|
|
|
// Open the log file
|
|
|
|
std::string fname = LogFileName(immutable_db_options_.wal_dir, log_number);
|
|
|
|
|
|
|
|
ROCKS_LOG_INFO(immutable_db_options_.info_log,
|
|
|
|
"Recovering log #%" PRIu64 " mode %d", log_number,
|
2019-04-04 21:05:42 +02:00
|
|
|
static_cast<int>(immutable_db_options_.wal_recovery_mode));
|
2017-04-06 02:14:05 +02:00
|
|
|
auto logFileDropped = [this, &fname]() {
|
|
|
|
uint64_t bytes;
|
|
|
|
if (env_->GetFileSize(fname, &bytes).ok()) {
|
|
|
|
auto info_log = immutable_db_options_.info_log.get();
|
|
|
|
ROCKS_LOG_WARN(info_log, "%s: dropping %d bytes", fname.c_str(),
|
|
|
|
static_cast<int>(bytes));
|
|
|
|
}
|
|
|
|
};
|
|
|
|
if (stop_replay_by_wal_filter) {
|
|
|
|
logFileDropped();
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2018-11-09 20:17:34 +01:00
|
|
|
std::unique_ptr<SequentialFileReader> file_reader;
|
2017-04-06 02:14:05 +02:00
|
|
|
{
|
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
|
|
|
std::unique_ptr<FSSequentialFile> file;
|
|
|
|
status = fs_->NewSequentialFile(fname,
|
|
|
|
fs_->OptimizeForLogRead(file_options_),
|
|
|
|
&file, nullptr);
|
2017-04-06 02:14:05 +02:00
|
|
|
if (!status.ok()) {
|
|
|
|
MaybeIgnoreError(&status);
|
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
} else {
|
|
|
|
// Fail with one log file, but that's ok.
|
|
|
|
// Try next one.
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
}
|
2019-07-19 20:54:38 +02:00
|
|
|
file_reader.reset(new SequentialFileReader(
|
2020-08-19 01:19:22 +02:00
|
|
|
std::move(file), fname, immutable_db_options_.log_readahead_size,
|
|
|
|
io_tracer_));
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
// Create the log reader.
|
|
|
|
LogReporter reporter;
|
|
|
|
reporter.env = env_;
|
|
|
|
reporter.info_log = immutable_db_options_.info_log.get();
|
|
|
|
reporter.fname = fname.c_str();
|
|
|
|
if (!immutable_db_options_.paranoid_checks ||
|
|
|
|
immutable_db_options_.wal_recovery_mode ==
|
|
|
|
WALRecoveryMode::kSkipAnyCorruptedRecords) {
|
|
|
|
reporter.status = nullptr;
|
|
|
|
} else {
|
|
|
|
reporter.status = &status;
|
|
|
|
}
|
|
|
|
// We intentially make log::Reader do checksumming even if
|
|
|
|
// paranoid_checks==false so that corruptions cause entire commits
|
|
|
|
// to be skipped instead of propagating bad information (like overly
|
|
|
|
// large sequence numbers).
|
|
|
|
log::Reader reader(immutable_db_options_.info_log, std::move(file_reader),
|
2019-03-27 00:41:31 +01:00
|
|
|
&reporter, true /*checksum*/, log_number);
|
2017-04-06 02:14:05 +02:00
|
|
|
|
|
|
|
// Determine if we should tolerate incomplete records at the tail end of the
|
|
|
|
// Read all the records and add to a memtable
|
|
|
|
std::string scratch;
|
|
|
|
Slice record;
|
|
|
|
WriteBatch batch;
|
|
|
|
|
2020-06-12 03:39:21 +02:00
|
|
|
TEST_SYNC_POINT_CALLBACK("DBImpl::RecoverLogFiles:BeforeReadWal",
|
|
|
|
/*arg=*/nullptr);
|
2017-04-06 02:14:05 +02:00
|
|
|
while (!stop_replay_by_wal_filter &&
|
|
|
|
reader.ReadRecord(&record, &scratch,
|
|
|
|
immutable_db_options_.wal_recovery_mode) &&
|
|
|
|
status.ok()) {
|
|
|
|
if (record.size() < WriteBatchInternal::kHeader) {
|
|
|
|
reporter.Corruption(record.size(),
|
|
|
|
Status::Corruption("log record too small"));
|
|
|
|
continue;
|
|
|
|
}
|
2020-08-25 01:41:42 +02:00
|
|
|
|
|
|
|
status = WriteBatchInternal::SetContents(&batch, record);
|
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
SequenceNumber sequence = WriteBatchInternal::Sequence(&batch);
|
|
|
|
|
|
|
|
if (immutable_db_options_.wal_recovery_mode ==
|
|
|
|
WALRecoveryMode::kPointInTimeRecovery) {
|
|
|
|
// In point-in-time recovery mode, if sequence id of log files are
|
|
|
|
// consecutive, we continue recovery despite corruption. This could
|
|
|
|
// happen when we open and write to a corrupted DB, where sequence id
|
|
|
|
// will start from the last sequence id we recovered.
|
2017-12-01 08:39:56 +01:00
|
|
|
if (sequence == *next_sequence) {
|
2017-04-06 02:14:05 +02:00
|
|
|
stop_replay_for_corruption = false;
|
|
|
|
}
|
|
|
|
if (stop_replay_for_corruption) {
|
|
|
|
logFileDropped();
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifndef ROCKSDB_LITE
|
|
|
|
if (immutable_db_options_.wal_filter != nullptr) {
|
|
|
|
WriteBatch new_batch;
|
|
|
|
bool batch_changed = false;
|
|
|
|
|
|
|
|
WalFilter::WalProcessingOption wal_processing_option =
|
|
|
|
immutable_db_options_.wal_filter->LogRecordFound(
|
|
|
|
log_number, fname, batch, &new_batch, &batch_changed);
|
|
|
|
|
|
|
|
switch (wal_processing_option) {
|
|
|
|
case WalFilter::WalProcessingOption::kContinueProcessing:
|
|
|
|
// do nothing, proceeed normally
|
|
|
|
break;
|
|
|
|
case WalFilter::WalProcessingOption::kIgnoreCurrentRecord:
|
|
|
|
// skip current record
|
|
|
|
continue;
|
|
|
|
case WalFilter::WalProcessingOption::kStopReplay:
|
|
|
|
// skip current record and stop replay
|
|
|
|
stop_replay_by_wal_filter = true;
|
|
|
|
continue;
|
|
|
|
case WalFilter::WalProcessingOption::kCorruptedRecord: {
|
|
|
|
status =
|
|
|
|
Status::Corruption("Corruption reported by Wal Filter ",
|
|
|
|
immutable_db_options_.wal_filter->Name());
|
|
|
|
MaybeIgnoreError(&status);
|
|
|
|
if (!status.ok()) {
|
|
|
|
reporter.Corruption(record.size(), status);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
default: {
|
|
|
|
assert(false); // unhandled case
|
|
|
|
status = Status::NotSupported(
|
|
|
|
"Unknown WalProcessingOption returned"
|
|
|
|
" by Wal Filter ",
|
|
|
|
immutable_db_options_.wal_filter->Name());
|
|
|
|
MaybeIgnoreError(&status);
|
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
} else {
|
|
|
|
// Ignore the error with current record processing.
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (batch_changed) {
|
|
|
|
// Make sure that the count in the new batch is
|
|
|
|
// within the orignal count.
|
|
|
|
int new_count = WriteBatchInternal::Count(&new_batch);
|
|
|
|
int original_count = WriteBatchInternal::Count(&batch);
|
|
|
|
if (new_count > original_count) {
|
|
|
|
ROCKS_LOG_FATAL(
|
|
|
|
immutable_db_options_.info_log,
|
|
|
|
"Recovering log #%" PRIu64
|
|
|
|
" mode %d log filter %s returned "
|
|
|
|
"more records (%d) than original (%d) which is not allowed. "
|
|
|
|
"Aborting recovery.",
|
2019-04-04 21:05:42 +02:00
|
|
|
log_number,
|
|
|
|
static_cast<int>(immutable_db_options_.wal_recovery_mode),
|
2017-04-06 02:14:05 +02:00
|
|
|
immutable_db_options_.wal_filter->Name(), new_count,
|
|
|
|
original_count);
|
|
|
|
status = Status::NotSupported(
|
|
|
|
"More than original # of records "
|
|
|
|
"returned by Wal Filter ",
|
|
|
|
immutable_db_options_.wal_filter->Name());
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
// Set the same sequence number in the new_batch
|
|
|
|
// as the original batch.
|
|
|
|
WriteBatchInternal::SetSequence(&new_batch,
|
|
|
|
WriteBatchInternal::Sequence(&batch));
|
|
|
|
batch = new_batch;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
#endif // ROCKSDB_LITE
|
|
|
|
|
|
|
|
// If column family was not found, it might mean that the WAL write
|
|
|
|
// batch references to the column family that was dropped after the
|
|
|
|
// insert. We don't want to fail the whole write batch in that case --
|
|
|
|
// we just ignore the update.
|
|
|
|
// That's why we set ignore missing column families to true
|
|
|
|
bool has_valid_writes = false;
|
|
|
|
status = WriteBatchInternal::InsertInto(
|
Refactor trimming logic for immutable memtables (#5022)
Summary:
MyRocks currently sets `max_write_buffer_number_to_maintain` in order to maintain enough history for transaction conflict checking. The effectiveness of this approach depends on the size of memtables. When memtables are small, it may not keep enough history; when memtables are large, this may consume too much memory.
We are proposing a new way to configure memtable list history: by limiting the memory usage of immutable memtables. The new option is `max_write_buffer_size_to_maintain` and it will take precedence over the old `max_write_buffer_number_to_maintain` if they are both set to non-zero values. The new option accounts for the total memory usage of flushed immutable memtables and mutable memtable. When the total usage exceeds the limit, RocksDB may start dropping immutable memtables (which is also called trimming history), starting from the oldest one.
The semantics of the old option actually works both as an upper bound and lower bound. History trimming will start if number of immutable memtables exceeds the limit, but it will never go below (limit-1) due to history trimming.
In order the mimic the behavior with the new option, history trimming will stop if dropping the next immutable memtable causes the total memory usage go below the size limit. For example, assuming the size limit is set to 64MB, and there are 3 immutable memtables with sizes of 20, 30, 30. Although the total memory usage is 80MB > 64MB, dropping the oldest memtable will reduce the memory usage to 60MB < 64MB, so in this case no memtable will be dropped.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5022
Differential Revision: D14394062
Pulled By: miasantreble
fbshipit-source-id: 60457a509c6af89d0993f988c9b5c2aa9e45f5c5
2019-08-23 22:54:09 +02:00
|
|
|
&batch, column_family_memtables_.get(), &flush_scheduler_,
|
|
|
|
&trim_history_scheduler_, true, log_number, this,
|
|
|
|
false /* concurrent_memtable_writes */, next_sequence,
|
|
|
|
&has_valid_writes, seq_per_batch_, batch_per_txn_);
|
2017-04-06 02:14:05 +02:00
|
|
|
MaybeIgnoreError(&status);
|
|
|
|
if (!status.ok()) {
|
|
|
|
// We are treating this as a failure while reading since we read valid
|
|
|
|
// blocks that do not form coherent data
|
|
|
|
reporter.Corruption(record.size(), status);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (has_valid_writes && !read_only) {
|
|
|
|
// we can do this because this is called before client has access to the
|
|
|
|
// DB and there is only a single thread operating on DB
|
|
|
|
ColumnFamilyData* cfd;
|
|
|
|
|
|
|
|
while ((cfd = flush_scheduler_.TakeNextColumnFamily()) != nullptr) {
|
2019-12-13 04:02:51 +01:00
|
|
|
cfd->UnrefAndTryDelete();
|
2017-04-06 02:14:05 +02:00
|
|
|
// If this asserts, it means that InsertInto failed in
|
|
|
|
// filtering updates to already-flushed column families
|
|
|
|
assert(cfd->GetLogNumber() <= log_number);
|
|
|
|
auto iter = version_edits.find(cfd->GetID());
|
|
|
|
assert(iter != version_edits.end());
|
|
|
|
VersionEdit* edit = &iter->second;
|
|
|
|
status = WriteLevel0TableForRecovery(job_id, cfd, cfd->mem(), edit);
|
|
|
|
if (!status.ok()) {
|
|
|
|
// Reflect errors immediately so that conditions like full
|
|
|
|
// file-systems cause the DB::Open() to fail.
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
flushed = true;
|
|
|
|
|
|
|
|
cfd->CreateNewMemtable(*cfd->GetLatestMutableCFOptions(),
|
|
|
|
*next_sequence);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!status.ok()) {
|
2017-11-11 20:23:43 +01:00
|
|
|
if (status.IsNotSupported()) {
|
|
|
|
// We should not treat NotSupported as corruption. It is rather a clear
|
|
|
|
// sign that we are processing a WAL that is produced by an incompatible
|
|
|
|
// version of the code.
|
|
|
|
return status;
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
if (immutable_db_options_.wal_recovery_mode ==
|
|
|
|
WALRecoveryMode::kSkipAnyCorruptedRecords) {
|
|
|
|
// We should ignore all errors unconditionally
|
|
|
|
status = Status::OK();
|
|
|
|
} else if (immutable_db_options_.wal_recovery_mode ==
|
|
|
|
WALRecoveryMode::kPointInTimeRecovery) {
|
2020-06-12 03:39:21 +02:00
|
|
|
if (status.IsIOError()) {
|
|
|
|
ROCKS_LOG_ERROR(immutable_db_options_.info_log,
|
|
|
|
"IOError during point-in-time reading log #%" PRIu64
|
|
|
|
" seq #%" PRIu64
|
|
|
|
". %s. This likely mean loss of synced WAL, "
|
|
|
|
"thus recovery fails.",
|
|
|
|
log_number, *next_sequence,
|
|
|
|
status.ToString().c_str());
|
|
|
|
return status;
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
// We should ignore the error but not continue replaying
|
|
|
|
status = Status::OK();
|
|
|
|
stop_replay_for_corruption = true;
|
2017-09-23 02:24:14 +02:00
|
|
|
corrupted_log_number = log_number;
|
2020-01-29 20:39:11 +01:00
|
|
|
if (corrupted_log_found != nullptr) {
|
|
|
|
*corrupted_log_found = true;
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
ROCKS_LOG_INFO(immutable_db_options_.info_log,
|
|
|
|
"Point in time recovered to log #%" PRIu64
|
|
|
|
" seq #%" PRIu64,
|
|
|
|
log_number, *next_sequence);
|
|
|
|
} else {
|
|
|
|
assert(immutable_db_options_.wal_recovery_mode ==
|
|
|
|
WALRecoveryMode::kTolerateCorruptedTailRecords ||
|
|
|
|
immutable_db_options_.wal_recovery_mode ==
|
|
|
|
WALRecoveryMode::kAbsoluteConsistency);
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
flush_scheduler_.Clear();
|
Refactor trimming logic for immutable memtables (#5022)
Summary:
MyRocks currently sets `max_write_buffer_number_to_maintain` in order to maintain enough history for transaction conflict checking. The effectiveness of this approach depends on the size of memtables. When memtables are small, it may not keep enough history; when memtables are large, this may consume too much memory.
We are proposing a new way to configure memtable list history: by limiting the memory usage of immutable memtables. The new option is `max_write_buffer_size_to_maintain` and it will take precedence over the old `max_write_buffer_number_to_maintain` if they are both set to non-zero values. The new option accounts for the total memory usage of flushed immutable memtables and mutable memtable. When the total usage exceeds the limit, RocksDB may start dropping immutable memtables (which is also called trimming history), starting from the oldest one.
The semantics of the old option actually works both as an upper bound and lower bound. History trimming will start if number of immutable memtables exceeds the limit, but it will never go below (limit-1) due to history trimming.
In order the mimic the behavior with the new option, history trimming will stop if dropping the next immutable memtable causes the total memory usage go below the size limit. For example, assuming the size limit is set to 64MB, and there are 3 immutable memtables with sizes of 20, 30, 30. Although the total memory usage is 80MB > 64MB, dropping the oldest memtable will reduce the memory usage to 60MB < 64MB, so in this case no memtable will be dropped.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5022
Differential Revision: D14394062
Pulled By: miasantreble
fbshipit-source-id: 60457a509c6af89d0993f988c9b5c2aa9e45f5c5
2019-08-23 22:54:09 +02:00
|
|
|
trim_history_scheduler_.Clear();
|
2017-04-06 02:14:05 +02:00
|
|
|
auto last_sequence = *next_sequence - 1;
|
|
|
|
if ((*next_sequence != kMaxSequenceNumber) &&
|
|
|
|
(versions_->LastSequence() <= last_sequence)) {
|
2017-11-11 02:18:01 +01:00
|
|
|
versions_->SetLastAllocatedSequence(last_sequence);
|
2017-12-01 08:39:56 +01:00
|
|
|
versions_->SetLastPublishedSequence(last_sequence);
|
2017-04-06 02:14:05 +02:00
|
|
|
versions_->SetLastSequence(last_sequence);
|
|
|
|
}
|
|
|
|
}
|
2017-09-23 02:24:14 +02:00
|
|
|
// Compare the corrupted log number to all columnfamily's current log number.
|
|
|
|
// Abort Open() if any column family's log number is greater than
|
|
|
|
// the corrupted log number, which means CF contains data beyond the point of
|
|
|
|
// corruption. This could during PIT recovery when the WAL is corrupted and
|
|
|
|
// some (but not all) CFs are flushed
|
2020-01-29 20:39:11 +01:00
|
|
|
// Exclude the PIT case where no log is dropped after the corruption point.
|
|
|
|
// This is to cover the case for empty logs after corrupted log, in which we
|
|
|
|
// don't reset stop_replay_for_corruption.
|
2017-09-23 02:24:14 +02:00
|
|
|
if (stop_replay_for_corruption == true &&
|
|
|
|
(immutable_db_options_.wal_recovery_mode ==
|
|
|
|
WALRecoveryMode::kPointInTimeRecovery ||
|
|
|
|
immutable_db_options_.wal_recovery_mode ==
|
|
|
|
WALRecoveryMode::kTolerateCorruptedTailRecords)) {
|
|
|
|
for (auto cfd : *versions_->GetColumnFamilySet()) {
|
|
|
|
if (cfd->GetLogNumber() > corrupted_log_number) {
|
|
|
|
ROCKS_LOG_ERROR(immutable_db_options_.info_log,
|
|
|
|
"Column family inconsistency: SST file contains data"
|
|
|
|
" beyond the point of corruption.");
|
|
|
|
return Status::Corruption("SST file is ahead of WALs");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
|
|
|
|
// True if there's any data in the WALs; if not, we can skip re-processing
|
|
|
|
// them later
|
|
|
|
bool data_seen = false;
|
|
|
|
if (!read_only) {
|
|
|
|
// no need to refcount since client still doesn't have access
|
|
|
|
// to the DB and can not drop column families while we iterate
|
|
|
|
auto max_log_number = log_numbers.back();
|
|
|
|
for (auto cfd : *versions_->GetColumnFamilySet()) {
|
|
|
|
auto iter = version_edits.find(cfd->GetID());
|
|
|
|
assert(iter != version_edits.end());
|
|
|
|
VersionEdit* edit = &iter->second;
|
|
|
|
|
|
|
|
if (cfd->GetLogNumber() > max_log_number) {
|
|
|
|
// Column family cfd has already flushed the data
|
|
|
|
// from all logs. Memtable has to be empty because
|
|
|
|
// we filter the updates based on log_number
|
|
|
|
// (in WriteBatch::InsertInto)
|
|
|
|
assert(cfd->mem()->GetFirstSequenceNumber() == 0);
|
|
|
|
assert(edit->NumEntries() == 0);
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2019-10-25 03:28:03 +02:00
|
|
|
TEST_SYNC_POINT_CALLBACK(
|
|
|
|
"DBImpl::RecoverLogFiles:BeforeFlushFinalMemtable", /*arg=*/nullptr);
|
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
// flush the final memtable (if non-empty)
|
|
|
|
if (cfd->mem()->GetFirstSequenceNumber() != 0) {
|
|
|
|
// If flush happened in the middle of recovery (e.g. due to memtable
|
|
|
|
// being full), we flush at the end. Otherwise we'll need to record
|
|
|
|
// where we were on last flush, which make the logic complicated.
|
|
|
|
if (flushed || !immutable_db_options_.avoid_flush_during_recovery) {
|
|
|
|
status = WriteLevel0TableForRecovery(job_id, cfd, cfd->mem(), edit);
|
|
|
|
if (!status.ok()) {
|
|
|
|
// Recovery failed
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
flushed = true;
|
|
|
|
|
|
|
|
cfd->CreateNewMemtable(*cfd->GetLatestMutableCFOptions(),
|
|
|
|
versions_->LastSequence());
|
|
|
|
}
|
|
|
|
data_seen = true;
|
|
|
|
}
|
|
|
|
|
2019-10-25 03:28:03 +02:00
|
|
|
// Update the log number info in the version edit corresponding to this
|
|
|
|
// column family. Note that the version edits will be written to MANIFEST
|
|
|
|
// together later.
|
2017-04-06 02:14:05 +02:00
|
|
|
// writing log_number in the manifest means that any log file
|
|
|
|
// with number strongly less than (log_number + 1) is already
|
|
|
|
// recovered and should be ignored on next reincarnation.
|
|
|
|
// Since we already recovered max_log_number, we want all logs
|
|
|
|
// with numbers `<= max_log_number` (includes this one) to be ignored
|
|
|
|
if (flushed || cfd->mem()->GetFirstSequenceNumber() == 0) {
|
|
|
|
edit->SetLogNumber(max_log_number + 1);
|
|
|
|
}
|
2019-10-25 03:28:03 +02:00
|
|
|
}
|
|
|
|
if (status.ok()) {
|
2017-04-06 02:14:05 +02:00
|
|
|
// we must mark the next log number as used, even though it's
|
|
|
|
// not actually used. that is because VersionSet assumes
|
|
|
|
// VersionSet::next_file_number_ always to be strictly greater than any
|
|
|
|
// log number
|
2017-10-10 22:07:00 +02:00
|
|
|
versions_->MarkFileNumberUsed(max_log_number + 1);
|
2019-10-25 03:28:03 +02:00
|
|
|
|
|
|
|
autovector<ColumnFamilyData*> cfds;
|
|
|
|
autovector<const MutableCFOptions*> cf_opts;
|
|
|
|
autovector<autovector<VersionEdit*>> edit_lists;
|
|
|
|
for (auto* cfd : *versions_->GetColumnFamilySet()) {
|
|
|
|
cfds.push_back(cfd);
|
|
|
|
cf_opts.push_back(cfd->GetLatestMutableCFOptions());
|
|
|
|
auto iter = version_edits.find(cfd->GetID());
|
|
|
|
assert(iter != version_edits.end());
|
|
|
|
edit_lists.push_back({&iter->second});
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
2019-10-25 03:28:03 +02:00
|
|
|
// write MANIFEST with update
|
|
|
|
status = versions_->LogAndApply(cfds, cf_opts, edit_lists, &mutex_,
|
|
|
|
directories_.GetDbDir(),
|
|
|
|
/*new_descriptor_log=*/true);
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-09-26 19:34:56 +02:00
|
|
|
if (status.ok() && data_seen && !flushed) {
|
|
|
|
status = RestoreAliveLogFiles(log_numbers);
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
event_logger_.Log() << "job" << job_id << "event"
|
|
|
|
<< "recovery_finished";
|
|
|
|
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2018-09-26 19:34:56 +02:00
|
|
|
Status DBImpl::RestoreAliveLogFiles(const std::vector<uint64_t>& log_numbers) {
|
|
|
|
if (log_numbers.empty()) {
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
Status s;
|
|
|
|
mutex_.AssertHeld();
|
|
|
|
assert(immutable_db_options_.avoid_flush_during_recovery);
|
|
|
|
if (two_write_queues_) {
|
|
|
|
log_write_mutex_.Lock();
|
|
|
|
}
|
|
|
|
// Mark these as alive so they'll be considered for deletion later by
|
|
|
|
// FindObsoleteFiles()
|
|
|
|
total_log_size_ = 0;
|
|
|
|
log_empty_ = false;
|
|
|
|
for (auto log_number : log_numbers) {
|
|
|
|
LogFileNumberSize log(log_number);
|
|
|
|
std::string fname = LogFileName(immutable_db_options_.wal_dir, log_number);
|
|
|
|
// This gets the appear size of the logs, not including preallocated space.
|
|
|
|
s = env_->GetFileSize(fname, &log.size);
|
|
|
|
if (!s.ok()) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
total_log_size_ += log.size;
|
|
|
|
alive_log_files_.push_back(log);
|
|
|
|
// We preallocate space for logs, but then after a crash and restart, those
|
|
|
|
// preallocated space are not needed anymore. It is likely only the last
|
|
|
|
// log has such preallocated space, so we only truncate for the last log.
|
|
|
|
if (log_number == log_numbers.back()) {
|
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
|
|
|
std::unique_ptr<FSWritableFile> last_log;
|
|
|
|
Status truncate_status = fs_->ReopenWritableFile(
|
|
|
|
fname,
|
|
|
|
fs_->OptimizeForLogWrite(
|
|
|
|
file_options_,
|
|
|
|
BuildDBOptions(immutable_db_options_, mutable_db_options_)),
|
|
|
|
&last_log, nullptr);
|
2018-09-26 19:34:56 +02:00
|
|
|
if (truncate_status.ok()) {
|
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
|
|
|
truncate_status = last_log->Truncate(log.size, IOOptions(), nullptr);
|
2018-09-26 19:34:56 +02:00
|
|
|
}
|
|
|
|
if (truncate_status.ok()) {
|
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
|
|
|
truncate_status = last_log->Close(IOOptions(), nullptr);
|
2018-09-26 19:34:56 +02:00
|
|
|
}
|
|
|
|
// Not a critical error if fail to truncate.
|
|
|
|
if (!truncate_status.ok()) {
|
|
|
|
ROCKS_LOG_WARN(immutable_db_options_.info_log,
|
|
|
|
"Failed to truncate log #%" PRIu64 ": %s", log_number,
|
|
|
|
truncate_status.ToString().c_str());
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (two_write_queues_) {
|
|
|
|
log_write_mutex_.Unlock();
|
|
|
|
}
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
Status DBImpl::WriteLevel0TableForRecovery(int job_id, ColumnFamilyData* cfd,
|
|
|
|
MemTable* mem, VersionEdit* edit) {
|
|
|
|
mutex_.AssertHeld();
|
|
|
|
const uint64_t start_micros = env_->NowMicros();
|
|
|
|
FileMetaData meta;
|
2019-10-08 23:18:48 +02:00
|
|
|
std::unique_ptr<std::list<uint64_t>::iterator> pending_outputs_inserted_elem(
|
|
|
|
new std::list<uint64_t>::iterator(
|
|
|
|
CaptureCurrentFileNumberInPendingOutputs()));
|
2017-04-06 02:14:05 +02:00
|
|
|
meta.fd = FileDescriptor(versions_->NewFileNumber(), 0, 0);
|
|
|
|
ReadOptions ro;
|
|
|
|
ro.total_order_seek = true;
|
|
|
|
Arena arena;
|
|
|
|
Status s;
|
|
|
|
TableProperties table_properties;
|
|
|
|
{
|
|
|
|
ScopedArenaIterator iter(mem->NewIterator(ro, &arena));
|
|
|
|
ROCKS_LOG_DEBUG(immutable_db_options_.info_log,
|
|
|
|
"[%s] [WriteLevel0TableForRecovery]"
|
|
|
|
" Level-0 table #%" PRIu64 ": started",
|
|
|
|
cfd->GetName().c_str(), meta.fd.GetNumber());
|
|
|
|
|
|
|
|
// Get the latest mutable cf options while the mutex is still locked
|
|
|
|
const MutableCFOptions mutable_cf_options =
|
|
|
|
*cfd->GetLatestMutableCFOptions();
|
|
|
|
bool paranoid_file_checks =
|
|
|
|
cfd->GetLatestMutableCFOptions()->paranoid_file_checks;
|
2017-06-28 02:02:20 +02:00
|
|
|
|
2017-07-05 21:02:00 +02:00
|
|
|
int64_t _current_time = 0;
|
2020-08-25 01:41:42 +02:00
|
|
|
env_->GetCurrentTime(&_current_time)
|
|
|
|
.PermitUncheckedError(); // ignore error
|
2017-06-28 02:02:20 +02:00
|
|
|
const uint64_t current_time = static_cast<uint64_t>(_current_time);
|
2019-11-23 01:01:21 +01:00
|
|
|
meta.oldest_ancester_time = current_time;
|
2017-06-28 02:02:20 +02:00
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
{
|
2017-11-10 18:25:26 +01:00
|
|
|
auto write_hint = cfd->CalculateSSTWriteHint(0);
|
2017-04-06 02:14:05 +02:00
|
|
|
mutex_.Unlock();
|
|
|
|
|
|
|
|
SequenceNumber earliest_write_conflict_snapshot;
|
|
|
|
std::vector<SequenceNumber> snapshot_seqs =
|
|
|
|
snapshots_.GetAll(&earliest_write_conflict_snapshot);
|
2017-10-18 18:09:31 +02:00
|
|
|
auto snapshot_checker = snapshot_checker_.get();
|
|
|
|
if (use_custom_gc_ && snapshot_checker == nullptr) {
|
|
|
|
snapshot_checker = DisableGCSnapshotChecker::Instance();
|
|
|
|
}
|
2018-12-17 22:12:22 +01:00
|
|
|
std::vector<std::unique_ptr<FragmentedRangeTombstoneIterator>>
|
|
|
|
range_del_iters;
|
|
|
|
auto range_del_iter =
|
|
|
|
mem->NewRangeTombstoneIterator(ro, kMaxSequenceNumber);
|
|
|
|
if (range_del_iter != nullptr) {
|
|
|
|
range_del_iters.emplace_back(range_del_iter);
|
|
|
|
}
|
Pass IOStatus to write path and set retryable IO Error as hard error in BG jobs (#6487)
Summary:
In the current code base, we use Status to get and store the returned status from the call. Specifically, for IO related functions, the current Status cannot reflect the IO Error details such as error scope, error retryable attribute, and others. With the implementation of https://github.com/facebook/rocksdb/issues/5761, we have the new Wrapper for IO, which returns IOStatus instead of Status. However, the IOStatus is purged at the lower level of write path and transferred to Status.
The first job of this PR is to pass the IOStatus to the write path (flush, WAL write, and Compaction). The second job is to identify the Retryable IO Error as HardError, and set the bg_error_ as HardError. In this case, the DB Instance becomes read only. User is informed of the Status and need to take actions to deal with it (e.g., call db->Resume()).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6487
Test Plan: Added the testing case to error_handler_fs_test. Pass make asan_check
Reviewed By: anand1976
Differential Revision: D20685017
Pulled By: zhichao-cao
fbshipit-source-id: ff85f042896243abcd6ef37877834e26f36b6eb0
2020-03-28 00:03:05 +01:00
|
|
|
IOStatus io_s;
|
2017-04-06 02:14:05 +02:00
|
|
|
s = BuildTable(
|
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
|
|
|
dbname_, env_, fs_.get(), *cfd->ioptions(), mutable_cf_options,
|
|
|
|
file_options_for_compaction_, cfd->table_cache(), iter.get(),
|
2018-12-17 22:12:22 +01:00
|
|
|
std::move(range_del_iters), &meta, cfd->internal_comparator(),
|
2017-04-06 02:14:05 +02:00
|
|
|
cfd->int_tbl_prop_collector_factories(), cfd->GetID(), cfd->GetName(),
|
2017-10-06 19:26:38 +02:00
|
|
|
snapshot_seqs, earliest_write_conflict_snapshot, snapshot_checker,
|
2017-04-06 02:14:05 +02:00
|
|
|
GetCompressionFlush(*cfd->ioptions(), mutable_cf_options),
|
2019-03-18 20:07:35 +01:00
|
|
|
mutable_cf_options.sample_for_compression,
|
2020-03-31 21:08:41 +02:00
|
|
|
mutable_cf_options.compression_opts, paranoid_file_checks,
|
Pass IOStatus to write path and set retryable IO Error as hard error in BG jobs (#6487)
Summary:
In the current code base, we use Status to get and store the returned status from the call. Specifically, for IO related functions, the current Status cannot reflect the IO Error details such as error scope, error retryable attribute, and others. With the implementation of https://github.com/facebook/rocksdb/issues/5761, we have the new Wrapper for IO, which returns IOStatus instead of Status. However, the IOStatus is purged at the lower level of write path and transferred to Status.
The first job of this PR is to pass the IOStatus to the write path (flush, WAL write, and Compaction). The second job is to identify the Retryable IO Error as HardError, and set the bg_error_ as HardError. In this case, the DB Instance becomes read only. User is informed of the Status and need to take actions to deal with it (e.g., call db->Resume()).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6487
Test Plan: Added the testing case to error_handler_fs_test. Pass make asan_check
Reviewed By: anand1976
Differential Revision: D20685017
Pulled By: zhichao-cao
fbshipit-source-id: ff85f042896243abcd6ef37877834e26f36b6eb0
2020-03-28 00:03:05 +01:00
|
|
|
cfd->internal_stats(), TableFileCreationReason::kRecovery, &io_s,
|
2020-09-08 19:49:01 +02:00
|
|
|
io_tracer_, &event_logger_, job_id, Env::IO_HIGH,
|
|
|
|
nullptr /* table_properties */, -1 /* level */, current_time,
|
|
|
|
0 /* oldest_key_time */, write_hint, 0 /* file_creation_time */,
|
|
|
|
db_id_, db_session_id_);
|
2017-04-06 02:14:05 +02:00
|
|
|
LogFlush(immutable_db_options_.info_log);
|
|
|
|
ROCKS_LOG_DEBUG(immutable_db_options_.info_log,
|
|
|
|
"[%s] [WriteLevel0TableForRecovery]"
|
|
|
|
" Level-0 table #%" PRIu64 ": %" PRIu64 " bytes %s",
|
|
|
|
cfd->GetName().c_str(), meta.fd.GetNumber(),
|
|
|
|
meta.fd.GetFileSize(), s.ToString().c_str());
|
|
|
|
mutex_.Lock();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
ReleaseFileNumberFromPendingOutputs(pending_outputs_inserted_elem);
|
|
|
|
|
|
|
|
// Note that if file_size is zero, the file has been deleted and
|
|
|
|
// should not be added to the manifest.
|
|
|
|
int level = 0;
|
|
|
|
if (s.ok() && meta.fd.GetFileSize() > 0) {
|
|
|
|
edit->AddFile(level, meta.fd.GetNumber(), meta.fd.GetPathId(),
|
|
|
|
meta.fd.GetFileSize(), meta.smallest, meta.largest,
|
2018-07-28 01:00:26 +02:00
|
|
|
meta.fd.smallest_seqno, meta.fd.largest_seqno,
|
2019-11-23 01:01:21 +01:00
|
|
|
meta.marked_for_compaction, meta.oldest_blob_file_number,
|
2020-02-11 00:42:46 +01:00
|
|
|
meta.oldest_ancester_time, meta.file_creation_time,
|
|
|
|
meta.file_checksum, meta.file_checksum_func_name);
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
|
2018-04-11 19:47:54 +02:00
|
|
|
InternalStats::CompactionStats stats(CompactionReason::kFlush, 1);
|
2017-04-06 02:14:05 +02:00
|
|
|
stats.micros = env_->NowMicros() - start_micros;
|
|
|
|
stats.bytes_written = meta.fd.GetFileSize();
|
|
|
|
stats.num_output_files = 1;
|
2019-03-20 01:24:09 +01:00
|
|
|
cfd->internal_stats()->AddCompactionStats(level, Env::Priority::USER, stats);
|
2018-12-17 22:12:22 +01:00
|
|
|
cfd->internal_stats()->AddCFStats(InternalStats::BYTES_FLUSHED,
|
|
|
|
meta.fd.GetFileSize());
|
2017-04-06 02:14:05 +02:00
|
|
|
RecordTick(stats_, COMPACT_WRITE_BYTES, meta.fd.GetFileSize());
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
|
|
|
Status DB::Open(const Options& options, const std::string& dbname, DB** dbptr) {
|
|
|
|
DBOptions db_options(options);
|
|
|
|
ColumnFamilyOptions cf_options(options);
|
|
|
|
std::vector<ColumnFamilyDescriptor> column_families;
|
|
|
|
column_families.push_back(
|
|
|
|
ColumnFamilyDescriptor(kDefaultColumnFamilyName, cf_options));
|
2019-06-18 00:17:43 +02:00
|
|
|
if (db_options.persist_stats_to_disk) {
|
|
|
|
column_families.push_back(
|
|
|
|
ColumnFamilyDescriptor(kPersistentStatsColumnFamilyName, cf_options));
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
std::vector<ColumnFamilyHandle*> handles;
|
|
|
|
Status s = DB::Open(db_options, dbname, column_families, &handles, dbptr);
|
|
|
|
if (s.ok()) {
|
2019-06-18 00:17:43 +02:00
|
|
|
if (db_options.persist_stats_to_disk) {
|
|
|
|
assert(handles.size() == 2);
|
|
|
|
} else {
|
|
|
|
assert(handles.size() == 1);
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
// i can delete the handle since DBImpl is always holding a reference to
|
|
|
|
// default column family
|
2019-06-18 00:17:43 +02:00
|
|
|
if (db_options.persist_stats_to_disk && handles[1] != nullptr) {
|
|
|
|
delete handles[1];
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
delete handles[0];
|
|
|
|
}
|
|
|
|
return s;
|
|
|
|
}
|
2017-05-24 20:25:38 +02:00
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
Status DB::Open(const DBOptions& db_options, const std::string& dbname,
|
|
|
|
const std::vector<ColumnFamilyDescriptor>& column_families,
|
|
|
|
std::vector<ColumnFamilyHandle*>* handles, DB** dbptr) {
|
2018-06-29 03:46:39 +02:00
|
|
|
const bool kSeqPerBatch = true;
|
|
|
|
const bool kBatchPerTxn = true;
|
2017-11-11 02:18:01 +01:00
|
|
|
return DBImpl::Open(db_options, dbname, column_families, handles, dbptr,
|
2018-06-29 03:46:39 +02:00
|
|
|
!kSeqPerBatch, kBatchPerTxn);
|
2017-11-11 02:18:01 +01:00
|
|
|
}
|
|
|
|
|
2020-07-15 20:02:44 +02:00
|
|
|
IOStatus DBImpl::CreateWAL(uint64_t log_file_num, uint64_t recycle_log_number,
|
|
|
|
size_t preallocate_block_size,
|
|
|
|
log::Writer** new_log) {
|
|
|
|
IOStatus io_s;
|
Introduce a new storage specific Env API (#5761)
Summary:
The current Env API encompasses both storage/file operations, as well as OS related operations. Most of the APIs return a Status, which does not have enough metadata about an error, such as whether its retry-able or not, scope (i.e fault domain) of the error etc., that may be required in order to properly handle a storage error. The file APIs also do not provide enough control over the IO SLA, such as timeout, prioritization, hinting about placement and redundancy etc.
This PR separates out the file/storage APIs from Env into a new FileSystem class. The APIs are updated to return an IOStatus with metadata about the error, as well as to take an IOOptions structure as input in order to allow more control over the IO.
The user can set both ```options.env``` and ```options.file_system``` to specify that RocksDB should use the former for OS related operations and the latter for storage operations. Internally, a ```CompositeEnvWrapper``` has been introduced that inherits from ```Env``` and redirects individual methods to either an ```Env``` implementation or the ```FileSystem``` as appropriate. When options are sanitized during ```DB::Open```, ```options.env``` is replaced with a newly allocated ```CompositeEnvWrapper``` instance if both env and file_system have been specified. This way, the rest of the RocksDB code can continue to function as before.
This PR also ports PosixEnv to the new API by splitting it into two - PosixEnv and PosixFileSystem. PosixEnv is defined as a sub-class of CompositeEnvWrapper, and threading/time functions are overridden with Posix specific implementations in order to avoid an extra level of indirection.
The ```CompositeEnvWrapper``` translates ```IOStatus``` return code to ```Status```, and sets the severity to ```kSoftError``` if the io_status is retryable. The error handling code in RocksDB can then recover the DB automatically.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5761
Differential Revision: D18868376
Pulled By: anand1976
fbshipit-source-id: 39efe18a162ea746fabac6360ff529baba48486f
2019-12-13 23:47:08 +01:00
|
|
|
std::unique_ptr<FSWritableFile> lfile;
|
2019-04-16 03:47:24 +02:00
|
|
|
|
|
|
|
DBOptions db_options =
|
|
|
|
BuildDBOptions(immutable_db_options_, mutable_db_options_);
|
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
|
|
|
FileOptions opt_file_options =
|
|
|
|
fs_->OptimizeForLogWrite(file_options_, db_options);
|
2019-04-16 03:47:24 +02:00
|
|
|
std::string log_fname =
|
|
|
|
LogFileName(immutable_db_options_.wal_dir, log_file_num);
|
|
|
|
|
|
|
|
if (recycle_log_number) {
|
|
|
|
ROCKS_LOG_INFO(immutable_db_options_.info_log,
|
|
|
|
"reusing log %" PRIu64 " from recycle list\n",
|
|
|
|
recycle_log_number);
|
|
|
|
std::string old_log_fname =
|
|
|
|
LogFileName(immutable_db_options_.wal_dir, recycle_log_number);
|
2020-02-18 22:52:09 +01:00
|
|
|
TEST_SYNC_POINT("DBImpl::CreateWAL:BeforeReuseWritableFile1");
|
|
|
|
TEST_SYNC_POINT("DBImpl::CreateWAL:BeforeReuseWritableFile2");
|
2020-07-15 20:02:44 +02:00
|
|
|
io_s = fs_->ReuseWritableFile(log_fname, old_log_fname, opt_file_options,
|
|
|
|
&lfile, /*dbg=*/nullptr);
|
2019-04-16 03:47:24 +02:00
|
|
|
} else {
|
2020-07-15 20:02:44 +02:00
|
|
|
io_s = NewWritableFile(fs_.get(), log_fname, &lfile, opt_file_options);
|
2019-04-16 03:47:24 +02:00
|
|
|
}
|
|
|
|
|
2020-07-15 20:02:44 +02:00
|
|
|
if (io_s.ok()) {
|
2019-04-16 03:47:24 +02:00
|
|
|
lfile->SetWriteLifeTimeHint(CalculateWALWriteHint());
|
|
|
|
lfile->SetPreallocationBlockSize(preallocate_block_size);
|
|
|
|
|
|
|
|
const auto& listeners = immutable_db_options_.listeners;
|
2020-09-08 19:49:01 +02:00
|
|
|
std::unique_ptr<WritableFileWriter> file_writer(new WritableFileWriter(
|
|
|
|
std::move(lfile), log_fname, opt_file_options, env_, io_tracer_,
|
|
|
|
nullptr /* stats */, listeners));
|
2019-04-16 03:47:24 +02:00
|
|
|
*new_log = new log::Writer(std::move(file_writer), log_file_num,
|
|
|
|
immutable_db_options_.recycle_log_file_num > 0,
|
|
|
|
immutable_db_options_.manual_wal_flush);
|
|
|
|
}
|
2020-07-15 20:02:44 +02:00
|
|
|
return io_s;
|
2019-04-16 03:47:24 +02:00
|
|
|
}
|
|
|
|
|
2017-11-11 02:18:01 +01:00
|
|
|
Status DBImpl::Open(const DBOptions& db_options, const std::string& dbname,
|
|
|
|
const std::vector<ColumnFamilyDescriptor>& column_families,
|
|
|
|
std::vector<ColumnFamilyHandle*>* handles, DB** dbptr,
|
2018-06-29 03:46:39 +02:00
|
|
|
const bool seq_per_batch, const bool batch_per_txn) {
|
2017-04-06 02:14:05 +02:00
|
|
|
Status s = SanitizeOptionsByTable(db_options, column_families);
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
|
|
|
s = ValidateOptions(db_options, column_families);
|
|
|
|
if (!s.ok()) {
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
|
|
|
*dbptr = nullptr;
|
|
|
|
handles->clear();
|
|
|
|
|
|
|
|
size_t max_write_buffer_size = 0;
|
|
|
|
for (auto cf : column_families) {
|
|
|
|
max_write_buffer_size =
|
|
|
|
std::max(max_write_buffer_size, cf.options.write_buffer_size);
|
|
|
|
}
|
|
|
|
|
2018-06-29 03:46:39 +02:00
|
|
|
DBImpl* impl = new DBImpl(db_options, dbname, seq_per_batch, batch_per_txn);
|
2017-04-06 02:14:05 +02:00
|
|
|
s = impl->env_->CreateDirIfMissing(impl->immutable_db_options_.wal_dir);
|
|
|
|
if (s.ok()) {
|
2018-04-06 04:49:06 +02:00
|
|
|
std::vector<std::string> paths;
|
|
|
|
for (auto& db_path : impl->immutable_db_options_.db_paths) {
|
|
|
|
paths.emplace_back(db_path.path);
|
|
|
|
}
|
|
|
|
for (auto& cf : column_families) {
|
|
|
|
for (auto& cf_path : cf.options.cf_paths) {
|
|
|
|
paths.emplace_back(cf_path.path);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
for (auto& path : paths) {
|
|
|
|
s = impl->env_->CreateDirIfMissing(path);
|
2017-04-06 02:14:05 +02:00
|
|
|
if (!s.ok()) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
Auto recovery from out of space errors (#4164)
Summary:
This commit implements automatic recovery from a Status::NoSpace() error
during background operations such as write callback, flush and
compaction. The broad design is as follows -
1. Compaction errors are treated as soft errors and don't put the
database in read-only mode. A compaction is delayed until enough free
disk space is available to accomodate the compaction outputs, which is
estimated based on the input size. This means that users can continue to
write, and we rely on the WriteController to delay or stop writes if the
compaction debt becomes too high due to persistent low disk space
condition
2. Errors during write callback and flush are treated as hard errors,
i.e the database is put in read-only mode and goes back to read-write
only fater certain recovery actions are taken.
3. Both types of recovery rely on the SstFileManagerImpl to poll for
sufficient disk space. We assume that there is a 1-1 mapping between an
SFM and the underlying OS storage container. For cases where multiple
DBs are hosted on a single storage container, the user is expected to
allocate a single SFM instance and use the same one for all the DBs. If
no SFM is specified by the user, DBImpl::Open() will allocate one, but
this will be one per DB and each DB will recover independently. The
recovery implemented by SFM is as follows -
a) On the first occurance of an out of space error during compaction,
subsequent
compactions will be delayed until the disk free space check indicates
enough available space. The required space is computed as the sum of
input sizes.
b) The free space check requirement will be removed once the amount of
free space is greater than the size reserved by in progress
compactions when the first error occured
c) If the out of space error is a hard error, a background thread in
SFM will poll for sufficient headroom before triggering the recovery
of the database and putting it in write-only mode. The headroom is
calculated as the sum of the write_buffer_size of all the DB instances
associated with the SFM
4. EventListener callbacks will be called at the start and completion of
automatic recovery. Users can disable the auto recov ery in the start
callback, and later initiate it manually by calling DB::Resume()
Todo:
1. More extensive testing
2. Add disk full condition to db_stress (follow-on PR)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4164
Differential Revision: D9846378
Pulled By: anand1976
fbshipit-source-id: 80ea875dbd7f00205e19c82215ff6e37da10da4a
2018-09-15 22:36:19 +02:00
|
|
|
|
|
|
|
// For recovery from NoSpace() error, we can only handle
|
|
|
|
// the case where the database is stored in a single path
|
|
|
|
if (paths.size() <= 1) {
|
|
|
|
impl->error_handler_.EnableAutoRecovery();
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
2020-03-12 02:36:43 +01:00
|
|
|
if (s.ok()) {
|
|
|
|
s = impl->CreateArchivalDirectory();
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
if (!s.ok()) {
|
|
|
|
delete impl;
|
|
|
|
return s;
|
|
|
|
}
|
2019-07-07 06:04:22 +02:00
|
|
|
|
2019-09-20 21:00:55 +02:00
|
|
|
impl->wal_in_db_path_ = IsWalDirSameAsDBPath(&impl->immutable_db_options_);
|
2019-07-07 06:04:22 +02:00
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
impl->mutex_.Lock();
|
|
|
|
// Handles create_if_missing, error_if_exists
|
2020-01-29 20:39:11 +01:00
|
|
|
uint64_t recovered_seq(kMaxSequenceNumber);
|
|
|
|
s = impl->Recover(column_families, false, false, false, &recovered_seq);
|
2017-04-06 02:14:05 +02:00
|
|
|
if (s.ok()) {
|
|
|
|
uint64_t new_log_number = impl->versions_->NewFileNumber();
|
2019-04-16 03:47:24 +02:00
|
|
|
log::Writer* new_log = nullptr;
|
|
|
|
const size_t preallocate_block_size =
|
|
|
|
impl->GetWalPreallocateBlockSize(max_write_buffer_size);
|
|
|
|
s = impl->CreateWAL(new_log_number, 0 /*recycle_log_number*/,
|
|
|
|
preallocate_block_size, &new_log);
|
2017-04-06 02:14:05 +02:00
|
|
|
if (s.ok()) {
|
2019-04-16 03:47:24 +02:00
|
|
|
InstrumentedMutexLock wl(&impl->log_write_mutex_);
|
|
|
|
impl->logfile_number_ = new_log_number;
|
|
|
|
assert(new_log != nullptr);
|
|
|
|
impl->logs_.emplace_back(new_log_number, new_log);
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
|
2019-04-16 03:47:24 +02:00
|
|
|
if (s.ok()) {
|
2017-04-06 02:14:05 +02:00
|
|
|
// set column family handles
|
|
|
|
for (auto cf : column_families) {
|
|
|
|
auto cfd =
|
|
|
|
impl->versions_->GetColumnFamilySet()->GetColumnFamily(cf.name);
|
|
|
|
if (cfd != nullptr) {
|
|
|
|
handles->push_back(
|
|
|
|
new ColumnFamilyHandleImpl(cfd, impl, &impl->mutex_));
|
|
|
|
impl->NewThreadStatusCfInfo(cfd);
|
|
|
|
} else {
|
|
|
|
if (db_options.create_missing_column_families) {
|
|
|
|
// missing column family, create it
|
|
|
|
ColumnFamilyHandle* handle;
|
|
|
|
impl->mutex_.Unlock();
|
|
|
|
s = impl->CreateColumnFamily(cf.options, cf.name, &handle);
|
|
|
|
impl->mutex_.Lock();
|
|
|
|
if (s.ok()) {
|
|
|
|
handles->push_back(handle);
|
|
|
|
} else {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
} else {
|
2020-08-07 00:16:50 +02:00
|
|
|
s = Status::InvalidArgument("Column family not found", cf.name);
|
2017-04-06 02:14:05 +02:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (s.ok()) {
|
2017-10-06 03:00:38 +02:00
|
|
|
SuperVersionContext sv_context(/* create_superversion */ true);
|
2017-04-06 02:14:05 +02:00
|
|
|
for (auto cfd : *impl->versions_->GetColumnFamilySet()) {
|
2017-10-06 03:00:38 +02:00
|
|
|
impl->InstallSuperVersionAndScheduleWork(
|
|
|
|
cfd, &sv_context, *cfd->GetLatestMutableCFOptions());
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
2017-10-06 03:00:38 +02:00
|
|
|
sv_context.Clean();
|
2017-11-11 02:18:01 +01:00
|
|
|
if (impl->two_write_queues_) {
|
2017-06-30 18:30:03 +02:00
|
|
|
impl->log_write_mutex_.Lock();
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
impl->alive_log_files_.push_back(
|
|
|
|
DBImpl::LogFileNumberSize(impl->logfile_number_));
|
2017-11-11 02:18:01 +01:00
|
|
|
if (impl->two_write_queues_) {
|
2017-06-30 18:30:03 +02:00
|
|
|
impl->log_write_mutex_.Unlock();
|
|
|
|
}
|
2020-01-29 20:39:11 +01:00
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
impl->DeleteObsoleteFiles();
|
2020-03-03 01:14:00 +01:00
|
|
|
s = impl->directories_.GetDbDir()->Fsync(IOOptions(), nullptr);
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
2020-01-29 20:39:11 +01:00
|
|
|
if (s.ok()) {
|
|
|
|
// In WritePrepared there could be gap in sequence numbers. This breaks
|
|
|
|
// the trick we use in kPointInTimeRecovery which assumes the first seq in
|
|
|
|
// the log right after the corrupted log is one larger than the last seq
|
|
|
|
// we read from the logs. To let this trick keep working, we add a dummy
|
|
|
|
// entry with the expected sequence to the first log right after recovery.
|
|
|
|
// In non-WritePrepared case also the new log after recovery could be
|
|
|
|
// empty, and thus missing the consecutive seq hint to distinguish
|
|
|
|
// middle-log corruption to corrupted-log-remained-after-recovery. This
|
|
|
|
// case also will be addressed by a dummy write.
|
|
|
|
if (recovered_seq != kMaxSequenceNumber) {
|
|
|
|
WriteBatch empty_batch;
|
|
|
|
WriteBatchInternal::SetSequence(&empty_batch, recovered_seq);
|
|
|
|
WriteOptions write_options;
|
|
|
|
uint64_t log_used, log_size;
|
|
|
|
log::Writer* log_writer = impl->logs_.back().writer;
|
|
|
|
s = impl->WriteToWAL(empty_batch, log_writer, &log_used, &log_size);
|
2020-02-14 03:39:38 +01:00
|
|
|
if (s.ok()) {
|
|
|
|
// Need to fsync, otherwise it might get lost after a power reset.
|
|
|
|
s = impl->FlushWAL(false);
|
|
|
|
if (s.ok()) {
|
|
|
|
s = log_writer->file()->Sync(impl->immutable_db_options_.use_fsync);
|
|
|
|
}
|
|
|
|
}
|
2020-01-29 20:39:11 +01:00
|
|
|
}
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
2019-06-18 00:17:43 +02:00
|
|
|
if (s.ok() && impl->immutable_db_options_.persist_stats_to_disk) {
|
|
|
|
// try to read format version but no need to fail Open() even if it fails
|
|
|
|
s = impl->PersistentStatsProcessFormatVersion();
|
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
|
|
|
|
if (s.ok()) {
|
|
|
|
for (auto cfd : *impl->versions_->GetColumnFamilySet()) {
|
|
|
|
if (cfd->ioptions()->compaction_style == kCompactionStyleFIFO) {
|
|
|
|
auto* vstorage = cfd->current()->storage_info();
|
|
|
|
for (int i = 1; i < vstorage->num_levels(); ++i) {
|
|
|
|
int num_files = vstorage->NumLevelFiles(i);
|
|
|
|
if (num_files > 0) {
|
|
|
|
s = Status::InvalidArgument(
|
|
|
|
"Not all files are at level 0. Cannot "
|
|
|
|
"open with FIFO compaction style.");
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (!cfd->mem()->IsSnapshotSupported()) {
|
|
|
|
impl->is_snapshot_supported_ = false;
|
|
|
|
}
|
|
|
|
if (cfd->ioptions()->merge_operator != nullptr &&
|
|
|
|
!cfd->mem()->IsMergeOperatorSupported()) {
|
|
|
|
s = Status::InvalidArgument(
|
|
|
|
"The memtable of column family %s does not support merge operator "
|
2018-12-17 22:12:22 +01:00
|
|
|
"its options.merge_operator is non-null",
|
|
|
|
cfd->GetName().c_str());
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
if (!s.ok()) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
TEST_SYNC_POINT("DBImpl::Open:Opened");
|
|
|
|
Status persist_options_status;
|
|
|
|
if (s.ok()) {
|
|
|
|
// Persist RocksDB Options before scheduling the compaction.
|
|
|
|
// The WriteOptionsFile() will release and lock the mutex internally.
|
2017-05-08 07:12:55 +02:00
|
|
|
persist_options_status = impl->WriteOptionsFile(
|
|
|
|
false /*need_mutex_lock*/, false /*need_enter_write_thread*/);
|
2017-04-06 02:14:05 +02:00
|
|
|
|
|
|
|
*dbptr = impl;
|
|
|
|
impl->opened_successfully_ = true;
|
|
|
|
impl->MaybeScheduleFlushOrCompaction();
|
|
|
|
}
|
|
|
|
impl->mutex_.Unlock();
|
|
|
|
|
|
|
|
#ifndef ROCKSDB_LITE
|
|
|
|
auto sfm = static_cast<SstFileManagerImpl*>(
|
|
|
|
impl->immutable_db_options_.sst_file_manager.get());
|
|
|
|
if (s.ok() && sfm) {
|
2020-06-05 18:41:03 +02:00
|
|
|
// Set Statistics ptr for SstFileManager to dump the stats of
|
|
|
|
// DeleteScheduler.
|
|
|
|
sfm->SetStatisticsPtr(impl->immutable_db_options_.statistics);
|
|
|
|
ROCKS_LOG_INFO(impl->immutable_db_options_.info_log,
|
|
|
|
"SstFileManager instance %p", sfm);
|
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
// Notify SstFileManager about all sst files that already exist in
|
2018-04-06 04:49:06 +02:00
|
|
|
// db_paths[0] and cf_paths[0] when the DB is opened.
|
2020-02-04 22:38:00 +01:00
|
|
|
|
|
|
|
// SstFileManagerImpl needs to know sizes of the files. For files whose size
|
|
|
|
// we already know (sst files that appear in manifest - typically that's the
|
|
|
|
// vast majority of all files), we'll pass the size to SstFileManager.
|
|
|
|
// For all other files SstFileManager will query the size from filesystem.
|
|
|
|
|
|
|
|
std::vector<LiveFileMetaData> metadata;
|
|
|
|
|
|
|
|
impl->mutex_.Lock();
|
|
|
|
impl->versions_->GetLiveFilesMetaData(&metadata);
|
|
|
|
impl->mutex_.Unlock();
|
|
|
|
|
|
|
|
std::unordered_map<std::string, uint64_t> known_file_sizes;
|
|
|
|
for (const auto& md : metadata) {
|
|
|
|
std::string name = md.name;
|
|
|
|
if (!name.empty() && name[0] == '/') {
|
|
|
|
name = name.substr(1);
|
|
|
|
}
|
|
|
|
known_file_sizes[name] = md.size;
|
|
|
|
}
|
|
|
|
|
2018-04-06 04:49:06 +02:00
|
|
|
std::vector<std::string> paths;
|
|
|
|
paths.emplace_back(impl->immutable_db_options_.db_paths[0].path);
|
|
|
|
for (auto& cf : column_families) {
|
|
|
|
if (!cf.options.cf_paths.empty()) {
|
|
|
|
paths.emplace_back(cf.options.cf_paths[0].path);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
// Remove duplicate paths.
|
|
|
|
std::sort(paths.begin(), paths.end());
|
|
|
|
paths.erase(std::unique(paths.begin(), paths.end()), paths.end());
|
|
|
|
for (auto& path : paths) {
|
|
|
|
std::vector<std::string> existing_files;
|
2020-08-21 04:16:56 +02:00
|
|
|
// TODO: Check for errors here?
|
|
|
|
impl->immutable_db_options_.env->GetChildren(path, &existing_files)
|
|
|
|
.PermitUncheckedError();
|
2018-04-06 04:49:06 +02:00
|
|
|
for (auto& file_name : existing_files) {
|
|
|
|
uint64_t file_number;
|
|
|
|
FileType file_type;
|
|
|
|
std::string file_path = path + "/" + file_name;
|
|
|
|
if (ParseFileName(file_name, &file_number, &file_type) &&
|
|
|
|
file_type == kTableFile) {
|
2020-08-25 01:41:42 +02:00
|
|
|
// TODO: Check for errors from OnAddFile?
|
2020-02-04 22:38:00 +01:00
|
|
|
if (known_file_sizes.count(file_name)) {
|
|
|
|
// We're assuming that each sst file name exists in at most one of
|
|
|
|
// the paths.
|
|
|
|
sfm->OnAddFile(file_path, known_file_sizes.at(file_name),
|
2020-08-25 01:41:42 +02:00
|
|
|
/* compaction */ false)
|
|
|
|
.PermitUncheckedError();
|
2020-02-04 22:38:00 +01:00
|
|
|
} else {
|
2020-08-25 01:41:42 +02:00
|
|
|
sfm->OnAddFile(file_path).PermitUncheckedError();
|
2020-02-04 22:38:00 +01:00
|
|
|
}
|
2018-04-06 04:49:06 +02:00
|
|
|
}
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
}
|
Auto recovery from out of space errors (#4164)
Summary:
This commit implements automatic recovery from a Status::NoSpace() error
during background operations such as write callback, flush and
compaction. The broad design is as follows -
1. Compaction errors are treated as soft errors and don't put the
database in read-only mode. A compaction is delayed until enough free
disk space is available to accomodate the compaction outputs, which is
estimated based on the input size. This means that users can continue to
write, and we rely on the WriteController to delay or stop writes if the
compaction debt becomes too high due to persistent low disk space
condition
2. Errors during write callback and flush are treated as hard errors,
i.e the database is put in read-only mode and goes back to read-write
only fater certain recovery actions are taken.
3. Both types of recovery rely on the SstFileManagerImpl to poll for
sufficient disk space. We assume that there is a 1-1 mapping between an
SFM and the underlying OS storage container. For cases where multiple
DBs are hosted on a single storage container, the user is expected to
allocate a single SFM instance and use the same one for all the DBs. If
no SFM is specified by the user, DBImpl::Open() will allocate one, but
this will be one per DB and each DB will recover independently. The
recovery implemented by SFM is as follows -
a) On the first occurance of an out of space error during compaction,
subsequent
compactions will be delayed until the disk free space check indicates
enough available space. The required space is computed as the sum of
input sizes.
b) The free space check requirement will be removed once the amount of
free space is greater than the size reserved by in progress
compactions when the first error occured
c) If the out of space error is a hard error, a background thread in
SFM will poll for sufficient headroom before triggering the recovery
of the database and putting it in write-only mode. The headroom is
calculated as the sum of the write_buffer_size of all the DB instances
associated with the SFM
4. EventListener callbacks will be called at the start and completion of
automatic recovery. Users can disable the auto recov ery in the start
callback, and later initiate it manually by calling DB::Resume()
Todo:
1. More extensive testing
2. Add disk full condition to db_stress (follow-on PR)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4164
Differential Revision: D9846378
Pulled By: anand1976
fbshipit-source-id: 80ea875dbd7f00205e19c82215ff6e37da10da4a
2018-09-15 22:36:19 +02:00
|
|
|
|
|
|
|
// Reserve some disk buffer space. This is a heuristic - when we run out
|
|
|
|
// of disk space, this ensures that there is atleast write_buffer_size
|
|
|
|
// amount of free space before we resume DB writes. In low disk space
|
|
|
|
// conditions, we want to avoid a lot of small L0 files due to frequent
|
|
|
|
// WAL write failures and resultant forced flushes
|
|
|
|
sfm->ReserveDiskBuffer(max_write_buffer_size,
|
|
|
|
impl->immutable_db_options_.db_paths[0].path);
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
2020-08-15 05:11:35 +02:00
|
|
|
|
2017-04-06 02:14:05 +02:00
|
|
|
#endif // !ROCKSDB_LITE
|
|
|
|
|
|
|
|
if (s.ok()) {
|
2019-02-16 01:56:58 +01:00
|
|
|
ROCKS_LOG_HEADER(impl->immutable_db_options_.info_log, "DB pointer %p",
|
|
|
|
impl);
|
2017-04-06 02:14:05 +02:00
|
|
|
LogFlush(impl->immutable_db_options_.info_log);
|
2018-05-14 19:53:32 +02:00
|
|
|
assert(impl->TEST_WALBufferIsEmpty());
|
|
|
|
// If the assert above fails then we need to FlushWAL before returning
|
|
|
|
// control back to the user.
|
2017-04-06 02:14:05 +02:00
|
|
|
if (!persist_options_status.ok()) {
|
2017-05-08 07:12:55 +02:00
|
|
|
s = Status::IOError(
|
|
|
|
"DB::Open() failed --- Unable to persist Options file",
|
|
|
|
persist_options_status.ToString());
|
2017-04-06 02:14:05 +02:00
|
|
|
}
|
|
|
|
}
|
move dump stats to a separate thread (#4382)
Summary:
Currently statistics are supposed to be dumped to info log at intervals of `options.stats_dump_period_sec`. However the implementation choice was to bind it with compaction thread, meaning if the database has been serving very light traffic, the stats may not get dumped at all.
We decided to separate stats dumping into a new timed thread using `TimerQueue`, which is already used in blob_db. This will allow us schedule new timed tasks with more deterministic behavior.
Tested with db_bench using `--stats_dump_period_sec=20` in command line:
> LOG:2018/09/17-14:07:45.575025 7fe99fbfe700 [WARN] [db/db_impl.cc:605] ------- DUMPING STATS -------
LOG:2018/09/17-14:08:05.643286 7fe99fbfe700 [WARN] [db/db_impl.cc:605] ------- DUMPING STATS -------
LOG:2018/09/17-14:08:25.691325 7fe99fbfe700 [WARN] [db/db_impl.cc:605] ------- DUMPING STATS -------
LOG:2018/09/17-14:08:45.740989 7fe99fbfe700 [WARN] [db/db_impl.cc:605] ------- DUMPING STATS -------
LOG content:
> 2018/09/17-14:07:45.575025 7fe99fbfe700 [WARN] [db/db_impl.cc:605] ------- DUMPING STATS -------
2018/09/17-14:07:45.575080 7fe99fbfe700 [WARN] [db/db_impl.cc:606]
** DB Stats **
Uptime(secs): 20.0 total, 20.0 interval
Cumulative writes: 4447K writes, 4447K keys, 4447K commit groups, 1.0 writes per commit group, ingest: 5.57 GB, 285.01 MB/s
Cumulative WAL: 4447K writes, 0 syncs, 4447638.00 writes per sync, written: 5.57 GB, 285.01 MB/s
Cumulative stall: 00:00:0.012 H:M:S, 0.1 percent
Interval writes: 4447K writes, 4447K keys, 4447K commit groups, 1.0 writes per commit group, ingest: 5700.71 MB, 285.01 MB/s
Interval WAL: 4447K writes, 0 syncs, 4447638.00 writes per sync, written: 5.57 MB, 285.01 MB/s
Interval stall: 00:00:0.012 H:M:S, 0.1 percent
** Compaction Stats [default] **
Level Files Size Score Read(GB) Rn(GB) Rnp1(GB) Write(GB) Wnew(GB) Moved(GB) W-Amp Rd(MB/s) Wr(MB/s) Comp(sec) Comp(cnt) Avg(sec) KeyIn KeyDrop
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4382
Differential Revision: D9933051
Pulled By: miasantreble
fbshipit-source-id: 6d12bb1e4977674eea4bf2d2ac6d486b814bb2fa
2018-10-09 07:52:58 +02:00
|
|
|
if (s.ok()) {
|
2020-08-15 05:11:35 +02:00
|
|
|
impl->StartStatsDumpScheduler();
|
|
|
|
} else {
|
2017-04-06 02:14:05 +02:00
|
|
|
for (auto* h : *handles) {
|
|
|
|
delete h;
|
|
|
|
}
|
|
|
|
handles->clear();
|
|
|
|
delete impl;
|
|
|
|
*dbptr = nullptr;
|
|
|
|
}
|
|
|
|
return s;
|
|
|
|
}
|
2020-02-20 21:07:53 +01:00
|
|
|
} // namespace ROCKSDB_NAMESPACE
|