2016-10-21 02:05:32 +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).
|
2016-10-21 02:05:32 +02:00
|
|
|
|
2017-02-28 20:05:08 +01:00
|
|
|
#ifndef ROCKSDB_LITE
|
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
#include "db/external_sst_file_ingestion_job.h"
|
|
|
|
|
|
|
|
#include <algorithm>
|
2019-06-21 19:12:29 +02:00
|
|
|
#include <cinttypes>
|
2016-10-21 02:05:32 +02:00
|
|
|
#include <string>
|
2019-06-21 19:12:29 +02:00
|
|
|
#include <unordered_set>
|
2016-10-21 02:05:32 +02:00
|
|
|
#include <vector>
|
|
|
|
|
2019-06-21 19:12:29 +02:00
|
|
|
#include "db/db_impl/db_impl.h"
|
2016-10-21 02:05:32 +02:00
|
|
|
#include "db/version_edit.h"
|
2019-05-30 05:44:08 +02:00
|
|
|
#include "file/file_util.h"
|
2019-09-16 19:31:27 +02:00
|
|
|
#include "file/random_access_file_reader.h"
|
2017-02-03 01:38:40 +01:00
|
|
|
#include "table/merging_iterator.h"
|
2016-10-21 02:05:32 +02:00
|
|
|
#include "table/scoped_arena_iterator.h"
|
|
|
|
#include "table/sst_file_writer_collectors.h"
|
|
|
|
#include "table/table_builder.h"
|
2019-05-31 02:39:43 +02:00
|
|
|
#include "test_util/sync_point.h"
|
2016-10-21 02:05:32 +02:00
|
|
|
#include "util/stop_watch.h"
|
|
|
|
|
2020-02-20 21:07:53 +01:00
|
|
|
namespace ROCKSDB_NAMESPACE {
|
2016-10-21 02:05:32 +02:00
|
|
|
|
|
|
|
Status ExternalSstFileIngestionJob::Prepare(
|
2018-07-27 23:02:07 +02:00
|
|
|
const std::vector<std::string>& external_files_paths,
|
Ingest SST files with checksum information (#6891)
Summary:
Application can ingest SST files with file checksum information, such that during ingestion, DB is able to check data integrity and identify of the SST file. The PR introduces generate_and_verify_file_checksum to IngestExternalFileOption to control if the ingested checksum information should be verified with the generated checksum.
1. If generate_and_verify_file_checksum options is *FALSE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enables the SST file checksum and the checksum function name matches the checksum function name in DB, we trust the ingested checksum, store it in Manifest. If the checksum function name does not match, we treat that as an error and fail the IngestExternalFile() call.
2. If generate_and_verify_file_checksum options is *TRUE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enable the SST file checksum, we will use the checksum generator from DB to calculate the checksum for each ingested SST files after they are copied or moved. Then, compare the checksum results with the ingested checksum information: _A)_ if the checksum function name does not match, _verification always report true_ and we store the DB generated checksum information in Manifest. _B)_ if the checksum function name mach, and checksum match, ingestion continues and stores the checksum information in the Manifest. Otherwise, terminate file ingestion and report file corruption.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6891
Test Plan: added unit test, pass make asan_check
Reviewed By: pdillinger
Differential Revision: D21935988
Pulled By: zhichao-cao
fbshipit-source-id: 7b55f486632db467e76d72602218d0658aa7f6ed
2020-06-11 23:25:01 +02:00
|
|
|
const std::vector<std::string>& files_checksums,
|
|
|
|
const std::vector<std::string>& files_checksum_func_names,
|
2018-07-28 01:00:26 +02:00
|
|
|
uint64_t next_file_number, SuperVersion* sv) {
|
2016-10-21 02:05:32 +02:00
|
|
|
Status status;
|
|
|
|
|
|
|
|
// Read the information of files we are ingesting
|
|
|
|
for (const std::string& file_path : external_files_paths) {
|
|
|
|
IngestedFileInfo file_to_ingest;
|
2018-05-21 23:33:55 +02:00
|
|
|
status = GetIngestedFileInfo(file_path, &file_to_ingest, sv);
|
2016-10-21 02:05:32 +02:00
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
files_to_ingest_.push_back(file_to_ingest);
|
|
|
|
}
|
|
|
|
|
2016-12-05 23:16:23 +01:00
|
|
|
for (const IngestedFileInfo& f : files_to_ingest_) {
|
|
|
|
if (f.cf_id !=
|
|
|
|
TablePropertiesCollectorFactory::Context::kUnknownColumnFamily &&
|
|
|
|
f.cf_id != cfd_->GetID()) {
|
|
|
|
return Status::InvalidArgument(
|
2020-02-28 23:10:51 +01:00
|
|
|
"External file column family id don't match");
|
2016-12-05 23:16:23 +01:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
const Comparator* ucmp = cfd_->internal_comparator().user_comparator();
|
|
|
|
auto num_files = files_to_ingest_.size();
|
|
|
|
if (num_files == 0) {
|
|
|
|
return Status::InvalidArgument("The list of files is empty");
|
|
|
|
} else if (num_files > 1) {
|
2020-02-28 23:10:51 +01:00
|
|
|
// Verify that passed files don't have overlapping ranges
|
2016-10-21 02:05:32 +02:00
|
|
|
autovector<const IngestedFileInfo*> sorted_files;
|
|
|
|
for (size_t i = 0; i < num_files; i++) {
|
|
|
|
sorted_files.push_back(&files_to_ingest_[i]);
|
|
|
|
}
|
|
|
|
|
|
|
|
std::sort(
|
|
|
|
sorted_files.begin(), sorted_files.end(),
|
|
|
|
[&ucmp](const IngestedFileInfo* info1, const IngestedFileInfo* info2) {
|
2019-08-15 05:58:59 +02:00
|
|
|
return sstableKeyCompare(ucmp, info1->smallest_internal_key,
|
|
|
|
info2->smallest_internal_key) < 0;
|
2016-10-21 02:05:32 +02:00
|
|
|
});
|
|
|
|
|
2020-06-03 00:02:44 +02:00
|
|
|
for (size_t i = 0; i + 1 < num_files; i++) {
|
2019-08-15 05:58:59 +02:00
|
|
|
if (sstableKeyCompare(ucmp, sorted_files[i]->largest_internal_key,
|
|
|
|
sorted_files[i + 1]->smallest_internal_key) >= 0) {
|
2019-09-13 23:48:18 +02:00
|
|
|
files_overlap_ = true;
|
|
|
|
break;
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-09-13 23:48:18 +02:00
|
|
|
if (ingestion_options_.ingest_behind && files_overlap_) {
|
|
|
|
return Status::NotSupported("Files have overlapping ranges");
|
|
|
|
}
|
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
for (IngestedFileInfo& f : files_to_ingest_) {
|
2018-07-14 07:40:23 +02:00
|
|
|
if (f.num_entries == 0 && f.num_range_deletions == 0) {
|
2016-10-21 02:05:32 +02:00
|
|
|
return Status::InvalidArgument("File contain no entries");
|
|
|
|
}
|
|
|
|
|
2019-09-20 21:00:55 +02:00
|
|
|
if (!f.smallest_internal_key.Valid() || !f.largest_internal_key.Valid()) {
|
2016-10-21 02:05:32 +02:00
|
|
|
return Status::Corruption("Generated table have corrupted keys");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// Copy/Move external files into DB
|
2019-06-21 19:12:29 +02:00
|
|
|
std::unordered_set<size_t> ingestion_path_ids;
|
2016-10-21 02:05:32 +02:00
|
|
|
for (IngestedFileInfo& f : files_to_ingest_) {
|
2018-07-27 23:02:07 +02:00
|
|
|
f.fd = FileDescriptor(next_file_number++, 0, f.file_size);
|
2019-05-24 06:54:23 +02:00
|
|
|
f.copy_file = false;
|
2016-10-21 02:05:32 +02:00
|
|
|
const std::string path_outside_db = f.external_file_path;
|
|
|
|
const std::string path_inside_db =
|
2018-04-06 04:49:06 +02:00
|
|
|
TableFileName(cfd_->ioptions()->cf_paths, f.fd.GetNumber(),
|
|
|
|
f.fd.GetPathId());
|
2016-10-21 02:05:32 +02:00
|
|
|
if (ingestion_options_.move_files) {
|
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
|
|
|
status =
|
|
|
|
fs_->LinkFile(path_outside_db, path_inside_db, IOOptions(), nullptr);
|
2019-06-21 19:12:29 +02:00
|
|
|
if (status.ok()) {
|
|
|
|
// It is unsafe to assume application had sync the file and file
|
|
|
|
// directory before ingest the file. For integrity of RocksDB we need
|
|
|
|
// to sync the file.
|
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_to_sync;
|
|
|
|
status = fs_->ReopenWritableFile(path_inside_db, env_options_,
|
|
|
|
&file_to_sync, nullptr);
|
2019-06-21 19:12:29 +02:00
|
|
|
if (status.ok()) {
|
|
|
|
TEST_SYNC_POINT(
|
|
|
|
"ExternalSstFileIngestionJob::BeforeSyncIngestedFile");
|
|
|
|
status = SyncIngestedFile(file_to_sync.get());
|
|
|
|
TEST_SYNC_POINT("ExternalSstFileIngestionJob::AfterSyncIngestedFile");
|
|
|
|
if (!status.ok()) {
|
|
|
|
ROCKS_LOG_WARN(db_options_.info_log,
|
|
|
|
"Failed to sync ingested file %s: %s",
|
|
|
|
path_inside_db.c_str(), status.ToString().c_str());
|
|
|
|
}
|
|
|
|
}
|
|
|
|
} else if (status.IsNotSupported() &&
|
|
|
|
ingestion_options_.failed_move_fall_back_to_copy) {
|
2019-05-24 06:54:23 +02:00
|
|
|
// Original file is on a different FS, use copy instead of hard linking.
|
2018-04-13 19:47:54 +02:00
|
|
|
f.copy_file = true;
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
} else {
|
2019-05-24 06:54:23 +02:00
|
|
|
f.copy_file = true;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (f.copy_file) {
|
|
|
|
TEST_SYNC_POINT_CALLBACK("ExternalSstFileIngestionJob::Prepare:CopyFile",
|
|
|
|
nullptr);
|
2019-06-21 19:12:29 +02:00
|
|
|
// CopyFile also sync the new file.
|
2020-08-13 02:28:10 +02:00
|
|
|
status = CopyFile(fs_.get(), path_outside_db, path_inside_db, 0,
|
2020-08-19 01:19:22 +02:00
|
|
|
db_options_.use_fsync, io_tracer_);
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
2018-04-13 19:47:54 +02:00
|
|
|
TEST_SYNC_POINT("ExternalSstFileIngestionJob::Prepare:FileAdded");
|
2016-10-21 02:05:32 +02:00
|
|
|
if (!status.ok()) {
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
f.internal_file_path = path_inside_db;
|
Ingest SST files with checksum information (#6891)
Summary:
Application can ingest SST files with file checksum information, such that during ingestion, DB is able to check data integrity and identify of the SST file. The PR introduces generate_and_verify_file_checksum to IngestExternalFileOption to control if the ingested checksum information should be verified with the generated checksum.
1. If generate_and_verify_file_checksum options is *FALSE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enables the SST file checksum and the checksum function name matches the checksum function name in DB, we trust the ingested checksum, store it in Manifest. If the checksum function name does not match, we treat that as an error and fail the IngestExternalFile() call.
2. If generate_and_verify_file_checksum options is *TRUE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enable the SST file checksum, we will use the checksum generator from DB to calculate the checksum for each ingested SST files after they are copied or moved. Then, compare the checksum results with the ingested checksum information: _A)_ if the checksum function name does not match, _verification always report true_ and we store the DB generated checksum information in Manifest. _B)_ if the checksum function name mach, and checksum match, ingestion continues and stores the checksum information in the Manifest. Otherwise, terminate file ingestion and report file corruption.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6891
Test Plan: added unit test, pass make asan_check
Reviewed By: pdillinger
Differential Revision: D21935988
Pulled By: zhichao-cao
fbshipit-source-id: 7b55f486632db467e76d72602218d0658aa7f6ed
2020-06-11 23:25:01 +02:00
|
|
|
// Initialize the checksum information of ingested files.
|
|
|
|
f.file_checksum = kUnknownFileChecksum;
|
|
|
|
f.file_checksum_func_name = kUnknownFileChecksumFuncName;
|
2019-06-21 19:12:29 +02:00
|
|
|
ingestion_path_ids.insert(f.fd.GetPathId());
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_SYNC_POINT("ExternalSstFileIngestionJob::BeforeSyncDir");
|
|
|
|
if (status.ok()) {
|
|
|
|
for (auto path_id : ingestion_path_ids) {
|
2020-03-03 01:14:00 +01:00
|
|
|
status = directories_->GetDataDir(path_id)->Fsync(IOOptions(), nullptr);
|
2019-06-21 19:12:29 +02:00
|
|
|
if (!status.ok()) {
|
|
|
|
ROCKS_LOG_WARN(db_options_.info_log,
|
|
|
|
"Failed to sync directory %" ROCKSDB_PRIszt
|
|
|
|
" while ingest file: %s",
|
|
|
|
path_id, status.ToString().c_str());
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
2019-06-21 19:12:29 +02:00
|
|
|
TEST_SYNC_POINT("ExternalSstFileIngestionJob::AfterSyncDir");
|
2016-10-21 02:05:32 +02:00
|
|
|
|
Ingest SST files with checksum information (#6891)
Summary:
Application can ingest SST files with file checksum information, such that during ingestion, DB is able to check data integrity and identify of the SST file. The PR introduces generate_and_verify_file_checksum to IngestExternalFileOption to control if the ingested checksum information should be verified with the generated checksum.
1. If generate_and_verify_file_checksum options is *FALSE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enables the SST file checksum and the checksum function name matches the checksum function name in DB, we trust the ingested checksum, store it in Manifest. If the checksum function name does not match, we treat that as an error and fail the IngestExternalFile() call.
2. If generate_and_verify_file_checksum options is *TRUE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enable the SST file checksum, we will use the checksum generator from DB to calculate the checksum for each ingested SST files after they are copied or moved. Then, compare the checksum results with the ingested checksum information: _A)_ if the checksum function name does not match, _verification always report true_ and we store the DB generated checksum information in Manifest. _B)_ if the checksum function name mach, and checksum match, ingestion continues and stores the checksum information in the Manifest. Otherwise, terminate file ingestion and report file corruption.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6891
Test Plan: added unit test, pass make asan_check
Reviewed By: pdillinger
Differential Revision: D21935988
Pulled By: zhichao-cao
fbshipit-source-id: 7b55f486632db467e76d72602218d0658aa7f6ed
2020-06-11 23:25:01 +02:00
|
|
|
// Generate and check the sst file checksum. Note that, if
|
|
|
|
// IngestExternalFileOptions::write_global_seqno is true, we will not update
|
|
|
|
// the checksum information in the files_to_ingests_ here, since the file is
|
|
|
|
// upadted with the new global_seqno. After global_seqno is updated, DB will
|
|
|
|
// generate the new checksum and store it in the Manifest. In all other cases
|
|
|
|
// if ingestion_options_.write_global_seqno == true and
|
|
|
|
// verify_file_checksum is false, we only check the checksum function name.
|
|
|
|
if (status.ok() && db_options_.file_checksum_gen_factory != nullptr) {
|
|
|
|
if (ingestion_options_.verify_file_checksum == false &&
|
|
|
|
files_checksums.size() == files_to_ingest_.size() &&
|
|
|
|
files_checksum_func_names.size() == files_to_ingest_.size()) {
|
|
|
|
// Only when verify_file_checksum == false and the checksum for ingested
|
|
|
|
// files are provided, DB will use the provided checksum and does not
|
|
|
|
// generate the checksum for ingested files.
|
|
|
|
need_generate_file_checksum_ = false;
|
|
|
|
} else {
|
|
|
|
need_generate_file_checksum_ = true;
|
|
|
|
}
|
|
|
|
FileChecksumGenContext gen_context;
|
|
|
|
std::unique_ptr<FileChecksumGenerator> file_checksum_gen =
|
|
|
|
db_options_.file_checksum_gen_factory->CreateFileChecksumGenerator(
|
|
|
|
gen_context);
|
|
|
|
std::vector<std::string> generated_checksums;
|
|
|
|
std::vector<std::string> generated_checksum_func_names;
|
|
|
|
// Step 1: generate the checksum for ingested sst file.
|
|
|
|
if (need_generate_file_checksum_) {
|
|
|
|
for (size_t i = 0; i < files_to_ingest_.size(); i++) {
|
|
|
|
std::string generated_checksum, generated_checksum_func_name;
|
|
|
|
IOStatus io_s = GenerateOneFileChecksum(
|
2020-08-13 02:28:10 +02:00
|
|
|
fs_.get(), files_to_ingest_[i].internal_file_path,
|
Ingest SST files with checksum information (#6891)
Summary:
Application can ingest SST files with file checksum information, such that during ingestion, DB is able to check data integrity and identify of the SST file. The PR introduces generate_and_verify_file_checksum to IngestExternalFileOption to control if the ingested checksum information should be verified with the generated checksum.
1. If generate_and_verify_file_checksum options is *FALSE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enables the SST file checksum and the checksum function name matches the checksum function name in DB, we trust the ingested checksum, store it in Manifest. If the checksum function name does not match, we treat that as an error and fail the IngestExternalFile() call.
2. If generate_and_verify_file_checksum options is *TRUE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enable the SST file checksum, we will use the checksum generator from DB to calculate the checksum for each ingested SST files after they are copied or moved. Then, compare the checksum results with the ingested checksum information: _A)_ if the checksum function name does not match, _verification always report true_ and we store the DB generated checksum information in Manifest. _B)_ if the checksum function name mach, and checksum match, ingestion continues and stores the checksum information in the Manifest. Otherwise, terminate file ingestion and report file corruption.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6891
Test Plan: added unit test, pass make asan_check
Reviewed By: pdillinger
Differential Revision: D21935988
Pulled By: zhichao-cao
fbshipit-source-id: 7b55f486632db467e76d72602218d0658aa7f6ed
2020-06-11 23:25:01 +02:00
|
|
|
db_options_.file_checksum_gen_factory.get(), &generated_checksum,
|
|
|
|
&generated_checksum_func_name,
|
|
|
|
ingestion_options_.verify_checksums_readahead_size,
|
|
|
|
db_options_.allow_mmap_reads);
|
|
|
|
if (!io_s.ok()) {
|
|
|
|
status = io_s;
|
|
|
|
ROCKS_LOG_WARN(db_options_.info_log,
|
|
|
|
"Sst file checksum generation of file: %s failed: %s",
|
|
|
|
files_to_ingest_[i].internal_file_path.c_str(),
|
|
|
|
status.ToString().c_str());
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (ingestion_options_.write_global_seqno == false) {
|
|
|
|
files_to_ingest_[i].file_checksum = generated_checksum;
|
|
|
|
files_to_ingest_[i].file_checksum_func_name =
|
|
|
|
generated_checksum_func_name;
|
|
|
|
}
|
|
|
|
generated_checksums.push_back(generated_checksum);
|
|
|
|
generated_checksum_func_names.push_back(generated_checksum_func_name);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
// Step 2: based on the verify_file_checksum and ingested checksum
|
|
|
|
// information, do the verification.
|
|
|
|
if (status.ok()) {
|
|
|
|
if (files_checksums.size() == files_to_ingest_.size() &&
|
|
|
|
files_checksum_func_names.size() == files_to_ingest_.size()) {
|
|
|
|
// Verify the checksum and checksum function name.
|
|
|
|
if (ingestion_options_.verify_file_checksum) {
|
|
|
|
for (size_t i = 0; i < files_to_ingest_.size(); i++) {
|
|
|
|
if (files_checksum_func_names[i] !=
|
|
|
|
generated_checksum_func_names[i]) {
|
|
|
|
status = Status::InvalidArgument(
|
|
|
|
"Checksum function name does not match with the checksum "
|
|
|
|
"function name of this DB");
|
|
|
|
ROCKS_LOG_WARN(
|
|
|
|
db_options_.info_log,
|
|
|
|
"Sst file checksum verification of file: %s failed: %s",
|
|
|
|
external_files_paths[i].c_str(), status.ToString().c_str());
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (files_checksums[i] != generated_checksums[i]) {
|
|
|
|
status = Status::Corruption(
|
|
|
|
"Ingested checksum does not match with the generated "
|
|
|
|
"checksum");
|
|
|
|
ROCKS_LOG_WARN(
|
|
|
|
db_options_.info_log,
|
|
|
|
"Sst file checksum verification of file: %s failed: %s",
|
|
|
|
files_to_ingest_[i].internal_file_path.c_str(),
|
|
|
|
status.ToString().c_str());
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
// If verify_file_checksum is not enabled, we only verify the
|
|
|
|
// checksum function name. If it does not match, fail the ingestion.
|
|
|
|
// If matches, we trust the ingested checksum information and store
|
|
|
|
// in the Manifest.
|
|
|
|
for (size_t i = 0; i < files_to_ingest_.size(); i++) {
|
|
|
|
if (files_checksum_func_names[i] != file_checksum_gen->Name()) {
|
|
|
|
status = Status::InvalidArgument(
|
|
|
|
"Checksum function name does not match with the checksum "
|
|
|
|
"function name of this DB");
|
|
|
|
ROCKS_LOG_WARN(
|
|
|
|
db_options_.info_log,
|
|
|
|
"Sst file checksum verification of file: %s failed: %s",
|
|
|
|
external_files_paths[i].c_str(), status.ToString().c_str());
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
files_to_ingest_[i].file_checksum = files_checksums[i];
|
|
|
|
files_to_ingest_[i].file_checksum_func_name =
|
|
|
|
files_checksum_func_names[i];
|
|
|
|
}
|
|
|
|
}
|
|
|
|
} else if (files_checksums.size() != files_checksum_func_names.size() ||
|
|
|
|
(files_checksums.size() == files_checksum_func_names.size() &&
|
|
|
|
files_checksums.size() != 0)) {
|
|
|
|
// The checksum or checksum function name vector are not both empty
|
|
|
|
// and they are incomplete.
|
|
|
|
status = Status::InvalidArgument(
|
|
|
|
"The checksum information of ingested sst files are nonempty and "
|
|
|
|
"the size of checksums or the size of the checksum function "
|
|
|
|
"names "
|
|
|
|
"does not match with the number of ingested sst files");
|
|
|
|
ROCKS_LOG_WARN(
|
|
|
|
db_options_.info_log,
|
|
|
|
"The ingested sst files checksum information is incomplete: %s",
|
|
|
|
status.ToString().c_str());
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2019-06-21 19:12:29 +02:00
|
|
|
// TODO: The following is duplicated with Cleanup().
|
2016-10-21 02:05:32 +02:00
|
|
|
if (!status.ok()) {
|
|
|
|
// We failed, remove all files that we copied into the db
|
|
|
|
for (IngestedFileInfo& f : files_to_ingest_) {
|
2018-04-13 19:47:54 +02:00
|
|
|
if (f.internal_file_path.empty()) {
|
2019-08-31 03:27:43 +02:00
|
|
|
continue;
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
Status s = env_->DeleteFile(f.internal_file_path);
|
|
|
|
if (!s.ok()) {
|
2017-03-16 03:22:52 +01:00
|
|
|
ROCKS_LOG_WARN(db_options_.info_log,
|
|
|
|
"AddFile() clean up for file %s failed : %s",
|
|
|
|
f.internal_file_path.c_str(), s.ToString().c_str());
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2018-02-28 02:08:34 +01:00
|
|
|
Status ExternalSstFileIngestionJob::NeedsFlush(bool* flush_needed,
|
|
|
|
SuperVersion* super_version) {
|
|
|
|
autovector<Range> ranges;
|
|
|
|
for (const IngestedFileInfo& file_to_ingest : files_to_ingest_) {
|
2019-08-15 05:58:59 +02:00
|
|
|
ranges.emplace_back(file_to_ingest.smallest_internal_key.user_key(),
|
|
|
|
file_to_ingest.largest_internal_key.user_key());
|
2018-02-28 02:08:34 +01:00
|
|
|
}
|
2016-10-21 02:05:32 +02:00
|
|
|
Status status =
|
2018-02-28 02:08:34 +01:00
|
|
|
cfd_->RangesOverlapWithMemtables(ranges, super_version, flush_needed);
|
2016-10-21 02:05:32 +02:00
|
|
|
if (status.ok() && *flush_needed &&
|
|
|
|
!ingestion_options_.allow_blocking_flush) {
|
|
|
|
status = Status::InvalidArgument("External file requires flush");
|
|
|
|
}
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2017-06-24 23:06:43 +02:00
|
|
|
// REQUIRES: we have become the only writer by entering both write_thread_ and
|
|
|
|
// nonmem_write_thread_
|
2016-10-21 02:05:32 +02:00
|
|
|
Status ExternalSstFileIngestionJob::Run() {
|
|
|
|
Status status;
|
2018-02-28 02:08:34 +01:00
|
|
|
SuperVersion* super_version = cfd_->GetSuperVersion();
|
2016-10-21 02:05:32 +02:00
|
|
|
#ifndef NDEBUG
|
|
|
|
// We should never run the job with a memtable that is overlapping
|
|
|
|
// with the files we are ingesting
|
|
|
|
bool need_flush = false;
|
2018-02-28 02:08:34 +01:00
|
|
|
status = NeedsFlush(&need_flush, super_version);
|
2016-10-21 02:05:32 +02:00
|
|
|
assert(status.ok() && need_flush == false);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
bool force_global_seqno = false;
|
2017-04-26 22:28:39 +02:00
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
if (ingestion_options_.snapshot_consistency && !db_snapshots_->empty()) {
|
|
|
|
// We need to assign a global sequence number to all the files even
|
2020-02-28 23:10:51 +01:00
|
|
|
// if the don't overlap with any ranges since we have snapshots
|
2016-10-21 02:05:32 +02:00
|
|
|
force_global_seqno = true;
|
|
|
|
}
|
2017-11-11 02:18:01 +01:00
|
|
|
// It is safe to use this instead of LastAllocatedSequence since we are
|
2017-06-24 23:06:43 +02:00
|
|
|
// the only active writer, and hence they are equal
|
2019-09-13 23:48:18 +02:00
|
|
|
SequenceNumber last_seqno = versions_->LastSequence();
|
2016-10-21 02:05:32 +02:00
|
|
|
edit_.SetColumnFamily(cfd_->GetID());
|
|
|
|
// The levels that the files will be ingested into
|
2017-04-26 22:28:39 +02:00
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
for (IngestedFileInfo& f : files_to_ingest_) {
|
2017-04-26 22:28:39 +02:00
|
|
|
SequenceNumber assigned_seqno = 0;
|
2017-05-17 20:32:26 +02:00
|
|
|
if (ingestion_options_.ingest_behind) {
|
|
|
|
status = CheckLevelForIngestedBehindFile(&f);
|
|
|
|
} else {
|
|
|
|
status = AssignLevelAndSeqnoForIngestedFile(
|
2019-09-13 23:48:18 +02:00
|
|
|
super_version, force_global_seqno, cfd_->ioptions()->compaction_style,
|
|
|
|
last_seqno, &f, &assigned_seqno);
|
2017-05-17 20:32:26 +02:00
|
|
|
}
|
2016-10-21 02:05:32 +02:00
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
2017-04-26 22:28:39 +02:00
|
|
|
status = AssignGlobalSeqnoForIngestedFile(&f, assigned_seqno);
|
|
|
|
TEST_SYNC_POINT_CALLBACK("ExternalSstFileIngestionJob::Run",
|
|
|
|
&assigned_seqno);
|
2019-09-13 23:48:18 +02:00
|
|
|
if (assigned_seqno > last_seqno) {
|
|
|
|
assert(assigned_seqno == last_seqno + 1);
|
|
|
|
last_seqno = assigned_seqno;
|
|
|
|
++consumed_seqno_count_;
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
2019-10-15 00:19:31 +02:00
|
|
|
|
Ingest SST files with checksum information (#6891)
Summary:
Application can ingest SST files with file checksum information, such that during ingestion, DB is able to check data integrity and identify of the SST file. The PR introduces generate_and_verify_file_checksum to IngestExternalFileOption to control if the ingested checksum information should be verified with the generated checksum.
1. If generate_and_verify_file_checksum options is *FALSE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enables the SST file checksum and the checksum function name matches the checksum function name in DB, we trust the ingested checksum, store it in Manifest. If the checksum function name does not match, we treat that as an error and fail the IngestExternalFile() call.
2. If generate_and_verify_file_checksum options is *TRUE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enable the SST file checksum, we will use the checksum generator from DB to calculate the checksum for each ingested SST files after they are copied or moved. Then, compare the checksum results with the ingested checksum information: _A)_ if the checksum function name does not match, _verification always report true_ and we store the DB generated checksum information in Manifest. _B)_ if the checksum function name mach, and checksum match, ingestion continues and stores the checksum information in the Manifest. Otherwise, terminate file ingestion and report file corruption.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6891
Test Plan: added unit test, pass make asan_check
Reviewed By: pdillinger
Differential Revision: D21935988
Pulled By: zhichao-cao
fbshipit-source-id: 7b55f486632db467e76d72602218d0658aa7f6ed
2020-06-11 23:25:01 +02:00
|
|
|
status = GenerateChecksumForIngestedFile(&f);
|
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2019-11-23 01:01:21 +01:00
|
|
|
// We use the import time as the ancester time. This is the time the data
|
|
|
|
// is written to the database.
|
|
|
|
int64_t temp_current_time = 0;
|
2019-11-27 06:38:38 +01:00
|
|
|
uint64_t current_time = kUnknownFileCreationTime;
|
|
|
|
uint64_t oldest_ancester_time = kUnknownOldestAncesterTime;
|
2019-11-23 01:01:21 +01:00
|
|
|
if (env_->GetCurrentTime(&temp_current_time).ok()) {
|
2019-11-27 06:38:38 +01:00
|
|
|
current_time = oldest_ancester_time =
|
|
|
|
static_cast<uint64_t>(temp_current_time);
|
2019-11-23 01:01:21 +01:00
|
|
|
}
|
|
|
|
|
Ingest SST files with checksum information (#6891)
Summary:
Application can ingest SST files with file checksum information, such that during ingestion, DB is able to check data integrity and identify of the SST file. The PR introduces generate_and_verify_file_checksum to IngestExternalFileOption to control if the ingested checksum information should be verified with the generated checksum.
1. If generate_and_verify_file_checksum options is *FALSE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enables the SST file checksum and the checksum function name matches the checksum function name in DB, we trust the ingested checksum, store it in Manifest. If the checksum function name does not match, we treat that as an error and fail the IngestExternalFile() call.
2. If generate_and_verify_file_checksum options is *TRUE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enable the SST file checksum, we will use the checksum generator from DB to calculate the checksum for each ingested SST files after they are copied or moved. Then, compare the checksum results with the ingested checksum information: _A)_ if the checksum function name does not match, _verification always report true_ and we store the DB generated checksum information in Manifest. _B)_ if the checksum function name mach, and checksum match, ingestion continues and stores the checksum information in the Manifest. Otherwise, terminate file ingestion and report file corruption.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6891
Test Plan: added unit test, pass make asan_check
Reviewed By: pdillinger
Differential Revision: D21935988
Pulled By: zhichao-cao
fbshipit-source-id: 7b55f486632db467e76d72602218d0658aa7f6ed
2020-06-11 23:25:01 +02:00
|
|
|
edit_.AddFile(f.picked_level, f.fd.GetNumber(), f.fd.GetPathId(),
|
|
|
|
f.fd.GetFileSize(), f.smallest_internal_key,
|
|
|
|
f.largest_internal_key, f.assigned_seqno, f.assigned_seqno,
|
|
|
|
false, kInvalidBlobFileNumber, oldest_ancester_time,
|
|
|
|
current_time, f.file_checksum, f.file_checksum_func_name);
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
|
|
|
void ExternalSstFileIngestionJob::UpdateStats() {
|
|
|
|
// Update internal stats for new ingested files
|
|
|
|
uint64_t total_keys = 0;
|
|
|
|
uint64_t total_l0_files = 0;
|
|
|
|
uint64_t total_time = env_->NowMicros() - job_start_time_;
|
2019-09-13 23:48:18 +02:00
|
|
|
|
|
|
|
EventLoggerStream stream = event_logger_->Log();
|
|
|
|
stream << "event"
|
|
|
|
<< "ingest_finished";
|
|
|
|
stream << "files_ingested";
|
|
|
|
stream.StartArray();
|
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
for (IngestedFileInfo& f : files_to_ingest_) {
|
2018-04-11 19:47:54 +02:00
|
|
|
InternalStats::CompactionStats stats(CompactionReason::kExternalSstIngestion, 1);
|
2016-10-21 02:05:32 +02:00
|
|
|
stats.micros = total_time;
|
2018-10-05 05:44:43 +02:00
|
|
|
// If actual copy occurred for this file, then we need to count the file
|
2018-04-13 19:47:54 +02:00
|
|
|
// size as the actual bytes written. If the file was linked, then we ignore
|
|
|
|
// the bytes written for file metadata.
|
|
|
|
// TODO (yanqin) maybe account for file metadata bytes for exact accuracy?
|
|
|
|
if (f.copy_file) {
|
|
|
|
stats.bytes_written = f.fd.GetFileSize();
|
|
|
|
} else {
|
|
|
|
stats.bytes_moved = f.fd.GetFileSize();
|
|
|
|
}
|
2016-10-21 02:05:32 +02:00
|
|
|
stats.num_output_files = 1;
|
2019-03-20 01:24:09 +01:00
|
|
|
cfd_->internal_stats()->AddCompactionStats(f.picked_level,
|
|
|
|
Env::Priority::USER, stats);
|
2016-10-21 02:05:32 +02:00
|
|
|
cfd_->internal_stats()->AddCFStats(InternalStats::BYTES_INGESTED_ADD_FILE,
|
|
|
|
f.fd.GetFileSize());
|
|
|
|
total_keys += f.num_entries;
|
|
|
|
if (f.picked_level == 0) {
|
|
|
|
total_l0_files += 1;
|
|
|
|
}
|
2017-03-16 03:22:52 +01:00
|
|
|
ROCKS_LOG_INFO(
|
|
|
|
db_options_.info_log,
|
2016-10-21 02:05:32 +02:00
|
|
|
"[AddFile] External SST file %s was ingested in L%d with path %s "
|
|
|
|
"(global_seqno=%" PRIu64 ")\n",
|
|
|
|
f.external_file_path.c_str(), f.picked_level,
|
|
|
|
f.internal_file_path.c_str(), f.assigned_seqno);
|
2019-09-13 23:48:18 +02:00
|
|
|
stream << "file" << f.internal_file_path << "level" << f.picked_level;
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
2019-09-13 23:48:18 +02:00
|
|
|
stream.EndArray();
|
|
|
|
|
|
|
|
stream << "lsm_state";
|
|
|
|
stream.StartArray();
|
|
|
|
auto vstorage = cfd_->current()->storage_info();
|
|
|
|
for (int level = 0; level < vstorage->num_levels(); ++level) {
|
|
|
|
stream << vstorage->NumLevelFiles(level);
|
|
|
|
}
|
|
|
|
stream.EndArray();
|
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
cfd_->internal_stats()->AddCFStats(InternalStats::INGESTED_NUM_KEYS_TOTAL,
|
|
|
|
total_keys);
|
|
|
|
cfd_->internal_stats()->AddCFStats(InternalStats::INGESTED_NUM_FILES_TOTAL,
|
|
|
|
files_to_ingest_.size());
|
|
|
|
cfd_->internal_stats()->AddCFStats(
|
|
|
|
InternalStats::INGESTED_LEVEL0_NUM_FILES_TOTAL, total_l0_files);
|
|
|
|
}
|
|
|
|
|
|
|
|
void ExternalSstFileIngestionJob::Cleanup(const Status& status) {
|
|
|
|
if (!status.ok()) {
|
|
|
|
// We failed to add the files to the database
|
|
|
|
// remove all the files we copied
|
|
|
|
for (IngestedFileInfo& f : files_to_ingest_) {
|
2019-08-31 03:27:43 +02:00
|
|
|
if (f.internal_file_path.empty()) {
|
|
|
|
continue;
|
|
|
|
}
|
2016-10-21 02:05:32 +02:00
|
|
|
Status s = env_->DeleteFile(f.internal_file_path);
|
|
|
|
if (!s.ok()) {
|
2017-03-16 03:22:52 +01:00
|
|
|
ROCKS_LOG_WARN(db_options_.info_log,
|
|
|
|
"AddFile() clean up for file %s failed : %s",
|
|
|
|
f.internal_file_path.c_str(), s.ToString().c_str());
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
}
|
2019-09-13 23:48:18 +02:00
|
|
|
consumed_seqno_count_ = 0;
|
|
|
|
files_overlap_ = false;
|
2016-10-21 02:05:32 +02:00
|
|
|
} else if (status.ok() && ingestion_options_.move_files) {
|
|
|
|
// The files were moved and added successfully, remove original file links
|
|
|
|
for (IngestedFileInfo& f : files_to_ingest_) {
|
|
|
|
Status s = env_->DeleteFile(f.external_file_path);
|
|
|
|
if (!s.ok()) {
|
2017-03-16 03:22:52 +01:00
|
|
|
ROCKS_LOG_WARN(
|
|
|
|
db_options_.info_log,
|
2016-10-21 02:05:32 +02:00
|
|
|
"%s was added to DB successfully but failed to remove original "
|
|
|
|
"file link : %s",
|
|
|
|
f.external_file_path.c_str(), s.ToString().c_str());
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
Status ExternalSstFileIngestionJob::GetIngestedFileInfo(
|
2018-05-21 23:33:55 +02:00
|
|
|
const std::string& external_file, IngestedFileInfo* file_to_ingest,
|
|
|
|
SuperVersion* sv) {
|
2016-10-21 02:05:32 +02:00
|
|
|
file_to_ingest->external_file_path = external_file;
|
|
|
|
|
|
|
|
// Get external file size
|
Introduce a new storage specific Env API (#5761)
Summary:
The current Env API encompasses both storage/file operations, as well as OS related operations. Most of the APIs return a Status, which does not have enough metadata about an error, such as whether its retry-able or not, scope (i.e fault domain) of the error etc., that may be required in order to properly handle a storage error. The file APIs also do not provide enough control over the IO SLA, such as timeout, prioritization, hinting about placement and redundancy etc.
This PR separates out the file/storage APIs from Env into a new FileSystem class. The APIs are updated to return an IOStatus with metadata about the error, as well as to take an IOOptions structure as input in order to allow more control over the IO.
The user can set both ```options.env``` and ```options.file_system``` to specify that RocksDB should use the former for OS related operations and the latter for storage operations. Internally, a ```CompositeEnvWrapper``` has been introduced that inherits from ```Env``` and redirects individual methods to either an ```Env``` implementation or the ```FileSystem``` as appropriate. When options are sanitized during ```DB::Open```, ```options.env``` is replaced with a newly allocated ```CompositeEnvWrapper``` instance if both env and file_system have been specified. This way, the rest of the RocksDB code can continue to function as before.
This PR also ports PosixEnv to the new API by splitting it into two - PosixEnv and PosixFileSystem. PosixEnv is defined as a sub-class of CompositeEnvWrapper, and threading/time functions are overridden with Posix specific implementations in order to avoid an extra level of indirection.
The ```CompositeEnvWrapper``` translates ```IOStatus``` return code to ```Status```, and sets the severity to ```kSoftError``` if the io_status is retryable. The error handling code in RocksDB can then recover the DB automatically.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/5761
Differential Revision: D18868376
Pulled By: anand1976
fbshipit-source-id: 39efe18a162ea746fabac6360ff529baba48486f
2019-12-13 23:47:08 +01:00
|
|
|
Status status = fs_->GetFileSize(external_file, IOOptions(),
|
|
|
|
&file_to_ingest->file_size, nullptr);
|
2016-10-21 02:05:32 +02:00
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
|
|
|
// Create TableReader for external file
|
|
|
|
std::unique_ptr<TableReader> table_reader;
|
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> sst_file;
|
2016-10-21 02:05:32 +02:00
|
|
|
std::unique_ptr<RandomAccessFileReader> sst_file_reader;
|
|
|
|
|
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
|
|
|
status = fs_->NewRandomAccessFile(external_file, env_options_,
|
|
|
|
&sst_file, nullptr);
|
2016-10-21 02:05:32 +02:00
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
2017-06-29 06:26:03 +02:00
|
|
|
sst_file_reader.reset(new RandomAccessFileReader(std::move(sst_file),
|
|
|
|
external_file));
|
2016-10-21 02:05:32 +02:00
|
|
|
|
|
|
|
status = cfd_->ioptions()->table_factory->NewTableReader(
|
2018-05-21 23:33:55 +02:00
|
|
|
TableReaderOptions(*cfd_->ioptions(),
|
|
|
|
sv->mutable_cf_options.prefix_extractor.get(),
|
|
|
|
env_options_, cfd_->internal_comparator()),
|
2016-10-21 02:05:32 +02:00
|
|
|
std::move(sst_file_reader), file_to_ingest->file_size, &table_reader);
|
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2019-01-30 01:16:53 +01:00
|
|
|
if (ingestion_options_.verify_checksums_before_ingest) {
|
2019-08-17 01:40:09 +02:00
|
|
|
// If customized readahead size is needed, we can pass a user option
|
|
|
|
// all the way to here. Right now we just rely on the default readahead
|
|
|
|
// to keep things simple.
|
2019-08-20 19:40:59 +02:00
|
|
|
ReadOptions ro;
|
|
|
|
ro.readahead_size = ingestion_options_.verify_checksums_readahead_size;
|
2019-08-17 01:40:09 +02:00
|
|
|
status = table_reader->VerifyChecksum(
|
2019-08-20 19:40:59 +02:00
|
|
|
ro, TableReaderCaller::kExternalSSTIngestion);
|
2019-01-30 01:16:53 +01:00
|
|
|
}
|
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
// Get the external file properties
|
|
|
|
auto props = table_reader->GetTableProperties();
|
|
|
|
const auto& uprops = props->user_collected_properties;
|
|
|
|
|
|
|
|
// Get table version
|
|
|
|
auto version_iter = uprops.find(ExternalSstFilePropertyNames::kVersion);
|
|
|
|
if (version_iter == uprops.end()) {
|
|
|
|
return Status::Corruption("External file version not found");
|
|
|
|
}
|
|
|
|
file_to_ingest->version = DecodeFixed32(version_iter->second.c_str());
|
|
|
|
|
|
|
|
auto seqno_iter = uprops.find(ExternalSstFilePropertyNames::kGlobalSeqno);
|
|
|
|
if (file_to_ingest->version == 2) {
|
|
|
|
// version 2 imply that we have global sequence number
|
|
|
|
if (seqno_iter == uprops.end()) {
|
|
|
|
return Status::Corruption(
|
|
|
|
"External file global sequence number not found");
|
|
|
|
}
|
|
|
|
|
|
|
|
// Set the global sequence number
|
|
|
|
file_to_ingest->original_seqno = DecodeFixed64(seqno_iter->second.c_str());
|
2018-07-07 01:07:29 +02:00
|
|
|
auto offsets_iter = props->properties_offsets.find(
|
2016-10-21 02:05:32 +02:00
|
|
|
ExternalSstFilePropertyNames::kGlobalSeqno);
|
2018-07-07 01:07:29 +02:00
|
|
|
if (offsets_iter == props->properties_offsets.end() ||
|
|
|
|
offsets_iter->second == 0) {
|
|
|
|
file_to_ingest->global_seqno_offset = 0;
|
2016-10-21 02:05:32 +02:00
|
|
|
return Status::Corruption("Was not able to find file global seqno field");
|
|
|
|
}
|
2018-09-06 03:07:53 +02:00
|
|
|
file_to_ingest->global_seqno_offset = static_cast<size_t>(offsets_iter->second);
|
2017-01-20 20:53:33 +01:00
|
|
|
} else if (file_to_ingest->version == 1) {
|
|
|
|
// SST file V1 should not have global seqno field
|
|
|
|
assert(seqno_iter == uprops.end());
|
2017-02-17 02:02:51 +01:00
|
|
|
file_to_ingest->original_seqno = 0;
|
2017-01-20 20:53:33 +01:00
|
|
|
if (ingestion_options_.allow_blocking_flush ||
|
|
|
|
ingestion_options_.allow_global_seqno) {
|
|
|
|
return Status::InvalidArgument(
|
|
|
|
"External SST file V1 does not support global seqno");
|
|
|
|
}
|
2016-10-21 02:05:32 +02:00
|
|
|
} else {
|
2017-01-20 20:53:33 +01:00
|
|
|
return Status::InvalidArgument("External file version is not supported");
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
// Get number of entries in table
|
|
|
|
file_to_ingest->num_entries = props->num_entries;
|
2018-07-14 07:40:23 +02:00
|
|
|
file_to_ingest->num_range_deletions = props->num_range_deletions;
|
2016-10-21 02:05:32 +02:00
|
|
|
|
|
|
|
ParsedInternalKey key;
|
2016-12-08 22:30:09 +01:00
|
|
|
ReadOptions ro;
|
|
|
|
// During reading the external file we can cache blocks that we read into
|
|
|
|
// the block cache, if we later change the global seqno of this file, we will
|
|
|
|
// have block in cache that will include keys with wrong seqno.
|
|
|
|
// We need to disable fill_cache so that we read from the file without
|
|
|
|
// updating the block cache.
|
|
|
|
ro.fill_cache = false;
|
2018-05-21 23:33:55 +02:00
|
|
|
std::unique_ptr<InternalIterator> iter(table_reader->NewIterator(
|
2019-06-20 23:28:22 +02:00
|
|
|
ro, sv->mutable_cf_options.prefix_extractor.get(), /*arena=*/nullptr,
|
|
|
|
/*skip_filters=*/false, TableReaderCaller::kExternalSSTIngestion));
|
2018-07-14 07:40:23 +02:00
|
|
|
std::unique_ptr<InternalIterator> range_del_iter(
|
|
|
|
table_reader->NewRangeTombstoneIterator(ro));
|
2016-10-21 02:05:32 +02:00
|
|
|
|
2018-07-14 07:40:23 +02:00
|
|
|
// Get first (smallest) and last (largest) key from file.
|
2019-09-20 21:00:55 +02:00
|
|
|
file_to_ingest->smallest_internal_key =
|
|
|
|
InternalKey("", 0, ValueType::kTypeValue);
|
|
|
|
file_to_ingest->largest_internal_key =
|
|
|
|
InternalKey("", 0, ValueType::kTypeValue);
|
2018-07-14 07:40:23 +02:00
|
|
|
bool bounds_set = false;
|
2016-10-21 02:05:32 +02:00
|
|
|
iter->SeekToFirst();
|
2018-07-14 07:40:23 +02:00
|
|
|
if (iter->Valid()) {
|
|
|
|
if (!ParseInternalKey(iter->key(), &key)) {
|
|
|
|
return Status::Corruption("external file have corrupted keys");
|
|
|
|
}
|
|
|
|
if (key.sequence != 0) {
|
|
|
|
return Status::Corruption("external file have non zero sequence number");
|
|
|
|
}
|
2019-08-15 05:58:59 +02:00
|
|
|
file_to_ingest->smallest_internal_key.SetFrom(key);
|
2018-07-14 07:40:23 +02:00
|
|
|
|
|
|
|
iter->SeekToLast();
|
|
|
|
if (!ParseInternalKey(iter->key(), &key)) {
|
|
|
|
return Status::Corruption("external file have corrupted keys");
|
|
|
|
}
|
|
|
|
if (key.sequence != 0) {
|
|
|
|
return Status::Corruption("external file have non zero sequence number");
|
|
|
|
}
|
2019-08-15 05:58:59 +02:00
|
|
|
file_to_ingest->largest_internal_key.SetFrom(key);
|
2016-10-21 02:05:32 +02:00
|
|
|
|
2018-07-14 07:40:23 +02:00
|
|
|
bounds_set = true;
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
2018-07-14 07:40:23 +02:00
|
|
|
|
|
|
|
// We may need to adjust these key bounds, depending on whether any range
|
|
|
|
// deletion tombstones extend past them.
|
|
|
|
const Comparator* ucmp = cfd_->internal_comparator().user_comparator();
|
|
|
|
if (range_del_iter != nullptr) {
|
|
|
|
for (range_del_iter->SeekToFirst(); range_del_iter->Valid();
|
|
|
|
range_del_iter->Next()) {
|
|
|
|
if (!ParseInternalKey(range_del_iter->key(), &key)) {
|
|
|
|
return Status::Corruption("external file have corrupted keys");
|
|
|
|
}
|
|
|
|
RangeTombstone tombstone(key, range_del_iter->value());
|
|
|
|
|
2019-08-15 05:58:59 +02:00
|
|
|
InternalKey start_key = tombstone.SerializeKey();
|
2019-09-20 21:00:55 +02:00
|
|
|
if (!bounds_set ||
|
|
|
|
sstableKeyCompare(ucmp, start_key,
|
|
|
|
file_to_ingest->smallest_internal_key) < 0) {
|
2019-08-15 05:58:59 +02:00
|
|
|
file_to_ingest->smallest_internal_key = start_key;
|
2018-07-14 07:40:23 +02:00
|
|
|
}
|
2019-08-15 05:58:59 +02:00
|
|
|
InternalKey end_key = tombstone.SerializeEndKey();
|
2019-09-20 21:00:55 +02:00
|
|
|
if (!bounds_set ||
|
|
|
|
sstableKeyCompare(ucmp, end_key,
|
|
|
|
file_to_ingest->largest_internal_key) > 0) {
|
2019-08-15 05:58:59 +02:00
|
|
|
file_to_ingest->largest_internal_key = end_key;
|
2018-07-14 07:40:23 +02:00
|
|
|
}
|
|
|
|
bounds_set = true;
|
|
|
|
}
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
|
2016-12-05 23:16:23 +01:00
|
|
|
file_to_ingest->cf_id = static_cast<uint32_t>(props->column_family_id);
|
|
|
|
|
2016-12-06 22:56:17 +01:00
|
|
|
file_to_ingest->table_properties = *props;
|
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2017-04-26 22:28:39 +02:00
|
|
|
Status ExternalSstFileIngestionJob::AssignLevelAndSeqnoForIngestedFile(
|
|
|
|
SuperVersion* sv, bool force_global_seqno, CompactionStyle compaction_style,
|
2019-09-13 23:48:18 +02:00
|
|
|
SequenceNumber last_seqno, IngestedFileInfo* file_to_ingest,
|
|
|
|
SequenceNumber* assigned_seqno) {
|
2017-04-26 22:28:39 +02:00
|
|
|
Status status;
|
|
|
|
*assigned_seqno = 0;
|
|
|
|
if (force_global_seqno) {
|
|
|
|
*assigned_seqno = last_seqno + 1;
|
2019-09-13 23:48:18 +02:00
|
|
|
if (compaction_style == kCompactionStyleUniversal || files_overlap_) {
|
2017-04-26 22:28:39 +02:00
|
|
|
file_to_ingest->picked_level = 0;
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
}
|
2016-10-21 02:05:32 +02:00
|
|
|
|
2017-04-26 22:28:39 +02:00
|
|
|
bool overlap_with_db = false;
|
2016-10-21 02:05:32 +02:00
|
|
|
Arena arena;
|
|
|
|
ReadOptions ro;
|
|
|
|
ro.total_order_seek = true;
|
|
|
|
int target_level = 0;
|
|
|
|
auto* vstorage = cfd_->current()->storage_info();
|
2017-05-17 20:32:26 +02:00
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
for (int lvl = 0; lvl < cfd_->NumberLevels(); lvl++) {
|
|
|
|
if (lvl > 0 && lvl < vstorage->base_level()) {
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (vstorage->NumLevelFiles(lvl) > 0) {
|
|
|
|
bool overlap_with_level = false;
|
2019-09-20 21:00:55 +02:00
|
|
|
status = sv->current->OverlapWithLevelIterator(
|
|
|
|
ro, env_options_, file_to_ingest->smallest_internal_key.user_key(),
|
|
|
|
file_to_ingest->largest_internal_key.user_key(), lvl,
|
|
|
|
&overlap_with_level);
|
2016-10-21 02:05:32 +02:00
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
if (overlap_with_level) {
|
|
|
|
// We must use L0 or any level higher than `lvl` to be able to overwrite
|
|
|
|
// the keys that we overlap with in this level, We also need to assign
|
|
|
|
// this file a seqno to overwrite the existing keys in level `lvl`
|
2017-04-26 22:28:39 +02:00
|
|
|
overlap_with_db = true;
|
2016-10-21 02:05:32 +02:00
|
|
|
break;
|
|
|
|
}
|
2017-04-26 22:28:39 +02:00
|
|
|
|
|
|
|
if (compaction_style == kCompactionStyleUniversal && lvl != 0) {
|
|
|
|
const std::vector<FileMetaData*>& level_files =
|
|
|
|
vstorage->LevelFiles(lvl);
|
|
|
|
const SequenceNumber level_largest_seqno =
|
|
|
|
(*max_element(level_files.begin(), level_files.end(),
|
|
|
|
[](FileMetaData* f1, FileMetaData* f2) {
|
2018-07-28 01:00:26 +02:00
|
|
|
return f1->fd.largest_seqno < f2->fd.largest_seqno;
|
2017-04-26 22:28:39 +02:00
|
|
|
}))
|
2018-07-28 01:00:26 +02:00
|
|
|
->fd.largest_seqno;
|
2018-02-15 23:11:08 +01:00
|
|
|
// should only assign seqno to current level's largest seqno when
|
|
|
|
// the file fits
|
|
|
|
if (level_largest_seqno != 0 &&
|
|
|
|
IngestedFileFitInLevel(file_to_ingest, lvl)) {
|
2017-04-26 22:28:39 +02:00
|
|
|
*assigned_seqno = level_largest_seqno;
|
|
|
|
} else {
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
} else if (compaction_style == kCompactionStyleUniversal) {
|
|
|
|
continue;
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
|
2020-02-28 23:10:51 +01:00
|
|
|
// We don't overlap with any keys in this level, but we still need to check
|
2016-10-21 02:05:32 +02:00
|
|
|
// if our file can fit in it
|
|
|
|
if (IngestedFileFitInLevel(file_to_ingest, lvl)) {
|
|
|
|
target_level = lvl;
|
|
|
|
}
|
|
|
|
}
|
2019-09-13 23:48:18 +02:00
|
|
|
// If files overlap, we have to ingest them at level 0 and assign the newest
|
|
|
|
// sequence number
|
|
|
|
if (files_overlap_) {
|
|
|
|
target_level = 0;
|
|
|
|
*assigned_seqno = last_seqno + 1;
|
|
|
|
}
|
2017-04-26 22:28:39 +02:00
|
|
|
TEST_SYNC_POINT_CALLBACK(
|
|
|
|
"ExternalSstFileIngestionJob::AssignLevelAndSeqnoForIngestedFile",
|
|
|
|
&overlap_with_db);
|
2016-10-21 02:05:32 +02:00
|
|
|
file_to_ingest->picked_level = target_level;
|
2017-04-26 22:28:39 +02:00
|
|
|
if (overlap_with_db && *assigned_seqno == 0) {
|
|
|
|
*assigned_seqno = last_seqno + 1;
|
|
|
|
}
|
2016-10-21 02:05:32 +02:00
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2017-05-17 20:32:26 +02:00
|
|
|
Status ExternalSstFileIngestionJob::CheckLevelForIngestedBehindFile(
|
|
|
|
IngestedFileInfo* file_to_ingest) {
|
|
|
|
auto* vstorage = cfd_->current()->storage_info();
|
|
|
|
// first check if new files fit in the bottommost level
|
|
|
|
int bottom_lvl = cfd_->NumberLevels() - 1;
|
|
|
|
if(!IngestedFileFitInLevel(file_to_ingest, bottom_lvl)) {
|
|
|
|
return Status::InvalidArgument(
|
|
|
|
"Can't ingest_behind file as it doesn't fit "
|
|
|
|
"at the bottommost level!");
|
|
|
|
}
|
|
|
|
|
|
|
|
// second check if despite allow_ingest_behind=true we still have 0 seqnums
|
|
|
|
// at some upper level
|
|
|
|
for (int lvl = 0; lvl < cfd_->NumberLevels() - 1; lvl++) {
|
|
|
|
for (auto file : vstorage->LevelFiles(lvl)) {
|
2018-07-28 01:00:26 +02:00
|
|
|
if (file->fd.smallest_seqno == 0) {
|
2017-05-17 20:32:26 +02:00
|
|
|
return Status::InvalidArgument(
|
|
|
|
"Can't ingest_behind file as despite allow_ingest_behind=true "
|
|
|
|
"there are files with 0 seqno in database at upper levels!");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
file_to_ingest->picked_level = bottom_lvl;
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
Status ExternalSstFileIngestionJob::AssignGlobalSeqnoForIngestedFile(
|
|
|
|
IngestedFileInfo* file_to_ingest, SequenceNumber seqno) {
|
|
|
|
if (file_to_ingest->original_seqno == seqno) {
|
|
|
|
// This file already have the correct global seqno
|
|
|
|
return Status::OK();
|
|
|
|
} else if (!ingestion_options_.allow_global_seqno) {
|
|
|
|
return Status::InvalidArgument("Global seqno is required, but disabled");
|
|
|
|
} else if (file_to_ingest->global_seqno_offset == 0) {
|
|
|
|
return Status::InvalidArgument(
|
2020-02-28 23:10:51 +01:00
|
|
|
"Trying to set global seqno for a file that don't have a global seqno "
|
2016-10-21 02:05:32 +02:00
|
|
|
"field");
|
|
|
|
}
|
|
|
|
|
2018-07-28 01:00:26 +02:00
|
|
|
if (ingestion_options_.write_global_seqno) {
|
|
|
|
// Determine if we can write global_seqno to a given offset of file.
|
|
|
|
// If the file system does not support random write, then we should not.
|
|
|
|
// Otherwise we should.
|
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<FSRandomRWFile> rwfile;
|
|
|
|
Status status =
|
|
|
|
fs_->NewRandomRWFile(file_to_ingest->internal_file_path, env_options_,
|
|
|
|
&rwfile, nullptr);
|
2018-07-28 01:00:26 +02:00
|
|
|
if (status.ok()) {
|
|
|
|
std::string seqno_val;
|
|
|
|
PutFixed64(&seqno_val, seqno);
|
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
|
|
|
status = rwfile->Write(file_to_ingest->global_seqno_offset, seqno_val,
|
|
|
|
IOOptions(), nullptr);
|
2019-06-21 19:12:29 +02:00
|
|
|
if (status.ok()) {
|
|
|
|
TEST_SYNC_POINT("ExternalSstFileIngestionJob::BeforeSyncGlobalSeqno");
|
|
|
|
status = SyncIngestedFile(rwfile.get());
|
|
|
|
TEST_SYNC_POINT("ExternalSstFileIngestionJob::AfterSyncGlobalSeqno");
|
|
|
|
if (!status.ok()) {
|
|
|
|
ROCKS_LOG_WARN(db_options_.info_log,
|
|
|
|
"Failed to sync ingested file %s after writing global "
|
|
|
|
"sequence number: %s",
|
|
|
|
file_to_ingest->internal_file_path.c_str(),
|
|
|
|
status.ToString().c_str());
|
|
|
|
}
|
|
|
|
}
|
2018-07-28 01:00:26 +02:00
|
|
|
if (!status.ok()) {
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
} else if (!status.IsNotSupported()) {
|
|
|
|
return status;
|
|
|
|
}
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
|
2018-07-28 01:00:26 +02:00
|
|
|
file_to_ingest->assigned_seqno = seqno;
|
|
|
|
return Status::OK();
|
2016-10-21 02:05:32 +02:00
|
|
|
}
|
|
|
|
|
Ingest SST files with checksum information (#6891)
Summary:
Application can ingest SST files with file checksum information, such that during ingestion, DB is able to check data integrity and identify of the SST file. The PR introduces generate_and_verify_file_checksum to IngestExternalFileOption to control if the ingested checksum information should be verified with the generated checksum.
1. If generate_and_verify_file_checksum options is *FALSE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enables the SST file checksum and the checksum function name matches the checksum function name in DB, we trust the ingested checksum, store it in Manifest. If the checksum function name does not match, we treat that as an error and fail the IngestExternalFile() call.
2. If generate_and_verify_file_checksum options is *TRUE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enable the SST file checksum, we will use the checksum generator from DB to calculate the checksum for each ingested SST files after they are copied or moved. Then, compare the checksum results with the ingested checksum information: _A)_ if the checksum function name does not match, _verification always report true_ and we store the DB generated checksum information in Manifest. _B)_ if the checksum function name mach, and checksum match, ingestion continues and stores the checksum information in the Manifest. Otherwise, terminate file ingestion and report file corruption.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6891
Test Plan: added unit test, pass make asan_check
Reviewed By: pdillinger
Differential Revision: D21935988
Pulled By: zhichao-cao
fbshipit-source-id: 7b55f486632db467e76d72602218d0658aa7f6ed
2020-06-11 23:25:01 +02:00
|
|
|
IOStatus ExternalSstFileIngestionJob::GenerateChecksumForIngestedFile(
|
|
|
|
IngestedFileInfo* file_to_ingest) {
|
|
|
|
if (db_options_.file_checksum_gen_factory == nullptr ||
|
|
|
|
need_generate_file_checksum_ == false ||
|
|
|
|
ingestion_options_.write_global_seqno == false) {
|
|
|
|
// If file_checksum_gen_factory is not set, we are not able to generate
|
|
|
|
// the checksum. if write_global_seqno is false, it means we will use
|
|
|
|
// file checksum generated during Prepare(). This step will be skipped.
|
|
|
|
return IOStatus::OK();
|
|
|
|
}
|
|
|
|
std::string file_checksum, file_checksum_func_name;
|
|
|
|
IOStatus io_s = GenerateOneFileChecksum(
|
2020-08-13 02:28:10 +02:00
|
|
|
fs_.get(), file_to_ingest->internal_file_path,
|
Ingest SST files with checksum information (#6891)
Summary:
Application can ingest SST files with file checksum information, such that during ingestion, DB is able to check data integrity and identify of the SST file. The PR introduces generate_and_verify_file_checksum to IngestExternalFileOption to control if the ingested checksum information should be verified with the generated checksum.
1. If generate_and_verify_file_checksum options is *FALSE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enables the SST file checksum and the checksum function name matches the checksum function name in DB, we trust the ingested checksum, store it in Manifest. If the checksum function name does not match, we treat that as an error and fail the IngestExternalFile() call.
2. If generate_and_verify_file_checksum options is *TRUE*: *1)* if DB does not enable SST file checksum, the checksum information ingested will be ignored; *2)* if DB enable the SST file checksum, we will use the checksum generator from DB to calculate the checksum for each ingested SST files after they are copied or moved. Then, compare the checksum results with the ingested checksum information: _A)_ if the checksum function name does not match, _verification always report true_ and we store the DB generated checksum information in Manifest. _B)_ if the checksum function name mach, and checksum match, ingestion continues and stores the checksum information in the Manifest. Otherwise, terminate file ingestion and report file corruption.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/6891
Test Plan: added unit test, pass make asan_check
Reviewed By: pdillinger
Differential Revision: D21935988
Pulled By: zhichao-cao
fbshipit-source-id: 7b55f486632db467e76d72602218d0658aa7f6ed
2020-06-11 23:25:01 +02:00
|
|
|
db_options_.file_checksum_gen_factory.get(), &file_checksum,
|
|
|
|
&file_checksum_func_name,
|
|
|
|
ingestion_options_.verify_checksums_readahead_size,
|
|
|
|
db_options_.allow_mmap_reads);
|
|
|
|
if (!io_s.ok()) {
|
|
|
|
return io_s;
|
|
|
|
}
|
|
|
|
file_to_ingest->file_checksum = file_checksum;
|
|
|
|
file_to_ingest->file_checksum_func_name = file_checksum_func_name;
|
|
|
|
return IOStatus::OK();
|
|
|
|
}
|
|
|
|
|
2016-10-21 02:05:32 +02:00
|
|
|
bool ExternalSstFileIngestionJob::IngestedFileFitInLevel(
|
|
|
|
const IngestedFileInfo* file_to_ingest, int level) {
|
|
|
|
if (level == 0) {
|
|
|
|
// Files can always fit in L0
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
auto* vstorage = cfd_->current()->storage_info();
|
2019-09-20 21:00:55 +02:00
|
|
|
Slice file_smallest_user_key(
|
|
|
|
file_to_ingest->smallest_internal_key.user_key());
|
2019-08-15 05:58:59 +02:00
|
|
|
Slice file_largest_user_key(file_to_ingest->largest_internal_key.user_key());
|
2016-10-21 02:05:32 +02:00
|
|
|
|
|
|
|
if (vstorage->OverlapInLevel(level, &file_smallest_user_key,
|
|
|
|
&file_largest_user_key)) {
|
|
|
|
// File overlap with another files in this level, we cannot
|
|
|
|
// add it to this level
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
if (cfd_->RangeOverlapWithCompaction(file_smallest_user_key,
|
|
|
|
file_largest_user_key, level)) {
|
|
|
|
// File overlap with a running compaction output that will be stored
|
|
|
|
// in this level, we cannot add this file to this level
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
// File did not overlap with level files, our compaction output
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2019-06-21 19:12:29 +02:00
|
|
|
template <typename TWritableFile>
|
|
|
|
Status ExternalSstFileIngestionJob::SyncIngestedFile(TWritableFile* file) {
|
|
|
|
assert(file != nullptr);
|
|
|
|
if (db_options_.use_fsync) {
|
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
|
|
|
return file->Fsync(IOOptions(), nullptr);
|
2019-06-21 19:12:29 +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
|
|
|
return file->Sync(IOOptions(), nullptr);
|
2019-06-21 19:12:29 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-02-20 21:07:53 +01:00
|
|
|
} // namespace ROCKSDB_NAMESPACE
|
2017-02-28 20:05:08 +01:00
|
|
|
|
|
|
|
#endif // !ROCKSDB_LITE
|