mirror of
https://github.com/scylladb/scylladb.git
synced 2026-04-21 09:00:35 +00:00
76cde84540
ignore_partial_runs() brings confusion because i__p__r() equal to true doesn't mean filter out partial runs from compaction. It actually means not caring about compaction of a partial run. The logic was wrong because any compaction strategy that chooses not to ignore partial sstable run[1] would have any fragment composing it incorrectly becoming a candidate for compaction. This problem could make compaction include only a subset of fragments composing the partial run or even make the same fragment be compacted twice due to parallel compaction. [1]: partial sstable run is a sstable that is still being generated by compaction and as a result cannot be selected as candidate whatsoever. Fix is about making sure partial sstable run has none of its fragments selected for compaction. And also renaming i__p__r. Fixes #4729. Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com> Message-Id: <20190807022814.12567-1-raphaelsc@scylladb.com>
104 lines
4.3 KiB
C++
104 lines
4.3 KiB
C++
/*
|
|
* Copyright (C) 2017 ScyllaDB
|
|
*/
|
|
|
|
/*
|
|
* This file is part of Scylla.
|
|
*
|
|
* Scylla is free software: you can redistribute it and/or modify
|
|
* it under the terms of the GNU Affero General Public License as published by
|
|
* the Free Software Foundation, either version 3 of the License, or
|
|
* (at your option) any later version.
|
|
*
|
|
* Scylla is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
* GNU General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU General Public License
|
|
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
|
|
*/
|
|
|
|
#pragma once
|
|
|
|
#include "cql3/statements/property_definitions.hh"
|
|
#include "compaction_backlog_manager.hh"
|
|
#include "compaction_strategy.hh"
|
|
#include "database_fwd.hh"
|
|
#include "db_clock.hh"
|
|
|
|
namespace sstables {
|
|
|
|
compaction_backlog_tracker& get_unimplemented_backlog_tracker();
|
|
|
|
class sstable_set_impl;
|
|
class compaction_descriptor;
|
|
class resharding_descriptor;
|
|
|
|
class compaction_strategy_impl {
|
|
static constexpr float DEFAULT_TOMBSTONE_THRESHOLD = 0.2f;
|
|
// minimum interval needed to perform tombstone removal compaction in seconds, default 86400 or 1 day.
|
|
static constexpr std::chrono::seconds DEFAULT_TOMBSTONE_COMPACTION_INTERVAL() { return std::chrono::seconds(86400); }
|
|
protected:
|
|
const sstring TOMBSTONE_THRESHOLD_OPTION = "tombstone_threshold";
|
|
const sstring TOMBSTONE_COMPACTION_INTERVAL_OPTION = "tombstone_compaction_interval";
|
|
|
|
bool _use_clustering_key_filter = false;
|
|
bool _disable_tombstone_compaction = false;
|
|
float _tombstone_threshold = DEFAULT_TOMBSTONE_THRESHOLD;
|
|
db_clock::duration _tombstone_compaction_interval = DEFAULT_TOMBSTONE_COMPACTION_INTERVAL();
|
|
public:
|
|
static std::optional<sstring> get_value(const std::map<sstring, sstring>& options, const sstring& name) {
|
|
auto it = options.find(name);
|
|
if (it == options.end()) {
|
|
return std::nullopt;
|
|
}
|
|
return it->second;
|
|
}
|
|
protected:
|
|
compaction_strategy_impl() = default;
|
|
explicit compaction_strategy_impl(const std::map<sstring, sstring>& options) {
|
|
using namespace cql3::statements;
|
|
|
|
auto tmp_value = get_value(options, TOMBSTONE_THRESHOLD_OPTION);
|
|
_tombstone_threshold = property_definitions::to_double(TOMBSTONE_THRESHOLD_OPTION, tmp_value, DEFAULT_TOMBSTONE_THRESHOLD);
|
|
|
|
tmp_value = get_value(options, TOMBSTONE_COMPACTION_INTERVAL_OPTION);
|
|
auto interval = property_definitions::to_long(TOMBSTONE_COMPACTION_INTERVAL_OPTION, tmp_value, DEFAULT_TOMBSTONE_COMPACTION_INTERVAL().count());
|
|
_tombstone_compaction_interval = db_clock::duration(std::chrono::seconds(interval));
|
|
|
|
// FIXME: validate options.
|
|
}
|
|
public:
|
|
virtual ~compaction_strategy_impl() {}
|
|
virtual compaction_descriptor get_sstables_for_compaction(column_family& cfs, std::vector<sstables::shared_sstable> candidates) = 0;
|
|
virtual compaction_descriptor get_major_compaction_job(column_family& cf, std::vector<sstables::shared_sstable> candidates);
|
|
virtual std::vector<resharding_descriptor> get_resharding_jobs(column_family& cf, std::vector<sstables::shared_sstable> candidates);
|
|
virtual void notify_completion(const std::vector<shared_sstable>& removed, const std::vector<shared_sstable>& added) { }
|
|
virtual compaction_strategy_type type() const = 0;
|
|
virtual bool parallel_compaction() const {
|
|
return true;
|
|
}
|
|
virtual int64_t estimated_pending_compactions(column_family& cf) const = 0;
|
|
virtual std::unique_ptr<sstable_set_impl> make_sstable_set(schema_ptr schema) const;
|
|
|
|
bool use_clustering_key_filter() const {
|
|
return _use_clustering_key_filter;
|
|
}
|
|
|
|
virtual bool can_compact_partial_runs() const {
|
|
return false;
|
|
}
|
|
|
|
// Check if a given sstable is entitled for tombstone compaction based on its
|
|
// droppable tombstone histogram and gc_before.
|
|
bool worth_dropping_tombstones(const shared_sstable& sst, gc_clock::time_point gc_before);
|
|
|
|
virtual compaction_backlog_tracker& get_backlog_tracker() = 0;
|
|
|
|
virtual uint64_t adjust_partition_estimate(const mutation_source_metadata& ms_meta, uint64_t partition_estimate);
|
|
|
|
virtual reader_consumer make_interposer_consumer(const mutation_source_metadata& ms_meta, reader_consumer end_consumer);
|
|
};
|
|
}
|