Skip to content

branch-4.1: [fix](filecache) skip redundant ttl scans for non-TTL tablets (pick #65434) #65429

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
yiguolei merged 1 commit into from
Jul 11, 2026
+131 −8
Merged

branch-4.1: [fix](filecache) skip redundant ttl scans for non-TTL tablets (pick #65434) #65429

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 10 additions & 3 deletions be/src/io/cache/block_file_cache_ttl_mgr.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,7 @@
#include "common/config.h"
#include "common/logging.h"
#include "common/status.h"
#include "cpp/sync_point.h"
#include "io/cache/block_file_cache.h"
#include "io/cache/cache_block_meta_store.h"
#include "io/cache/file_block.h"
Expand Down Expand Up @@ -134,6 +135,7 @@ void BlockFileCacheTtlMgr::run_background_tablet_id_flush() {

FileBlocks BlockFileCacheTtlMgr::get_file_blocks_from_tablet_id(int64_t tablet_id) {
FileBlocks result;
TEST_SYNC_POINT_CALLBACK("BlockFileCacheTtlMgr::get_file_blocks_from_tablet_id", tablet_id);

// Use meta store to get all blocks for this tablet
auto iterator = _meta_store->range_get(tablet_id);
Expand Down Expand Up @@ -169,8 +171,12 @@ FileBlocks BlockFileCacheTtlMgr::get_file_blocks_from_tablet_id(int64_t tablet_i
void BlockFileCacheTtlMgr::run_backgroud_update_ttl_info_map() {
Thread::set_self_name("ttl_mgr_update");

static constexpr uint64_t kFullReconcileIntervalRounds = 20;
uint64_t update_round = 0;

while (!_stop_background.load(std::memory_order_acquire)) {
try {
const bool need_full_reconcile = (++update_round % kFullReconcileIntervalRounds) == 0;
std::unordered_set<int64_t> tablet_ids_to_process;
{
std::lock_guard<std::mutex> lock(_tablet_id_mutex);
Expand Down Expand Up @@ -238,9 +244,10 @@ void BlockFileCacheTtlMgr::run_backgroud_update_ttl_info_map() {
}
}
} else {
// Remove from TTL map if TTL is 0
_ttl_info_map.erase(tablet_id);
need_convert_from_ttl = true;
// Periodically reconcile blocks restored from persisted TTL metadata,
// because _ttl_info_map is rebuilt only in memory after restart.
need_convert_from_ttl =
_ttl_info_map.erase(tablet_id) > 0 || need_full_reconcile;
Comment thread
freemandealer marked this conversation as resolved.
}
}

Expand Down
126 changes: 121 additions & 5 deletions be/test/io/cache/block_file_cache_ttl_mgr_test.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@
#include <gtest/gtest.h>

#include <algorithm>
#include <atomic>
#include <chrono>
#include <filesystem>
#include <memory>
Expand All @@ -29,6 +30,7 @@

#include "common/config.h"
#include "common/status.h"
#include "cpp/sync_point.h"
#include "io/cache/block_file_cache.h"
#include "io/cache/cache_block_meta_store.h"
#include "io/cache/file_block.h"
Expand Down Expand Up @@ -132,6 +134,7 @@ class FakeStorageEngine : public BaseStorageEngine {

Status get_tablet_meta(int64_t tablet_id, TabletMetaSharedPtr* tablet_meta,
bool /*force_use_only_cached*/ = false) override {
_get_tablet_meta_call_count.fetch_add(1, std::memory_order_relaxed);
auto tablet_res = get_tablet(tablet_id);
if (!tablet_res.has_value()) {
return tablet_res.error();
Expand All @@ -149,9 +152,14 @@ class FakeStorageEngine : public BaseStorageEngine {
_tablets[tablet_id] = tablet;
}

int64_t get_tablet_meta_call_count() const {
return _get_tablet_meta_call_count.load(std::memory_order_relaxed);
}

private:
std::mutex _mutex;
std::unordered_map<int64_t, BaseTabletSPtr> _tablets;
std::atomic<int64_t> _get_tablet_meta_call_count {0};
};

std::vector<FileBlockSPtr> blocks_from_holder(const FileBlocksHolder& holder) {
Expand Down Expand Up @@ -237,7 +245,9 @@ class BlockFileCacheTtlMgrTest : public testing::Test {
}

FileBlockSPtr create_block(int64_t tablet_id, const std::string& cache_key, size_t offset,
size_t size, UInt128Wrapper* out_hash) {
size_t size, UInt128Wrapper* out_hash,
FileCacheType cache_type = FileCacheType::NORMAL,
uint64_t expiration_time = 0) {
auto hash = BlockFileCache::hash(cache_key);
if (out_hash != nullptr) {
*out_hash = hash;
Expand All @@ -246,7 +256,8 @@ class BlockFileCacheTtlMgrTest : public testing::Test {
CacheContext context;
ReadStatistics stats;
context.stats = &stats;
context.cache_type = FileCacheType::NORMAL;
context.cache_type = cache_type;
context.expiration_time = expiration_time;
context.tablet_id = tablet_id;

auto holder = _cache->get_or_set(hash, offset, size, context);
Expand All @@ -269,7 +280,7 @@ class BlockFileCacheTtlMgrTest : public testing::Test {
}
auto block = *it;
EXPECT_TRUE(block);
EXPECT_EQ(FileCacheType::NORMAL, block->cache_type());
EXPECT_EQ(cache_type, block->cache_type());
EXPECT_EQ(FileBlock::get_caller_id(), block->get_or_set_downloader());
// Only append up to the selected file block's size. The requested
// range may be split into multiple file blocks, so appending the
Expand All @@ -282,9 +293,10 @@ class BlockFileCacheTtlMgrTest : public testing::Test {
}

void persist_block_meta(int64_t tablet_id, const UInt128Wrapper& hash, size_t offset,
size_t size) {
size_t size, FileCacheType cache_type = FileCacheType::NORMAL,
uint64_t expiration_time = 0) {
BlockMetaKey key {tablet_id, hash, offset};
BlockMeta meta {FileCacheType::NORMAL, size, 0};
BlockMeta meta {cache_type, size, expiration_time};
_meta_store->put(key, meta);
ASSERT_TRUE(wait_for_condition([this, &key]() { return _meta_store->get(key).has_value(); },
std::chrono::seconds(2)));
Expand Down Expand Up @@ -345,4 +357,108 @@ TEST_F(BlockFileCacheTtlMgrTest, ExpiredTabletMovesBlocksBackToNormal) {
std::chrono::seconds(5)));
}

TEST_F(BlockFileCacheTtlMgrTest, NonTtlTabletWithoutPriorTtlInfoSkipsBlockScan) {
config::file_cache_background_ttl_info_update_interval_ms = 100;

constexpr int64_t kTabletId = 3003;
auto tablet = std::make_shared<FakeTablet>(UnixSeconds(), 0);
fake_engine()->add_tablet(kTabletId, tablet);

UInt128Wrapper hash;
auto block = create_block(kTabletId, "non-ttl-tablet", 0, 1024, &hash);
persist_block_meta(kTabletId, hash, block->range().left, block->range().size());

std::atomic<int64_t> block_scan_count {0};
auto* sync_point = SyncPoint::get_instance();
sync_point->clear_all_call_backs();
sync_point->clear_trace();
SyncPoint::CallbackGuard guard;
sync_point->set_call_back(
"BlockFileCacheTtlMgr::get_file_blocks_from_tablet_id",
[&block_scan_count](std::vector<std::any>&& args) {
if (doris::try_any_cast<int64_t>(args[0]) == kTabletId) {
block_scan_count.fetch_add(1, std::memory_order_relaxed);
}
},
&guard);
sync_point->enable_processing();

_ttl_mgr = std::make_unique<BlockFileCacheTtlMgr>(_cache.get(), _meta_store.get());
_ttl_mgr->register_tablet_id(kTabletId);

bool update_thread_observed = wait_for_condition(
[this]() { return fake_engine()->get_tablet_meta_call_count() >= 2; },
std::chrono::seconds(5));
std::this_thread::sleep_for(std::chrono::milliseconds(100));
sync_point->disable_processing();
sync_point->clear_trace();

EXPECT_TRUE(update_thread_observed);
EXPECT_EQ(0, block_scan_count.load(std::memory_order_relaxed));
EXPECT_EQ(FileCacheType::NORMAL, block->cache_type());
}

TEST_F(BlockFileCacheTtlMgrTest, PeriodicReconcileDemotesTtlBlockWithoutPriorTtlInfo) {
constexpr int64_t kTabletId = 4004;
auto tablet = std::make_shared<FakeTablet>(UnixSeconds(), 0);
fake_engine()->add_tablet(kTabletId, tablet);

const uint64_t expiration_time = UnixSeconds() + 3600;
UInt128Wrapper hash;
auto block = create_block(kTabletId, "ttl-block-without-info", 0, 1024, &hash,
FileCacheType::TTL, expiration_time);
persist_block_meta(kTabletId, hash, block->range().left, block->range().size(),
FileCacheType::TTL, expiration_time);
ASSERT_EQ(FileCacheType::TTL, block->cache_type());

std::atomic<int64_t> block_scan_count {0};
auto* sync_point = SyncPoint::get_instance();
sync_point->clear_all_call_backs();
sync_point->clear_trace();
SyncPoint::CallbackGuard guard;
sync_point->set_call_back(
"BlockFileCacheTtlMgr::get_file_blocks_from_tablet_id",
[&block_scan_count](std::vector<std::any>&& args) {
if (doris::try_any_cast<int64_t>(args[0]) == kTabletId) {
block_scan_count.fetch_add(1, std::memory_order_relaxed);
}
},
&guard);
sync_point->enable_processing();

_ttl_mgr = std::make_unique<BlockFileCacheTtlMgr>(_cache.get(), _meta_store.get());
_ttl_mgr->register_tablet_id(kTabletId);

bool demoted =
wait_for_condition([&]() { return block->cache_type() == FileCacheType::NORMAL; },
std::chrono::seconds(5));
sync_point->disable_processing();
sync_point->clear_trace();

EXPECT_TRUE(demoted);
EXPECT_GE(block_scan_count.load(std::memory_order_relaxed), 1);
}

TEST_F(BlockFileCacheTtlMgrTest, TabletTtlRemovedMovesBlocksBackToNormal) {
constexpr int64_t kTabletId = 5005;
auto tablet = std::make_shared<FakeTablet>(UnixSeconds(), 120);
fake_engine()->add_tablet(kTabletId, tablet);

UInt128Wrapper hash;
auto block = create_block(kTabletId, "ttl-remove", 0, 1024, &hash);
persist_block_meta(kTabletId, hash, block->range().left, block->range().size());

_ttl_mgr = std::make_unique<BlockFileCacheTtlMgr>(_cache.get(), _meta_store.get());
_ttl_mgr->register_tablet_id(kTabletId);

ASSERT_TRUE(wait_for_condition([&]() { return block->cache_type() == FileCacheType::TTL; },
std::chrono::seconds(5)));

tablet->set_ttl_seconds(0);
_ttl_mgr->register_tablet_id(kTabletId);

ASSERT_TRUE(wait_for_condition([&]() { return block->cache_type() == FileCacheType::NORMAL; },
std::chrono::seconds(5)));
}

} // namespace doris::io
Loading