[Fix](bloom filter) Fix bloom filter memory leak (#34871)

* Issue: Doris occasionally encounters an issue where memory usage becomes exceptionally high and does not decrease. The leaked memory is occupied by Bloom filters stored in memory.

Reason: The segment cache stores segment objects read from files into memory. It functions as an LRU cache with an eviction strategy: when the number of segments exceeds the maximum number, or the total memory size of segment objects in the cache exceeds the maximum usage, it evicts the older segments. However, there is a piece of logic in the code that first reads the segment object into memory, assuming it occupies memory size A, then places the read segment object into the cache (at this point, the cache considers the segment object size to be A). It then reads the segment's Bloom filter from the file and assigns it to the segment's Bloom filter member variable, assuming the Bloom filter occupies memory size B. Thus, the total size of the segment object at this point is A+B. However, the cache does not update this size, leading to the actual size of the segment object stored in the cache (A+B) being larger than the size considered by the cache (A). When the number of segment objects in the cache increases to a certain extent, the used memory will surge dramatically. However, the cache does not perceive the size as reaching the eviction limit, so it does not evict the segment objects. In such cases, a memory leak issue arises.

Solution: Since each segment object only reads the Bloom filter once, the issue can be resolved by changing the logic from reading the segment, placing it into the cache, and then reading the Bloom filter to reading the segment, reading the Bloom filter, and then placing it into the cache.

be/src/olap/base_tablet.cpp

@@ -541,7 +541,7 @@ Status BaseTablet::lookup_row_key(const Slice& encoded_key, bool with_seq_col,
         if (UNLIKELY(segment_caches[i] == nullptr)) {
             segment_caches[i] = std::make_unique<SegmentCacheHandle>();
             RETURN_IF_ERROR(SegmentLoader::instance()->load_segments(
-                    std::static_pointer_cast<BetaRowset>(rs), segment_caches[i].get(), true));
+                    std::static_pointer_cast<BetaRowset>(rs), segment_caches[i].get(), true, true));
         }
         auto& segments = segment_caches[i]->get_segments();
         DCHECK_EQ(segments.size(), num_segments);

be/src/olap/rowset/segment_v2/segment.cpp

@@ -283,6 +283,15 @@ Status Segment::_parse_footer(SegmentFooterPB* footer) {
 }
 
 Status Segment::_load_pk_bloom_filter() {
+#ifdef BE_TEST
+    if (_pk_index_meta == nullptr) {
+        // for BE UT "segment_cache_test"
+        return _load_pk_bf_once.call([this] {
+            _meta_mem_usage += 100;
+            return Status::OK();
+        });
+    }
+#endif
     DCHECK(_tablet_schema->keys_type() == UNIQUE_KEYS);
     DCHECK(_pk_index_meta != nullptr);
     DCHECK(_pk_index_reader != nullptr);
@@ -312,6 +321,7 @@ Status Segment::load_pk_index_and_bf() {
     RETURN_IF_ERROR(_load_pk_bloom_filter());
     return Status::OK();
 }
+
 Status Segment::load_index() {
     auto status = [this]() { return _load_index_impl(); }();
     if (!status.ok()) {

be/src/olap/segment_loader.cpp

@@ -18,6 +18,7 @@
 #include "olap/segment_loader.h"
 
 #include "common/config.h"
+#include "common/status.h"
 #include "olap/olap_define.h"
 #include "olap/rowset/beta_rowset.h"
 #include "util/stopwatch.hpp"
@@ -50,7 +51,8 @@ void SegmentCache::erase(const SegmentCache::CacheKey& key) {
 }
 
 Status SegmentLoader::load_segments(const BetaRowsetSharedPtr& rowset,
-                                    SegmentCacheHandle* cache_handle, bool use_cache) {
+                                    SegmentCacheHandle* cache_handle, bool use_cache,
+                                    bool need_load_pk_index_and_bf) {
     if (cache_handle->is_inited()) {
         return Status::OK();
     }
@@ -61,9 +63,13 @@ Status SegmentLoader::load_segments(const BetaRowsetSharedPtr& rowset,
         }
         segment_v2::SegmentSharedPtr segment;
         RETURN_IF_ERROR(rowset->load_segment(i, &segment));
+        if (need_load_pk_index_and_bf) {
+            RETURN_IF_ERROR(segment->load_pk_index_and_bf());
+        }
         if (use_cache && !config::disable_segment_cache) {
             // memory of SegmentCache::CacheValue will be handled by SegmentCache
             auto* cache_value = new SegmentCache::CacheValue();
+            _cache_mem_usage += segment->meta_mem_usage();
             cache_value->segment = std::move(segment);
             _segment_cache->insert(cache_key, *cache_value, cache_handle);
         } else {

be/src/olap/segment_loader.h

@@ -114,15 +114,20 @@ class SegmentLoader {
     // Load segments of "rowset", return the "cache_handle" which contains segments.
     // If use_cache is true, it will be loaded from _cache.
     Status load_segments(const BetaRowsetSharedPtr& rowset, SegmentCacheHandle* cache_handle,
-                         bool use_cache = false);
+                         bool use_cache = false, bool need_load_pk_index_and_bf = false);
 
     void erase_segment(const SegmentCache::CacheKey& key);
 
     void erase_segments(const RowsetId& rowset_id, int64_t num_segments);
 
+    // Just used for BE UT
+    int64_t cache_mem_usage() const { return _cache_mem_usage; }
+
 private:
     SegmentLoader();
     std::unique_ptr<SegmentCache> _segment_cache;
+    // Just used for BE UT
+    int64_t _cache_mem_usage = 0;
 };
 
 // A handle for a single rowset from segment lru cache.