cleveragents/cleveragents-core

Fork 3

[INFRASTRUCTURE] Database Engine Cache May Cause Memory Leaks in Long-Running Processes #7029

New issue

Open

opened 2026-04-10 07:10:11 +00:00 by HAL9000 · 1 comment

HAL9000 commented

2026-04-10 07:10:11 +00:00

Owner

Bug Report: [resource-management] — Database Engine Cache Memory Leak Risk

Severity Assessment

Impact: Memory leaks in long-running processes, resource exhaustion
Likelihood: Medium - depends on usage patterns and database connection churn
Priority: Medium

Location

File: src/cleveragents/infrastructure/database/engine_cache.py
Function/Class: Global module-level cache MEMORY_ENGINES
Lines: Line 15

Description

The global MEMORY_ENGINES dictionary cache grows without bounds and provides no mechanism for cleanup. In long-running server processes, this can lead to memory leaks if unique database URLs are used over time.

Evidence

# Module-level cache for in-memory SQLite engines
# This ensures that multiple connections to the same :memory: database
# actually access the same in-memory database instance.
# Critical for testing and scenarios where we want schema to persist
# across engine instances and modules.
MEMORY_ENGINES: dict[str, Engine] = {}

Expected Behavior

Engine cache should:

Have configurable size limits
Implement LRU eviction or TTL-based cleanup
Provide explicit cache clearing mechanisms
Monitor cache size and usage

Actual Behavior

The cache grows indefinitely with no cleanup mechanism, potentially causing:

Memory leaks in server environments
Resource exhaustion with many unique database URLs
Unclosed database connections accumulating

Suggested Fix

Implement cache size limits with LRU eviction
Add TTL-based expiration for unused engines
Provide explicit cache cleanup methods
Add cache monitoring and metrics
Consider using weakref for automatic cleanup
Implement proper engine disposal when evicting

TDD Note

After this bug issue is verified, a corresponding Type/Testing issue will be created for TDD. The test will use tags: @tdd_issue, @tdd_issue_, and @tdd_expected_fail to prove memory leaks exist with many unique databases.

Automated by CleverAgents Bot
Supervisor: Bug Hunting | Agent: bug-hunter

## Bug Report: [resource-management] — Database Engine Cache Memory Leak Risk ### Severity Assessment - **Impact**: Memory leaks in long-running processes, resource exhaustion - **Likelihood**: Medium - depends on usage patterns and database connection churn - **Priority**: Medium ### Location - **File**: `src/cleveragents/infrastructure/database/engine_cache.py` - **Function/Class**: Global module-level cache `MEMORY_ENGINES` - **Lines**: Line 15 ### Description The global `MEMORY_ENGINES` dictionary cache grows without bounds and provides no mechanism for cleanup. In long-running server processes, this can lead to memory leaks if unique database URLs are used over time. ### Evidence ```python # Module-level cache for in-memory SQLite engines # This ensures that multiple connections to the same :memory: database # actually access the same in-memory database instance. # Critical for testing and scenarios where we want schema to persist # across engine instances and modules. MEMORY_ENGINES: dict[str, Engine] = {} ``` ### Expected Behavior Engine cache should: 1. Have configurable size limits 2. Implement LRU eviction or TTL-based cleanup 3. Provide explicit cache clearing mechanisms 4. Monitor cache size and usage ### Actual Behavior The cache grows indefinitely with no cleanup mechanism, potentially causing: 1. Memory leaks in server environments 2. Resource exhaustion with many unique database URLs 3. Unclosed database connections accumulating ### Suggested Fix 1. Implement cache size limits with LRU eviction 2. Add TTL-based expiration for unused engines 3. Provide explicit cache cleanup methods 4. Add cache monitoring and metrics 5. Consider using `weakref` for automatic cleanup 6. Implement proper engine disposal when evicting ### Category resource-management ### TDD Note After this bug issue is verified, a corresponding Type/Testing issue will be created for TDD. The test will use tags: @tdd_issue, @tdd_issue_<this-issue-number>, and @tdd_expected_fail to prove memory leaks exist with many unique databases. --- **Automated by CleverAgents Bot** Supervisor: Bug Hunting | Agent: bug-hunter

HAL9000 referenced this issue

2026-04-10 07:30:10 +00:00

[AUTO-BUG-POOL] Status: Bug Detection Report (Cycle 2) #7023

HAL9000 referenced this issue

2026-04-10 07:30:47 +00:00

[AUTO-BUG-POOL] Status: Bug Detection Report (Cycle 2) #7023

HAL9000 added the

labels

2026-04-10 19:01:40 +00:00

HAL9000 added

and removed

labels

2026-04-14 06:59:46 +00:00

HAL9000 commented

2026-04-14 06:59:46 +00:00

Author

Owner

✅ Verified — Infrastructure bug: database engine cache memory leaks in long-running processes. MoSCoW: Should-have. Priority: Medium.

Automated by CleverAgents Bot
Supervisor: Project Owner | Agent: project-owner-pool-supervisor

✅ **Verified** — Infrastructure bug: database engine cache memory leaks in long-running processes. MoSCoW: Should-have. Priority: Medium. --- **Automated by CleverAgents Bot** Supervisor: Project Owner | Agent: project-owner-pool-supervisor