[Bug] Memory is not working at all

[datrinh]

Still have no working memory in v2.0.0-alpha.78
It's pretty unusable right now

After trying with ~5 hive-mind spawns i still get this output:

✅ Memory Bank Statistics:
   Total Entries: 0
   Namespaces: 0
   Size: 0.00 KB

Furthermore all my hive-mind spawns show the same status basically regardless of their progress:

Status: active
Created: 7/30/2025, 6:18:10 AM

Agents:
  👑 Queen: Queen Coordinator (active)
  🐝 Workers:
    - Researcher Worker 1 (researcher) idle
    - Coder Worker 2 (coder) idle
    - Analyst Worker 3 (analyst) idle
    - Tester Worker 4 (tester) idle

Tasks:
  Total: 0
  Completed: null
  In Progress: null
  Pending: null

Collective Memory:
  Entries: 4

During the runs I can sometimes see successful outputs like this though:

claude-flow - memory_usage (MCP)(action: "store", key: "hive/quality-assessment", value:
                                  "Code-Analyzer confirms excellent architecture (8.5/10), strong
                                  TypeScript implementation, production-ready foundations")
  ⎿  {
       "success": true,
       "action": "store",
       "key": "hive/quality-assessment",
       "namespace": "default",
       "stored": true,
       "size": 118,
       "id": 1578,
       "storage_type": "sqlite",
       "timestamp": "2025-07-30T06:46:24.849Z"
     }

⏺ claude-flow - memory_usage (MCP)(action: "store", key: "hive/testing-status", value: "QA-Engineer
                                  reports MVP validation complete with success criteria mostly met,
                                  configuration issues identified")
  ⎿  {
       "success": true,
       "action": "store",
       "key": "hive/testing-status",
       "namespace": "default",
       "stored": true,
       "size": 109,
       "id": 1579,
       "storage_type": "sqlite",
       "timestamp": "2025-07-30T06:46:25.908Z"
     }

When I quit (Ctrl+C) claude-flow after a session I get this:

✓ Claude Code completed successfully
/Users/xxx/.local/share/mise/installs/node/22.14.0/lib/node_modules/claude-flow/node_modules/better-sqlite3/lib/methods/wrappers.js:5
        return this[cppdb].prepare(sql, this, false);
                           ^

TypeError: The database connection is not open
    at Database.prepare (/Users/xxx/.local/share/mise/installs/node/22.14.0/lib/node_modules/claude-flow/node_modules/better-sqlite3/lib/methods/wrappers.js:5:21)
    at HiveMindSessionManager.logSessionEvent (file:///Users/xxx/.local/share/mise/installs/node/22.14.0/lib/node_modules/claude-flow/src/cli/simple-commands/hive-mind/session-manager.js:644:28)
    at process.processTicksAndRejections (node:internal/process/task_queues:105:5)

Node.js v22.14.0

Is there any workaround? Any previous version I should downgrade to?

[MegaBalls-byte]

What OS? Seems like a typical Windows problem for which there is an extra explanation

[datrinh]

What OS? Seems like a typical Windows problem for which there is an extra explanation

It's MacOS

[vuln3r]

you need to install node 22 and than better-sqlite3. for me worked on that way

[datrinh]

you need to install node 22 and than better-sqlite3. for me worked on that way

Install better-sqlite3 globally? Isn't it a dev dep from claude-flow?

Edit: I tried with installing better-sqlite3 both globally and as devDependency without any improvement

[dylansproule]

Same thing here:

/Users/dylansproule/.nvm/versions/node/v22.15.0/lib/node_modules/claude-flow/node_modules/better-sqlite3/lib/methods/wrappers.js:5
return this[cppdb].prepare(sql, this, false);
^

TypeError: The database connection is not open
at Database.prepare (/Users/dylansproule/.nvm/versions/node/v22.15.0/lib/node_modules/claude-flow/node_modules/better-sqlite3/lib/methods/wrappers.js:5:21)
at HiveMindSessionManager.logSessionEvent (file:///Users/dylansproule/.nvm/versions/node/v22.15.0/lib/node_modules/claude-flow/src/cli/simple-commands/hive-mind/session-manager.js:644:28)
at process.processTicksAndRejections (node:internal/process/task_queues:105:5)

[Cris9870]

Same

After using the run command with the memory name, when checking memory stats it does not appear.

The flow I am using is quite simple:

I install Claude and Claude-flow globally via npm.

Then I launch the project
$claude-flow hive-mind spawn "Develop a to-do list application"
--agents 10
--strategy parallel
--memory-namespace todolist
--claude

Checking $claude-flow memory stats I get:
Memory Bank Statistics:
Total Entries: 0
Namespaces: 0
Size: 0.00 KB

Similarly $ claude-flow memory list
⚠️ No namespaces found

I only get a result if I explicitly create the namespace with a command like:
$ claude-flow memory store "project-context" "Full-stack app requirements"
✅ Stored successfully
📝 Key: project-context
📦 Namespace: default
💾 Size: 27 bytes

I am very interested in the memory issue, since many projects become large and consume all my session tokens and when they are close to finish I would like to be able to pause the development and continue it in my next session and I had understood that with the persistent memory it would be easier to resume the progress made.

I'm using Ubuntu 22.04 with VS Code Server.

[PhilflowIO]

Same here on arch. Please fix it

[ruvnet]

Thank you for reporting this critical issue. The memory system is indeed broken in alpha.78-82, and this is a Claude Flow bug that we need to fix urgently.

Immediate Workaround

Until we fix this properly, you can try using the file-based memory fallback:

# Force file-based memory storage
export CLAUDE_FLOW_MEMORY_TYPE=file

# Or manually store/retrieve using files
npx claude-flow@alpha hooks notify --message "your data" > .swarm/memory/manual-backup.json

Root Cause

The error you're seeing:

TypeError: The database connection is not open
at Database.prepare (/Users/xxx/.local/share/mise/installs/node/22.14.0/lib/node_modules/claude-flow/node_modules/better-sqlite3/lib/methods/wrappers.js:5:21)

This indicates that:

1. The SQLite database connection is being closed prematurely
2. The connection lifecycle isn't properly managed during async operations
3. Session cleanup is happening before memory operations complete

Temporary Fix

For now, you can downgrade to a more stable version:

# Version 2.0.0-alpha.75 had working memory
npm install -g claude-flow@2.0.0-alpha.75

Planned Fix for Alpha 84

We'll implement:

1. Connection Pool Management

   • Maintain persistent database connections
   • Proper connection lifecycle management
   • Graceful shutdown handling

2. Fallback Memory System

   • Automatic fallback to file-based storage when SQLite fails
   • Memory export/import commands for backup

3. Better Error Handling

   • Catch and recover from database errors
   • Retry logic for failed operations
   • Clear error messages for debugging

Debugging Commands

To help diagnose the issue:

# Check if database file exists
ls -la .swarm/memory.db

# Try manual database operations
npx claude-flow@alpha memory list
npx claude-flow@alpha memory export > memory-backup.json

# Check for database locks
lsof | grep memory.db

Status in Alpha 83

Alpha 83 includes some improvements but doesn't fully fix the memory system yet. This is our top priority for Alpha 84.

Thank you for your patience. The memory system is critical for swarm coordination and we're committed to fixing it properly.

[pierrecorsini]

No "immediate" warkaround working for me ( export CLAUDE_FLOW_MEMORY_TYPE=file or NODE_OPTIONS='--max-old-space-size=4096'), I'm on windows wsl.

[KingInYellow18]

Same issue Ubuntu server -- hasn't seemed to fix on latest release but may need to do more testing and reinstall on a clone from scratch. Seems I continue to get the better-sql3 error

[Koroqe]

Same issue here. Session state just vanishes between runs. I gave up on built-in memory and moved to a scratchpad file that persists context between agent sessions — each agent reads/writes to it so nothing gets lost. Works way more reliably than any DB-backed memory I tried.

Built it into this if anyone is looking for an alternative approach: https://github.com/Koroqe/claude-code-sdlc