mastra-ai/mastra · error · HTTPException

Memory storage is not configured for agent "${agent.id}"

Error message

Memory storage is not configured for agent "${agent.id}"

What it means

Beyond having a Memory instance, the agent's memory must have a storage backend for conversations to persist threads and messages. The handler calls getAgentMemoryStore and throws HTTP 400 when the memory's storage is not configured.

Source

Thrown at packages/server/src/server/handlers/conversations.ts:68

  responseSchema: conversationObjectSchema,
  summary: 'Create a conversation',
  description: 'Creates a new thread-backed conversation for agent-backed Responses API requests',
  tags: ['Responses'],
  requiresAuth: true,
  requiresPermission: MastraFGAPermissions.AGENTS_CREATE,
  handler: async ({ mastra, requestContext, agent_id, conversation_id, resource_id, title, metadata }) => {
    try {
      if (!mastra) {
        throw new HTTPException(500, { message: 'Mastra instance is required for conversations' });
      }

      const agent = await getAgentFromSystem({ mastra, agentId: agent_id });
      const memory = await agent.getMemory({ requestContext });
      if (!memory) {
        throw new HTTPException(400, { message: `Agent "${agent.id}" does not have memory configured` });
      }
      if (!(await getAgentMemoryStore({ agent, requestContext }))) {
        throw new HTTPException(400, { message: `Memory storage is not configured for agent "${agent.id}"` });
      }

      const threadId = conversation_id ?? randomUUID();
      const resourceId = getEffectiveResourceId(requestContext, resource_id) ?? threadId;
      const thread = await memory.createThread({
        threadId,
        resourceId,
        title,
        metadata,
      });

      return buildConversationObject({ thread });
    } catch (error) {
      return handleError(error, 'Error creating conversation');
    }
  },
});

View on GitHub (pinned to 75dd419e61)

Solutions

  1. Construct Memory with a storage backend: new Memory({ storage: new LibSQLStore({ url: ... }) }) or your preferred storage adapter
  2. Verify the storage adapter is correctly instantiated and its connection (DB URL/token) is valid
  3. Share one storage instance across agents if conversations should be portable, and confirm it is the one wired into the agent
  4. Check env vars for the storage connection are present at server start

Example fix

// before
memory: new Memory({ options: { lastMessages: 10 } })
// after
memory: new Memory({ storage: new LibSQLStore({ url: process.env.DATABASE_URL }), options: { lastMessages: 10 } })
Defensive patterns

Strategy: validation

Validate before calling

const memory = await agent.getMemory();
if (!memory) throw new Error('No memory configured');
if (!memory.constructor.name.includes('Memory')) throw new Error('Unexpected memory type');
// storage is required for conversations:
const mem = new Memory({ storage: resolveStorageFromEnv() }); // ensure storage present at construction

Type guard

function memoryHasStorage(m: Memory | undefined): boolean {
  return !!m && 'storage' in m && m.storage != null;
}

Try / catch

try {
  const conv = await client.createConversation({ agent_id });
} catch (e) {
  if (e.status === 400 && /storage is not configured/.test(e.message)) {
    console.error('Wire a storage adapter into Memory for this agent');
  }
  throw e;
}

Prevention

When it happens

Trigger: POSTing to create-conversation for an agent whose Memory was constructed without a `storage` option (in-memory only or none), so getAgentMemoryStore resolves to undefined.

Common situations: Memory created only with options (working memory/lastMessages) but no storage in dev setups; storage intentionally omitted for tests then the agent is used with the conversations API; switching deployments and forgetting to wire storage (e.g. LibSQL/Postgres/Upstash) into Memory.

Related errors

AI-assisted analysis of mastra-ai/mastra@75dd419e61 (2026-08-30). Data as JSON: /api/errors/6c0ac57c4c16ddb3. Report an issue: GitHub.