MetaApi Not Working? Common MT4/MT5 API Issues and When to Switch

When you are managing a live algorithmic trading system, every second of downtime costs money. If you are reading this, you are likely staring at a console error, a dropped WebSocket, or a stalled copy trading queue, wondering why MetaApi is not working.

Because cloud wrappers involve several moving parts (your backend, the provider's cloud, the account's session and the broker), diagnosing a MetaApi error can be frustrating.

Before you assume it is total MetaApi downtime, run through this technical troubleshooting checklist. We will cover the most common points of failure, how to fix them, and how to know when it is time to consider another API.

---

1. Check Your Broker Connection & IP Allowlisting

The most common cause of a MetaApi connection issue is not an API outage; it is a broker rejection. Retail brokers frequently change their server IPs, update their domain names, or deploy strict regional firewalls to prevent DDoS attacks.

• The Fix: Verify that the broker server string you passed during provisioning is exactly what appears in the MT4/MT5 login dialog. If the broker recently changed infrastructure, the stored server address may no longer answer, so reconnect with the current server name.
• A Note on Prop Firms: If you are connecting to a prop firm, ensure their specific server name hasn't changed during an equity migration. (See our guide on MT5 API for Prop Trading Firms for handling dynamic prop firm servers).

2. Check Account Authorization (Investor vs Main Passwords)

If the API connects but you receive an "Invalid Request" or "Authorization Failed" error when attempting to /OrderSend, you likely have a permission mismatch.

• The Fix: Ensure you provisioned the account using the main (trading) password, not the investor (read-only) password. If you are only streaming data, an investor password works fine, but any execution requires the main password. Also, confirm the broker has not forced a password reset due to inactivity.

3. Check the MetaTrader Instance/Session Status

A hosted account session can occasionally stall or become unresponsive. If your dashboard shows the account as active but trades are not mirroring, the instance session might be stalled.

• The Fix: You need to forcefully redeploy or restart the provisioned account via your SDK or their dashboard. If you find yourself restarting sessions frequently, raise it with the provider's support and keep a record of each incident.

4. Check API Rate Limits

If your application works perfectly during the Asian session but throws errors when the New York session opens, you are likely hitting an API rate limit. Heavy polling or a mass copy-trading broadcast across many accounts through one key can trigger throttling.

• The Fix: Audit your code. If you are polling /AccountSummary every second to check for trade closures, stop immediately. Switch to an event-driven architecture using WebSockets.

5. Check WebSocket Disconnects

A frequently searched problem is finding MetaApi disconnected from its WebSocket stream without triggering a clean error catch in your Node.js or Python backend. This leaves your trading bot flying blind.

• The Fix: Implement aggressive heartbeat monitoring. If your WebSocket does not receive a ping within a designated timeframe (e.g., 5 seconds), programmatically kill the connection, request a new session state, and reconnect to ensure you did not miss a critical Stop Loss trigger during the blackout.

6. Check Server Region & Latency

Sometimes the API is technically "working," but the latency is so severe that trades execute with massive slippage. This happens when your broker's servers are in one region (e.g., London LD4) but your cloud API node is hosted in a different region.

• The Fix: Check whether your provider lets you choose a region close to your broker, and measure round-trip times before and after a change.

---

What to Do When MetaApi Support is Not Responding

One of the most stressful scenarios for a SaaS founder is facing a production outage while MetaApi support is not responding quickly enough to save client accounts. When the SDK hides the connection, you depend on the provider's support desk to restart it. (With MetaTrader API, /Reconnect?id= restarts an account's connection from your own code, and /ConnectState?id= tells you why it failed.)

If you have run through the checklist above, the issue is likely a failure on the infrastructure side.

When Is It Time to Migrate?

Troubleshooting occasional bugs is part of software development. However, if you are experiencing repeated outages, unpredictable latency spikes, or instance freezes during high-volume trading hours, your architecture is restricting your growth.

You should consider an alternative such as MetaTrader API if:

1. You want a known price per connected account: $14 per seat per month, graduated down to $12 from the 20th seat (up to 50 seats per customer, more on request).
2. You want every call as a plain HTTP request you can reproduce with one curl command.
3. You want nothing to install on your side, no EA for you to run and no MetaTrader window to keep open.

Read our MetaApi migration guide to see how the endpoints map. There is no free trial; test with a broker demo account on a single seat.

Ready to Fix Your Infrastructure?

Send us your current setup and we'll tell you if MetaTrader API can replace it.

👉 Contact MetaTrader API Engineering Today

---