# Connector Status Abnormal Alert

The Connector Status Abnormal alert indicates that one or more data integration connectors in an EMQX Cloud deployment have remained outside the normal state for a period of time.

This usually means that the connection between EMQX Cloud and the target external system is unstable or unavailable. As a result, messages may not be delivered to the target service.

## External Service Unreachable or Network Failure

### Symptom

EMQX Cloud cannot establish or maintain a stable connection to the target external system.

Errors such as `connection refused`, `timeout`, `unreachable`, or `econnrefused` appear in **Deployment Logs**, and the connector status is `Disconnected`, `Error`, or inconsistent across cluster nodes.

### Common Causes

- The target system is unavailable or not running normally.
- The network is interrupted or unstable.
- The target service port is closed.
- Network policies restrict access between EMQX Cloud and the target service.

### Resolution

- Verify that the target system is running and its listening port is open.
- Work with EMQX Cloud Technical Support to use tools such as `ping`, `telnet`, or `curl` to test connectivity from EMQX Cloud to the target service.
- Verify that VPC, allowlist, firewall, or security group rules permit EMQX to access the target service.

## Authentication Failure or Insufficient Permissions

### Symptom

The connector connects briefly and then disconnects because its account, credentials, or authentication configuration is invalid, or because the account lacks the required permissions.

Errors such as `authentication failed`, `unauthorized`, or `permission denied` appear in **Deployment Logs**.

### Common Causes

- The username, password, API key, or certificate is incorrect.
- The credential has expired or been revoked.
- The target system has not granted the connector the required access or write permissions.

### Resolution

- Verify the username, password, API key, or certificate in the connector configuration.
- Verify that the target system grants the account all required permissions, such as database write access, Kafka producer permissions, or cloud IAM permissions.

## Target System Overloaded or Out of Resources

### Symptom

An overloaded or resource-constrained target system cannot respond to connection or write requests in time, causing an unstable connector state.

The target system logs errors such as `timeout` or `server busy`, while the connector repeatedly changes between `Connected` and `Disconnected` or appears inconsistent across cluster nodes.

### Common Causes

- The target system has exhausted its CPU, memory, or connection capacity.
- Request concurrency exceeds the target system's processing capacity.

### Resolution

- Check CPU, memory, and connection usage on the target system.
- Configure appropriate connection pool sizes, timeouts, buffering, and retry policies in EMQX.
- Scale the target system to support the required throughput.

## Incorrect Connector Configuration

### Symptom

EMQX cannot access the target resource because one or more connector parameters are incorrect. The connector never establishes a stable connection.

Errors such as `invalid config`, `unknown topic`, or `table not found` appear in **Deployment Logs**.

### Common Causes

- The database address, port, or instance information is incorrect.
- The Kafka topic does not exist or its name is misspelled.
- The database table does not exist or has an incompatible schema.

### Resolution

- Verify that the target address, port, topic, and table name match the target environment.
- Correct the configuration and save the connector again.

## Troubleshooting

1. Log in to the EMQX Cloud Console and enter the affected deployment.

2. Go to **Data Integration** -> **Connectors** and check the status of the connector that triggered the alert.

   ![Connector status on the Data Integration page](./_assets/connector_abnormal_status.png)

3. Open **Deployment Logs** and set **Error Type** to **Data Integration** to locate the relevant error.

   ![Connector errors in Deployment Logs](./_assets/connector_abnormal_logs.png)

4. Use the log details to identify whether the cause is a network failure, authentication failure, overloaded target system, or incorrect configuration.

5. Open the connector configuration, click **Edit**, and correct the parameters or update the credentials.

6. Save the configuration, restart the connector, and verify that its status changes to `Connected`.