- Release Notes and Announcements
- Product Introduction
- Purchase Guide
- Getting Started
- EMR on CVM Operation Guide
- Planning Cluster
- Configuring Cluster
- Managing Cluster
- Instance Information
- Node Specification Management
- Checking and Updating Public IP
- Cluster Scale-Out
- Cluster Scale-in
- Auto Scaling
- Repairing Disks
- Graceful Scale-In
- Disk Update Check
- Scaling up Cloud Disks
- Changing Configurations
- Automatic Replacement
- Exporting Software Configuration
- Cluster Scripts
- Cluster Termination
- Operation Logs
- Task Center
- Managing Service
- Managing Users
- Adding Components
- Restarting Service
- Starting/Stopping Services
- WebUI Access
- Resetting WebUI Password
- Software WebUI Entry
- Operation Guide for Access to WebUI over Private Network
- Role Management
- Client Management
- Configuration Management
- YARN Resource Scheduling
- HBase RIT Fixing
- Component Port Information
- Service Operation
- HBase Table-Level Monitoring
- Component Health Status
- Monitoring and Alarms
- Cluster Overview
- Node Status
- Service Status
- Cluster Event
- Log
- Application Analysis
- Cluster Inspection
- Monitoring Metrics
- Node Monitoring Metrics
- HDFS Monitoring Metrics
- YARN Monitoring Metrics
- ZooKeeper Monitoring Metrics
- HBase Monitoring Metrics
- Hive Monitoring Metrics
- Spark Monitoring Metrics
- Presto Monitoring Metrics
- Trino Monitoring Metrics
- ClickHouse Monitoring Metrics
- Druid Monitoring Metrics
- Kudu Monitoring Metrics
- Alluxio Monitoring Metrics
- PrestoSQL Monitoring Metrics
- Impala Monitoring Metrics
- Ranger Monitoring Metrics
- COSRanger Monitoring Metrics
- Doris Monitoring Metrics
- Kylin Monitoring Metrics
- Zeppelin Monitoring Metrics
- Oozie Monitoring Metrics
- Storm Monitoring Metrics
- Livy Monitoring Metrics
- Kyuubi Monitoring Metrics
- StarRocks Monitoring Metrics
- Kafka Monitoring Metrics
- Alarm Configurations
- Alarm Records
- Container-Based EMR
- EMR Development Guide
- Hadoop Development Guide
- HDFS Common Operations
- HDFS Federation Management Development Guide
- HDFS Federation Management
- Submitting MapReduce Tasks
- Automatically Adding Task Nodes Without Assigning ApplicationMasters
- YARN Task Queue Management
- Practices on YARN Label Scheduling
- Hadoop Best Practices
- Using API to Analyze Data in HDFS and COS
- Dumping YARN Job Logs to COS
- Spark Development Guide
- Hbase Development Guide
- Phoenix on Hbase Development Guide
- Hive Development Guide
- Presto Development Guide
- Sqoop Development Guide
- Hue Development Guide
- Oozie Development Guide
- Flume Development Guide
- Kerberos Development Guide
- Knox Development Guide
- Alluxio Development Guide
- Kylin Development Guide
- Livy Development Guide
- Kyuubi Development Guide
- Zeppelin Development Guide
- Hudi Development Guide
- Superset Development Guide
- Impala Development Guide
- ClickHouse Development Guide
- Druid Development Guide
- TensorFlow Development Guide
- Jupyter Development Guide
- Kudu Development Guide
- Ranger Development Guide
- Doris Development Guide
- Kafka Development Guide
- Iceberg Development Guide
- StarRocks Development Guide
- Flink Development Guide
- RSS Development Guide
- Hadoop Development Guide
- Best Practices
- API Documentation
- FAQs
- Service Level Agreement
- Contact Us
- Release Notes and Announcements
- Product Introduction
- Purchase Guide
- Getting Started
- EMR on CVM Operation Guide
- Planning Cluster
- Configuring Cluster
- Managing Cluster
- Instance Information
- Node Specification Management
- Checking and Updating Public IP
- Cluster Scale-Out
- Cluster Scale-in
- Auto Scaling
- Repairing Disks
- Graceful Scale-In
- Disk Update Check
- Scaling up Cloud Disks
- Changing Configurations
- Automatic Replacement
- Exporting Software Configuration
- Cluster Scripts
- Cluster Termination
- Operation Logs
- Task Center
- Managing Service
- Managing Users
- Adding Components
- Restarting Service
- Starting/Stopping Services
- WebUI Access
- Resetting WebUI Password
- Software WebUI Entry
- Operation Guide for Access to WebUI over Private Network
- Role Management
- Client Management
- Configuration Management
- YARN Resource Scheduling
- HBase RIT Fixing
- Component Port Information
- Service Operation
- HBase Table-Level Monitoring
- Component Health Status
- Monitoring and Alarms
- Cluster Overview
- Node Status
- Service Status
- Cluster Event
- Log
- Application Analysis
- Cluster Inspection
- Monitoring Metrics
- Node Monitoring Metrics
- HDFS Monitoring Metrics
- YARN Monitoring Metrics
- ZooKeeper Monitoring Metrics
- HBase Monitoring Metrics
- Hive Monitoring Metrics
- Spark Monitoring Metrics
- Presto Monitoring Metrics
- Trino Monitoring Metrics
- ClickHouse Monitoring Metrics
- Druid Monitoring Metrics
- Kudu Monitoring Metrics
- Alluxio Monitoring Metrics
- PrestoSQL Monitoring Metrics
- Impala Monitoring Metrics
- Ranger Monitoring Metrics
- COSRanger Monitoring Metrics
- Doris Monitoring Metrics
- Kylin Monitoring Metrics
- Zeppelin Monitoring Metrics
- Oozie Monitoring Metrics
- Storm Monitoring Metrics
- Livy Monitoring Metrics
- Kyuubi Monitoring Metrics
- StarRocks Monitoring Metrics
- Kafka Monitoring Metrics
- Alarm Configurations
- Alarm Records
- Container-Based EMR
- EMR Development Guide
- Hadoop Development Guide
- HDFS Common Operations
- HDFS Federation Management Development Guide
- HDFS Federation Management
- Submitting MapReduce Tasks
- Automatically Adding Task Nodes Without Assigning ApplicationMasters
- YARN Task Queue Management
- Practices on YARN Label Scheduling
- Hadoop Best Practices
- Using API to Analyze Data in HDFS and COS
- Dumping YARN Job Logs to COS
- Spark Development Guide
- Hbase Development Guide
- Phoenix on Hbase Development Guide
- Hive Development Guide
- Presto Development Guide
- Sqoop Development Guide
- Hue Development Guide
- Oozie Development Guide
- Flume Development Guide
- Kerberos Development Guide
- Knox Development Guide
- Alluxio Development Guide
- Kylin Development Guide
- Livy Development Guide
- Kyuubi Development Guide
- Zeppelin Development Guide
- Hudi Development Guide
- Superset Development Guide
- Impala Development Guide
- ClickHouse Development Guide
- Druid Development Guide
- TensorFlow Development Guide
- Jupyter Development Guide
- Kudu Development Guide
- Ranger Development Guide
- Doris Development Guide
- Kafka Development Guide
- Iceberg Development Guide
- StarRocks Development Guide
- Flink Development Guide
- RSS Development Guide
- Hadoop Development Guide
- Best Practices
- API Documentation
- FAQs
- Service Level Agreement
- Contact Us
Before using EMR, carefully read and understand the following use limits:
To ensure the cluster network security, new clusters will be placed in the same VPC. Do not change the VPC of an existing cluster or node; otherwise, the cluster network may fail.
When you create a cluster, EMR can help you create a new security group, or you can also manually select an existing EMR security group. Make sure that the selected security group has the necessary inbound/outbound rules for EMR. Do not delete or change the security group in use after creating the cluster; otherwise, cluster communication may fail, thus affecting the service.
Plan the storage space of nodes in advance based on your business needs, and promptly add storage nodes, as an insufficient storage space may cause data and node risks. Currently, core, task, and router nodes in the EMR cluster support mounting multiple cloud disks, and a node can have up to 15 cloud disks. Clusters in BM 2.0 and local disk models (IO and D series) don't support mounting multiple cloud disks.
When using EMR, do not perform operations in the CVM console, such as shutdown, restart, VPC switch, security group rule adjustment, so as to avoid cluster exceptions. OS reinstallation, instance termination, configuration adjustment, renewal, and billing mode change are also not recommended. You can perform necessary cluster maintenance operations in the EMR console.
Public IPs can increase the possibility of master nodes being attacked, so you need to manage and monitor the risks. EIPs (including the IPs on the secondary ENI) will be retained after the cluster is terminated, and the idle IPs will continue to incur fees. If you don't need to retain them, release them on the corresponding resource management page.
When you create a cluster, EMR provides component initialization parameters for general scenarios. Before you use components such as HDFS and HBase, we recommend you check whether the component parameters meet the needs of your business scenarios. To get the component initialization guide, contact us.
Keep the host login password of your EMR cluster secure. If you configure passwordless login for cross-node access, Tencent Cloud security services may detect vulnerability risks and prompt you.
Even if an exception occurs in your cluster, the cluster will continue to be billed. In this case, we recommend you promptly contact us for assistance. If we need to log in to your cluster for troubleshooting, we will gain your consent to request your account and password.
When you use or maintain your EMR cluster, some unexpected operations may render it unavailable or unstable, and you will receive a risk warning before performing such operations in the console. This document lists some of the prohibited and risky operations:
Prohibited Operations
Operation | Risk |
Change the private IP of an EMR node in CVM | Node communication exception or cluster unavailability |
Modify the security group of a CVM node while the cluster is running | Node communication exception or component service unavailability |
Delete an existing process, application, or file from a node | Cluster/Component service unavailability |
Delete or modify the `hosts` file in the `/etc` directory | Service exception, as the cluster cannot be associated with the service on the node |
Delete or modify the HDFS metadata file `edit log` | HDFS cluster unavailability |
Manually modify the data in the Hive metadatabase | Service exception caused by a Hive data parsing error |
Delete a ZooKeeper data directory | Failure of dependent components |
Risky Operations
Operation | Risk | Suggestion |
Shut down or restart an EMR cluster node in CVM | Service unavailability | Check whether the operation is necessary and read the CVM operation limits |
Mount a disk to an EMR node in the CVM console | Disk unavailability caused by EMR's failure to recognize and initialize the disk | Do this through core node scaling or under technical guidance |
Unmount a disk from an EMR node in the CVM console | Data loss or cluster unavailability | Do this through core node scaling or under technical guidance |
Modify the component configuration file parameters in CVM | Modified parameter overwrite after service restart | Modify the parameter configuration in the EMR console and seek technical guidance in special circumstances |
Delete or modify the `resolv.conf` file in the `/etc` directory | Service exception, as the cluster cannot be associated with the service on the node | Check whether the operation is necessary and perform it under technical guidance |
Modify the MetaDB password | EMR depends on the password configured in MetaDB, and after the password is changed, services such as Hive and Ranger will become unavailable | Modify the configuration in the EMR console under technical guidance |
Modify the MetaDB floating IP | Hive and Ranger service unavailability, as EMR depends on the IP configured in MetaDB | Modify the configuration in the EMR console under technical guidance |
Modify the MetaDB security group | Interruption of the communication between MetaDB and the cluster, or Hive and Ranger service unavailability | Perform the operation under technical guidance |
Yes
No
Was this page helpful?