Service operation is a collection of convenient Ops and management tools provided for components, including general service restart and command-based Ops operations specific to certain components, such as HDFS NameNode active/standby switch, HDFS data rebalance, YARN ResourceManager active/standby switch, and YARN queue refresh.
Note
1. YARN ResourceManager active/standby switch requires you to disable yarn.resourcemanager.ha.automatic-failover.enabled. If it is not displayed in the operation drop-down list on the YARN block, you need to find yarn.resourcemanager.ha.automatic-failover.enabled in the yarn-site.xml configuration file of YARN configuration management and disable it.
2. Do not delete the queues that have already taken effect in the scheduler's configuration file when refreshing YARN queues.
3. Ranger metadatabase change is currently supported only for MySQL databases, and only the connections of admin users will be tested.
Specification Items
HDFS NameNode active/standby switch: This operation switches the active NameNode to standby status and the standby NameNode to active status.
HDFS data rebalance: This operation is usually required when a new DataNode is added. It will make the data evenly distributed to avoid hotspotting and make the cluster read/write load more balanced.
YARN ResourceManager active/standby switch: This operation switches the active ResourceManager to the standby status and the standby ResourceManager to active status. It is allowed only when yarn.resourcemanager.ha.automatic-failover.embedded is disabled.
YARN queue refresh: This operation enables additions or updates to capacity-scheduler.xml or fair-scheduler.xml to take effect in the ResourceManager. Do not delete queues defined in them.
Ranger metadatabase change: To change the underlying database of Ranger, you need to modify the conf/install.properties file and then run the setup.sh script locally. This script can synchronize the database information to the local ranger-admin-site.xml configuration file without modifying the content of ranger-admin-site.xml in Configuration Management. When you modify and deliver ranger-admin-site.xml based on your needs, the database information will be overwritten, resulting in an exception. This operation allows you to quickly configure a Ranger metadatabase, so as to avoid exceptions caused by missing configuration items when you change the metadatabase address.
Directions
1. Log in to the EMR console and click the ID/Name of the corresponding cluster in the cluster list to go to the cluster details page.
2. On the cluster details page, select Cluster Service and click the target component block.
3. Taking HDFS NameNode active/standby switch as an example, in Cluster Service, select HDFS > Operation > NN Failover to perform the active/standby switch operation.
Yes
No
Was this page helpful?