In FusionInsight,How to handle Service Node Faulty


1.FusionInsight Manager regularly checks the running status of each node. In case one node is lost due to network or hardware problems, FusionInsight Manager generates the alarm "ALM-12006 Node Fault". Handle the alarm based on the handling procedure.
2.If the alarm persists after step 1, see 10.2.3 Replacing a Faulty Nodeto resolve the problem.

Other related questions:
In FusionInsight,How does Service Node Faulty impact on the system.
?The DataNode service is installed on service nodes by default. If less than three service nodes are lost, the service is not affected, but the service response time prolongs. You need to rectify the fault promptly. ?If more than three service nodes are lost, files stored in the Hadoop distributed file system (HDFS) may be lost and the service can be affected. You are advised to stop all service reading operations and rectify the fault immediately.

In FusionInsight,How to handle Management Node fault
1.If only one management node is faulty, handle the alarm based on the handling procedure of "ALM-12006 Node Fault." 2.If two management nodes are faulty but FusionInsight Manager can be accessed properly, the following alarms are reported:a.ALM-12006 Node Fault b.Active/standby Component Failover c.Service Unavailable Handle alarms a, b, and c in sequence. If the alarm persists after handling a and b,replacing a Faulty Node to resolve the problem. 3.If FusionInsight Manager cannot be accessed properly, contact the administrator to rectify hardware, network, and OS faults. If the fault persists, contact technical support; otherwise, go to 2.

Retaining online services of clients connected to a faulty node
The InfoEqualizer feature introduces dynamic IP addresses, which each corresponds to a dynamic domain name. If domain names are not resolved, when clients access a node using the node's IP address and the node becomes faulty later, this IP address can automatically "float" to another node.

FusionInsight node types
The FusionInsight system has three types of nodes: management, control, and data nodes.A management node is a node where FusionInsight Manager is installed. FusionInsight Manager is the management system of a FusionInsight HD cluster and provides a unified access portal to centrally manage nodes and services. A control node controls and monitors how data nodes store and receive data, and send process status, and provides other public functions. A data node executes instructions sent by management nodes, reports task status, stores data, and provides other public functions.

How Can I Handle NDMP Backup of a Node When I Delete the Node?
Before deleting the node, disable InfoEqualizer to allow the floating IP address of the node to be assigned to another node and then stop the NDMP service. There is no need to determine whether NDMP service is running or online. After the floating IP address floats to a node, the node takes over the backup or recovery services of the node to be deleted.

If you have more questions, you can seek help from following ways:
To iKnow To Live Chat
Scroll to top