ALM-12089 Inter-Node Network Is Abnormal

Alarm Description

The alarm module checks the network health status of nodes in the cluster every 10 seconds. This alarm is generated when the network between two nodes is unreachable or the network status is unstable.

This alarm is cleared when the network recovers.

Alarm Attributes

Alarm Parameters

Impact on the System

• Data transmission becomes slow or interrupted. Data may be lost or incomplete.
• Task scheduling is affected. For example, Yarn tasks cannot be executed properly or fail to be executed due to timeout.
• Data processing is affected. For example, HDFS data synchronization fails or the data is inaccurate.
• System performance deteriorates. The efficiency and quality of data processing is low.

Possible Causes

• The node breaks down.
• The network is faulty.

Handling Procedure

Check the network health status.

1. In the alarm list on FusionInsight Manager, locate the row that contains the alarm, click to view the description in additional information. Record the source IP address and destination IP address of the node for which the alarm is reported.
2. Log in to the node that the alarm is generated for. On the node, run ping the destination node to check whether the network between the two nodes is normal.

   For details about how to log in to a cluster node, see Logging In to an MRS Cluster Node.

   ping IP address of the destination node

   • If yes, go to 6.
   • If no, go to 3.

Check the node status.

3. On FusionInsight Manager, click Host and check whether the host list contains the faulty node to determine whether the faulty node has been removed from the cluster.
   • If yes, go to 5.
   • If no, go to 4.

4. Check whether the faulty node is powered off.
   • If yes, start the faulty node and go to Step 2.
   • If no, contact related personnel to find root cause. If you need to remove the faulty nodes from the cluster, go to Step 5. Otherwise, go to Step 6.

5. Remove the faulty node from the $NODE_AGENT_HOME/etc/agent/hosts.ini file on all nodes in the cluster, clear the /var/log/Bigdata/unreachable/unreachable_ip_info.log file, and clear the alarm.
6. Wait 30 seconds and check whether the alarm is automatically cleared.
   • If yes, no further action is required.
   • If no, go to 7.

Collect fault information.

7. On the FusionInsight Manager portal, choose O&M > Log > Download.
8. Select OmmAgent from the Service and click OK.
9. Click in the upper right corner to set the log collection time range. Generally, the time range is 10 seconds before and after the alarm generation time. Click Download.
10. Contact the O&M personnel and send the collected log information.

Alarm Clearance

After the fault is rectified, the system automatically clears this alarm.

Related Information

None