Other hardware and network issues

This topic covers other potential hardware and network issues, excluding node failure, network jitter, and ODP failure. It also provides the corresponding emergency handling procedure.

Emergency handling procedure

In addition to common issues like node failure, network jitter, and ODP failure, there are other hardware and network issues, such as disk I/O exceptions, disk recognition failures, network card exceptions, and node CPU/memory exceptions. These issues often result in increased response time (RT) and transaction commit failures.

• For any suspected hardware issues, you can start by running the stop server command. This is a safe operation without data loss and is the most effective method for immediate fault isolation. However, please note that this operation will cause an error when a client attempts to access the node. If possible, run the isolate server command first. Here is an example:

  ALTER SYSTEM ISOLATE SERVER 'xxx.xxx.xxx.xxx:2882';
  ALTER SYSTEM  STOP SERVER 'xxx.xxx.xxx.xxx:2882';
  

• For any hardware issues, simply replace the faulty server.