Hitachi

JP1 Version 12 JP1/Automatic Job Management System 3 Messages


KAVU2227-E

A connection error occurred during TCP/IP communication. (connection destination host name: destination-host-name, IP address: IP-address, port number: port-number, system error number: system-error-number)

A connection error occurred during TCP/IP communication.

(S)

Cancels the requested processing.

(O)

Probable causes are listed below. Check whether any of them following probable causes are present.

  1. JP1/AJS3 at the connection destination has not started yet.

  2. The port numbers used for submitting a job do not match between the manager and the host where the job was executed.

  3. You changed the default gateway while JP1/AJS3 was running.

  4. The IP address of the logical host manager and that of the agent belong to different networks.

  5. The network, including hardware, has a problem.

  6. No more socket ports can be used in the entire system.

  7. Memory was insufficient.

  8. For a cluster configuration, the system is set up so that the logical IP address is released before JP1/AJS3 stops when a failover occurs.

  9. A large number of job-start and job-end notifications are output in a short period.

Take one of the following actions according to the cause of the error, and then reissue the request.

  1. Start JP1/AJS3.

  2. If you changed the port number for JP1/AJS3 or if JP1/AJS3 links to another system such as JP1/NQSEXEC or JP1/OJE for VOS3, correct the port number in the services file.

  3. Before changing the network configuration, stop the JP1/AJS3 service.

  4. Correct the setting so that the IP addresses belong to the same network. Alternatively, see 2.3 Determining the network configuration in the JP1/Automatic Job Management System 3 System Design (Configuration) Guide to specify the setting for communication using multiple network addresses.

  5. Recheck the setting of the network including hardware. For example, execute the ping command to check whether communication is possible.

  6. Check the socket status and wait until no more sockets remains in the TIME_WAIT status. You can also prevent a communication error by reducing the recovery time for the TIME_WAIT ports managed by the OS.

  7. Check the memory status and re-estimate the memory requirements.

  8. Modify the setting so that the logical IP address is released after JP1/AJS3 has stopped.

  9. If there is no problem with the network, and this error occurred because a large number of jobs to be executed are concentrated in a short period, take action such as distributing the number of jobs to be executed during the period. Such an action will prevent a large number of job-start and job-end notifications from being output in a short period. Generally, this error might occur when 50 or more jobs are executed per second for the physical or logical host of the manager.

    To check the number of jobs that were executed in the short period during which a large number of job-start and job-end notifications were output, check the number of KAVS0263-I messages that were output to the scheduler log file during the corresponding period.