Understanding Docker deployment parameters

  • Mapping directories for data persistence

    In production environment, cfg, secure+, security and work directories inside the container should be mapped to directories on Storage Volume. It enables saving the content of these directories even if the container is destroyed. Use -v option at docker run command to complete this action.
    Note: cfg, secure+, security and work directories are created at fresh installation. Users must not create these directories on Storage Volume's mounted path.
  • Mapping ports

    IBM Connect:Direct for Unix client and server ports, default 1363 and 1364 respectively, can be mapped to available host ports. This is required to communicate with Connect:Direct for Unix nodes running on different hosts and to configure the node using Connect:Direct Web Services. Use -p option at docker run command to complete this action.

  • Adding Linux capabilities to the Docker container
    Following capabilities are required in a container for IBM Connect:Direct for Unix application to attempt a successful file transfer:
    • FOWNER
    • SETUID
    • SETGID
    • DAC_OVERRIDE
    • CHOWN
    • IPC_OWNER
    • IPC_LOCK
    • SYS_CHROOT
  • Container Name

    Container name must be unique for containers running on same host machine. This helps identify docker instances. Use --name option to set a container name.

  • Host Name

    Host name should be set for the container or container-id is set as hostname.

    Note: Hostname is required for container recovery. It is recommended you provide a hostname when IBM Connect:Direct for UNIX is deployed using a Docker container.
  • To transfer files present on host machines, user must map the source and destination paths inside the container.