Using the server push (applet) method
Note:
The server push (applet) method is deprecated.
While this method is currently supported, IBM® reserves the right to remove
this function in a future release. You are advised to use the client
pull (AJAX) method instead.
There are two functions performed by the server
push (applet) implementation method, also known as the asynchronous
update applet:
- Pushing asynchronous host updates to the browser:
- This enables the user to see host updates automatically without the user clicking the Refresh button
- Notification of browser close to HATS server:
- This allows the HATS server to clean up resources associated with the browser session and reduces the load on the server
When using the asynchronous update applet, in addition
to the connection used to send transformed terminal screens from the
HATS server to the user's browser, a separate socket connection is
established between the applet running on the client and the HATS
server. This separate socket connection is used to pass small messages,
between the client and server, relating to asynchronous screen refreshes
from the host and notification to the HATS server that the browser
has been closed. Typically, the amount of traffic over this separate
socket connection is very small. The sequence of events which occurs
to set up this separate socket connection is as follows:
- The user enters the HATS application URL in the browser (using a link or typing in the address bar).
- The HATS server receives the request and sends a response to the browser, which includes the asynchronous update applet Java™ code.
- Once the applet is downloaded to the client, it attempts to communicate
directly back to the HATS server by opening a separate socket connection.
The local port used on the client for making this socket connection
is chosen at random from the available unused ports on the client,
and is not configurable. The remote HATS server IP name and port information
that the applet attempts to connect to are supplied by the server
when the applet is downloaded to the client. If a value of 0 is specified
for the server's port number in the HATS application configuration
on the server, then one random, unused port number above 1024 is chosen
on the HATS server and supplied to the applet as a parameter. The
port number that has been chosen by the server can be seen by viewing
the page source of the HATS transformation in the browser. Search
the page source for the value specified for
PARAM NAME="port". For more information about the applet parameter settings, see Server push (applet) settings.
In order to run the applet, a Java-enabled browser is needed. See "System Requirements
for Host Access Transformation Services" at https://www.ibm.com/support/pages/node/320035
for the current list of supported browsers.
Note:
The applet does not work with HATS portlets. For more information, see
WebSphere Portal and HATS.