System Health and Program Use
As part of configuring a Workflow system, you indicate which servers should host which product options, and designate servers and product options as either essential or nonessential. These designations in combination with server status determine site health and affect the ability to log in and use this software.
If you designate a server as essential when configuring the system, all associated product options are automatically designated as essential. If you designate any product option as essential, the server that is hosting that option is automatically designated as essential, but the essential status of other product options hosted by the server remains unchanged.
Note: The servers hosting the Core and Workflow product options are designated as essential by default, and this designation cannot be changed.
The designation of servers as either essential or nonessential and the state and accessibility of servers has implications for the health of the system and, therefore, the ability to log in and use the Workflow program, as follows:
- When all servers, both essential and nonessential, are running and reachable, system health is considered Complete. In this scenario, the site health indicator in the toolbar is green, users can log in to the program, and all program and product option functionality is available.
- When one or more nonessential servers are not running or are unreachable, system health is considered Partial. In this scenario, the site health indicator in the toolbar is yellow, active users remain logged in to the program, and inactive users can log in, but program functionality tied to product options hosted on compromised servers is unavailable.Note: Specifically, related displays become inaccessible, binding to related external data faults, and related events stop firing. In addition, workflows that consume related data or monitor related events fail (making it a recommended practice to wrap such operations in fault handlers). Upon restoration of server availability, display inaccessibility and data-binding, event-firing, and workflow failure are automatically reversed. However, any work that was being performed using a display when a related server became unavailable may be lost.
- When one or more essential servers that do not host the Core product option are not running or are unreachable, system health is considered Waiting. In this scenario, the site health indicator in the toolbar is yellow, active users are logged out of the program, and inactive users cannot log in.Note: In this case, upon restoration of the essential server(s), users active at the time of interruption are automatically logged back in to the program, and functionality is restored to the level in accordance with the current health of the site.
- When the server that hosts the Core product option (typically, SOAServer) is not running or is unreachable, system health is considered Unavailable. In this scenario, the site health indicator in the toolbar is gray, active users are logged out of the program, and inactive users cannot log in. Note: In this case, upon restoration of the server that hosts the Core product option, users active at the time of interruption are automatically logged back in to the program, and functionality is restored to the level in accordance with the current health of the site.