TIBCO Deployment

Have two deployment methods i.e One-Click deployment and Scripting deployment

These are the general steps in process.

1. Create Project.
2. Build EAR from project.
3. Save EAR to Disk.
4. Upload EAR via HTTP.

• Domain Configuration
• Application Runtime RepoInstance

5. Configure Application
6. Deploy Application
7. Contact TSM for Deploy with Hawk
8. Execute Deployment Plugin
9. Save .tra file to disk (substituting from template .tra)
10. Contact TSM for Starting Engines with Hawk
11. Launch Engine
12. BW/Adapter reads configuration with RV or HTTP(s).
13. Monitor Application.
14. Gather stats with Hawk
15. Return runtime stats.


steps 5-11 are “One-Click Deployment” 

Command-Line Deployment

This is Full command-line deployment (scripting deployment) and available in Add-on pack on top of BW 5.x. and also Redeployment of an existing application is supported via command line.

However, with BW 5.x, can manually generate scripts to create an automated deployment process and can use  prepareDeployment.exe, redeploy.exe and deployment.xml, tibco.xml (global variables)


Redeployment

When a new archive is loaded, a diff of the new archive with the existing
archive is performed.New services are added, services not present in the new archive are removed.
 

If the shared configuration in the application archive or application level deployment descriptors are changed, all deployed service instances will be stopped (whether they are affected by the new archive or not), the repo instance is updated, and the client side .tra files are regenerated.
 

If the binary content of the service archive or the service level deployment descriptors are changed, then ONLY the affected service instances are stopped, and the client side .tra files for those affected service instances are regenerated.

The following configuration items must be performed in the Administrator:
– Binding services to machines.
– Hawk rulebase uploading.
– Service instance configuration (e.g. JVM properties).
– Fault-Tolerance.