Tibco Hawk Deployment Checklist and Rulebase Monitoring

Deployment Planning and Highlights of TIBCO Hawk

This Post provides several checklists for TIBCO Hawk deployment planning and highlights some of the best practices. Hawk is designed to monitor your applications locally. With the appropriate rulebase, you can start the application if it is not running. If your application writes status, warning, or error messages into a logfile, Hawk can parse the logfile and take appropriate actions.

Deployment CheckLists

1. Identify applications, systems, platforms, and networks to be monitored.
2. Identify pre-requisite software and operating system patches for Hawk (RV, JVM).
3. Verify availability and compatible versions of RV, Hawk, and JVM for the platforms required.
4. Review network topology and identify systems to run RVRD for multiple subnets.
5. Gather or review existing system and network administration policy, operation, and maintenance procedures.
6. Request and obtain the latest GA versions of RV, Tibco Hawk, JVM, documentation, and any required software license.
7. Determine strategy and steps for installation, configuration, verification, version control, upgrade, and maintenance.
8. Determine release content and deployment details based on esp. 1, 4, and 5.
9. Prepare a project plan with the above information plus budget, resource, and time estimates

Pre-Installation Checklist

1. Pre-requisites: RV, JVM, and patches.
2. Installation: Native install vs. untar; remote/mass installation/update using rdist/SMS.
3. Agent Configuration: autoconfig vs. repository; Hawk AMI RV session parameters; cluster names; repository names; logfile parameters;
4. Event Service Configuration: script for notification
5. RV Configuration: RVD, RVRD
6. Security Configuration: Trusted model – Trusted.txt, Agent/Display
7. Configuration objects: Rulebasemap, Schedule, and Rulebases
8. Auxiliary scripts, programs, utilities, files
9. Autostart configuration: hawk2; NT services.

Rulebase Monitoring

1. Process existence monitoring: determine and start one instance or n number of instances of an application.
2. Application log file parsing: pattern match for status, warning, or error messages on each new line and take actions accordingly.
3. Process resource monitoring: check for CPU usage, memory usage, file system usage, etc. with upper/lower bounds/thresholds.