#================================================================================

#================================================================================

Git repo for CSM automated regression testing

1. Create a file named "csm_fvt.csm_test.cfg" based on the csm_test.cfg.template file shipped with the CAST repo in the CAST/csm_fvt/ directory
2. Run tools/csm_uninstall.sh to uninstall any rpms currently installed on the nodes specified in the csm_fvt.csm_test.cfg file
3. Run tools/csm_install.sh to install the RPMs specificed in the csm_fvt.csm_test.cfg file on to the appropriate nodes as specified in the csm_fvt.csm_test.cfg file

The LOG_PATH parameter of csm_fvt.csm_test.cfg will indicate the parent results directory. Within that directory, the following subdirectory structure is required buckets/ buckets/basic/ buckets/advanced/ buckets/error_injection/ buckets/timing/ buckets/BDS/ setup/

The sql files in the CAST/csm_fvt/include/sql/ directory are used to insert dummy data in to the database for several regression buckets. These files must be copied to a postgres user accessible directory, specified by the SQL_DIR parameter in the csm_fvt.csm_test.cfg file. Additionally ssd.sql must be editted so that node_name is the same as the SINGLE_COMPUTE parameter specified in the csm_fvt.csm_test.cfg file.

Running tools/complete_fvt.sh will start a complete regression run. Additionally, individual buckets can be run by simply running buckets//.sh. NOTE: Certain buckets are currently dependent on the exit conditions of other buckets, which can lead to false-negative failures. Refer to the tools/complete_fvt.sh script for the "correct" ordering of buckets