cancel
Showing results for 
Search instead for 
Did you mean: 

Thresholds that monitor Database Collections for Prognosis Reporting Advanced

Anon
Community Manager

Thresholds that monitor Database Collections for Prognosis Reporting Advanced

Threshold to monitor BDI odbc database collections.
A pre-Prognosis 11 threshold for BDI solution is replaced with the new one. The old threshold is removed from the GUI packaging where it was residing on a permanent basis.
A new threshold is created on the fly during the BDI Setup operation. The Setup uses a template 'Template_BDI_Alert.pm' (found under the 'Setup' folder) and the set of the odbc collections associated with a model to generate and to start up the Threshold.
The threshold is named after the deployed model, e.g. BDIUC for 'UC' model. It monitors the ETL job status by polling the 'AutomatedSummarySchedulerLog' record every minute and the set of odbc database collections by polling each of the odbc process every 10 seconds.
If the automated summary log has a 'Failed' row for the automated summary job then the threshold:
raised an Error Alert
sends an SNMP Trap (destination host needs to be configured)
sends a message to the Dispatch Manager (dispatch manager profile needs to be configured)
If a process is not found then the threshold:
raises a Critical Alert
sends an SNMP Trap (destination host needs to be configured)
sends a message to the Dispatch Manager (dispatch manager profile needs to be configured)
attempts to re-start the collection
NB! this new functionality of re-starting an absent odbc database collection needs to be taken into account during the upgrade or manual re-setting of the deployed solution.
NB! also keep in mind that the threshold rule that re-starts a db collection uses the original db start script cached under PROGNOSIS_HOME\Server\Configuration\BigDataInsight\MODEL_DEPLOYMENT_DIR\PrognosisScripts\backup. It might create a pitfall when you try manually setup dbi db collections and the corresponding threshold.​
Tags (1)