KBI 310959 Slow Alerting And 200,000 Work Order Files

Version

Argent Advanced Technology all versions

Date

Tuesday, 10 June 2014

Summary

Environments with frequent Relator executions may experience delayed Alerts and over 200,000 files in the PENDING_EVENTS folder i.e. (X:\Argent\ArgentGuardianUltra\PENDING_EVENTS)

Each product has its own PENDING_EVENTS folder, troubleshooting the correct product is important

Easiest way to determine Alerts are being delayed apart from customer manually checking their mail is to review the “Events Tab” under “Events And Alerts” menu:

Event Time differs from Time Recorded:

Technical Background

Monitoring and saving ALL of the running processes in a Windows environment can be resource intensive to any system since the instances are dynamic and numerous

Argent tries to save each of the dynamic performance instance into the Database and this creates numerous Work Order files which cannot be processed in a timely fashion and backlog of the Work Order files starts

Illustration of a setting that will cause slow alerting (Check box “Save Performance Data Into The Argent Predictor” has been selected for ALL processes being monitored)

Resolution

Consider to use “Save Performance Data Into The Argent Predictor” only for named instances of the process being monitored i.e. mcshield or sqlservr or beremote but NOT all of the running instances like in the example above

A good illustration of hand-picking a process to monitor and saving the data into Argent Predictor:

General monitoring of ALL the processes is accepted as long as they’re not all being saved into Argent Predictor

Please review performance troubleshooting KBI:

KBI_310516