Version

Argent Predictor 8.0A-0701 and below

Date

29 Mar 2007

Summary

An error exists in the logic used for generating the SLA Up Down Reports

Technical Background

An issue exists with the logic employed for the SLA UPDOWN TIME REPORTS in ALL Argent XT Products.

The following use case applies:

• For prolonged entries of metric=0 in the PRD_SLAVE tables, the Argent Predictor locates the last date & time when the metric=1.

• This causes the report to include a date & time which is significantly outside the specified date and time range (in the Definition Tab in the XT GUI).

• This causes uptimes of negative numbers or extremely high numbers in excess of 100%.

Screenshots as follows:

ARGENT_SAMPLE -> master_key=2

An example of a report received by the Customer is as follows:

The issue is magnified since they have been collecting SLA Downs (metric=0) due to Remote Monitoring Engines failing for an extended period of time.

Resolution

Issue Report Submitted under Ticket AAS497