This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Custom Alert - Oracle RMAN Backup Failed

Introduction

This alert will check the status of RMAN backups by looking at the V$RMAN_BACKUP_JOB_DETAILS tables and return any row from the last hour (based on frequency of the alert) where the status is 'COMPLETED WITH WARNINGS', 'COMPLETED WITH ERRORS', 'FAILED'.


ALERT DEFINITION

To create this alert, click on Alerts > Manage Alerts tab and create a Custom SQL Alert with Multiple Numeric Return type. Configure the alert similar to below:


Alert Name: RMAN Backups Failed

Execution Interval: 1 Hour

Notification Text: The following RMAN backup jobs failed within the last hour.

SQL Statement:

select 'Backup: ' || input_type || ' failed with status: ' || status, 1

from v$rman_backup_job_details

where end_time >= current_timestamp - #FREQUENCY#

and status in ('COMPLETED WITH WARNINGS', 'COMPLETED WITH ERRORS', 'FAILED')

Execute Against: Monitored Instance

Units: <blank>

High Threshold: MIN 1, MAX empty

Medium Threshold: <none>

  • There seems to be a problem with the math (Version 9.2.110)

    %FREQUENCY% appears to be in minutes.  The above code is subtracting 60 (1 hour) from the current timestamp which means in the last 60 days.

    Something like:

    where end_time >= current_timestamp - (1/(24*60) * #FREQUENCY#)

    should convert it to minutes and work as expected.

  • If that works, great. This alert was intended to be used with the DPA application and we *should* be converting the value used for interval for the #frequency# (it gets parsed by the java code and should be in the same increment as how the interval is defined for the alert). Let me know if that's not quite clear...