How can we configure the sql_response probe to monitor the total QOS collected by the snmpcollector probe?
The goal is to generate sql_reponse probe alarms when the expected snmpcollector QOS drops.
1. Create an sql file using query below that returns a row count (total number of QOS present) for the last hour for snmpcollector and hub origin from the S_QOS_SNAPSHOT table.
Note: Replace ca_uim with your UIM database name and origin with your origin name
SELECT *
FROM [ca_uim].[dbo].[S_QOS_DATA] AS def
JOIN [ca_uim].[dbo].[S_QOS_SNAPSHOT] AS snap ON snap.table_id = def.table_id
WHERE [nim_origin] = 'origin'
AND snap.[sampletime] > dateadd(hour, -1, getdate())
AND probe = 'pollagent'
2. Save the *sql file in the sql_response probe home directory.
3. Deploy the sql_response to a robot.
4. Open the sql_response probe configuration in IM and add a new Connection for the UIM back-end database.
Example:
5. Click on the Profiles menu and add a new Profile.
6. Click on the General Tab and select the new Connection.
Example:
7. Click on SQL Query tab, click on the "From file" option and the *sql file name should display.
8. Click on the "Read File" option and the query should display. Ensure Test is successful.
Example of SQL Query:
9. Click on the Row count tab, Enable Alarm , QOS, Operator and Threshold (rows returned in the sql query).
Example
10. Save the profile and sql_response probe alarms should generate when the row count threshold is met.
As a test, you may want to set the threshold to a lower value to test the SQL query profile.
Alarm example: