Rule Monitoring
- Ben Clark
Alterian CX is central to the success of CX and marketing campaigns run by our customers. When Rules are not working correctly your customers are getting a lessened CX experience with your brands.
Rule health monitoring is already available in the Rules screen as a visual check as you work.
If a Rule fails when users are not in the Alterian CX you still want to be able to quickly identify the issue to expedite resolution.
The Rule monitoring Public Template and Rule provides this functionality, allowing you to build notifications around Rules when they fall below previously configured Health thresholds.
A Rule Monitoring Rule should be run in each Client and Target where you want alerts generated for Rules running within them.
For most customers this would be the Production Target only with Rules in Development and Test not being monitored.
Rule Montoring runs at the same cadence as the Health Score checks and therefore will run and once per hour once configured.
A method for notification should be linked to the Notify Output. Examples would be text message sends via Sendgrid or emails sent through Email Manager or SES. The output values like current health score can add details to the messages to indicate the severity of the issue.
Input Data
Field | Description | Example |
|---|---|---|
N/A |
|
|
Configuration
The configuration items set the threshold to alert and the frequency.
Parameter | Description | Example |
|---|---|---|
Minutes Between Notifications | When we find a Rule that had failed the prescribed threshold test, we will alert. This parameter allows you to configure how often the alert is triggered while the Rule is in an alerting state. 0 is the default and will notify just once when the alert is first triggered. To add further notifications, add a number that indicates the number of minutes you would like between notifications, 120 is two hours for example. Be aware that these notifications will continue until the Rule is no longer in a failed state and therefore will notify day and night if it is not resolved. | 0 = default 120 = every two hours |
Monitoring Threshold Percentage | The failure threshold at which to start to notify | 0.5 Default equates to 0.5% |
Rule Monitoring Table Name | Where the notification stats are stored | rule_monitoring_stats |
Rule Monitoring | Required Parameter to mark rule as a monitoring rule |
|
Processing
This tile template will Provides an output on an hourly basis highlighting whether any of your Rules have fallen below the previously configured Monitoring Threshold Percentage.
Output Data
Field | Description | Example |
|---|---|---|
errorCount | Number of Errors over the monitoring period | 30 |
healthScore | Current Health Score for the Rule | 99.8 |
invokeCount | Number of Invokes over the monitoring period | 567 |
ruleName | Name of the rule |
|
totalTimeouts | Total number of Timeouts over the monitoring period | 77 |
monitoringThresholdPrct | Threshold set in the Parameters screen highlighting the failure level as a where you would like to start notifying | 0.5 |
healthPercentage | Current health percentage for this Rule | 99.3 |
thresholdPrct | Threshold set in the Parameters screen highlighting the failure level as a percentage where you would like to start notifying | 0.5 |
|
|
|