Skip to main content

​
Monitoring Modules

Apinizer Monitoring modules are used to monitor the general health status of the system, detect anomalies, and trigger actions when specified threshold values are exceeded.

API Health Check

Automatically checks whether APIs and web services are continuously running

Anomaly Detector

Automatically detects abnormal behaviors in API traffic

Alarm

Sends notifications when specified threshold values are exceeded

​
API Health Check

The API Health Check module enables monitoring the uptime status of the Proxy within the specified time interval using the Uptime Monitor. Actions can be triggered when expectations are not met.
  • Basic Purpose: Monitor uptime status and trigger actions
  • Configuration: Monitor Name, Description, Active status, and execution frequency (Trigger Every) are determined
  • Request Information: Method type and URL of the URL to be tested are entered
  • Assertions: Used to validate monitor results
  • Action: Selection of the action to be performed when the monitor is executed is provided
Used to validate monitor results:
  • Result Status Code: Default 200
  • Result Body: Response body check
  • XPath Result: XML response check
  • JsonPath Result: JSON response check

​
Anomaly Detector

The Anomaly Detector is a technique that enables finding unexpected situations or patterns by analyzing time-based data in log records. In the literature, these unexpected situations are called outliers or exceptions.
  • Definition: Detection of situations (anomalies) that do not conform to expected behaviors of data (e.g., unexpected spending increase)
  • Data Sources: Query created/selected to analyze log records and Filter information to be added to it are required
  • Conditions: Multiple conditions can be used to determine an anomaly, and these are connected with the ‘and’ operator
  • Action: Selection of the action to be performed when an anomaly is detected is provided
There are four different types of metric measurements:
  1. Metric Value Check: Situations such as whether the average value of the metric is above/below a threshold are checked
  2. Metric Increase/Decrease Rate Check: Comparison of the metric’s behavior (e.g., increase rate) relative to the previous value is specified
  3. EMA with Bollinger Bands Usage: EMA formula is applied on the selected metric value, and it is checked whether the result is above the Upper Bollinger Band value
  4. Query/Filter Rate Check: The ratio of the result from the filter and query at runtime to only the result from the filter is expressed as a percentage

​
Alarm

The Alarm module enables adding alarms that will occur when a specific condition is triggered. Alarm works based on a specified Threshold value.
  • Basic Purpose: Create warnings based on specified threshold value
  • Configuration: Name, Description, and Threshold value for the alarm are determined
  • Trigger Type: Alarm’s Trigger Type is selected
  • Execution Frequency: Alarm’s execution frequency is determined with the Trigger Every field
  • Actions: Selection of the action to be performed when the alarm is triggered is provided
  • Application Logs Count: Log Type must be selected
  • Elasticsearch Types: Connection Definition must be selected
  • Kubernetes: Pod and Node health status monitoring
  • Certificate: SSL and JWK certificate duration tracking