Alerts

The Alerts page in ADOC is your central hub for tracking issues across pipelines, compute, and data reliability. An alert is automatically generated whenever a monitoring rule or policy condition is breached. These alerts allow you to catch problems early, before they impact operations, budgets, or decision-making.

By reviewing alerts, you can:

Detect pipeline failures or performance slowdowns in real time.
Identify data quality issues such as missing values, schema changes, or anomalies.
Monitor compute usage and costs to avoid overruns.
Maintain a complete audit trail of all issues across your environment.

Alerts vs Notifications

While closely related, alerts and notifications serve different purposes in ADOC:

Alert:
- An event created every time a rule is triggered.
- Logged in ADOC to ensure a full audit trail.
- Examples: A pipeline fails, sales data arrives late, or compute costs exceed budget.
Notification:
- The message that informs you about an alert.
- Delivered via Slack, Teams, email, or other channels.
- Configurable—so you can choose when and how often you’re notified.
- Example: “Notify me only the first time a freshness policy fails.”

This separation ensures that all issues are tracked, but teams only get the right notifications at the right time, reducing noise while improving accountability.

How Alerts are Created in ADOC

You don’t create alerts directly. Alerts are automatically generated whenever a monitoring rule or policy condition is breached. This could be from:

Policies (data quality, freshness, anomaly, drift, reconciliation)
Pipelines (failures, SLA breaches, long runtimes)
Compute/Budgets (cost thresholds, warehouse usage, suspend times)

Example: Creating a Data Quality Alert

Go to Data Reliability > Manage Policies and click Add New.
Choose Data Quality and set up your rule (e.g., email IS NOT NULL).
In the Alerts & Notifications step:
1. Pick a severity level (Critical, High, Medium, Low).
2. Choose a Notification Group (e.g., Slack, Teams, Email).
Save the policy and run it. If the rule fails, ADOC creates an alert and sends a notification to your chosen channel.

Types of Alerts

ADOC generates alerts across three key domains:

Pipeline Alerts
- Track execution and performance of pipelines.
- Example: Alerts for failed jobs, long runtimes, or critical errors.
Reliability Alerts
- Triggered by Data Reliability policies to ensure trustworthy data.
- Example: Alerts for duplicate rows, missing values, or unusual statistical patterns.
Compute Alerts
- Monitor usage, health, and costs of compute environments.
- Example: Alerts for warehouse suspends, high resource utilization, or budget threshold breaches.

Alert Lifecycle

Each alert progresses through a lifecycle, indicated by its status. Understanding these statuses helps your team coordinate effectively and maintain accountability.

Status	Description	How It's Set
Open	Default state of a newly generated alert. The issue has been detected but not yet reviewed.	Automatically when a monitoring condition is breached.
Acknowledged	Indicates the alert has been reviewed and assigned for follow-up.	Manually by a user.
In Progress	Work to investigate or resolve the issue is actively underway.	Manually by a user.
Dismissed	The alert does not require action (e.g., known issue, low priority, or automatically cleared after a monitor update).	Manually by a user, or automatically by ADOC if the parent monitor changes.
Resolved	The issue has been fixed or the condition is no longer met.	Automatically if the monitor rerun succeeds, or manually by a user once a fix is applied.

Alert Lists

The Listing tab on the Alerts page provides a complete, filterable table of all alerts generated in ADOC. This is your central workspace for investigating, assigning, and managing alerts.

Finding and Filtering Alerts

The toolbar at the top of the page makes it easy to narrow down alerts and focus on what matters most:

Filter	Description	Example
Date Range	By default, alerts from the last 7 days are shown. Click the calendar icon to adjust the time frame.	Last 30 days
Datasource	Limit alerts to a specific data source.	Snowflake, MongoDB
Type	Filter by where the alert originated.	Pipeline, Compute, Reliability
Status	Show alerts by lifecycle state.	Show only alerts in a certain state, such as Open or Resolved.
Severity	Focus on alerts by priority level.	Critical, High, Medium, Low
Assignee	View alerts assigned to a particular user.	User A
Search	Find alerts by name.	“ETL Failure”
Refresh	Reload alerts with latest data without clearing filters.
Reset	Clear all filters and return to default view.	Default 7-day view

Alerts List Details

Each row in the Alerts List represents a single alert, providing quick visibility into its context:

Column	Description
Name	The alert name, often tied to the policy or pipeline that generated it.
Severity	Assigned impact level (Critical, High, Medium, Low).
Type	The alert’s category (e.g., Pipeline, Reliability, Compute).
Status	Where the alert is in its lifecycle (Open or Resolved).
Raised	Date and time the alert was first triggered.
Tags	Any custom tags used for categorization.
Occurrence Count	How many times the same alert has been triggered.
Assignee	The user or team responsible for resolving it.
Updated By	The user who last changed the alert’s state.
Last Updated At	Timestamp of the most recent change.

Last updated on

Was this page helpful?