Alerting with Prometheus Alertmanager in Kubernetes - Mini Project: Build & Apply

Alerting with Prometheus Alertmanager

📖 Scenario: You are managing a Kubernetes cluster and want to monitor the health of your applications. You will set up Prometheus Alertmanager to send alerts when your application metrics cross certain thresholds.

🎯 Goal: Build a simple Prometheus alerting rule and configure Alertmanager to send notifications when the alert fires.

📋 What You'll Learn

Create a Prometheus alerting rule for high CPU usage

Configure Alertmanager with a basic alert receiver

Connect Prometheus to Alertmanager

Test the alert and display the alert firing message

💡 Why This Matters

🌍 Real World

Monitoring applications in Kubernetes clusters is critical to detect issues early. Prometheus and Alertmanager help automate alerting to notify teams.

💼 Career

DevOps engineers and SREs use Prometheus and Alertmanager daily to maintain system reliability and respond quickly to problems.

1

Create Prometheus Alerting Rule

Create a file called cpu_alert_rule.yaml with a Prometheus alerting rule named HighCpuUsage that triggers when rate(container_cpu_usage_seconds_total[5m]) > 0.5 for 5 minutes.

Kubernetes

# Your code here

Hint

Use the alert, expr, and for fields inside a rules list.

2

Configure Alertmanager Receiver

Create a file called alertmanager.yaml with a receiver named team-email that sends alerts to the email address [e-mail].

Kubernetes

groups:
  - name: example
    rules:
      - alert: HighCpuUsage
        expr: rate(container_cpu_usage_seconds_total[5m]) > 0.5
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: "High CPU usage detected"
          description: "CPU usage is above 0.5 for more than 5 minutes."

# Your code here

Hint

Define receivers with a name and email_configs including the to email address.

3

Connect Prometheus to Alertmanager

Add the Alertmanager configuration to Prometheus by creating a prometheus.yaml file that includes the Alertmanager URL http://alertmanager:9093 under the alerting section.

Kubernetes

groups:
  - name: example
    rules:
      - alert: HighCpuUsage
        expr: rate(container_cpu_usage_seconds_total[5m]) > 0.5
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: "High CPU usage detected"
          description: "CPU usage is above 0.5 for more than 5 minutes."

receivers:
  - name: team-email
    email_configs:
      - to: [e-mail]

# Your code here

Hint

Under alerting, add alertmanagers with static_configs and targets listing the Alertmanager URL.

4

Test Alert and Display Output

Print the message Alert fired: HighCpuUsage to simulate the alert firing.

Kubernetes

groups:
  - name: example
    rules:
      - alert: HighCpuUsage
        expr: rate(container_cpu_usage_seconds_total[5m]) > 0.5
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: "High CPU usage detected"
          description: "CPU usage is above 0.5 for more than 5 minutes."

receivers:
  - name: team-email
    email_configs:
      - to: [e-mail]

alerting:
  alertmanagers:
    - static_configs:
        - targets:
            - 'alertmanager:9093'

# Your code here

Hint

Use a print statement to show the alert firing message exactly as given.

Practice

1. What is the main role of Prometheus Alertmanager in Kubernetes monitoring?

easy

A. To collect metrics from Kubernetes nodes

B. To send notifications when Prometheus detects alerts

C. To store logs from containers

D. To deploy applications automatically

Solution

1. Step 1: Understand Prometheus and Alertmanager roles

   Prometheus collects metrics and detects alerts based on rules.

2. Step 2: Identify Alertmanager's function

   Alertmanager receives alerts from Prometheus and sends notifications to users or systems.

3. Final Answer:

   To send notifications when Prometheus detects alerts -> Option B

4. Quick Check:

   Alertmanager = Notification sender [OK]

Hint: Alertmanager handles alert notifications, not metric collection [OK]

Common Mistakes:

• Confusing Alertmanager with Prometheus server
• Thinking Alertmanager collects metrics
• Assuming Alertmanager deploys apps

2. Which of the following is the correct YAML snippet to define an email receiver named 'team-email' in Alertmanager?

easy

A. receivers:
- name: team-email
email_configs:
- to: '[e-mail]'

B. receivers:
- team-email:
email: '[e-mail]'

C. receiver:
name: team-email
email: '[e-mail]'

D. receivers:
- name: team-email
slack_configs:
- channel: '#alerts'

Solution

1. Step 1: Review Alertmanager receiver syntax

   Receivers are defined under 'receivers' list with 'name' and config type like 'email_configs'.

2. Step 2: Match correct YAML structure

   receivers: - name: team-email email_configs: - to: '[e-mail]' correctly uses 'receivers', 'name', and 'email_configs' with 'to' field.

3. Final Answer:

   Correct YAML with 'receivers', 'name', and 'email_configs' -> Option A

4. Quick Check:

   Receiver YAML uses 'name' and 'email_configs' [OK]

Hint: Receiver configs use 'name' and specific config like 'email_configs' [OK]

Common Mistakes:

• Using 'receiver' instead of 'receivers'
• Incorrect nesting of email fields
• Confusing slack_configs with email_configs

3. Given this Alertmanager config snippet, what will happen when multiple alerts fire simultaneously?

route:
  group_by: ['alertname']
  receiver: 'team-email'
receivers:
  - name: 'team-email'
    email_configs:
      - to: '[e-mail]'

medium

A. Alerts with the same 'alertname' will be grouped into one notification

B. Each alert will send a separate email regardless of grouping

C. No alerts will be sent because 'group_wait' is missing

D. Alerts will be sent only to Slack, not email

Solution

1. Step 1: Understand 'group_by' in Alertmanager route

   'group_by' groups alerts by specified labels; here, alerts with same 'alertname' are grouped.

2. Step 2: Check receiver and notification method

   Receiver 'team-email' uses email_configs, so grouped alerts send one email per alertname.

3. Final Answer:

   Alerts with the same 'alertname' will be grouped into one notification -> Option A

4. Quick Check:

   'group_by' controls alert grouping [OK]

Hint: 'group_by' label controls alert grouping in notifications [OK]

Common Mistakes:

• Assuming each alert sends separate email
• Thinking 'group_wait' is required to send alerts
• Confusing receiver type with Slack

4. You configured Alertmanager but no notifications are sent. Which of these is a likely cause based on this snippet?

receivers:
  - name: 'team-email'
    email_configs:
      - to: '[e-mail]'
route:
  receiver: 'team-email'
  group_by: ['alertname']
  group_wait: 30s
  group_interval: 5m
  repeat_interval: 1h

medium

A. Alertmanager does not support email notifications

B. Incorrect 'group_by' label causes no alerts

C. Receiver name does not match route receiver

D. Missing SMTP server configuration in Alertmanager

Solution

1. Step 1: Check email notification requirements

   Email notifications require SMTP server settings in Alertmanager config, not shown here.

2. Step 2: Verify receiver and route match

   Receiver name 'team-email' matches route receiver, so routing is correct.

3. Final Answer:

   Missing SMTP server configuration in Alertmanager -> Option D

4. Quick Check:

   Email needs SMTP setup to send alerts [OK]

Hint: Email alerts need SMTP server configured in Alertmanager [OK]

Common Mistakes:

• Assuming 'group_by' label stops alerts
• Thinking receiver name mismatch causes no alerts here
• Believing Alertmanager can't send emails

5. You want to avoid alert spam by grouping alerts by both 'alertname' and 'severity', and send notifications to Slack channel '#alerts'. Which Alertmanager route and receiver config is correct?

hard

A. route:
group_by: ['severity']
receiver: 'email-team'
receivers:
- name: 'email-team'
slack_configs:
- channel: '#alerts'

B. route:
group_by: ['alertname']
receiver: 'slack-notifications'
receivers:
- name: 'slack-notifications'
email_configs:
- to: '#alerts'

C. route:
group_by: ['alertname', 'severity']
receiver: 'slack-notifications'
receivers:
- name: 'slack-notifications'
slack_configs:
- channel: '#alerts'
send_resolved: true

D. route:
group_by: ['alertname', 'severity']
receiver: 'email-team'
receivers:
- name: 'email-team'
email_configs:
- to: '[e-mail]'

Solution

1. Step 1: Set grouping labels in route

   To group alerts by 'alertname' and 'severity', list both in 'group_by'.

2. Step 2: Configure Slack receiver correctly

   Receiver named 'slack-notifications' uses 'slack_configs' with channel '#alerts' and 'send_resolved' true.

3. Final Answer:

   Route groups by alertname and severity; receiver sends Slack messages to #alerts -> Option C

4. Quick Check:

   Group by multiple labels and use correct receiver config [OK]

Hint: Group by multiple labels and match receiver type to notification [OK]

Common Mistakes:

• Using email_configs for Slack notifications
• Grouping by only one label when two needed
• Mismatch between route receiver and receiver name