drop-observability-azure-research

Drop — Observability Stack on Azure (Research)

Client: Drop (Digital Banking) Type: Research — Azure equivalent of AWS observability plan Reference: ~/system/specs/drop-observability-plan.md (AWS version) Created: 2026-02-20

AWS → Azure Mapping

Azure-Specific Prednosti

1. Application Insights — Ubija 3 alata odjednom (APM + error tracking + metrics). Auto-instrument za Node.js. Zamjenjuje Sentry + Prometheus + dio Grafane.
2. Log Analytics Workspace — KQL query language (moćniji od CloudWatch Insights). Centralni log store.
3. Action Groups — Jedna konfiguracija za SVE alerte (email + SMS + Slack webhook + Teams + Azure Function).
4. Azure Workbooks — Besplatni interaktivni dashboardi (zamjena za Grafana ako ne treba multi-cloud).

Terraform (AzureRM)

# 1. Log Analytics Workspace
resource "azurerm_log_analytics_workspace" "main" {
  name                = "drop-logs"
  resource_group_name = azurerm_resource_group.main.name
  location            = azurerm_resource_group.main.location
  sku                 = "PerGB2018"    # Pay-as-you-go, 5GB/mo free
  retention_in_days   = 30
}

# 2. Application Insights
resource "azurerm_application_insights" "main" {
  name                = "drop-appinsights"
  resource_group_name = azurerm_resource_group.main.name
  location            = azurerm_resource_group.main.location
  workspace_id        = azurerm_log_analytics_workspace.main.id
  application_type    = "Node.JS"
}

# 3. Action Group (notifikacije)
resource "azurerm_monitor_action_group" "ops" {
  name                = "drop-ops-alerts"
  resource_group_name = azurerm_resource_group.main.name
  short_name          = "dropops"

  email_receiver {
    name          = "ceo"
    email_address = "[email protected]"
  }
  # Slack: webhook_receiver { ... service_uri = "https://hooks.slack.com/..." }
}

# 4. Alert Rules (primjeri)
resource "azurerm_monitor_metric_alert" "high_error_rate" {
  name                = "drop-high-error-rate"
  resource_group_name = azurerm_resource_group.main.name
  scopes              = [azurerm_application_insights.main.id]
  severity            = 1
  frequency           = "PT1M"
  window_size         = "PT5M"

  criteria {
    metric_namespace = "microsoft.insights/components"
    metric_name      = "requests/failed"
    aggregation      = "Count"
    operator         = "GreaterThan"
    threshold        = 5
  }
  action { action_group_id = azurerm_monitor_action_group.ops.id }
}

# 5. Availability Test (uptime monitoring)
resource "azurerm_application_insights_standard_web_test" "health" {
  name                    = "drop-health-check"
  resource_group_name     = azurerm_resource_group.main.name
  location                = azurerm_resource_group.main.location
  application_insights_id = azurerm_application_insights.main.id
  frequency               = 300  # 5 min
  enabled                 = true

  request { url = "https://drop.example.com/api/health" }
}

Koraci za implementaciju

1. Kreiraj Azure Resource Group za monitoring resurse
2. Deploy Log Analytics Workspace (Terraform)
3. Deploy Application Insights connected to Workspace (Terraform)
4. Dodaj APPLICATIONINSIGHTS_CONNECTION_STRING u app environment
5. npm install applicationinsights u app + appinsights.setup().start() na boot
6. Kreiraj Action Group sa email + Slack webhook (Terraform)
7. Dodaj Metric Alerts: error rate, response time p99, availability (Terraform)
8. Dodaj Availability Test za health endpoint (Terraform)
9. Opciono: BetterStack kao external monitor (isti setup kao AWS varijanta)
10. Verifikuj: App Insights prima telemetriju, alarmi rade, logi vidljivi u Log Analytics

Cost (Azure)

Reference

• AWS plan: ~/system/specs/drop-observability-plan.md
• Azure Monitor docs: https://learn.microsoft.com/en-us/azure/azure-monitor/
• Application Insights Node.js: https://learn.microsoft.com/en-us/azure/azure-monitor/app/nodejs
• Terraform AzureRM Monitor: https://registry.terraform.io/providers/hashicorp/azurerm/latest/docs/resources/monitor_metric_alert