Monitoring Azure Communication Services¶

Monitoring ensures your ACS application is healthy, messages are delivered, and communication latency is within acceptable limits.

graph TD
    ACS[ACS Resource] --> Metrics[Azure Metrics Explorer]
    ACS --> Logs[Log Analytics]
    ACS --> Events[Azure Event Grid]

    Metrics --> Alerts[Azure Monitor Alerts]
    Logs --> Dashboards[Azure Dashboards]
    Events --> Functions[Azure Functions Handler]

Azure Monitor Integration¶

ACS integrates with Azure Monitor to provide key metrics and diagnostic logs for troubleshooting.

Log Analytics Workspace Setup¶

1. Create a Log Analytics workspace.
2. In the Azure Portal, go to your ACS resource > Diagnostic settings.
3. Select "Add diagnostic setting" and choose your workspace.
4. Select the logs and metrics you want to collect (e.g., SMS, Email, Chat, Recording).

Alternatively, use Azure CLI:

# Create Log Analytics Workspace
az monitor log-analytics workspace create \
  --resource-group rg-acs-email-lab \
  --workspace-name law-acs-email-lab \
  --location koreacentral

# Create Diagnostic Settings for ACS resource
az monitor diagnostic-settings create \
  --name "acs-diag-all" \
  --resource "/subscriptions/{subscription-id}/resourceGroups/rg-acs-email-lab/providers/Microsoft.Communication/communicationServices/acs-email-lab" \
  --workspace "/subscriptions/{subscription-id}/resourceGroups/rg-acs-email-lab/providers/Microsoft.OperationalInsights/workspaces/law-acs-email-lab" \
  --logs '[{"categoryGroup":"allLogs","enabled":true}]' \
  --metrics '[{"category":"AllMetrics","enabled":true}]'

Key Metrics for ACS¶

Diagnostic Settings Configuration¶

To capture granular data, enable the following categories in Diagnostic settings:

• SMS logs: Detailed delivery and status information.
• Email logs: Delivery, bounce, and spam report tracking.
• Chat logs: Message events and participant updates.
• Calling logs: Call summary and call diagnostic details.

Verified Setup (April 2026)¶

The CLI commands shown in the Log Analytics Workspace Setup section above were used to provision the test environment.

Actual log table discovered: ACSEmailStatusUpdateOperational

Schema: | Column | Type | Description | |---|---|---| | TimeGenerated | datetime | Event timestamp | | CorrelationId | string | Maps to SDK message ID | | DeliveryStatus | string | "", "OutForDelivery", "Delivered", "Bounced", etc. | | SmtpStatusCode | string | SMTP response code | | EnhancedSmtpStatusCode | string | Extended SMTP code | | SenderDomain | string | Verified sender domain | | SenderUsername | string | Sender username (e.g., DoNotReply) | | RecipientMailServerHostName | string | Target mail server | | IsHardBounce | string | "True"/"False" | | FailureReason | string | Error category | | FailureMessage | string | Detailed error message |

Verified monitoring results: - Log ingestion delay: < 5 minutes from email send to Log Analytics availability - Retention: 30 days (default PerGB2018 tier) - All 9 test emails appeared in logs with full lifecycle tracking - Each email generates 3-4 log events (status transitions: "" → "OutForDelivery" → "Delivered")

Alert Rules and Action Groups¶

Set up alerts for critical thresholds:

• SMS Delivery Alert: Trigger when SmsDeliveryRate drops below 95%.
• Email Bounce Alert: Trigger when bounce rate exceeds 5%.
• Action Groups: Notify SRE teams via email, SMS, or webhook when an alert is fired.

See Also¶

• Monitoring ACS using Azure Monitor
• How to: Create diagnostic settings in Azure Monitor

Sources¶

• ACS Metrics Reference