Troubleshoot cloud platforms

If you encounter trouble with AWS, Azure, or GCP cloud platforms data collection, try the following resolutions:

Data not displaying in SolarWinds Observability SaaS

Depending on the way your cloud platform or its services and resources were configured, some metrics may not be available. Use the Azure Resource Manager to gather metrics for an Azure resource. Metrics for an AWS service are only available if CloudWatch metrics have been enabled in the AWS Console.

The metrics available may also depend on the subscription pricing tier of your cloud platform or its services and resources. For some Azure metrics, the premium pricing tier would need to be selected when configuring your Azure namespaces. To gather additional AWS metrics, enable detailed CloudWatch metrics for your AWS services.

See Overview of Event Hubs Premium for details about the Premium pricing tier for the Event Hub resources, and Configure processing units for a premium tier Azure Event Hubs namespace for how to set up your Event Hub namespace with the premium tier.

The exact pricing tier requirements and instructions may vary by Azure resource.

This issue may also occur due to one or more configuration problems in the cloud account. Review the common causes and recommended solutions below.

Incorrect region selection during onboarding

Example: you select Region A, but you are trying to view metrics or entities for Region B.

Solution: Update the cloud account settings and include the correct region for monitoring.

Required services were not selected during onboarding

Example: You haven't enabled AWS Lambda in the services selection step, but you are attempting to view Lambda-related data in SolarWinds Observability SaaS.

Solution: Modify the cloud account configuration to enable the necessary services for monitoring.

Monitored cloud resources don't have any metrics

Every cloud resource must have at least one metric, otherwise SolarWinds Observability SaaS does not generate an entity. Make sure your cloud resources generate metric data.

Missing Azure permissions

In some cases, Azure metric data may not be displayed in SolarWinds Observability SaaS because of missing or incorrect permissions.

To resolve the issue:

Verify the Service Principal is assigned the Monitoring Reader role at the subscription scope.

To assign the role, execute the command az role assignment create --assignee $SP_OBJECT_ID --role "Monitoring Reader" --scope "/subscriptions/$SUBSCRIPTION_ID"
Verify App registration has a valid Client ID and a valid and active Secret/Certificate.
Verify Microsoft.Insights and Microsoft.Monitor providers are registered using commands:
- az provider show --namespace Microsoft.Insights --query registrationState
- az provider show --namespace Microsoft.Monitor --query registrationState
To register the providers, use the commands:
- az provider register --namespace Microsoft.Insights
- az provider register --namespace Microsoft.Monitor

Metric data points are not consistent, some entities are not showing up in SolarWinds Observability SaaS

This issue can be caused by multiple data collection or configuration-related factors. Review the possible causes and recommended solutions below to identify and resolve the problem.

AWS data collection stopped due to throttling

There is a specific limit set for the number of API calls sent to AWS accounts to fetch data. When API thresholds are crossed, AWS creates a throttling exception and the AWS poller stops fetching data for the resource that throttled for the next ten minutes. This provides enough time for throttling situations to settle down. However, if you have too many resources to be monitored in a specific region, or if you have integrated your AWS account with multiple vendors, you might want to consider increasing your API request quota.

Create custom dashboards to view the API call and throttling rates

The steps below describe how to create a dashboard to view the number of CloudWatch polling calls made to your AWS account against the number of calls that are throttled by AWS. This helps you understand the limits, may help avoiding future throttling issues, or provide a clear justification for the API quota increase request.

In the AWS Management Console, click CloudWatch.
In the left pane, select Dashboards.
Click Create dashboard in the upper right.
Enter the dashboard name and click Create dashboard.
Under Data sources types - new, select CloudWatch. Under Widget type, select Line. Click Next.
Under Metrics, type GetMetricData in the search field. Click Usage > By AWS Resource, and select both metrics listed.
Open the Graphed metrics tab. For both the ThrottleCount and the CallCount metrics, select Sum in the Statistic column. Click Create widget.
If necessary, resize the metric tile and click Save in the top-right corner.

Hovering over the line chart on the dashboard provides information on the call count and throttling rate.

Request a service limit increase

Use the instructions below to request a service limit increase for your API request quota.

In the AWS Management Console, click Service Quotas > Dashboard.
Search for CloudWatch to find the corresponding quotas.
Select the quota you want to increase. For example, GetMetricData API requests.
Click Request quota increase.
Fill out the form specifying your new limit and the justification for the increase.

AWS IAM policy is outdated

This section explains how to proceed when SolarWinds Observability SaaS displays the following error message about outdated IAM policies.

The error message is displayed when:

You modify the IAM policy by deleting some permissions.
The IAM policy permissions are modified from the SolarWinds Observability side.
New IAM policy permissions are added on the SolarWinds Observability side.

To update policies and resolve the issue, you can copy the policy permissions list provided in SolarWinds Observability and paste it in the AWS Management Console.

In SolarWinds Observability, click Settings > Cloud Accounts > AWS.

The accounts with outdated policies show Policy Outdated in the Status column and Update Policy in the Status Description column.
Click Update Policy.
In the Update Policy wizard, copy the provided policy by clicking Copy to clipboard, and use it to update the policy in the AWS Management Console. For details, follow the instructions provided in the Update Policy wizard.
After you update the policy in the AWS Management Console, return to SolarWinds Observability and click Confirm updated in the Update Policy wizard.

If you do not confirm the update in SolarWinds Observability, the Status will show Policy Outdated and you will continue to lose data even if you have updated the policy in the AWS Management Console.

Cloud account status displays error state

This issue indicates that the cloud account integration is not functioning correctly, typically due to configuration or permission problems. Review the common causes and recommended solutions below.

Expired permissions

This issue may be caused by expired API token or client credentials.

To resolve the issue, regenerate API token or client credentials, delete the old integrations, and integrate the accounts with newly generated token or client credentials.

Insufficient resources in cloud account

In some cases, IAM roles or policies associated with the integration are deleted from the cloud account, preventing SolarWinds Observability SaaS from connecting to the account.

To resolve the issue, delete the affected integration in SolarWinds Observability SaaS and recreate it.

Metrics data of some resources doesn’t match with cloud account metrics

Some cloud resources include dimensions, and their metrics are generated with those dimensions. When SolarWinds Observability SaaS displays aggregated metric data, the values may not match those shown in the cloud provider’s console.

To view accurate metric data points, use the “Group By Attribute” option in SolarWinds Observability SaaS. This allows metrics to be grouped based on specific dimensions, ensuring consistency with the cloud console data.

IP addresses shown in AWS or Azure logs

Some AWS or Azure logs may show the IP addresses instead of resource names. This occurs because the current log ingestion design captures IP addresses only and does not resolve them to resource names.

Displaying resource names instead of IP addresses is not supported at this time. An enhancement to add resource name resolution is planned.

Troubleshoot issues related to Azure Regions

Issues related to Azure regions typically result from missing permissions or region-specific limitations. Review the following use cases and solutions to resolve region visibility and access problems.

Service Principal missing Monitoring Reader role

Make sure you assign the Monitoring Reader role to the Service Principal at the subscription scope.

To assign the role, execute the following command:

az role assignment create --assignee $SP_OBJECT_ID --role "Monitoring Reader" --scope "/subscriptions/$SUBSCRIPTION_ID"

Azure regions are not visible in the Add Azure Cloud Account wizard

Azure regions may not appear during the SolarWinds Observability SaaS onboarding process.

Make sure at least one resource is deployed and running in the required region.

This is a current limitation in SolarWinds Observability SaaS. Improvements are planned in a future release.

Missing permissions errors in the Add GCP Cloud Account wizard

This issue can be caused by several permission-related configuration problems. Review the possible causes and recommended solutions below to resolve onboarding errors.

SolarWinds Observability SaaS displays `IAM bindings missing` error when you select projects in the Add/Edit GCP Cloud Account wizard

This error is displayed when the GCP service account used for onboarding doesn't have sufficient permissions to list or access projects. This means the service account lacks the Viewer role (roles/viewer) IAM binding on the target GCP project or organization.