Installing the Splunk OTel Collector on an EC2 instance is not populating Splunk Observability Cloud with the expected EC2 resource attributes consistent with AWS EC2 metadata.
Attempts to query EC2 metadata from the affected instance result in the error: "Unable to connect to remote server". This indicates a potential routing issue and suggests verifying the EC2 instance's routing configuration following the migration.
User can see the following resource attributes in the Splunk OTel Collector agent logs:
-> cloud.provider: Str()
-> cloud.platform: Str()
-> cloud.region: Str()
-> cloud.account.id: Str()
-> cloud.availability_zone: Str()
-> host.id: Str()
-> host.image.id: Str()
-> host.type: Str()
-> host.name: Str()
The resource attributes user shared are consistent with AWS EC2 metadata. The EC2 detector collects the following resource attributes:
cloud.provider (Value: aws)
cloud.platform (Value: aws_ec2)
cloud.account.id
cloud.region
cloud.availability_zone
host.id
host.image.id
host.name
host.type
agent_config.yaml seems correctly configured for resource detection when compared to sample config here and sample config in Resource Detection Processor documentation here.
85 processors:
90 resourcedetection:
91 detectors: [gce, ecs, ec2, azure, system]
92 override: true
129 service:
131 pipelines:
132 traces:
137 - resourcedetection
138 - attributes
140 metrics:
142 processors: [memory_limiter, batch, resourcedetection, resourcedetection/system, attributes]
To investigate the resource detection errors, user is asked to provide the Splunk OTel Collector debug logs. Ensure the agent is restarted before collecting the logs. Attach the debug logs, including the support bundle, to the support case. Instructions for enabling debug logging and gathering support bundles on Windows can be found here.
This issue is not affecting other EC2 instances.
Resource attributes were not working previously.
The user initially encountered the issue with resource attributes not working after installing the Splunk OTel Collector.
When the affected EC2 instance attempts to query EC2 metadata, as described in this article, the user gets response: Unable to connect to remote server.
After user migrated to AWS, the route tables were not updated. This prevented the EC2 instance metadata from syncing with Splunk Observability Cloud.
User was able to resolve this issue by adding this route, and infrastructure metrics are now visible.
| Revision | Publish Date | Comments |
|---|---|---|
1.0 |
29-Sep-2026
|
Initial Release |