4 Troubleshooting Scenarios

This section provides information on troubleshooting the common errors encountered during the deployment and creation of the OCI Adaptor infrastructure.

• Problem: The resource limit exceeded and the following error displayed: "Error: 400-LimitExceeded, The following service limits were exceeded: <OCI Resource name>. Request a service limit increase from the service limits page in the console".

  Solution: The user must request for the additional resources.

• Problem: The user encounters the following errors while running the Terraform: "Failed to read ssh private key: no key found" or "Failed to read ssh public key: no key found"

  Solution: Verify the format of the public and private keys and ensure there are no blank spaces either at the beginning or at the end of the keys.

• Problem: The user gets the following error: "NotAuthorizedOrNotFound, Authorization failed or requested resource not found"

  Solution: Verify the following:

  • The user is a tenency Admin.
  • To create the OCI infrastructure, it is recommended that first-level subcompartments be created in the root compartment. The creation of OCI infrastructure at the second-level subcompartment is not supported.

• Problem: At present, management agent has a non-configurable scraping interval of five minutes. By default all the metrics gets populated after five minutes.

  Solution:

  Follow the below steps to update the scraping interval to one minute:

  • Run the following command to enter the management agent's pod:

    kubectl exec -it <Management-agent pod name> bash -n oci-adaptor

  • Open the following file:

    /opt/oracle/mgmt_agent/agent_inst/config/destinations/OCI/services/Agent/1/types/PrometheusEmitter.json

  • Update the key interval associated with metricStream block to one minute, and save and close using :wq command.

    "metricStream":{
    	"metricEndpoint":"TelemetryStreaming",
    	"streamDescriptor":{
    	    "collectorId":"Prometheus",
    	    "credentialName":"RestCreds",
    	    "properties":[
    	       {  "name":"url", "scope":"INSTANCE", "content":"url"},
    	       {  "name":"streamNamespace","scope":"INSTANCE", "content":"namespace"},
    	       {  "name":"compartmentId", "scope": "INSTANCE", "content":"compartmentId","optional":true},
    	       {  "name":"additionalDimensions", "scope": "INSTANCE", "content":"metricDimensions","optional":true},
    	       {  "name":"allowMetrics", "scope": "INSTANCE", "content":"allowMetrics","optional":true},
    	       {  "name":"read-timeout","scope": "INSTANCE","content":"read-timeout","optional":true},
    	       {  "name":"connection-timeout","scope": "INSTANCE","content":"connection-timeout","optional":true},
    	       {  "name":"proxy-url","scope": "INSTANCE","content":"proxy-url","optional":true},
    	       {  "name":"dimensionFilterFilePath","scope":"INSTANCE","content": "dimensionFilterFilePath","optional":true},
    	       {  "name":"dimensionFilter","scope":"INSTANCE","content": "dimensionFilter","optional":true}
    	    ]
    	},
    	"schedule": {
    	"interval":1,
     	"timeUnit":"Min"
       }
    }

  • Exit from management agent pod.
  • Restart the management agent pod by deleting the existing management agent pod.

  Note:

  Updating the scraping interval using the above procedure updates the metrics data only for the Metric Namespace specified by the user while deploying the OCI Adaptor.