Import a CPU and Memory Probe in Juniper Apstra for Apstra Cloud Services
SUMMARY Follow these steps to import a CPU and Memory probe in Juniper Apstra. Note that you must import a probe per blueprint on your Apstra Cloud Services instance for CPU and Memory metrics streaming.
-
From within your blueprint in the Juniper Apstra GUI, navigate to Analytics >
Probes > Create Probe > Import Probes.
Note:
Before proceeding, you must create a blueprint if you don't already have one. For more information, see Create Blueprint.
-
Upload a JSON file with the following content and click Import.
{ "label": "Stream CPU and Memory Utilization", "description": "Probe used to stream cpu and memory utilization from a blueprint in Apstra to the Cloud", "processors": [ { "name": "System cpu utilization data", "type": "extensible_data_collector", "properties": { "execution_count": "-1", "service_name": "resource_util", "query_expansion": {}, "service_interval": "10", "enable_streaming": true, "system_id": "system.system_id", "query_tag_filter": { "filter": {}, "operation": "and" }, "graph_query": "node('system', name='system', role=is_in(['leaf', 'access', 'spine', 'superspine']), deploy_mode=is_in(['deploy', 'drain']))", "keys": [ "metric" ], "query_group_by": [], "ingestion_filter": {}, "data_type": "number", "service_input": "''", "metric": "'system_cpu_utilization'" }, "inputs": {}, "outputs": { "out": "System cpu utilization data" } }, { "name": "System memory utilization data", "type": "extensible_data_collector", "properties": { "execution_count": "-1", "service_name": "resource_util", "query_expansion": {}, "service_interval": "10", "enable_streaming": true, "system_id": "system.system_id", "query_tag_filter": { "filter": {}, "operation": "and" }, "graph_query": "node('system', name='system', role=is_in(['leaf', 'access', 'spine', 'superspine']), deploy_mode=is_in(['deploy', 'drain']))", "keys": [ "metric" ], "query_group_by": [], "ingestion_filter": {}, "data_type": "number", "service_input": "''", "metric": "'system_memory_utilization'" }, "inputs": {}, "outputs": { "out": "System memory utilization data" } } ], "stages": [ { "description": "System memory utilization percentage", "enable_metric_logging": false, "retention_size": 0, "name": "System memory utilization data", "graph_annotation_properties": {}, "retention_duration": 2592000, "units": { "value": "%" } }, { "description": "System cpu utilization percentage", "enable_metric_logging": false, "retention_size": 0, "name": "System cpu utilization data", "graph_annotation_properties": {}, "retention_duration": 2592000, "units": { "value": "%" } } ] }The new probe appears in the Probes view.
The probe automatically starts streaming metrics data to the Juniper Apstra Edge. The Edge forwards that information for use by Apstra Cloud Services.The CPU and memory metrics should now be visible in the Apstra Cloud Services GUI in the Service Aware topology view. Select a topology device to view CPU and memory metrics.