Configure the Prometheus receiver to collect VAST DataStore metrics

Configure the Splunk Distribution of the OpenTelemetry Collector to send VAST DataStore metrics to Splunk Observability Cloud.

You can monitor the performance of your VAST DataStore by configuring the Splunk Distribution of the OpenTelemetry Collector to send VAST DataStore metrics to Splunk Observability Cloud.

This solution uses the Prometheus receiver to collect metrics from VAST DataStore, which exposes the Prometheus-compatible /metrics endpoint.

To configure the Prometheus receiver to collect VAST DataStore metrics, you must meet the following requirements.

  • A running VAST Cluster (VMS) with access credentials (username and password).

  • Network line-of-sight from the Splunk OpenTelemetry Collector pods to the VAST Management IP/VIP on port 443.

  • SSL certificates configured or proper tls_config exemptions applied if VAST uses a self-signed certificate.

  1. Deploy the Splunk Distribution of the OpenTelemetry Collector to your host or container platform:
  2. To manually activate the Prometheus receiver for VAST DataStore, make the following changes to your Collector values.yaml configuration file.
    1. Add prometheus/vast to the receivers section. For example:
      JSON 
      agent:
  config:
    receivers:
      prometheus/vast:
        config:
          scrape_configs:
          # 1. Base Cluster Performance
          - job_name: 'vast-datastore-core'
            scrape_interval: 15s 
            metrics_path: /api/prometheusmetrics/
            scheme: https
            tls_config:
              insecure_skip_verify: true
            basic_auth:
              username: '${VAST_USER}'
              password: '${VAST_PASSWORD}'
            static_configs:
            - targets: ['<VAST_MANAGEMENT_IP>:443']

          # 2. View Performance & QoS
          - job_name: 'vast-datastore-views'
            scrape_interval: 60s 
            metrics_path: /api/prometheusmetrics/views
            scheme: https
            tls_config:
              insecure_skip_verify: true
            basic_auth:
              username: '${VAST_USER}'
              password: '${VAST_PASSWORD}'
            static_configs:
            - targets: ['<VAST_MANAGEMENT_IP>:443']

          # 3. Storage Quotas & Limits
          - job_name: 'vast-datastore-quotas'
            scrape_interval: 60s 
            metrics_path: /api/prometheusmetrics/quotas
            scheme: https
            tls_config:
              insecure_skip_verify: true
            basic_auth:
              username: '${VAST_USER}'
              password: '${VAST_PASSWORD}'
            static_configs:
            - targets: ['<VAST_MANAGEMENT_IP>:443']

          # 4. Critical Hardware Alarms
          - job_name: 'vast-datastore-alarms'
            scrape_interval: 15s 
            metrics_path: /api/prometheusmetrics/alarms
            scheme: https
            tls_config:
              insecure_skip_verify: true
            basic_auth:
              username: '${VAST_USER}'
              password: '${VAST_PASSWORD}'
            static_configs:
            - targets: ['<VAST_MANAGEMENT_IP>:443']
    2. (Optional) Add metric paths to your Collector values.yaml configuration file:
      CODE 
      ##/api/prometheusmetrics/ – Core cluster performance, protocol totals (NFS, S3), and compute node hardware metrics.

##/api/prometheusmetrics/views – IOPS, throughput, latency, and data reduction ratios broken down by specific storage views.

##/api/prometheusmetrics/quotas – Capacity usage, inode counts, and threshold limit violations per configured storage path.

##/api/prometheusmetrics/alarms – Active, unresolved system faults and hardware alerts marked as a binary state.

##/api/prometheusmetrics/devices – NVMe SSD and NVRAM drive wear cycles, operating temperatures, and disk health metrics.

##/api/prometheusmetrics/vips – Network traffic balance, throughput speeds, and client load distribution across Virtual IPs.

##/api/prometheusmetrics/users – Specialized client storage performance tracking mapped to individual user IDs.

##/api/prometheusmetrics/all – Combined dump of all underlying resource paths; should be avoided to prevent performance overhead
    3. Add prometheus/vast to the metrics pipeline of the service section. For example:
      CODE 
      service:
  pipelines:
    metrics:
      receivers:
      - prometheus/vast
  3. Restart the Splunk Distribution of the OpenTelemetry Collector.

Configuration settings

To view the configuration options for the Prometheus receiver, see Settings.

Metrics

The following metrics are available for VAST DataStore. For more information, see Monitor AI LLM metrics in the VAST documentation.

These metrics are considered custom metrics in Splunk Observability Cloud.

Metric name Metric type Description
vast_cluster_read_iops Gauge Total read input/output operations per second for the cluster.
vast_cluster_write_iops Gauge Total write input/output operations per second for the cluster.
vast_cluster_read_bandwidth_bytes Gauge Total data volume read from the cluster per second, in bytes.
vast_cluster_write_bandwidth_bytes Gauge Total data volume written to the cluster per second, in bytes.
vast_cluster_read_latency_usec Gauge Average read latency experienced across cluster protocols, in microseconds.
vast_logical_space_in_use_bytes Gauge Total logical storage space consumed by workloads before data reduction.
vast_physical_space_in_use_bytes Gauge Total raw physical capacity consumed on disks after compression and de-duplication.
vast_cnode_cpu_utilization Gauge Percentage of CPU resource utilization across internal VAST Compute Nodes.
vast_alarm Gauge Binary state representation of active, unresolved cluster alerts (1 = Active, 0 = Resolved).
vast_quota_used_capacity Gauge Current capacity in bytes consumed by a configured logical path or project view.

Attributes

The following resource attributes are available for VAST.
Attribute name Description
storage.provider Identifies the vendor infrastructure source (Value: vastdata).
storage.layer Distinguishes between physical layer devices (devices) and logical namespace layouts (views, quotas).
cluster The specific human-readable namespace string identifying the target VAST storage cluster.
severity Captured exclusively from the /alarms endpoint to parse critical, major, or warning status in Splunk dashboards.
tenant The specific logical sub-tenant segment using the storage pool (useful for cloud service provider multi-tenancy).
view The file system path or S3 bucket namespace bound to a workload data pipeline.
uid User ID identifier matching target Active Directory/LDAP tracking records.