Metrics and Dashboards

Explore this Page

Overview

This page lists the Prometheus metrics the exporter publishes, and describes the prebuilt Grafana dashboards used to visualize them.

Metrics Reference

All metrics use the ssv_ prefix and are generated live on each scrape. The full, current list is always available at http://<host>:9876/metrics.

Conventions

  • Counters end in _total and are best read using rate(). Gauges represent a point in time value.
  • Latency and time values are expressed in seconds.
  • State and status values are the raw SANsymphony values. The Grafana dashboards map them to readable labels.
  • Which performance metrics appear can vary depending on your SANsymphony PSP version.

Scrape Health

Metric Type Meaning
ssv_up{collector} gauge 1 if that collector's scrape succeeded, otherwise 0.
ssv_scrape_duration_seconds{collector} gauge How long that collector's scrape took.

Inventory Metrics

Object Example metrics
Server group ssv_server_group_state, ssv_server_group_storage_used_bytes, ssv_server_group_license_expires_seconds
Server ssv_server_state, ssv_server_cache_state, ssv_server_storage_used_bytes, ssv_server_memory_bytes, ssv_server_info
Pool ssv_pool_status, ssv_pool_type, ssv_pool_chunk_size_bytes
Virtual disk ssv_virtual_disk_status, ssv_virtual_disk_size_bytes, ssv_virtual_disk_offline
Host ssv_host_state, ssv_host_connection_state, ssv_host_info
Port ssv_port_connected, ssv_port_role_capability, ssv_port_info
Physical disk ssv_physical_disk_status, ssv_physical_disk_size_bytes, ssv_physical_disk_free_bytes, ssv_physical_disk_pool, ssv_physical_disk_info

Health Metrics

ssv_monitor_state, ssv_alerts_total, ssv_alert_info{alert_id,level,machine,message}, and ssv_alert_age_seconds{alert_id}.

Performance Metrics

Object Example metrics
Server ssv_server_read_bytes_total, ssv_server_write_ops_total, ssv_server_cache_size_bytes, per class ssv_server_class_io_operations_total and _time_seconds_total
Pool ssv_pool_read_bytes_total, ssv_pool_write_time_seconds_total, ssv_pool_capacity_bytes, ssv_pool_used_bytes, ssv_pool_oversubscribed_bytes
Virtual disk ssv_virtual_disk_read_ops_total, ssv_virtual_disk_cache_total, ssv_virtual_disk_io_time_seconds_total
Host ssv_host_read_bytes_total, ssv_host_provisioned_bytes, ssv_host_max_read_op_size_bytes
Port ssv_port_read_ops_total, ssv_port_pending_commands, target and initiator metrics, link error counters
Physical disk ssv_physical_disk_read_ops_total, ssv_physical_disk_io_time_seconds_total, ssv_physical_disk_avg_queue_length

These metrics are visualized in the prebuilt dashboards described in Grafana Dashboards.

Scraping the Exporter with Prometheus

Once the exporter is installed and running, Prometheus does not pick it up automatically. Add it as a scrape target in your Prometheus server's own configuration file, prometheus.yml, so Prometheus starts polling /metrics on a regular interval. Add an entry under scrape_configs similar to the example below, replacing <exporter-host> with the exporter's hostname or IP address and <exporter-port> with the port it listens on, as set during installation or configuration.

Copy
Command 
scrape_configs:
  - job_name: sansymphony
    static_configs:
      - targets: ['<exporter-host>:<exporter-port>']
        labels:
          group: production

Save the file and reload or restart Prometheus for the new target to take effect. Add one entry per SANsymphony server group you monitor, each with its own targets value.

The group label shown above is not optional. The sample dashboards use it to filter which server group's metrics they display, so set a different value for each target if you monitor more than one SANsymphony server group. Otherwise the dashboards cannot tell the groups apart.

Grafana Dashboards

Five sample Grafana dashboards are provided to visualize the metrics described in Metrics Reference. They are publicly available on Grafana Labs: import each one into Grafana using a Prometheus data source that scrapes the exporter, or browse all five at DataCore Dashboards | Grafana Labs.

Overview Dashboard

This dashboard displays a single view of overall health across the server group, including scrape status, active alerts, servers online, license expiry, total IOPS and throughput, average pool latency, and capacity used.

Refer to the DataCore SANsymphony Overview dashboard.

Servers Dashboard

This dashboard displays per-server detail, including state, version, cache usage and hit ratios, IOPS and throughput, and latency by I/O type.

Refer to the DataCore SANsymphony Servers dashboard.

Storage Dashboard

This dashboard displays per-pool capacity, used space, free space, and oversubscription, along with IOPS and latency. It also shows physical disks by pool and per-virtual-disk status, size, cache hit rate, latency, and IOPS.

Refer to the DataCore SANsymphony Storage dashboard.

Hosts Dashboard

This dashboard displays an inventory of connected hosts, including IOPS and bandwidth, maximum I/O size, provisioned capacity, and the ports associated with each host.

Refer to the DataCore SANsymphony Hosts dashboard.

Ports Dashboard

This dashboard displays SCSI, iSCSI, and Fibre Channel port inventory, including IOPS and bandwidth, target latency, pending commands, and link, CRC, and sync errors.

Refer to the DataCore SANsymphony Ports dashboard.

Learn More

Troubleshooting

Configuring Prometheus Exporter

Logging and Upgrade