Storage device IO
1readings
1conditions
2themes fed
Livestatus
Per-device IO posture, including a device that is busy but moving almost nothing.
Topic id: storage_device_io.
Reported every 5 minutes on clock boundaries for the window ending at t. sparklogs.window_coverage_pct is below 100 when collection covered only part of the window. Use chart buckets at least 5 minutes wide.
Fields
Read/write activity, response times and stalled-I/O indicators for each storage device.
| Field | Type | Unit | Meaning |
|---|---|---|---|
sparklogs.data.storage_device_io.device | string | The device's stable identity, the same identity storage_devices uses. | |
sparklogs.data.storage_device_io.display_name | string | The device's friendly name, for display. | |
sparklogs.data.storage_device_io.transport | string | The bus this device is attached over. | |
sparklogs.data.storage_device_io.media | string | The device's media: hdd or ssd. Omitted when not read. | |
sparklogs.data.storage_device_io.total_iops_avg | float | iops | Average combined read and write operations per second over the window. |
sparklogs.data.storage_device_io.total_mb_per_s_avg | float | megabytes_per_second | The average combined read-plus-write throughput over the window, in 1024-based MB per second. |
sparklogs.data.storage_device_io.bus_permanent | bool | Whether this device uses a permanent bus, excluding USB. | |
sparklogs.data.storage_device_io.carries_writeable_volume | bool | Whether this device backs at least one mounted volume that allows writes. | |
sparklogs.data.storage_device_io.carries_os_volume | bool | Whether this device backs the volume Windows runs from. False when it does not, including when that volume's role was not established. | |
sparklogs.data.storage_device_io.busy_pct_avg | float | percent | The average share of the window this device spent busy. |
sparklogs.data.storage_device_io.busy_pct_p90_10s | float | percent | The 90th percentile, across the window's 10-second samples, of busy share. |
sparklogs.data.storage_device_io.busy_pct_max_10s | float | percent | The highest 10-second busy share the window observed. |
sparklogs.data.storage_device_io.read_iops_avg | float | iops | The average read rate over the window. |
sparklogs.data.storage_device_io.read_iops_max_10s | float | iops | The highest 10-second read rate the window observed. |
sparklogs.data.storage_device_io.write_iops_avg | float | iops | The average write rate over the window. |
sparklogs.data.storage_device_io.write_iops_max_10s | float | iops | The highest 10-second write rate the window observed. |
sparklogs.data.storage_device_io.read_mb_per_s_avg | float | megabytes_per_second | The average read throughput over the window, in 1024-based MB per second. |
sparklogs.data.storage_device_io.read_mb_per_s_max_10s | float | megabytes_per_second | The highest 10-second read throughput the window observed, in 1024-based MB per second. |
sparklogs.data.storage_device_io.write_mb_per_s_avg | float | megabytes_per_second | The average write throughput over the window, in 1024-based MB per second. |
sparklogs.data.storage_device_io.write_mb_per_s_max_10s | float | megabytes_per_second | The highest 10-second write throughput the window observed, in 1024-based MB per second. |
sparklogs.data.storage_device_io.read_latency_ms_avg | float | milliseconds | The count-weighted average read latency over the window. |
sparklogs.data.storage_device_io.write_latency_ms_avg | float | milliseconds | The count-weighted average write latency over the window. |
sparklogs.data.storage_device_io.latency_ms_p90_10s | float | milliseconds | The 90th percentile, across the window's 10-second mean-latency samples, of latency. |
sparklogs.data.storage_device_io.latency_ms_max_10s | float | milliseconds | The highest 10-second mean latency the window observed. |
sparklogs.data.storage_device_io.avg_read_kb | float | kilobytes | The average size of a read over the window. |
sparklogs.data.storage_device_io.avg_write_kb | float | kilobytes | The average size of a write over the window. |
sparklogs.data.storage_device_io.queue_depth_avg | float | The average outstanding IO queue depth over the window. | |
sparklogs.data.storage_device_io.queue_depth_max_10s | float | The highest 10-second queue depth the window observed. | |
sparklogs.data.storage_device_io.topology_segment_count | integer | count | How many distinct topology signatures the window's samples held. More than two in one window marks topology_mixed and holds until every sample shares one topology again. |
sparklogs.data.storage_device_io.topology_mixed | bool | Whether the topology changed too many times within one window to trust a single reduction. Absent when the topology held steady. | |
sparklogs.data.storage_device_io.disk_unresponsive_age_basis | string | onset: witnessed start. observed: already present when first seen, making age a lower bound. unknown_ongoing: no meaningful onset time. | |
sparklogs.data.storage_device_io.disk_unresponsive_age_h | float | hours | How long this condition has been open, in hours. |
Conditions
A condition is a state that holds for a while. The agent opens it when the host enters it, keeps it open while it lasts, and closes it when the host comes back out, so one episode answers for the whole stretch instead of one alert per sample.
| Condition | Severity | How an episode ends |
|---|---|---|
disk unresponsive (disk_unresponsive) | Minor to Critical | It closes when any one of the several recovery conditions is met. |
Example
Inventory (every 5 minutes)
1 device; "Disk 0" nvme ssd busy 12%, 0.5 ms p90.
sparklogs.data.storage_device_io.device: storage_device:eui.0025385c91b5a1e2
sparklogs.data.storage_device_io.busy_pct_avg: 12.0
sparklogs.data.storage_device_io.total_iops_avg: 1400.0
sparklogs.data.storage_device_io.total_mb_per_s_avg: 10.87
SparkLogs: CONTEXT, Info, storage_device_io: INVENTORY: 1 device; "Disk 0" nvme ssd busy 12%, 0.5 ms p90.
Selected conditions
disk_unresponsive
A storage device is busy but moving almost no data.
Also reported by: Storage device IO
Impact: IO to this device may be stalled, and a writeable volume on it can stop responding.
Example
started; disk "Disk 1" busy 99
sparklogs.instance: storage_device:eui.0025385c91b5a1e3
sparklogs.data.storage_device_io.device: storage_device:eui.0025385c91b5a1e3
sparklogs.data.storage_device_io.busy_pct_avg: 99.0
sparklogs.data.storage_device_io.total_iops_avg: 0.4
sparklogs.data.storage_device_io.total_mb_per_s_avg: 0.01
sparklogs.data.storage_device_io.disk_unresponsive_age_h: 0.0
SparkLogs: disk_unresponsive, Critical, storage_device_io: disk_unresponsive: NOTABLE: started; disk "Disk 1" busy 99
| Case | Severity | Ticket class |
|---|---|---|
onset | Trace to Fatal | storage |
held | Trace to Fatal | storage |
recovered | Trace to Fatal | storage |