Live monitoring
Live monitoring
These views show what StorONE is doing right now. In an interactive CLI session they stream continuously — refreshing on the configured interval — until you press a key to stop; the web interface shows the same data as a live-updating chart. When called non-interactively (for example, from a script or through the REST API) they instead return a single sample and exit immediately.
While a live CLI monitor is running, you can usually press a letter key to change how it's sorted (for example N for name, I for IOPS, L for latency) — press any other key to stop.
Node utilization
monitoring utilization live shows how busy each node's disks, CPU, and memory are. Filter to a single node, or leave it unfiltered to see every node.
Using the web interface
- Open Monitoring > Utilization.
- Under Node Selection, optionally choose a specific Node — leave it blank to see all nodes.
Using the CLI
monitoring utilization live [--node=<name>] [--interval=<interval>] [--singleSample]
| Option | Description |
|---|---|
--node=<name> |
Filter results for a specific node. |
--interval=<interval> |
The sampling interval in seconds. |
--singleSample |
Get a single sample instead of monitoring continuously. |
Examples
Continuously monitor utilization on every node:
monitoring utilization live
Single utilization sample for node s1-node-01 (useful for scripting):
monitoring utilization live --node s1-node-01 --singleSample
Volume/application performance
monitoring performance live shows IOPS, throughput, latency, and I/O size, split into read and write. Scope it to a single application (optionally further filtered to specific volumes within it), or leave it unfiltered to see every volume individually. Turn on Aggregate to combine the selected scope (the whole system, or the application) into one rolled-up set of figures instead of one row per volume.
Using the web interface
- Open Monitoring > Performance.
- Under Entity Selection, optionally choose an Application (and, within it, specific Volumes), and turn on Aggregate to see combined figures instead of one row per volume.
- Under Counter Selection, choose which counters to show — IOPS, Throughput, Latency, I/O Size. All four are shown by default.
- Under Read/Write Selection, choose Read and/or Write. Both are shown by default.
Using the CLI
monitoring performance live [--application=<name>] [--volumes=<name...>] [--aggr] [--interval=<interval>] [--iops] [--throughput] [--latency] [--iosize] [--read] [--write] [--interactiveChart]
| Option | Description |
|---|---|
--application=<name> |
Filter results for a specific application only. |
--volumes=<name...> |
Filter results for specific volumes. |
--aggr |
Aggregate results for the entire system / specific application. |
--interval=<interval> |
The sampling interval in seconds. |
--iops |
Show input/output operations per second. |
--throughput |
Show throughput. |
--latency |
Show latency. |
--iosize |
Show I/O size. |
--read |
Show read statistics. |
--write |
Show write statistics. |
--interactiveChart |
Show a chart instead of textual information. Windows CLI only. |
Examples
Live performance for every volume, all counters:
monitoring performance live
Live IOPS and latency for the emails application, aggregated into one line:
monitoring performance live --application emails --aggr --iops --latency
Live read/write throughput for two specific volumes:
monitoring performance live --application training --volumes videos,thumbnails --throughput
Drive performance
monitoring drives live shows the same kind of counters (IOPS, throughput, latency, I/O size, read/write) but at the individual drive level, which is the view to reach for when you suspect a specific disk rather than a volume. Filter by node, by pool, or by specific drive numbers, and optionally aggregate the selection into one line.
Using the web interface
- Open Monitoring > Drives.
- Under Node Selection, optionally filter to a specific Node.
- Under Pools and Drives Selection, optionally filter to specific Pools and/or Drives, and turn on Aggregate to combine the selection into one line.
- Under Counter Selection and Read/Write Selection, choose which counters and directions to show (all are shown by default).
Using the CLI
monitoring drives live [--interval=<interval>] [--node=<name>] [--pools=<name...>] [--drives=<numbers...>] [--aggr] [--iops] [--throughput] [--latency] [--iosize] [--read] [--write] [--singleSample]
| Option | Description |
|---|---|
--interval=<interval> |
The sampling interval in seconds. |
--node=<name> |
Filter results for a specific node only. |
--pools=<name...> |
Filter results for specific pools only. |
--drives=<numbers...> |
Filter results for specific drives. |
--aggr |
Aggregate results for the entire pool / specific drives. |
--iops |
Show input/output operations per second. |
--throughput |
Show throughput. |
--latency |
Show latency. |
--iosize |
Show I/O size. |
--read |
Show read statistics. |
--write |
Show write statistics. |
--singleSample |
Get a single sample instead of monitoring continuously. |
Examples
Live per-drive statistics for the HDD-1 pool:
monitoring drives live --pools HDD-1
Single-sample latency and IOPS for specific drives, aggregated:
monitoring drives live --drives 3,4,5 --iops --latency --aggr --singleSample
ALUA path state
On a high-availability pair, each volume has one node that owns it (the "optimized" path) and a peer node that can also serve I/O for it but at a cost (the "unoptimized" path) — this is ALUA (Asymmetric Logical Unit Access). Hosts using multipathing should be sending the bulk of their I/O down the optimized path; if a host's multipathing configuration is wrong, you'll see meaningful IOPS on the unoptimized path instead, which costs extra latency.
monitoring alua live samples optimized vs. unoptimized IOPS per volume, once a second, for the number of iterations you specify (10 by default), and reports the average. This command requires a high-availability system.
Using the web interface
Open Monitoring > ALUA.
Using the CLI
monitoring alua live [--iterations=<numbers>]
| Option | Description |
|---|---|
--iterations=<numbers> |
The number of one-second samples to collect and average (default: 10). |
Examples
Average optimized/unoptimized IOPS over the default 10 samples:
monitoring alua live
Average over a longer, 60-sample window:
monitoring alua live --iterations 60
Next: Historical monitoring — performance and utilization trends over time, top-volume rankings, and reserve space. See also: Health and capacity · Tiering monitoring