Skip to main content
Version: 0.0.15

Monitoring Overview

MASS provides built-in monitoring through the Dashboard. No external tools are required to observe system health, storage performance, and hardware status.

Where to Look​

PageWhat it shows
DashboardSystem state and uptime, capacity, active alerts, cluster-wide performance trend
Analytics → PerformanceThroughput, IOPS, and latency over a selectable time range
Analytics → CapacityCapacity usage and its trend
Analytics → Usage & RankingVolumes, hosts, or volume groups ranked by IOPS, bandwidth, or latency
HardwareNode and drive health, link status, device temperatures
System → LogsPlatform log with level filter and keyword search
System → AlertsActive and recently solved alerts

System State​

The sidebar status card, the health card, and the status ticker all report the same system state, so they never disagree.

StateMeaning
Not ConfiguredInitial setup has not been completed
InitializingNodes are joining; shows how many are ready
ReconfiguringA live reconfiguration is in progress
ReadyNormal
DegradedSome components are experiencing issues
FailedThe system failed to reach a healthy state
OfflineThe management service cannot be reached

When the state is Degraded or Failed, a banner at the top of the page lists every cause with a suggested remedy, and offers recovery actions where one exists. See Troubleshooting.

Capacity​

The storage header presents capacity as a single stacked bar — Used, Allocated, Available, and Reserved — sized against total raw capacity, with an 80% allocation marker. The capacity donut on the Dashboard shows the same figures.

Allocated capacity comes from the requested sizes of your volume groups, so the value is stable from the moment a volume group is created. Failed and deleted volume groups are excluded.

Performance​

Performance views split read and write into separate series for both IOPS and bandwidth, with latency in ms. Time ranges run from Last 1 minute to Last 90 days.

Detail charts update about once a second and scroll smoothly, with selectable live windows (1m / 5m / 15m).

Top consumers can be ranked by volume, host, or volume group. The averaging window is selectable, and an instantaneous peak column is shown alongside the average so short bursts are not smoothed away.

Usage & Ranking ranks entities by average or maximum IOPS, average or maximum bandwidth, or read/write latency (lower is better for latency).

info

An idle or brand-new volume shows a clear "no data" message rather than an invented graph.

Connected Hosts​

Identity & Access → Connected Hosts lists the clients currently connected. Clients connected to an open-access volume appear too, marked as anonymous.

System Logs​

System → Logs supports a level filter (showing All when no filter is applied) and case-insensitive keyword search.

Log levels map to visibility: informational, warning, and error entries are meant for operators. Internal operations are not surfaced here.