Network UPS Tools / Troubleshooting

NUT Reports Data Stale for a USB UPS

Confirm the current stale response, then separate driver freshness, USB visibility, access and transport evidence without changing shutdown protection.

UPS hardware supplies the driver, the driver supplies upsd, and clients read the server's data. DATA-STALE means the connected driver is not delivering regular updates or has marked them stale; it does not uniquely identify a cable, permission or hardware fault.[3]

Stable manuals support the operational checks here. The separately labelled development protocol document is used only for compatible error semantics, not development CLI options.[1][2][3]

Applicability

Products
Network UPS Tools
Scope
For Linux with stable NUT 2.8.5, a previously working local USB UPS supported by usbhid-ups, and a current local query that reaches the intended upsd but reports stale data. Excludes remote-server connection failures, other drivers, unsupported hardware, appliance-specific wrappers and shutdown configuration. Identify the exact model and installed package before interpreting its logs.[1][2]
Last verified

Symptoms

  • A current local query returns Data stale for the intended UPS. An old dashboard notification, connection refusal or DRIVER-NOT-CONNECTED response is a different observation.[5][3]

Quick diagnosis

Gather current evidence before choosing a branch. Unknown observations do not establish a failed component.

Safe / low risk

Read current data from the intended server

Run locally with authorized access. Replace <upsname> with the configured UPS name; localhost must be the intended upsd. This example assumes its default local listener. If the installation uses another listener/port, use its documented query target rather than changing the server.[5][4]

upsc <upsname>@localhost ups.status
Read-only

Record version, model and driver

Use existing package/configuration records to identify NUT, the UPS model and its driver. This page does not start a second driver instance or test UPS power state.[1][6]

Read-only

Keep monitoring protection in place

If current power conditions are uncertain, involve the responsible operator before any intervention. Reading data does not test automatic shutdown or prove the UPS battery is healthy.[7]

Interactive troubleshooter

Your answers stay in this browser tab. The tool does not connect to your systems or send diagnostic results.

Guided check / step 1

Safe diagnostic guidance

Is this the supported local USB UPS installation?

Confirm Linux, stable 2.8.5, a previously working supported usbhid-ups model and the intended local server.[1][6]

Read the complete diagnostic tree without using the controls
  1. Is this the supported local USB UPS installation?

    Confirm Linux, stable 2.8.5, a previously working supported usbhid-ups model and the intended local server.[1][6]

    • Yes, the installation is identifiedWhat does the current local query return?
    • No, or applicability is unknownInvestigate the actual error or installation
  2. What does the current local query return?

    Use the intended UPS/server. Do not classify an old warning as the current response.[5][3]

    • Data stale from reachable upsdDo current logs still describe the same freshness problem?
    • Current data is availableCurrent data has recovered
    • Different error or server unreachableInvestigate the actual error or installation
  3. Do current logs still describe the same freshness problem?

    Correlate timestamps and driver connection evidence. Do not start a duplicate driver to inspect it.[2]

    • Yes, same current stale conditionIs the expected USB UPS visible to the OS?
    • Driver disconnected or error changedInvestigate the actual error or installation
    • Driver evidence is unavailableCollect a driver-freshness evidence packet
  4. Is the expected USB UPS visible to the OS?

    Use model/device identity, not a guessed serial pathname.[6]

    • Yes, the expected UPS is visibleIs there actual access or competing-driver evidence?
    • No, the expected device is absentHand off missing USB visibility
    • Visibility is unknownCollect a driver-freshness evidence packet
  5. Is there actual access or competing-driver evidence?

    Distinguish a logged denial or duplicate owner from a theory based only on staleness.[2][6]

    • Yes, access or ownership conflict is evidencedRequest access and driver-ownership review
    • No such conflict is establishedAre driver transport failures repeating?
    • Access evidence is unavailableCollect a driver-freshness evidence packet
  6. Are driver transport failures repeating?

    Correlate the messages with this model and current timestamps.[6]

    • Yes, repeated communication errorsRequest model-specific transport investigation
    • No repeated transport evidenceIs there an exact documented model-specific timing match?
    • The evidence is insufficientCollect a driver-freshness evidence packet
  7. Is there an exact documented model-specific timing match?

    A generic stale warning is not a timing diagnosis.[6][4]

    • Yes, model and pattern match documentationRequest a bounded timing assessment
    • No exact match or still uncertainCollect a driver-freshness evidence packet
  8. Investigate the actual error or installation

    Record the precise response and environment. Server reachability, identity, authentication and DRIVER-NOT-CONNECTED need their own investigation.[3]

  9. Current data has recovered

    Retain timestamps and observe recurrence with the owner. Do not claim that a successful query tests battery health or shutdown behavior.[5][7]

  10. Hand off missing USB visibility

    Provide expected identity and current hardware evidence to the operator; no generic unplug or reset test.[6]

  11. Request access and driver-ownership review

    Provide the actual denial or duplicate instance evidence. Any intervention must meet the monitoring-safety boundary below.[2][7]

  12. Request model-specific transport investigation

    Supply current query and repeated driver messages. Communication evidence does not uniquely identify a defective cable or UPS.[6]

  13. Request a bounded timing assessment

    Include the exact model note and observation. Do not tune freshness or shutdown thresholds from this result.[4][6]

  14. Collect a driver-freshness evidence packet

    Keep the exact query, server identity, version, model and time-correlated logs. Ask the operator for missing evidence without disabling protection.[2][5]

Detailed diagnosis

Read-only

1. Classify the actual response

A reached server reporting DATA-STALE is distinct from DRIVER-NOT-CONNECTED, UNKNOWN-UPS, ACCESS-DENIED or a network timeout. If the error changes, leave this freshness tree and retain the exact response.[3]

Read-only

2. Correlate existing driver and server logs

Compare the current query timestamp with existing driver/process and server connection evidence. Distribution service names and paths vary. A now-disconnected driver needs a different diagnosis, not an assertion that every stale response means the driver stopped.[2]

Read-only

3. Inspect OS visibility and runtime access

Identify the expected USB UPS from existing hardware evidence and compare access/ownership with the installed package's requirements. usbhid-ups does not select its USB hardware by a conventional tty port path; do not import a serial-driver path fix.[6]

Read-only

4. Separate access conflict from repeated transport errors

Look for actual denied access or duplicate-driver ownership before considering permissions. If communication errors repeat, record their timestamps and model/driver context; one error does not establish a bad cable.[2][6]

Read-only

5. Reserve timing changes for exact model review

Compare the observed pattern with the applicable driver documentation. Models can have different polling constraints. Do not increase MAXAGE or change polling values simply to remove the warning.[6][4]

Supported scenarios

These observations narrow the next investigation; they do not establish a unique cause.

The driver is not supplying current data

The stale response describes data freshness, not a unique failure of a physical component.[3]

How to check: Match a fresh local query with driver/server evidence for that UPS.[5][2]

USB visibility or access needs investigation

Missing hardware evidence, denied access and competing drivers are distinct observations that warrant different owner checks.[6][2]

How to check: Record device identity and current access/ownership evidence without restarting services.[2]

A documented model-specific pattern may apply

A timing-related hypothesis needs the exact model and matching driver documentation; no universal freshness timeout is established.[4][6]

How to check: Compare repeated timestamps and driver messages with the applicable model notes.[6]

Next actions and procedure boundaries

Read-only

Device absent: hardware-owner handoff

Provide the expected UPS identity and current OS observations. Ask the responsible operator to investigate the USB path without an automatic unplug, reset or power-cycle test.[6]

Read-only

Access or duplicate ownership: package-owner review

Provide the actual access denial or competing instance evidence. The package/service owner must establish least-privilege access and the intended driver instance before considering intervention.[2]

Read-only

Repeated transport errors: driver/model investigation

Send sanitized timestamps, current query result, version and exact model to the driver/package owner. Keep the observed communication failure separate from an unproven hardware diagnosis.[6]

Read-only

Documented timing match: specialist review only

Provide the exact matching model note and observed pattern. This outcome does not authorize changing MAXAGE, DEADTIME, polling or shutdown policy.[4][6]

Potentially disruptive

Before any separately scoped service or permission intervention

Planning boundary only. The responsible UPS/system operator must consider current power conditions and affected protected systems. This page does not restart services, disconnect USB, disable monitoring or alter shutdown behavior.[7]

Before changing anything

Risk
Disrupting NUT communication can impair the information used by shutdown monitoring.[7]
Safer check
First confirm the current response and read existing logs. Do not perform a battery test or change power state.[5][7]
Expected result
A later owner-approved recovery must restore current data without altering the protection policy; fresh data alone does not validate shutdown behavior.[7]
Backup / recovery access
The operator must preserve package/service configuration, identify protected systems, assess current power conditions and establish independent recovery access before intervention.[7]
Rollback
No universal service or permission rollback is supplied. Require an installation-specific restoration plan from the owner; if monitoring cannot be recovered safely, do not proceed.[7]
Read-only

Unresolved freshness: preserve evidence

Keep the UPS name, intended server, current error and timestamped driver/USB observations. Redact credentials and identifiers before a targeted support handoff; do not suppress the warning to claim recovery.[2][5]

Warnings and boundaries

Do not change DEADTIME/MAXAGE as trial-and-error or treat missing telemetry as a verified safe power state. Freshness and monitoring policy are separate concerns.[4][7]
No cable disconnection, USB reset, forced shutdown, battery test or blind restart is part of this page.[6][7]

Sources

Links were reviewed on 2026-09-05. Reachability and automated validation do not replace editorial verification of each claim.

  1. Network UPS Tools - Welcome
    Network UPS Tools · Tier A · accessed 2026-09-05
  2. FAQ
    Network UPS Tools · Tier A · accessed 2026-09-05
  3. 10. Network protocol information
    Network UPS Tools · Tier A · accessed 2026-09-05
  4. UPSD.CONF(5)
    Network UPS Tools · Tier A · accessed 2026-09-05
  5. UPSC(8)
    Network UPS Tools · Tier A · accessed 2026-09-05
  6. USBHID-UPS(8)
    Network UPS Tools · Tier A · accessed 2026-09-05
  7. UPSMON(8)
    Network UPS Tools · Tier A · accessed 2026-09-05