For an existing backup arrangement
Use the genuine tool and the responsible owner’s instructions. Keep existing recovery copies while diagnosing; local disks, cloud services, sync and managed business backups have different coverage and controls.
Step 1 of 9
In this guide
Identify the backup and what actually failed
Identify the backup and what actually failed
Open the genuine backup app or agreed service status through its normal route. Record the exact backup name, source/device or account, destination, error and last successful completion date. Was the job missed, paused, still running, partly completed or failed? A sync icon and a backup job do not necessarily describe the same protection.
Identify the data you expect to recover: particular folders, photos, email, a business application or a whole device. A completed job covering one folder does not prove another account or workload was included. If you do not know which service protects the data, ask the responsible owner before creating a competing job or changing accounts.
If files are already missing, keep the existing backup and use our backup and restore guide for recovery choices. This page focuses on diagnosing a failed or unreliable backup arrangement while preserving the copies you may need.
Identify the actual tool, expected data, destination, error and last completed run. A status icon or one successful folder is not proof of complete coverage.
Next: Protect existing copies and check for an incidentProtect existing copies and check for an incident
Protect existing copies and check for an incident
Keep existing generations and any known useful recovery copy. Do not format, initialise, delete the backup set, clear its catalogue or accept “start again” prompts as routine troubleshooting. If the source data is already lost, the backup may now be the only copy and needs especially careful handling.
For a damaged/noisy source or backup drive, repeated scans and reconnects can be the wrong next step; use our damaged-drive guide. If malware or ongoing encryption is suspected, keep offline backup media separate and follow incident containment. Do not attach the remaining good copy to test an infected system.
For shared/business backups, follow the responsible administrator’s retention and evidence instructions. Agree preservation before destination changes, repairs, migration or a new backup. Continue ordinary checks only when the source/destination are stable and it is appropriate to access them.
Preserve existing generations and sole-copy backups. Active incidents, failing storage and destructive reset prompts need assessment before routine reconnection or new runs.
Read the run history and schedule
Read the run history and schedule
Read the relevant job’s details rather than rely on its overall colour or a scheduled time. Note whether it completed, failed, skipped items, requested a sign-in or is still progressing. Record the real error/time privately. A job scheduled for last night is not evidence it actually ran.
Check whether the computer, destination and required connection were available during the intended window. A portable drive left elsewhere, sleeping source or unavailable network can prevent work. Use the tool’s supported scheduling/power guidance; do not permanently disable screen locking or leave a shared device unlocked unattended to make a backup run.
For an initial large job, consult the actual progress and product guidance before repeatedly cancelling it as “stuck”. If there is no progress or repeated error, preserve the record and investigate that cause. Business jobs need a named person to receive and review failures; an unread alert is not a monitoring process.
Check actual completion, skipped items and availability at the scheduled time. Keep the error record and a responsible alert owner rather than assume the schedule protects the data.
Next: Check the intended destination and connectionMore checks6 detailed checks
Check the intended destination and connection
Check the intended destination and connection
On a stable environment, confirm the actual destination shown by the tool matches the intended device/account/network location. If a local drive should be connected, check its normal supported cable and power arrangement while avoiding disruption to an active transfer. Do not guess by colour, drive letter alone or an unrelated disk with enough free space.
Microsoft’s File History guidance distinguishes an external drive from a network target. Reconnect/reselect the correct existing target only when its identity, condition and authority are clear. For network storage, confirm the appropriate approved connection and service availability with its owner; do not restart shared servers or change permissions blindly.
For Time Machine connection troubleshooting, use a supported destination and the actual device/macOS guidance. Apple describes direct connection for its local-disk check and a suitable network connection for network backups. Do not disable security/VPN requirements or accept erase/repair prompts merely to make a disk appear.
Confirm the real destination and appropriate supported connection. Preserve active transfers and existing data; unknown disks, failed storage or shared-network changes need the owner’s direction.
Next: Check capacity without deleting needed historyCheck capacity without deleting needed history
Check capacity without deleting needed history
Read the tool’s actual storage warning and identify which space it means: source temporary space, backup destination capacity or a cloud-account quota. A laptop with free space does not resolve a full cloud account, and a large external disk does not fix a full source used for temporary work.
Time Machine’s current troubleshooting guidance notes that a nearly full startup disk can prevent temporary backup data. Use a preservation plan for ordinary source-space cleanup and keep valuable files backed up before deleting them. Do not delete unknown backup folders manually.
For a full destination, review the actual retention policy, growth and available options with the owner/provider. A larger supported destination or quota change may be appropriate, but preserve useful older generations and understand costs/migration before committing. Excluding the important folder just to obtain a green result changes coverage; it does not repair that folder’s protection.
Identify the actual space constraint and preserve required history. Capacity, retention and exclusions need a deliberate plan, not deletion of the only recoverable copy.
Next: Confirm coverage, account and permissionsConfirm coverage, account and permissions
Confirm coverage, account and permissions
Check that the expected folders/accounts are selected and not excluded, and that the tool can read them through its intended supported permissions. A renamed/moved folder, newly created account or online-only content may need a coverage decision. Do not grant broad administrator/full-disk access to an unknown utility or remove management/protection to bypass a warning.
Microsoft’s consumer Windows Backup guidance describes personal Microsoft-account and OneDrive-backed folder/settings behaviour; it is not a general work/school-account or whole-system backup recipe. Check the genuine app’s specific error, correct account, OneDrive sign-in and quota when that is the tool in use. Do not switch business files into a personal account to make a consumer workflow fit.
For Time Machine, cloud-only content or protected files may have availability requirements; use the actual error and supported method. For managed business/cloud backups, ask the administrator which data/workloads, accounts, retention and restore permissions are actually included. A Microsoft 365 licence or synchronised folder alone does not establish a separate backup.
Confirm actual data coverage and intended access/account. Fix the relevant cause with its owner; do not call exclusions, personal-account migration or bypassed safeguards a successful backup.
Next: Retry only after the relevant cause is addressedRetry only after the relevant cause is addressed
Retry only after the relevant cause is addressed
Once the environment is stable and a relevant connection, quota, availability or authorised configuration issue is resolved, use the backup tool’s normal supported retry/manual-run control. Ensure you understand any prompt affecting existing generations. Stop if it proposes erasing history, replacing a destination or changing the protection scope unexpectedly.
For configured File History, Microsoft describes Control Panel, System and Security, File History, Run now after the intended drive/network target is appropriately available. For other tools, follow their current controls; do not assume one Windows service command restarts every backup. If a Time Machine check requires an unlocked Mac, perform it while present and appropriate rather than leave it unsecured.
Allow the supported operation to finish and read the completed result, time, coverage and remaining skipped items. If the same failure recurs, save the error and stop repeated resets/retries. A second failure needs diagnosis, not clearing the history or disabling the error notification.
Retry the supported job after its cause is addressed and confirm the real result. Repeated failure, changed scope or destructive prompts need assessment rather than more resets.
Verify a selected restore and the next run
Verify a selected restore and the next run
With appropriate authority and a healthy environment, select a harmless known file that was actually in scope before the completed backup. Use the supported restore method and a separate suitable location where available. Understand replacement behaviour first and preserve current originals; never delete a valuable file to create a test.
Open the selected restored content and check the expected version and useful content. Record its backup point, result and limitations. A successful small-file test does not prove an entire application, mailbox, business system or every older generation can be restored; wider recovery testing needs its own plan.
Australian business guidance recommends testing restoration as well as taking backups. Check that the next scheduled run completes and alerts reach the agreed owner. Once coverage, operation and the limited restore are verified, stop routine changes and retain the agreed recovery copies and follow-up record.
Verify an appropriate selected restore without deleting originals, then check the next scheduled run and alert owner. Record the test’s scope; one file is not a complete disaster-recovery test.
Back to the symptom checkerGet help with recurring or uncertain backups
Get help with recurring or uncertain backups
Ask for help when there is no known useful copy, repeated job/restore failure, damaged storage, uncertain account ownership or a change that could erase required history. Privately provide the tool, source/destination type, last successful date, exact error, important excluded/missing data and attempts made. Do not attach passwords, recovery keys or an entire backup archive to an initial enquiry.
Our data-protection support can discuss a scoped assessment of the arrangement. For business systems, agree the responsible administrator/provider, acceptable data-loss window, recovery needs and any extra costs before changing policy or buying a replacement. Do not assume every support plan includes every workload or round-the-clock backup monitoring.
Keep the remaining risk explicit until protection is verified: what data is not covered, the last usable recovery point and who will follow up. No tool or technician can promise all history can be rebuilt after it is lost. Preserve useful copies while deciding how to close the gap.
Get a scoped assessment for persistent failures or uncertain recovery copies. Agree the owner and protection gap before destructive changes, migrations or new spending.
