Jobs
Follow every backup, restore, download, directory sync and maintenance run with its live speed, see why a queued job waits, read its log, see resumed parts, restarts and skipped items, and cancel it if needed.
Everything backup does is a job: each backup run, each restore and download, each Microsoft 365 directory read and each repository maintenance. The Jobs list is the one place to see all of it, and the customer page, device page and overview all read from it.
The Jobs list
Open Backup > Jobs. Running jobs update every few seconds.
The list opens with running jobs first (the longest-running at the top, so a long job stays in sight), then queued jobs in the order they will start, then everything else by when it started, newest first. Running first on the Status column header switches that order off and on, and a saved view keeps your choice. Clicking Started sorts by start time either way.
Each row shows:
| Column | What it holds |
|---|---|
| Status | Queued, running, Completed, Completed with errors, Failed, interrupted or cancelled |
| What | Backup, Restore, Download, Directory sync, Maintenance and the other kinds below |
| Device or tenant | The device or Microsoft 365 tenant it ran for |
| Target | What it worked on: a volume or folder, a tenant, or the repository for maintenance |
| Customer | Whose it is |
| Started and Took | When it started and how long it ran |
| Sent | How much data went to storage after deduplication |
Filter by Status, What, Source (devices or Microsoft 365) or Started, or type a device, tenant, target or customer in the filter box. A sandbox host that cannot run boot checks would otherwise fill the list with the same check over and over: those repeats are hidden until you press Show repeats. Load older jobs at the bottom fetches more. Recent restores above the list opens each restore's own progress page.
The Views picker has three ready-made views:
- Running now: jobs in progress.
- Failed and errors: jobs that failed or finished with errors.
- Restores and downloads: everything you or your team restored.
What the statuses mean
| Status | Meaning |
|---|---|
| Queued | Waiting to start. The job says why: "Waiting for the device to come online", "Waiting for bandwidth at Head office, 3 ahead" when spreading holds it to the site's upload budget, "Waiting for repository maintenance", its turn among device backups ("3 device backups run at once on the backup server and 5 are ahead of this one"), or for Microsoft 365 its place in line ("Waiting, number 2 in line"), the tenant it is waiting for, or memory on the backup server. Where a setting decides, Change the limit links to it. A Microsoft 365 job stays queued until its first part really starts |
| Running | In progress, with a percentage |
| Completed | Everything that should have been backed up was. A job with a few skipped items is still completed, with an amber "3 skipped" beside it. Accepted and excluded items are not in the amber count |
| Completed with errors | The job finished, but some units failed or some files could not be read. What did back up is a usable recovery point |
| Failed | Every unit that ran failed, or the job could not start |
| Interrupted | A restart or update of Tenvara, a reboot or sleep cut it short. It is not a failure: it carries on by itself and is left out of the success figures. See Backups through restarts and updates |
| Cancelled | Someone cancelled it, named on the job, even when it was cancelled before it started |
| Missed, device offline | A scheduled backup whose device was still offline when its next backup fell due. It never ran, so it is not a failure and raises no alert; the device's own overdue alert covers it once it passes Overdue after |
A unit that cannot be backed up at all (for example an unsupported volume, or a Microsoft 365 user with no OneDrive) is skipped as not applicable. That is not a failure of the run.
A few single items the source refused (a damaged email, a locked file) are skipped items: the job stays completed with a quiet note, and only becomes completed with errors or alerts when there are more than a few. Items someone has accepted or excluded are known skips: they leave the job's warning line, and only a new or changed skip brings it back. See Skipped items.
A job that carried on parts an earlier run did not finish has a Resumed badge in the list. A Microsoft 365 backup that was running when the backup service restarted carries on in the same job, and the list says "Carried on after a restart at 13:54" beside its status.
Job details
Click a job to open its details in a side panel.
The panel shows:
- The kind, target, status, device or tenant, and job number.
- Customer and Started by: a person, or Scheduled.
- Queued, Started, Finished and Took.
- Items, Read and Sent after deduplication: how many files or items it went through, how much it read, and how much was new enough to send.
- For a running job, what it is doing (a backup, a boot check, a test restore or a verification) with its real percentage, its progress and each unit's state. A job held back by the backup server's memory, its turn or the site's bandwidth says so in a callout, with how fast it is reading and sending right now. A finished unit shows its average. A running OneDrive, site or Google Drive part shows a progress bar by bytes, files done of the total, files a second and an estimate to finish.
- Units: each volume, folder or Microsoft 365 part with how it ended. A part that carried on an interrupted run says so: "Resumed from 64%, 38 GB already stored (try 3)".
- Skipped items, with the reason for each, Try again now, and Accept or Exclude for one or several. Known skips are listed once under a collapsed Known skips (N).
- A note when the job carried on after a restart, with how many restarts, and "Waiting for the install (it may be updating)" while a device backup waits out an update.
- Scripts: any script that ran before or after the backup, as whom, and how it ended. Its output is in the log.
- Log: every step with its time. For Microsoft 365 the last line sums up the run: parts backed up, failed, with errors, skipped and cancelled, and how many items were new, changed or deleted.
Restore from this backup opens the restore browser at the recovery point this job made. A failed backup has Back up again, and a Microsoft 365 backup with failed parts has Back up the failed parts again and Back up the whole tenant again. The log can be narrowed with Only warnings and errors.
Downloads
A Download job prepares a file to save, for example a zip of files or a PST of mail. When it completes, the panel has the file to save. When a mail download holds emails rebuilt from Microsoft's message data, a PST or EML download's log says how many of them were rebuilt, and a zip carries a manifest.csv with the note on each rebuilt message. See Restoring. Download links work for seven days by default (Download links last on the Restores tab of Settings > Backup).
Cancelling a job
Open a queued or running job and cancel it.
- A queued job is cancelled at once.
- A running device job is told to stop and stops shortly after.
- A running Microsoft 365 job stops at the next item.
Whatever was stored before the cancel is kept as a partial recovery point, and the next scheduled run carries on from there rather than starting again. The cancel dialog says what is kept.
Interrupted jobs
An interruption is not a failure. What happens depends on what was interrupted:
- A device backup whose device restarts, or whose agent is updated, ends with "Interrupted:" and the reason, and a run carrying it on is queued straight away, linked from the job. A device that goes to sleep mid-backup shows Paused while the device sleeps and carries on when it wakes.
- A device backup or restore that loses Tenvara itself, for example while it updates, waits: the job says "Waiting for the install (it may be updating)" and carries on where it stopped when the install answers. Past the wait it ends interrupted and carries on by itself once the install is back. A restore that ended this way has Carry on in its panel.
- A Microsoft 365 backup carries on in the same job after a restart of the backup service, with nothing cancelled and no new job.
See Backups through restarts and updates and Resuming large and interrupted backups.
Maintenance jobs
Each repository gets quick maintenance daily and full maintenance weekly, which is also when retention removes old recovery points. Maintenance waits while a backup of that repository is running and tries again later, without leaving a cancelled job behind each time. These jobs show as Maintenance with the target Repository maintenance (quick) or (full). A running one shows its step and counts, and Cancel stops it within a minute, keeping the steps already finished. See Schedules and retention.
Other kinds of job
Besides backups, restores and downloads, the What filter lists Second copy (a repository copied to its second location), Test restore, Boot check, Machine from backup, Standby, Bare metal restore, Purge, Directory sync, Maintenance, Pack verification (stored data read back and checked) and Repository repair (damaged data put right). See Repository health, healing and verification. Restores also have their own progress page, opened from Recent restores at the top of the Jobs list. See Restoring.
Jobs elsewhere
- The device page lists its recent jobs in the Backup section of its Overview tab, with All to open the Jobs list filtered to that device.
- The customer's Backup tab lists that customer's recent jobs, with All to filter the list to them.
- The Backup overview shows Recent jobs, and failed jobs in Needs attention with See the log.
- An alert raised by a failed backup links to the job under Raised by, and offers to retry the backup from the alert.
Was this page helpful?
Thanks for the feedback.