## TL;DR
The protection job is paused, and after an upgrade it usually stays paused because the upgrade paused it and nothing resumed it, or a policy change left it in a paused state the job respects silently. Check the job's state and the cluster's post-upgrade tasks, resume the job explicitly, and watch its next run start; if it re-pauses, read the pause reason the job records rather than resuming blindly.

```text
Cohesity backup failed "protection job is paused" error after cluster upgrade
```

## Use this when
- The job or restore fails with this symptom: Cohesity backup failed with protection job is paused after a cluster upgrade.
- You are working on a Cohesity protection job that stopped running after a cluster upgrade and the failure repeats on the same target or schedule.
- You need the failing job, stage, or component isolated before changing retention or infrastructure.

## Not for this skill when
- A job paused deliberately by an administrator for maintenance, where the fix is the maintenance window ending, not a manual resume.
- The error text names a different product, stage, or root cause. Match the verbatim error first and use that skill.

## Steps
1. Open the protection job and confirm its current state and when it entered the paused state relative to the upgrade. Verify: the pause timing lines up with the upgrade or another event
2. Check the cluster's upgrade and maintenance task list for incomplete steps that hold jobs paused. Verify: post-upgrade tasks are complete or identified
3. Resume the job from the job's controls and watch for its next scheduled run to actually start. Verify: a run starts and progresses past protection start
4. If the job pauses again, read its recorded pause reason and fix that cause, such as a paused policy or source registration. Verify: the re-pause cause is named and addressed
5. Verify the missed protection window is covered by running an on-demand backup for the gap period. Verify: a fresh recovery point exists for the previously paused objects

## Variant phrasings
### cohesity protection job is paused after upgrade
The verbatim phrasing.
### cohesity backup not running job paused resume fix
The resume phrasing.
### cohesity job paused cluster upgrade missed backups
The missed-window phrasing.

Compatibility: Cohesity DataProtect protection jobs after cluster upgrades

## Why it happens
Upgrades pause protection to keep the cluster consistent, and resumption is a separate step that automation, a missed admin task, or a policy state can quietly own. The job is behaving correctly by honouring the pause; the failure is that nothing lifted it.

## Edge cases / pitfalls
- Policies can pause at the source level too; a resumed job with a paused source still protects nothing.
- Note the gap in the backup report; auditors need the paused window acknowledged, not silently backfilled.

## Provenance

Resolved from the public thread: https://vectle.com/posts/pst_4B75pqttziskX3JsJ6i4YQ