> ## Documentation Index
> Fetch the complete documentation index at: https://docs.assetinfinity.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Operations

> Whether the product's scheduled work is running, and what it tried to send and could not.

**Administration → Operations** answers two questions: is the work the product does on a schedule — marking deadlines as breached, raising planned maintenance, chasing overdue returns, sending reminders — actually running, and which notifications did not reach anybody. It is the screen to open when something that should have happened by itself did not.

<img className="block dark:hidden" src="https://mintcdn.com/assetinfinity/yJwYjua5PLquNk6z/images/admin/platform-operations-light.png?fit=max&auto=format&n=yJwYjua5PLquNk6z&q=85&s=837c271ffceab45459987b747725a5f2" alt="The Operations screen listing the scheduled jobs with their state and next run" width="2880" height="1800" data-path="images/admin/platform-operations-light.png" />

<img className="hidden dark:block" src="https://mintcdn.com/assetinfinity/yJwYjua5PLquNk6z/images/admin/platform-operations-dark.png?fit=max&auto=format&n=yJwYjua5PLquNk6z&q=85&s=af05bff027aeb1a7835d857b70a264a6" alt="The Operations screen listing the scheduled jobs with their state and next run" width="2880" height="1800" data-path="images/admin/platform-operations-dark.png" />

The screen is available on every plan. It takes **Configure integrations** — System Administrator by default. See [roles](/setup/roles).

## The header

| Figure | What it shows |
| - | - |
| **jobs** | How many scheduled jobs exist |
| **not running** | Jobs that are failing, stalled or switched off |
| **did not arrive** | Notifications that were not delivered |

## Scheduled jobs

Each row is one piece of scheduled work, described in a sentence — *Mark work orders whose response or resolution deadline has passed*, *Raise the work for due occurrences whose plan generates automatically*, *Remind custodians of assets coming up on their return date*, *Expire lapsed certifications and warn about ones expiring soon*, and so on.

| Column | What it shows |
| - | - |
| **Job** | What the job does |
| **State** | **healthy**, **failing** (its last run went wrong), **stalled** (it is overdue and has not run), **never run**, or **switched off** |
| **Last success** | When it last completed |
| **Next run** | When it is next due — *due 47m ago* means it is late |
| **Failures today** | How many runs failed today |
| **Run now** | Starts the job immediately rather than waiting for its schedule |

Click a job to see **Recent runs of \[job]**: when each run **Started**, how long it **Took**, what it **Did**, its **Outcome** (**ran** or **failed**, with the error) and who **Started by** — the schedule or a person who pressed **Run now**. *This job has not run here* means it has never run for your organisation.

<Tip>
  Many jobs stalled at once usually means the background service that runs them has stopped — not that each job is broken. If a job shows **failing**, its recent runs carry the error to pass on to your support contact.
</Tip>

Jobs that are commonly worth watching:

| If this is not happening… | Look for the job that… |
| - | - |
| Work orders are not marked as breached, or not escalated | Marks work orders whose deadline has passed; raises work orders up their priority's escalation ladder |
| Approved requests are not carried out, or approvers are not reminded | Performs the actions approval has cleared; nudges whoever a step is waiting on; brings in a second approver |
| Planned maintenance does not appear | Generates maintenance occurrences; moves occurrences to due; raises the work for due occurrences |
| Reorder suggestions and purchase requests do not appear | Recomputes what needs buying; raises purchase requests for shortages |
| Scheduled reports and dashboards are not sent | Writes a delivery for every report or dashboard schedule whose period has come |
| Reliability, performance or energy figures are stale | The monthly and nightly roll-ups for reliability, technicians, vendors and energy |

## Did not arrive

<img className="block dark:hidden" src="https://mintcdn.com/assetinfinity/yJwYjua5PLquNk6z/images/admin/platform-operations-failed-light.png?fit=max&auto=format&n=yJwYjua5PLquNk6z&q=85&s=696c09d96ee45e55602c9d28b0040710" alt="The Did not arrive tab listing notifications that reached nobody and what to do about each" width="2880" height="1800" data-path="images/admin/platform-operations-failed-light.png" />

<img className="hidden dark:block" src="https://mintcdn.com/assetinfinity/yJwYjua5PLquNk6z/images/admin/platform-operations-failed-dark.png?fit=max&auto=format&n=yJwYjua5PLquNk6z&q=85&s=6c0fd69f3f686273183f0aea62a5901a" alt="The Did not arrive tab listing notifications that reached nobody and what to do about each" width="2880" height="1800" data-path="images/admin/platform-operations-failed-dark.png" />

Every notification the product tried to send and could not. *Everything sent, arrived* means the list is empty.

| Column | What it shows |
| - | - |
| **When** | When it was raised |
| **About** | Which event it was, and the channel (email, push, text message, Telegram, WhatsApp) |
| **To** | Who it was meant for, where known |
| **What to do** | The reason in plain words, for example *The subscription matched nobody — check the target* or *Nobody has signed in to the field app on a handset for this person* |
| **Send again** | Tries the delivery again once you have fixed the cause. Not offered where nobody could receive it — such as a subscription that matched nobody — because retrying would fail the same way |

Most rows are fixed on [Notifications](/config/notifications): an audience naming a role nobody holds, a channel that has not been set up, or a person with no reachable handset.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.