Two things keep work moving: knowing when it has waited too long, and making sure nothing that was switched off stays off by accident. This article covers both.
1. Set a service level on a trigger
A service level (SLA) is how long an item may wait before it is late. You can set one on a mailbox, folder or queue trigger.
- Save the trigger first. The fields only show on a saved trigger.
- Open the trigger again.
- Fill in SLA (minutes): how long an item may wait before it is late.
- Fill in At-risk window (minutes): how long before the SLA an item counts as at risk. For example, an SLA of 120 and an at-risk window of 30 marks an item at risk after 90 minutes.
Enter 0 to turn either off. The longest SLA you can set is 90 days.
A web queue has its own SLA (minutes) on the queue's form. See Working with queues.
How it is measured
The bot reports when the oldest item waiting on the trigger arrived: the oldest unread email that passes the trigger's filter, the oldest new file in the folder, or the oldest new queue item. It sends the time only, never the subject, sender or file name. Automize compares that time with the SLA every minute.
When a trigger becomes late or at risk, Automize sends a notification.
2. The Service levels page
Company admins can open Service levels from the menu, or from the button on the Live Runs page. It shows every mailbox, folder, timer, schedule and web queue that starts a process, one row each, late ones first.
| Column | What it shows |
|---|---|
| Process | The process, with High care when it is marked so, and the trigger's name. |
| Source | Mailbox, Folder, Queue trigger, Timer, Schedule or Web queue, and when it last started a run. |
| Waiting | How many items are waiting. |
| Oldest waited | How long the oldest item has waited. |
| SLA | The SLA you set. |
| State | Late, At risk, On time, Switched off (with when it comes back on), No reading, No SLA set or Runs on a timer. |
| Done within SLA | How much of the last 7 days' work was done in time, counted from when each item arrived. |
The line at the top sums it up: how many sources have work past their SLA, how many are at risk, and how many are switched off. Click Refresh to read it again.
If a row says Bot cannot read this trigger, the trigger was saved in a form the bot does not understand, so it never starts its process. Open the trigger and save it again.
3. Switch a trigger or process off for a set time
Triggers that were switched off and forgotten are a common reason a process quietly stops running. So whenever you switch a trigger or a process off, Automize asks how long for, and switches it back on by itself at that time.
- On a trigger, choose Pause (on its menu, or its row in the bot page's Triggers tab). On a process card, open its menu and choose Switch off. You can also switch several off at once from the library.
- Under Switch it back on, choose For 30 minutes, For 1 hour, For 4 hours, Until 07:00, Until a time I choose or No end time. The longest end time is 30 days.
- For No end time, fill in Why is it staying off?. The reason is shown on the bot page and in the pause history, so the next person knows.
- Click Switch off.
To switch it back on early, choose Enable. To pause every trigger on one bot, use Pause triggers on the bot's page. See Triggers and webhooks.
Reminders
Anything switched off for more than 3 days gets a reminder: to the person who switched it off, the process owner, the people it is shared with, and the company's admins. It repeats every 7 days while it stays off.
Company rules for pausing
A company admin sets these in Settings › Automation › Pausing:
- Only administrators may switch something off with no end time. Off by default: anyone may, as long as they give a reason.
- Longest a High care process may be switched off (hours). 8 hours by default. A High care process, its triggers, and a bot that runs one can only be switched off with an end time, and for no longer than this.
To mark a process as High care, open it and choose Edit › Care level.