Jobs
Most of Synacl is continuous: dashboards show now, events scroll past, charts cover whatever range you pick. A job is the opposite. It has a start, an end, and a self-contained record of everything in between.
Use jobs when the unit of work matters more than the calendar — a drilling operation, a genset run, a compaction pass, a production batch, a service visit. At the end you get a job report: one document, identified by an automatically assigned job number, that you can export and hand to a customer, a supervisor, or an auditor.
What a job records
| Identity | Job number (assigned automatically), machine, operator, start and end time, duration |
| Location | Latitude and longitude, captured when the job starts, if the machine has a GPS device |
| Before and after | Chosen readings taken once at the start and once at the end, with the change between them |
| Throughout | Chosen readings recorded at a fixed interval for the whole job, as charts and a table |
| Alarms | Every alarm raised during the job, when it cleared, and how long it lasted |
| Rule activity | Which rules fired, how often, and how many actions were dispatched |
Job profiles
A job profile describes one machine, so you configure it once rather than per job. It says which devices belong to the machine, which readings to capture at the boundaries, which to record throughout, and how job numbers are formatted.
Go to Jobs → Job profiles → New profile. See Setting up a job profile for a walk-through.
Starting and ending a job
From the Jobs page, pick the machine, optionally type the operator's name, and press Start job. The job number is assigned for you — there is nothing to type or remember.
While a job runs it appears at the top of the Jobs page with a live elapsed timer. Press End job to close it. The report is produced the first time you open it.
Only one job can be running per machine at a time. If a second start arrives — someone else pressing the button, or the machine reporting one — it is folded into the job already running rather than opening a duplicate.
If a job is never ended
A job that runs past its profile's maximum length, or whose machine goes offline for longer than the profile's grace period, is closed automatically. When that happens the report says so: it shows the window the job could have ended in rather than presenting a made-up exact time, and the job is flagged as approximate in the list.
This is deliberate. An automatically closed job is weaker evidence than one you ended yourself, and the report never hides the difference.
What the report will not do
- It will not invent readings. If a value was not recorded at the start or end of the job, the report shows a dash, not a zero. A zero would read as a measurement.
- It will not paper over an outage. If the machine was offline mid-job, the chart breaks across the gap instead of drawing a line through it, and the coverage figure drops accordingly.
- It will not silently change later. Alarm and rule counts are frozen when the job ends, so a report you read next year says the same thing it said the day it was produced — even after the underlying events have aged out of your retention window.