JAMS Scheduler: Routing alerts by folder or job

Routing JAMS Scheduler alerts by folder or job

Most teams attach their JAMS Scheduler Notification Job high in the tree, often at the root, so it fires for every job beneath it. Routing lets one JAMS Connection decide which of those jobs page anyone, and who each one pages.

Example: failures in \Finance\ page the Finance on-call, failures in \Warehouse\ page Operations, and nothing else pages at all.

Routing is off by default. With it off, the Connection behaves as it always has: every job it receives pages its own recipients, and you can attach the Notification Job wherever you like.

Turn on routing

Go to /integrations/jams and click Edit on the Connection.
Below the Connection's settings, find Route by JAMS folder or job and tick Route alerts by folder or job.
Click + Add a rule and give it a name, such as Finance.
Choose its folders and jobs (see below), its recipients, its channels and its escalation time.
Click Save routing.

With routing on, only the folders and jobs your rules select page anyone. A failed job that no rule covers is recorded, but nobody is paged. To catch everything else, tick All of JAMS on one rule. More specific rules still win.

Choosing folders and jobs

The picker reads your JAMS Scheduler folders and jobs through the JAMS Incident Connector. It needs connector 0.9.13 or later. The list refreshes by itself when it's more than 15 minutes old, and Rescan refreshes it on demand.

Folders: tick a folder to cover it, every subfolder beneath it, and any job added there later. Use Find a folder to search a large tree.
Jobs: click a folder's name to see the jobs directly in it. All jobs is the same as ticking the folder. Untick it to choose individual jobs, which then covers only those jobs, not jobs added later.

To add a job JAMS doesn't have yet, use Or type a folder or job path.

Each folder or job can belong to one rule. When rules overlap, the most specific one wins: a job beats a folder, and a deeper folder beats a shallower one. The order you list the rules in doesn't matter.

Sequences and Workflows cover all their steps

A Sequence or Workflow step runs the job its step points at, and that job can live in a completely different folder from the Sequence itself. This is common when a Sequence was copied from another folder, because the copy still points at the original jobs.

A rule that covers a Sequence or Workflow covers every step it runs, wherever each step's job lives. It covers it if:

the Sequence or Workflow is ticked as a job, or
it sits in a ticked folder.

Example: \Payroll-Test\Nightly Sequence has a step that runs \Payroll\Upload Payroll. A rule on \Payroll-Test\ does cover that step's failure, because the Sequence that ran it is in \Payroll-Test\.

When more than one rule could take a step:

The Sequence's rule wins over a rule for the step job's own folder, so one Sequence always pages one team.
The nearest container wins when Sequences and Workflows are nested. A Sequence inside a Workflow pages the Sequence's rule if it has one, otherwise the Workflow's.
Otherwise the job's own folder or job decides, as for any job. A rule matches the folder of the job that failed, not the folder of the Notification Job.

This needs the current Notification Job script

Covering steps through their Sequence or Workflow needs the Notification Job script from 21 October 2026 or later, which sends the folder each Sequence and Workflow lives in. To update it:

On /integrations/jams, click Show Snippet on the Connection.
Copy it into the source of your Notification Job in JAMS Scheduler, replacing the old script.

With an older script, routing still works, but a step is matched only by its own job's folder. The routing card says so next to any step it missed. If you don't use routing, you don't need to update the script.

Seeing what was not paged

While routing is on, the routing card lists recent jobs no rule covered. It shows each job's full path, how many times it failed in the last 7 days, and when it last failed. If a job there should have paged someone, add its folder, or the job itself, to a rule.

This list is the quickest way to spot a rule on the wrong folder. When a missed job was a Sequence or Workflow step, the list names the Sequence and its folder, so you can tick it.

Turning routing off, and removing rules

Untick the switch and save to go back to the Connection's own recipients for every job. Your rules are kept for when you switch routing back on.
Remove every rule and save to turn routing off and delete the rules.
Switching the Connection off, or a maintenance window on the Connection, still silences every job, whether a rule covers it or not.

Each rule has its own flow, shown on /flows as Managed by Integrations → JAMS (routing rule). Edit there opens the Connection. Deleting that flow deletes the rule.

Good to know

Routing by a job's own folder works with any Notification Job script. JAMS Scheduler already sends that folder with every alert. Only covering steps through their Sequence or Workflow needs the current script (above).
Rules inherit the Connection's settings: the actions recipients can take, what happens after a restart, which connector runs remediation, and the cooldown.
An alert JAMS sends without a readable folder or job goes to the Connection's own recipients, so it is never lost.