Frequently Asked Questions — Automation Monitor for Jira

How fast will I find out a rule died?

Worst case ≈ the monitor's interval × 1.5, plus up to 5 minutes for the next sweep. Example: an hourly rule is flagged within ~95 minutes of its last heartbeat. A rule that reports failure (?status=fail) alerts immediately on that heartbeat.

The alert issue exists, but the email came late (or not at all).

Two separate things: the app creates the issue (that part you can see timestamped); Jira delivers notifications. Jira batches notification emails for up to ~10 minutes by default — each user can switch to individual emails in Personal settings → Email.

Also check who the project notifies on Issue Created: out of the box Jira emails watchers, assignee, and reporter — for app-created issues that's nobody, so set the project's Default assignee to your on-call person (Project settings → Details → Default assignee = Project lead).

No alert issue was created at all.

Open the app page — if you see a delivery warning, press "Send test alert" in settings. If the message mentions "the target project doesn't exist or you don't have permission to create issues in it" (Jira returns this as HTTP 400, sometimes 403), it is one of two things: the project was deleted or archived, or the app lacks permission on it.

Fix: pick an existing project in the settings and save, or grant the atlassian-addons-project-access role the Create Issues and Add Comments permissions in that project's permission scheme. Saving the settings retries delivery straight away — automatic background retries stop after 5 attempts so the app never hammers a broken project.

I resumed a paused monitor and it shows WAITING instead of HEALTHY.

By design. Resume clears the heartbeat history so a stale timestamp can't cause an instant false alarm. The monitor re-arms on its next real heartbeat.

Does this slow my automation rules down?

No, if you keep the web request as the last action and leave "Delay execution until response" off — it's fire-and-forget.

What data does the app store, and where?

Monitor names, intervals, timestamps, and counts — nothing else. No issue content, no personal data. Everything is stored inside your own Atlassian site (Runs on Atlassian); nothing leaves it.

Limits?

10 monitors per site on the free version.

The heartbeat URL says "atlassian-dev.net" — is this a test URL?

No. That's the standard address Atlassian gives every Forge app's web endpoint, in every environment including production. The "hello" and "dev" in it are part of Atlassian's infrastructure naming, not an indication of a test or development build.

How do I change a monitor's URL if it gets shared by mistake?

Delete the monitor and create a new one, then update the URL in your rule. The old URL stops working immediately. Heartbeat history resets; the rule itself is unaffected.