Actions FAQ
Answers to common questions about the five action categories. Pick the tab for your category.
Configuration
Why doesn't my datastore appear in the Datastore dropdown?
Only initialized datastores are listed, across all five operation types. A source datastore becomes initialized after successful Sync, Profile, and Scan runs, while an enrichment datastore needs only Sync and Profile. Run those operations first, then it appears in the dropdown. Export and Materialize ask for one thing more: the source datastore also needs an enrichment destination connected.
How many Operation actions can a Flow contain?
A single Flow accepts up to 10 Operation actions. Once the limit is reached, the Operation category is disabled in the List of Actions panel, and the Clone option is unavailable on Operation actions, until one of the existing Operation actions is removed.
Why is the Operation category disabled in the List of Actions panel?
Either the Flow already has 10 Operation actions, or the node you are chaining from is a Notification or Workflow action, which ends its branch and accepts no children.
Do the settings match running the operation directly on the datastore?
Yes. Each operation type's panel carries the same configuration options as running that operation on the datastore itself, including source datastore, table or file selection, read settings, and the type-specific options.
How can I tell my Operation nodes apart on the canvas?
Action nodes show inline summaries with key details, such as the name of the datastore the operation targets. Export nodes also display the asset type in their titles (for example, "Export Anomalies") so you can identify the exported content at a glance.
Saving and Publishing
I clicked Save on the settings panel. Why hasn't the action taken effect?
Save only stores the configuration as an unpublished change. The node shows a CHANGED badge and the You have unpublished changes banner appears. The Flow keeps running its last published version until you click Publish. See How a Flow Works.
Executions
Where can I follow the operations a Flow started?
On the Flows page's Executions tab, or on the Executions tab of the Flow itself. Opening an execution shows the Flow's graph read-only, with each node colored by its state, and actions that started an operation carry a shortcut to view the operation's results. See Flow Executions.
What happens to a running operation when I abort the execution?
Aborting the execution also stops the operations it started. Results captured before the abort are preserved. See Abort an Execution.
Archive and Delete
What's the difference between Archive and Delete?
Archive moves the anomaly out of the active queue but keeps the record, so it stays available for audits, compliance, or future reference. Delete removes the anomaly permanently, along with its comments and its links to tags, assignees, and external tickets. See How Anomaly Actions Work for the full list of effects.
Is Delete reversible?
No. Anomalies removed by a Delete action cannot be restored. If you might need the record later, use Archive instead.
Is Delete the same as the Discarded status?
No. Discarded is an anomaly status. The anomaly stays in the platform with a "no longer relevant" label. Delete removes the anomaly record entirely.
Inactivity and Filters
What counts as an inactive anomaly?
An anomaly is inactive when it has no updates or comments during the full Inactivity Period you set on the action. Any update or comment restarts the clock.
Do filter conditions replace the Inactivity Period?
No. Filter conditions add on top of the Inactivity Period, not in place of it, and all filters must match (AND logic). An anomaly is affected only when it meets every filter you fill in and has been inactive for the full period.
Announcing the Run
Can the Flow tell me how many anomalies were archived or deleted?
Yes. Chain a notification after the anomaly action; its message can carry the {{anomaly_count}} and {{inactivity_timeframe}} tokens with the run's numbers. See Message Variables.
Why isn't Duplicate offered as an archive reason?
The Duplicate reason requires manually linking a primary anomaly, which an automated action cannot do. Pick one of the other reasons in the Options dropdown.
General
Which tokens can I use in the message?
The available tokens depend on the Flow's trigger type, and the same set applies to every channel. The message editor's autocomplete only offers the valid ones, and a token outside that list is refused when you publish or test, with a message naming it. See Message Variables for the full reference.
Can I test a notification before publishing the Flow?
Yes, on Email, Slack, Microsoft Teams and PagerDuty, which offer a Test Notification button that sends a sample message with dummy data, and on the Webhook and HTTP Action, which offer Test Workflow. In App and n8n have no test: publish the Flow and check the bell icon or the receiving workflow after a real event.
Why was my publish refused because of a variable?
Because the message, or the Email subject, names a variable the Flow's trigger does not provide. A variable that is not provided would reach the recipient as literal text, so the platform refuses it instead: the Failed to publish dialog lists each refused action and the variables at fault. Open the {{ menu in the editor to see what the trigger does provide, or click Reset to defaults to return to a message written for it. A common cause is changing the Flow's trigger type after writing the message, since a token one trigger provides another may not.
Does Test Notification catch template errors?
Yes. A test runs the same check as publishing before it sends anything, so a message that names an unsupported variable or has a malformed conditional is refused with the same wording, shown under the button. What a test cannot tell you is how long a real value will be, because it fills variables with short sample values.
Are notifications sent in real time?
Yes. Each notification is dispatched while the Flow executes, as soon as the action runs, with no batching or waiting period in between. How long it takes to arrive after that is up to the destination, so an email or a Slack message can land a moment later than the execution shows it sent.
Can a Flow send more than one notification?
Yes. Add one notification action per destination, as sibling branches. Keep in mind that no action can be chained after a notification; it ends its branch.
Per Channel
Pick the channel for its specific questions.
Who receives In App notifications?
Every Admin user, plus every user belonging to a team associated with the event's target object (such as the datastore involved), regardless of role or team permission. When the target object has no assigned teams, or the trigger has no target object (Schedule and Manual), the users of every team receive it.
Can I target specific users with In App notifications?
No. Recipients are determined by the Admin role and by the teams of the event's target object; there is no per-user selection. To notify specific individuals, use Email or Slack notifications instead.
Where do In App notifications appear?
A new notification pops up as a toast in the bottom-right corner of the screen, and it stays available in the notification bell icon on the Qualytics navigation bar, where each user can view and manage their notifications.
Do In App notifications support all trigger types?
Yes, every trigger type can send them. On Schedule and Manual Flows the message carries only the {{flow_name}} token; the other triggers offer their full token sets.
Can I send notifications to multiple email addresses?
Yes. Enter them in the Email Addresses field, separated by ; or , (for example, user1@example.com, user2@example.com).
Is the email subject customizable?
Yes. The Email Subject field already comes filled in with the trigger's own default subject, and you can replace it with anything you like.
Can I use variables in the email subject?
Yes. The subject takes the same tokens as the message body, and the trigger's default subject is itself built from them. Publishing validates the subject's tokens along with the message's.
Is there a limit on the subject length?
Yes, 150 characters. A subject is a mail header, so it is always sent on one line, with any line breaks collapsed to spaces, and anything past 150 characters is cut with a trailing .... The trigger defaults stay well inside that; a subject reaches the limit after adding a token whose value has no length limit, such as {{operation_message}}.
Why does the email look different from before?
Notification emails are now formatted in the Qualytics layout, with the logo, a single reading column, and your Markdown rendered as headings, lists, and links, instead of arriving as plain characters. Nothing about how you write the message changed. A plain-text copy still travels with every email for clients that do not render HTML.
What happens if an email address is invalid?
The delivery may fail for that specific address; other valid addresses in the list still receive the notification. Use the Test Notification button to verify the addresses.
Do I need to set up the Slack integration first?
Yes. Connect the Slack integration in Settings > Integrations before selecting Slack channels in Flow notification actions. See the Slack Integration guide for setup instructions.
How do I choose which Slack channel receives the notifications?
The Channel dropdown shows all available channels from your connected Slack workspace. Select the channel where you want notifications to appear.
Can I send notifications to multiple Slack channels from the same Flow?
Yes. Add multiple Slack notification actions to the Flow, each configured with a different channel.
Do different trigger types generate different Slack messages?
Yes. Each trigger generates its own notification format. Operation completions include status and result details, anomaly detections include container and check information, and anomalous table or file detections include the scan target and anomaly count.
Is the Slack message customizable?
No. Slack composes the message itself, with one layout per event type, so the settings panel offers no message field to write in. The Preview section shows the layout the Flow's trigger will use.
What actions can I take on Slack notifications?
Depending on the event, the message carries View Operation, View Results, View Anomaly, or View Ticket, which open the matching page in Qualytics. Anomaly messages add an Acknowledge button and a menu (the ... control next to it) holding Comment and Archive, which act on the anomaly without leaving Slack.
Do I need to set up the Teams integration first?
Yes. Connect the Microsoft Teams integration in Settings > Integrations before selecting Teams channels in Flow notification actions. See the Microsoft Teams Integration guide for setup instructions.
How do I choose which Teams channel receives the notifications?
The Channel dropdown shows all available channels from your connected Teams workspace. Select the channel where you want notifications to appear.
Can I send notifications to multiple Teams channels from the same Flow?
Yes. Add multiple Microsoft Teams notification actions to the Flow, each configured with a different channel.
What format are Teams notifications?
Microsoft Teams notifications are sent as Adaptive Cards, a rich, structured format with sections, facts, and action buttons. The card format varies with the trigger type.
Can I customize the message?
No. Like Slack, Microsoft Teams composes the card itself, with one layout per event type, so there is no message field to write in. The Preview section shows the card the Flow's trigger will use.
How do I send PagerDuty notifications from a Flow?
Add a notification action and select PagerDuty. Configure the message and severity, and optionally add custom details or a Routing Key override. The PagerDuty integration must be connected first; see the PagerDuty Integration guide.
What severity levels are available?
Four levels: Info, Warning, Error, and Critical, with Info as the default. Each maps directly to PagerDuty's severity classification, which controls urgency and notification behavior based on your service's configuration.
Can I send different events to different PagerDuty services?
Yes. Each PagerDuty notification action supports a Routing Key Override. When set, that action's events route to a different PagerDuty service than the integration's default key.
What information is included in the PagerDuty incident?
The rendered message as the summary, the configured severity, your Qualytics instance as the source, the datastore as the component, the event type as the class, custom details (system-generated metadata plus your key-value pairs), and a direct link back to the relevant resource in Qualytics.
What reaches PagerDuty as the alert title?
The Summary you write. PagerDuty builds the alert title from the event summary, so whatever that field renders to is the headline on the incident. Each trigger starts you off with a one-line default written for that purpose, and Qualytics sends what you leave there without adding or replacing anything.
What happens if the Summary is too long?
PagerDuty caps the summary at 1024 characters and refuses an event that exceeds it, so no incident is opened. The notification is recorded as a failure in the Flow's execution logs. The trigger defaults stay well inside the limit; a summary usually crosses it only after adding a token with no length limit, such as {{anomaly_message}}. Test Notification cannot catch that case, because it fills tokens with short sample values rather than the real event's, so remove the unbounded token instead of relying on a passing test.
What happens if the PagerDuty integration is disconnected?
The notification step fails with an error, but the Flow itself keeps executing and the other actions are not affected.
Does the Test Notification button create a real incident?
Yes. Test notifications will create an incident in your PagerDuty service (unlike connection validation, which uses Change Events). A test is easy to spot: its variables are filled with sample values, and its summary is prefixed with [Test Notification]. A real Flow run sends neither.
Chaining
Can another action run after a workflow action?
No. A workflow action ends its branch of the Flow, so no child action can be attached after it. To run several things from the same trigger, add each action as a sibling branch. See How Workflow Actions Work.
Configuration
What's the difference between the Webhook and the HTTP Action?
A Webhook sends the event to a URL you provide, with an optional secret. The HTTP Action gives you more control over the request, including the authentication scheme, for endpoints that expect a specific call.
Messages, Testing, and Payload
What variables can I use in the message?
The Message field supports dynamic tokens such as {{flow_name}}, {{operation_type}}, and {{datastore_name}}, replaced with real event data when the Flow runs. As you type, the autocomplete suggests the valid ones, and a token the trigger does not provide is refused when you publish, with a message naming it. See Message Variables.
How do I check that my endpoint works before publishing?
Click the Test Workflow button in the Webhook or HTTP Action settings panel. It sends a test notification to the URL you entered, and a confirmation message appears when the endpoint is reachable.
What do I need in n8n before configuring the n8n action?
Install the Qualytics community node in your n8n instance with npm install @qualytics/n8n-nodes-qualytics, then create the n8n webhook whose URL you will paste into the action. See Configure an n8n Action.
Why are some fields missing from the payload?
The datastore, operation, and target_link fields are only sent when the trigger provides them. The context lists (anomalies, containers, quality checks) are always present, but arrive empty when the event carries nothing for them. See Payload Structure.
Which authentication methods does the HTTP Action support?
Three auth types are available: Basic (username and password sent with each request), Bearer (a token in the request header), and Digest (a hashed combination of the username, password, and request details). The Username and Secret fields carry the credentials.
Availability
Which ticketing actions are available?
Two: Create Ticket, which opens a new ticket when the Flow runs, and Update Ticket Status, which moves existing tickets to a target status. See Create a Ticket and Update Ticket Status.
Which triggers support ticketing actions?
Only Anomaly Detected and Anomaly Status Changed. With any other trigger type, the Ticketing category appears in the List of Actions panel but stays disabled.
Do ticketing actions require any setup outside the Flow?
Yes. They require a connected ticketing integration, such as Jira or ServiceNow. See Ticketing Integrations.
Working with Tickets
Can I create tickets in more than one tracker from the same Flow?
Yes. Add one ticketing action per connected integration; each action names the tracker it targets.
Can the ticket's details appear in a notification?
Yes. A notification chained after Create Ticket can carry the {{ticket_number}}, {{ticket_url}}, and {{ticket_status}} tokens, and one chained after Update Ticket Status can carry {{tickets_updated_count}} and {{ticket_status}}. See Message Variables.
What happens if my ticketing integration is unavailable during execution?
The ticketing action may not complete successfully. The execution's details record the failure to assist with troubleshooting; the rest of the Flow is not affected.