View Runtime Logs

Runtime logs record backend activity in your app. When an Actionflow fails, data changes unexpectedly, or an API call returns an error, start with the relevant log category and then narrow the results by time, level, field, or traceId.

Open runtime logs

You can open runtime logs from either of these locations:

Click Runtime Logs on the project details page.
Click the log icon in the upper-right corner of the editor.

Open runtime logs from the Momen project details page

Log retention depends on the project plan:

Plan	Retention
Free	1 day
Basic	7 days
Pro	30 days

Before troubleshooting, make sure the incident occurred within the retention period.

Choose a log category

Select a category based on where the issue occurred. This removes unrelated records before you begin searching.

Category	What it records	Use it to investigate
Actionflow	Actionflow and node execution	Node inputs, outputs, execution order, and errors
Database operations	Inserts, updates, and deletes, including their source and details	Why a record was created, changed, or deleted
AI	AI Agent activity, including model and message information	AI Agent calls and conversations
API	API request headers, bodies, and related details	Third-party API calls
Trigger	Trigger time, type, and related information	Whether an Actionflow started from the expected trigger
Gateway	Frontend requests sent to the backend and requests received through webhooks	Whether a request reached the backend and what it contained
Deployment and release	Records related to syncing changes and publishing	Whether backend configuration was updated successfully
GraphQL	GraphQL requests received by the backend	Data query and mutation requests

Actionflow logs

Each Actionflow node typically generates one log when it starts and another when it finishes. Calling context.log() in a Run code node also creates a log entry.

View Actionflow log fields and node types

Node logs include these common fields:

Field	Description
`traceId`	Connects entries generated by the same request
`nodeType`	Node type
`nodeName`	Node name
`version`	Actionflow version
`input`	Node input and data available to the node
`output`	Node execution result
`startAt`	Node start time
`endAt`	Node end time

Common nodes use the following type values in logs:

Node	`nodeType`
Input	`FLOW_START`
Get data	`QUERY_RECORD`
Update data	`UPDATE_RECORD`
Add data	`INSERT_RECORD`
Delete data	`DELETE_RECORD`
Run code	`CUSTOM_CODE`
Assign role	`ADD_ROLE_TO_ACCOUNT`
Remove role	`REMOVE_ROLE_FROM_ACCOUNT`
Assign variable	`UPDATE_GLOBAL_VARIABLES`
Branch start	`BRANCH_SEPARATION`
Branch end	`BRANCH_MERGE`

Gateway logs

Coverage: Frontend database requests, APIs, AI, and most other requests that pass through the gateway.
Visible data: Gateway logs show outgoing requests but not their results.
Next step: To inspect the execution result, switch to the relevant category and query the same traceId.

Filter and query logs

Time and level: Set the time range for the incident and filter by log level.
Bar chart: Review log volume over time and select a segment to update the time range.
Log details: Open an entry and select a field value to add it to the query.
Query statement: Narrow the results by keyword or field.

Filter and inspect runtime logs in Momen

Common query patterns include:

Goal	Example query
Find logs containing a keyword	`account`
Match an exact field value	`requestMethod: POST`
Use `*` as a text wildcard	`traceId: 20241024*`
Match a numeric range	`responseSize: [5000 to 10000]`
Combine conditions	`account and status: 200 and not requestMethod: GET`
Query a nested field	`requestBody.query: mutation*`
Match a complete `traceId`	`traceId: "c2ecac0b-cce8-463b-b5d2-7ee7aa8f703020241128114815"`

Use and, or, and not to combine conditions. Available fields and values depend on the fields shown in each log entry.

Trace a request

Use traceId to connect log entries generated by the same request.

Reproduce the issue

Note the time and entry point, then reproduce the issue once if possible. This avoids creating many similar log entries.

Choose the category closest to the issue, set a short time range, and filter by level, Actionflow name, endpoint, or another known field.

Copy the trace ID

Open the log details and copy the value of traceId.

Run a traceId: "complete value" query. To inspect another category, keep the query and switch categories.

Review the execution in order

Compare the inputs, outputs, status, and errors in chronological order to identify where the request diverged from the expected result.

⚠️

Logs can contain request parameters, node inputs, API headers, and other business data. Remove secrets, tokens, and personal data before sharing a screenshot or complete log entry.

To analyze logs together with browser requests and page state, see Troubleshoot an App.