Anypoint tells you something is stuck. nJAMS tells you where.
Executions are captured straight from the Mule flow, correlated across applications and made searchable. No log parsing, no changes to your application code.
| Process | Start | Duration |
|---|---|---|
| completed: experienceOrderStatusFlow | 08:44:03.117 | 88 ms |
| aborted: systemSapIdocFlow | 08:41:12.812 | 30.0 s |
| completed: systemCustomerLookupFlow | 08:41:12.560 | 410 ms |
| aborted, selected: processOrderToCashFlow | 08:41:12.402 | 30.6 s |
| completed: experienceCreateOrderFlow | 08:41:12.318 | 142 ms |
| completed: systemPriceLookupFlow | 08:38:41.902 | 96 ms |
| completed: processOrderValidationFlow | 08:38:41.774 | 128 ms |
| completed: experienceCreateOrderFlow | 08:38:41.690 | 155 ms |
Total events: 2
| Name | Event time | Message | Code |
|---|---|---|---|
| sap-idoc-sender | 08:41:42.917 | Connection timed out after 30000 ms | MULE:CONNECTIVITY |
| processOrderToCashFlow | 08:41:42.918 | Flow execution stopped | MULE:UNKNOWN |
java.util.concurrent.TimeoutException: connect timed out · order-to-cash/processOrderToCashFlow/processors/3
- {
- "message": {
- "payload": {
- "auftragsNr": "WS-2026-004711",
- "kunde": "Muster Industrieteile GmbH",
- "betrag": 12480.00,
- "waehrung": "EUR"
- },
- "attributes": {
- "x-correlation-id": "48fadc50-17b7-11f1-a6a7",
- "method": "POST",
- "requestUri": "/order-to-cash/orders"
- }
- }
- }
- Process name
- processOrderToCashFlow
- Correlation log ID
- 48fadc50-17b7-11f1
- Runtime version
- 4.9.1 - Mule EE
- Payload
- in full, replayable
Reconstructed nJAMS interface, sample data.
From Integration Platform to Business Insights: nJAMS Observability for MuleSoft
Playing the video opens a connection to Vimeo. No tracking cookies are set.
Two layers are covered. The third one decides.
Anypoint Monitoring reliably answers whether the infrastructure is running. It does not answer which business process is stuck, which data is affected and who needs to act.
The same run, from your role’s point of view.
The cause sits with the run, not scattered across log files.
Automatically detected flows, visual drill-downs and full-text search across process data, instead of correlating logs by hand across several applications.
- Automatic flow detection
- Full-text search across process data
- Drill-down to step level
See deviations before the phone rings.
A rule set detects missing or delayed events early. React to SLA-relevant events instead of reviewing them afterwards.
- Rule set for missing events
- SLA-relevant alerts
- Early deviation detection
One view across distributed deployments.
Hybrid and distributed landscapes in a single view, across CloudHub, Runtime Fabric and your own servers.
- CloudHub / Runtime Fabric / on-premise
- One view across environments
Payload context right at the flow.
Within the configured retention period, input and output data can be inspected in full for each flow run. Tracing can be switched on at runtime, data mapping is visible for each activity, and individual executions can be repeated with the original data.
- Transparency, input and output data fully visible
- Dynamic tracing, can be enabled at runtime
- Data mapping, visible for each activity
- Replay of individual executions with the original input data
Real runs as your test basis.
Replay for functional tests, performance tests and unit tests of integration processes and whole API chains.
- Functional tests
- Performance tests
- API chains end-to-end
Including what did not happen.
A rule set evaluates each individual message and reports what follows from it. Including an absence: the nightly batch run that did not start at half past one.
- Rule set per individual message
- Alert on events that fail to arrive
- Automatic flow detection
Fewer questions for IT.
Users search by order number, customer name or account number themselves and see whether an order has arrived in SAP.
- Search by business fields
- Status without a ticket
- Self-service for business users
Two components, no changes to your application code.
nJAMS Client
Runs inside the Mule application and captures transaction events straight from the flow executions. Related executions are correlated automatically.
nJAMS Server
Receives and persists the events, correlates them across application boundaries and keeps them available for search. Displays each individual run in the process model, opens payload and context, and allows replay of individual messages and batches.
Effect on operations.
Fewer blind spots in the process run
- Automatic flow detection
- Full-text search across process data
Lower risk on business-critical transactions
- Rule set for deviations and missing events
- React to SLA-relevant events
Greater resilience of the integration landscape
- Alert on events that fail to arrive
- Automatic detection of new and changed flows
An order was stuck between shop and SAP. Three departments looked for where.
An international group runs its business on MuleSoft as the central integration layer. The platform moved the data reliably. The question was what happens the one time it does not.
Before
- Log data spread across CloudHub and their own servers
- Whether an order had arrived in SAP could not be answered in one step
- Errors surfaced when a user called
Since then
- One process view across all Mule flows
- The step where it breaks is marked
- The payload that triggers the rejection sits next to it
Connected systems in the case described
- Integration
- MuleSoft
- ERP
- SAP
- CRM
- Microsoft Dynamics 365
- Commerce
- Shop
- MDM
- Master data
- WMS
- Warehouse
Your AI knows your systems. nJAMS records what happens inside them.
nJAMS captures process runs: which step, when, how long, carrying which business data. That is the context an AI system gets nowhere else. Log files hold lines, not runs.
nJAMS captures for your enterprise AI:
- how your processes actually run
- how often each path occurs
- which orders, customers and amounts depend on it
In the log file
- 08:41:42.917 MULE:CONNECTIVITY Connection timed out after 30000 ms
- 08:41:42.918 MULE:UNKNOWN Flow execution stopped
In nJAMS, the same moment
- Order no.
- WS-2026-004711
- Process
- processOrderToCashFlow
- Start
- 08:41:12.402
- Duration
- 30.6 s
- Stopped at
- sap-idoc-sender, timeout 30 s
- Not reached
- confirm-order
One view across all Mule deployments.
Client inside the Mule application, all deployment models, automatic flow detection, replay at message and batch level.
- CloudHub
- Runtime Fabric
- On-premise
- Hybrid
Common questions
How much effort is the rollout?
There are two parts, and the second one is easy to miss. The client is configured inside the Mule application, without changing your application code. The server is installed or rolled out in containers, which is a step of its own and needs an environment for it.
How is nJAMS sized?
Sizing depends on process volume, number of teams and deployment topology. That is settled in a scoping conversation, not through a blanket figure.
Who is behind nJAMS?
Salesfive, Salesforce and MuleSoft partner.
Beyond the Black Box: Unlocking 100% Visibility in MuleSoft with nJAMS
The larger a MuleSoft landscape grows, the harder it becomes to answer where exactly a business process is stuck and which data is affected. The white paper by Salesfive describes where standard tooling reaches that limit, and shows through an anonymised case study what changes when runs are captured together with their payload.
- The visibility limit in MuleSoft
- Capturing process steps and payloads
- Replay for development and test
- Anonymised case study, hybrid landscape
Show us your system landscape.
One conversation is enough to judge whether nJAMS changes anything for your Mule system landscape. Everything after that is your decision.