Eventstream topics

Capacity overview events and Timepoints details

rgsalido — Tue, 11 Aug 2026 11:04:27 GMT

Good morning,

is there any update on when the information about the operations provided by the 'Capacity overview events' connector from an eventstream will be included?

thanks in advance

Fabric Mirrored Database Change Feed Eventstream source connector stuck in a restart loop

philhansen — Wed, 22 Jul 2026 14:56:46 GMT

I tried using the database change feed with EventStream to trigger a notebook on changes, but ran into a problem. Has anyone else successfully used the change feed with an EventStream? Any advice on solving the error below?

We built open mirroring (Mirrored Database) → Delta Change Data Feed → Eventstream

Mirrored Database Change Feed source → Filter operator (keep specific table) → Activator.

The Change Feed source connector repeatedly logs, in the Eventstream Runtime logs:
Warning Validate connector: '<connector-id>'.
Warning Connector runtime config validation is unhealthy. The connector will be restarted.
Failed reason: Connector validation failed, please retry the request.

This recurs several times per hour. Because each restart appears to re-read from the snapshot rather than resume from a committed offset, the stream is continuously flooded with duplicate read events, all carrying the same (latest) EXPORT_TIME, arriving hours after the source batch was actually processed.

2026-07-13 03:59:39 — "Connector runtime config validation is unhealthy. The connector will be restarted."
2026-07-13 11:18:16 — "Failed reason: Connector validation failed, please retry the request."
2026-07-13 11:20:07 — "Connector runtime config validation is unhealthy. The connector will be restarted."
2026-07-13 12:19:35 — "Failed reason: Connector validation failed, please retry the request."
2026-07-13 12:21:15 — "Connector runtime config validation is unhealthy. The connector will be restarted."

Disable multiple eventstreams

RichCross — Wed, 01 Jul 2026 09:43:15 GMT

Hello

Is there a way to disable multiple evenstreams in one go rather than having to go into each one and deactivate manually?

I'm basically looking to get some code to run if at all possible.

Thanks

Heartbeat Action Query Not Sent to Source

lyancy — Wed, 24 Jun 2026 06:11:13 GMT

I created an Eventstream with Postgre SQL CDC as source. I want to use heartbeat action query to ensure that Eventstream to keep the replication slot size low in source. There is no issue/error in the Eventstream but the heartbeat query is never been sent to source. Is there a configuration I missed?

This is the table that I created and the initial value:

CREATE TABLE public.debezium_heartbeat (

id SERIAL PRIMARY KEY,

timestamp TIMESTAMP WITH TIME ZONE

);

INSERT INTO public.debezium_heartbeat (id, timestamp) VALUES (1, now());

This is the heartbeat action query that I put on the Eventstream source:

UPDATE public.debezium_heartbeat SET timestamp = now() WHERE id = 1

Spark Notebook termination after ~14 days (Eventstream Destination)

Hamidr — Thu, 30 Apr 2026 19:08:27 GMT

I have a realtime ingestion process as the following :

In an eventstream ingesting events from an eventhub
Doing some operations (filter events , ...)
Set a notebook as destination of the eventstream
In the notebook : apply upsert operation and append or updata data into a Lakehouse table

when I add the notebook as destination an instance of notebook starts to run (I can see the activity in monitor) and working fine (the events are ingested and in Lakehouse table, evrything looks great) but after almost 14 days the notebook suddenly stops and I need to manually delete the notebook from eventstrteam and add it again as destination.
Does it seem the native behaviour of Fabric to stop spending resources or there is a setting I can set it to let it run forever?
Thanks

SQL Operator with multiple destinations

msturzen — Thu, 30 Apr 2026 13:27:11 GMT

Hi all,
a few weeks back I was happy to read the SQL Operator for Eventstream is GA AND supports multiple destinations.

I tried to add a second destination to an existing SQL Operator. That does not work (for me).
I keep getting thr error: "Failed to create data source: <output-alias> due to the following error: Long running operation failed with status 'Failed'. Additional Info:'Stream Analytics job has validation errors: The output <output-alias> used in the query was not defined.'"
The error occurs after publishing, when activating source (works fine) and destinations (fail).

And I have run out of ideas what it could be. I tried:
* Create a new Eventstream
* Use different kinds of destinations (EH, LH)
* Naming things: with and wihtout "-", "_", camelCase, lowercase
* The order when and where and how to define the destinations (in the SQL editor first, in the Eventstream "main" view first, ....)
* End each of the "Select into. ..." with and without ";"
* There are no problems shown in the SQL Editor, Test query didn't complain either.

One destination works perfectly fine. Two not at all (for me).
What I observed: If I change the order of the "select into ..."s the other output-alias is the one failing. ==> It's always the first "select into ..." the error is coming from. (Probably trying to connect to this first, second probably not even tried after first fails).
Maybe important: I use several CTEs to prepare the events before the final Selects. Not both destinations use the same CTE as basis for their tables.

The fallback might be to have two SQL Operators, duplicating some code and stuff. I'm not looking forward to this, esp. since in ASA it (as I read) works.

Any help, hints, fixes, education what I made wrong are highly appreciated.

thx
Martin

Deserialization error when publishing events to Eventstream via Event Grid after initial success

AditiPattnaik — Thu, 30 Apr 2026 06:41:59 GMT

Hi,

I’m trying to ingest events into Microsoft Fabric Eventstream using Azure Event Grid and ran into an issue that I’m unable to understand.

Here’s what I did step by step:

I created an Event Grid Namespace, followed by a custom Topic.
In Fabric Eventstream, I configured a Custom Source by selecting the Event Grid namespace and topic.
From the Azure portal, I noted down the topic endpoint and access key.
Using Azure Cloud Shell, I started publishing events to the Event Grid topic.

Initially, my plan was to ingest 10 events at once, but that didn’t work. So I changed the approach and started publishing the events one by one.
This is where the issue started, the first two events were published and consumed by Eventstream successfully.When I tried to publish the third event, Eventstream threw a deserialization error.Despite that, only the third event (and subsequent ones) fail with a deserialization issue. What’s confusing is that the schema is exactly the same for all events and I followed the same steps, used the same format, and didn’t change anything between events.

Has anyone faced a similar issue when publishing events from Event Grid to Fabric Eventstream?
What could cause deserialization? Any known limitations or best practices for publishing events via Event Grid to Eventstream?

Thank You

Deleted workspace still showing CU usage in Capacity Metrics (Eventstream)

Shreya_Barhate — Wed, 29 Apr 2026 06:15:16 GMT

Hi all,
We deleted a workspace, but the Fabric Capacity Metrics app still shows CU usage for that workspace, and the top consumer listed is an Eventstream.
What’s confusing is that Eventstream doesn’t appear in the supported item types for Item Recovery, so it seems like a non‑recoverable item.
https://learn.microsoft.com/en-us/fabric/admin/retention-recovery#supported-item-types

So why would it still show up as consuming CUs after the workspace was deleted?
Is this expected because a deleted workspace stays in workspace retention for a period before it’s permanently removed?
If yes, what’s the correct way to stop the CU usage, do we need to permanently delete the workspace from the admin portal during the retention window?

Thanks,
Shreya

Event Schema Inside Eventstream Never Matching

Kelderic — Mon, 06 Apr 2026 10:43:25 GMT

I am exploring the new Event Schemas as a way to pre-define my schemas and replace the multiple-schema inferencing. I have created a Event Schema Set, with three Event Schemas, all very simple with five or less columns each.

In the screenshot above, one schema shows two fields."TermsCode" and "Description", both string.

I've created a new Event Stream, and associated it with the event schemas:

I then pushed in data, which in the preview, shows the correct matching headers. But the data is only seen as "Unknown schema":

Has anyone gotten event schemas to work?

Edit: I am using the new Event Schemas system, rather than Multiple Schema Inferencing. These are different and exclusionary. I have event streams working with inferred multi-schema and I'm trying to switch to defined event schemas.

Different schema into different destination

Srisakthi — Fri, 27 Mar 2026 13:28:45 GMT

Hello Everyone,

I have a scenario where I'm getting events with two diffferent schema flowing into event stream and i want to push those events into two different destination tables or two different destination(event house and lakehouse). Is there any best way to achieve this?

Do we need to create two different event stream?

What should be the best practise to handle this situation?

Regards,

Srisakthi

GA availability for streaming connector virtual network

akshatthakar1 — Wed, 25 Mar 2026 09:22:28 GMT

We understand that Event Stream setup for Private network source database CDC needs streaming connector virtual network. When will Streaming Virtual Network data gateway for event stream SQL CDC source be GA available, as it is in preview right now.

Options for High‑Frequency, Low‑Latency API Ingestion

prathijp — Tue, 17 Mar 2026 10:37:18 GMT

I’m looking for guidance on options for invoking external APIs(with dynamic date parameters in specific format like '202602152359') at very high frequency (every 2 minutes or less) and making the data available with minimal end‑to‑end latency.

Requirements:

Invoke external APIs at 2‑minute intervals (or less).
Parse incoming XML/JSON data into structured tables.
Refresh user reports/streamlit dashboards quickly, alongside other enterprise data, to support near‑real‑time decision‑making wihtout the delay of sql endpoint refresh.

We currently use a Fabric medallion architecture with payload based ingestions. However, this approach is not well‑suited for this use case, as running pipelines/notebooks on 2–5 minute schedules is triggering 70% Fabric capacity alerts.

Creating Fabric Eventstream via API: How to complete the KQL Data Connection configuration?

rgsalido — Mon, 16 Mar 2026 16:24:59 GMT

Hi everyone,

I'm trying to create a Fabric Eventstream programmatically using Python by calling the Microsoft Fabric REST API.

My current process is the following:

Create the Eventstream via the API.
Update the Eventstream definition.
Configure the destination as an Eventhouse with Direct Ingestion.

With this approach, I can successfully:

Create the Eventstream
Map the database
Map the table in the Eventhouse

However, I'm stuck on the final configuration step.

When doing this through the Fabric UI, after publishing the topology there is an additional step called "Configure KQL Data Connection", which finalizes the connection between the Eventstream and the Eventhouse table.

I haven't been able to find how to perform this last configuration step through the API.

So my question is:

Is there a way to programmatically configure the equivalent of "Configure KQL Data Connection" for an Eventhouse destination using the Fabric API?

Any documentation, example, or guidance would be greatly appreciated.

Thanks in advance.

Azure IotHub connection fails with Managed Private endpoint

aswary — Fri, 13 Feb 2026 12:32:24 GMT

I am trying to pull live data from Azure IotHub by following these 2 documentations for establishing connection between Fabric and Azure:

- https://learn.microsoft.com/en-us/fabric/security/security-managed-private-endpoints-create

- https://learn.microsoft.com/en-us/fabric/real-time-intelligence/event-streams/add-source-azure-iot-hub?pivots=enhanced-capabilities

The Azure is behind a Vnet and firewall and doesn't allow public access.

I have the private endpoint created and approved on IotHUB. When I try to create an eventstream and create a new connection to this IoTHub, it fails with this error:

{

 "code":"AsaDataPreviewFailure",
 "message":"Unexpected status code when previewing input data. Expected Accepted or OK, but received BadRequest. Error message: {status":"ErrorConnectingToInput",
 "eventsDownloadUrl":null,
 "error":{
 "code":"BadArgument",
 "message":"Unable to connect to IoT Hub 'azure-iot-hub' using connection string. Please make sure you have configured the IoT Hub input correctly. Exception details System.IO.IOException: Transport is closed ---> System.ObjectDisposedException: Cannot access a disposed object.Object name: 'sasl6285'. --- End of inner exception stack trace --- at Microsoft.Azure.Amqp.AsyncResult.End[TAsyncResult](IAsyncResult result) at Microsoft.Azure.Amqp.AmqpObject.OpenAsyncResult.End(IAsyncResult result) at Microsoft.Azure.Amqp.AmqpObject.EndOpen(IAsyncResult result) at Microsoft.Azure.Amqp.Transport.AmqpTransportInitiator.HandleTransportOpened(IAsyncResult result) at Microsoft.Azure.Amqp.Transport.AmqpTransportInitiator.OnTransportOpenCompete(IAsyncResult result)--- End of stack trace from previous location --- at Microsoft.Azure.Amqp.AsyncResult.End[TAsyncResult](IAsyncResult result) at Microsoft.Azure.Amqp.Transport.AmqpTransportInitiator.<>c.<ConnectAsync>b__17_1(IAsyncResult r) at System.Threading.Tasks.TaskFactory`1.FromAsyncCoreLogic(IAsyncResult iar, Func`2 endFunction, Action`1 endAction, Task`1 promise, Boolean requiresSynchronization)--- End of stack trace from previous location --- at Microsoft.Streaming.Services.TestConnectionAndSampleDataService.EventHubClient.IotHubConnection.CreateAndOpenConnectionAsync(Uri serviceEndpoint, String iotHubName, String sharedAccessKeyName, String sharedAccessKey, TimeSpan timeout) in C:__w1sProductSourceServicesTestConnectionAndSampleDataEventHubClientIotHubConnection.cs:line 196 at Microsoft.Streaming.Services.TestConnectionAndSampleDataService.EventHubClient.IotHubConnection.RequestEventHubsConnectionStringAsync(String iotHubConnectionString, Nullable`1 timeout) in C:__w1sProductSourceServicesTestConnectionAndSampleDataEventHubClientIotHubConnection.cs:line 76 at Microsoft.Streaming.Services.TestConnectionAndSampleDataService.JobAnnotationService.JobAnnotator.GetIotHubConfig(String iotHubConnectionString, String iotHubName) in C:__w1sProductSourceServicesTestConnectionAndSampleDataJobAnnotationServiceJobAnnotator.cs:line 389",
 "details":null},
 "diagnostics":null,
 "lastArrivalTime":null}"}

This is how I have created the IOThub connection (as suggested in the doc):

and then:

Error:

Eventstream vs Eventhouse

Srisakthi — Wed, 11 Feb 2026 10:00:55 GMT

Hello Everyone,

I'm doing analysis on when to use event stream and when to eventhouse for ingestion of streaming data. As i understand event stream can be used for continuous streaming data and event house for both batch and streaming data which doesnt require event stream. Is there any other recommendations?

Regards,

Srisakthi

Amazon Kinesis Firehose as source for Event Stream

Srisakthi — Fri, 06 Feb 2026 12:44:46 GMT

Hello Everyone,

Is there anyway we could use Kinesis Firehose as source to event stream of RTI? I see Kinesis Data Stream as source but i could not find Kinesis Firehose. It would b great if there is any option to use Kinesis as Source.

Few more architectural question

1. Is it good idea to have both AWS kinesis and Event Stream together ?

2. Is it good to have events consumed from AWS Eventbridge directly into Fabric Event Stream?

Regards,

Srisakthi

Defining Event Grid as a source - Namespace doesn't appear

TePe — Mon, 05 Jan 2026 16:28:04 GMT

Hi,

I want to define Event Grid as a source for my Eventstream to consume MQTT events (the MQTT connector doesn't work for me). However, after selecting my subscription the Event Grid Namespace list is empty. I recreated the Event Grid Namespace, I checked the settings in the Fabric Tenant Admin, I enabled MQTT and Routing, I'm not sure what I missed. Any idea?

Thanks,

Thomas

MQTT Events debugging

TePe — Fri, 02 Jan 2026 16:29:37 GMT

Hi,

I have a MQTT Broker (HiveMQ) which is receiving data from Home Assistant (Mosquitto).

I see data flowing into the MQTT source, at least in the "Data Insights" the chart shows events arriving. Runtime Logs are empty so I assume that there are no errors.

In the Message Stream I see in the "Data Insights" tab again a chart, so there should be data arriving. However, the Data Preview tab is empty, I can't see any events.

I added an Eventhouse destination. The Data Preview is again empty. But there I get Data Deserialization Errors:

{
    "ErrorCode": {
        "value": "Diagnostic: Diagnostic Error",
        "type": "string"
    },
    "TimeGenerated": {
        "value": "2026-01-02 16:18:40Z",
        "type": "datetime"
    },
    "Level": {
        "value": "Warning",
        "type": "string"
    },
    "Category": {
        "value": "Warning",
        "type": "string"
    },
    "ResourceId": {
        "value": "",
        "type": "string"
    },
    "SubscriptionId": {
        "value": "",
        "type": "string"
    },
    "ResourceGroup": {
        "value": "",
        "type": "string"
    },
    "JobName": {
        "value": "",
        "type": "string"
    },
    "OperationName": {
        "value": "",
        "type": "string"
    },
    "Status": {
        "value": "Warning",
        "type": "string"
    },
    "Source": {
        "value": "",
        "type": "string"
    },
    "TypeProperty": {
        "value": "Warning",
        "type": "string"
    },
    "DataErrorType": {
        "value": "",
        "type": "string"
    },
    "ErrorCategory": {
        "value": "Warning",
        "type": "string"
    },
    "BriefMessage": {
        "value": "",
        "type": "string"
    },
    "Message": {
        "value": "First Occurred: 1/2/2026 4:18:38 PM UTC | Resource Name: es-embeddedeventhub | Message: Source 'EventHubInputAdapter' had 14 occurrences of kind 'InputDeserializerError.InvalidData' between processing times '2026-01-02T16:18:36.6674295Z' and '2026-01-02T16:18:38.3290880Z'. Json input stream should either be an array of objects or line separated objects. Found token type: Float\r\n",
        "type": "string"
    },
    "FromTimestamp": {
        "value": null,
        "type": "datetime"
    },
    "ToTimestamp": {
        "value": null,
        "type": "datetime"
    },
    "EventCount": {
        "value": 0,
        "type": "long"
    },
    "GroupSize": {
        "value": 1,
        "type": "long"
    }
}

Here's my question: How should I debug these errors when I have no chance to see the data flowing in? From what I can see in "external" tools everything is looking great....

THANKS

Thomas

Publish error for Bicycles sample data

nsnandrey1 — Wed, 31 Dec 2025 11:20:39 GMT

The error message is :
"error":"invalid_grant","error_description":"AADSTS50076: Due to a configuration change made by your administrator, or because you moved to a new location, you must use multi-factor authentication to access '797f4846-ba00-4fd7-ba43-dac1f8f63013'. Trace ID: 5ff358cc-

Please help in resolving this error.

Getting data from Azure Event Grid backwards

foks — Tue, 02 Dec 2025 13:53:18 GMT

Hi,

I have Azure Event Grid Namespace. The retention period there is 7 days. However, I'm not able to get data from previous days with my evenstream. When I'm changing show results from: 7 days, I see nothing. So how is the retention period really working? Azure Event Grid has 7 days retention period. As far as I can see, the connection from Fabric to Azure Event Grid creates an "Push" subscription. But what if I would like to use "Queue" subscription? How I can connect there?