Developer Community

We have an app deployed in abp-sandbox for “offshore regulatory assistance” (Flows + KG-RAG agent) on the Agent API (`az-ams-sp-002`, `cdf-version: beta`). While developing document export, I've encountered a reproducible failure pattern:Generation calls with complex instructions and/or long input intermittently return HTTP 200 with empty `content.text` — no error signal. The failure threshold shifted within hours on identical configuration: a 28.7K prompt succeeded at 18:15, 25K failed 12/12 an hour later, 15K succeeded at 21:49. Compression/summarization calls of similar size succeeded in the same window where generation calls failed.A report with the full test matrix is attached.Questions for Cognite (prepared with AI assistance):1. Could the endpoint return an explicit error/finish-reason instead of an empty 200?2. Are there documented practical limits for input size × instruction complexity?3. What is the roadmap status for streaming or async chat?4. Does a plain LLM endpoint with

Hi Team,We are working as a data platform team and there are multiple data product teams using our application. Each data product team has a different space to isolate their data. Is there a possibility to see the storage statistics restricted to space level? Clients also want to see “max number of concurrent requests”.

I’m getting the below error details when trying to access a new sandbox environment that has been setup.  What details do I need to share w/my team to unblock access?Your network administrator has blocked accessExternal access is blocked by policy. Contact your IT department for access. Troubleshooting detailsIf you contact your administrator, send this info to them.Copy info to clipboardRequest Id: f3b6e072-9289-427c-97e4-766a40af0700Correlation Id: e24e2dc7-9a3a-488c-92a3-fb333dc35281Timestamp: 2026-06-01T21:34:21ZMessage: AADSTS500021: Access to 'XXXXXXXX' tenant is denied. App name: Cognite Data FusionDevice platform: Windows 10Device state: UnregisteredIP address: 35.199.59.75Signed in as: XXXXXXXXXXFlag sign-in errors for review: Enable flaggingIf you plan on getting help for this problem, enable flagging and try to reproduce the error within 20 minutes. Flagged events make diagnostics available and are raised to admin attention.

Hi experts, In order to decide whether to use an aggregation strategy (or no aggregation at all), I need to know the number of data points between two timestamps without aggregation (server-side calculation). I’m looking for this code of call :client.time_series.data.count(external_id=<str>, start=<int | datetime>, end=<int | datetime>) -> intI'm trying to avoid having to perform a granularity calculation Thanks !Regards,Pierre

Hello, We have around ~10/15 CDF functions. We deploy these functions via cdfToolkit.We notice that deployment of new version of a function takes around ~10min in order for it to be fully available. In the worst case where we modify a common module used by all function, redeploying of all functions can reach ~30-40mins. During this time these functions are not available and results in a failure in the dependent high frequency workflows. We wonder if there is a way to accelerate deployment of new function versions. Thank you!  

Hi Team,Our observation while storing duplicate entries in Cognite Streams:Mutable Streams - It discards the duplicate entries consistently.Immutable Streams - It sometimes allow exact duplicate entries (externalId and other fields) and sometimes it doesn’t.Even when we checked in the Cognite AI, it says it usually doesn’t allow duplicate entries but in some rare scenarios, it allows.Could you please let us know the behavior. If it allows in rare scenarios, please let us know the exact scenarios.Thanks,Rahul

Help! I am currently facing an issue where the dataset data type is set as string = true, whereas it should be false (numeric). As a result, the chart is not displaying any data due to the field being treated as a string.However, when I download the dataset as a CSV file, the data is present. Could you advise if there is a quick fix on your side to convert the data type from string to numeric?  

I am currently facing an issue where the dataset data type is set as string = true, whereas it should be false (numeric). As a result, the chart is not displaying any data due to the field being treated as a string.However, when I download the dataset as a CSV file, the data is present. Could you advise if there is a quick fix on your side to convert the data type from string to numeric?  

Hi,We're using the Cognite DB extractor (version 3.9.2) with a MongoDB database (Azure Cosmos DB via the MongoDB API) and are trying to set up incremental loading.The documentation for mongodb (https://docs.cognite.com/cdf/integration/guides/extraction/configuration/db#databases.mongodb) doesn't mention the use of start_at and incremental_field but after testing them and looking at the debug logs, we can see the extractor sends the {start_at} placeholder literally without substituting the state value. My questions are:    1. Is start_at substitution supported at all for MongoDB JSON queries, or is it SQL-only?    2. If incremental loading is not supported for MongoDB, is there a recommended workaround?Thanks.

When we’ve tried to update to python sdk v8 we get a 400 validation issue when we try to create a container.We’re currently using v7.92.0, where we have no problems. We’ve tested versions v8.0.7 and v8.2.0.client.config.api_subversion is ‘20230101’ for both 7.92.0 and 8.2.0.Here is an example of a request that fails with: Unexpected field - items[0].properties..constraintState.constraintState seems to be a new property for v8 of the api.ContainerApply( space="Space", external_id="Comp", name="Comp", used_for="node", properties={ "manufacturer_name": ContainerProperty( type=Text(), nullable=True, ) })I notice also that the CDF Jupyter notebook in browser is still on v7.92.0. Is there potentially any other issues preventing an upgrade to v8? Best RegardsDaniel Rasmussen

With Tableau 2026.x introduced the native REST API connector, I wanted to check if anyone has explored using it to connect Cognite Data Fusion (CDF) APIs.Has anyone successfully connected CDF data via the Tableau REST API connector, or evaluated this approach?Any insights, limitations, or best practices would be greatly appreciated. Thanks in advance!

Hello,  Is it planned to support RAW querying via CDF Grafana Datasource ?  Thanks ! 

Hello, In our project, we need to use TIMESTAMP_LTZ data type for our timestamps. This data type is only available in Spark starting from 3.4 version. Is it possible to upgrade your version (which is 3.3 I believe) to 3.4 at least please? Thank you

I am new to extending CogniteCore data model to further extend. How to Import CogniteCore datamodel to further extend in a new datamodel space. 

cdf version is 0.7.236.i cannot find the same quickstart mentioned in the training material. below is the screenshot from my environment.   the instruction mentions two times about “would you like to make changes to the selection”. the action is inconsistent.  The second question should be updated to be the real one.  

We have requirement to update capabilities in existing groups in CDF using Python SDK/API, we can add capabilities while creating group using python SDK, but if we want to updated that created group no provision for that. we tried client.iam.groups.create(group) but it creates new group with same name, in this case how we can update capabilities in group?

I have used the db extractor before, but with version 4.0.0 it now comes with a connect.yml and a config.yml. The documentation still lists one config file to be used. I’m not sure how to use the connection file, which looks like it used to be the cognite portion of the of the config file.I tried combining the two into one file, but when running “ dbextract.exe -c config.yml “ I am getting errors that its not expecting inputs such as version, log-handlers, database, and queries.How am I supposed to use these 2 different files when I run as an executable?Thank you

Hello experts,I'm interested in the following function: the sum of all data points in a time series server-side.I know there are aggregation operations for a specific time interval, but I'm looking for either a cumulative or total sum.I'd like to be able to perform this operation on either a standard time series or a synthetic time series, but unfortunately, I haven't found what I'm looking for yet.Please see below for the operations I am performing; I would like a discrete summation between the multiplication and the output.Does anyone have a server-side solution? Regards,Pierre Rambourg

We have a Cognite Function in which we are retrieving rows from staging tables using below sdk commandclient.raw.rows.retrieve_dataframe(db_name, tbl_name, limit=-1)The problem we are facing is the above command returns an empty Dataframe (0 columns, 0 rows) even though the table exists with data. The code execution does not fail but returns an empty Dataframe.When I manually/locally run the above command on the same database and tables, I get the required data.However, when running inside a CDF function it returns a Dataframe with shape (0,0) 

Hello experts,I have a question about Cognite Functions. I know there’s a limit on the size of the input data, as discussed here : Cognite Hub and documented there : Cloud provider limitations - Cognite Docs But is there a limit on the size of the function’s output? I have one that works perfectly fine but crashes without any explicit error message when handling a large output payload (>500 KB). I’m on Azure provider.Is there any information on this? Thank you,Best regards,Pierre Rambourg edit : the initial error was on my side, but i’m still interested for output limits !

I'm experiencing an issue with the Workflows editor where the bottom toolbar buttons (zoom in, zoom out, pan to center, auto layout) stop responding in certain conditions.Steps to reproduce:1. Open Workflows in CDF2. Create or edit a workflow3. Open the right panel (e.g., click on a task/agent to configure it)4. Try clicking the zoom in/out/pan buttons in the bottom toolbar5. Observe: Buttons don't respond when right panel is close to the left toolbar6. Maximize browser window or close right panel → buttons work againExpected: Toolbar buttons should work regardless of panel positionsActual: Buttons appear to be blocked, clicks don't register when panels are close together Workaround: Maximize browser window to create space between panels before using toolbar buttonsHas anyone else experienced this? Is there a better workaround or is this a known issue?  

I have 3 Views, ViewA, ViewB & ViewC. View A has direct relations to ViewB & ViewC via properties DefectEventID & AssetId, respectively. Pagination is expected and with Instance.Query end point, looping via cursors, it is not giving results for ViewB  & ViewC, apart from the first 100 records from the first call of instances.query(query). What is the best way to populate all_ViewA_Nodes,all_ViewB_Nodes ,all_ViewC_Nodes  Pagination via multiple cursors from Instance.Query?query = dm.query.Query(            with_={                "ViewA_Nodes": dm.query.NodeResultSetExpression(                    filter=filters.HasData(views=[ViewA]),                    limit=100,                ),                "ViewB_Nodes": dm.query.NodeResultSetExpression(                    from_="ViewA_Nodes",                    through=dm.PropertyId(                        source=ViewA,                        property="DefectEventID"                    ),                    direction="outwards", 

In addition to using config.[env].yaml, is it possible to create a custom file for configuring variables and then reference those placeholders in data models, transformations, workflows, etc ?For instance, I have several common SQL blocks that I would like to share across different transformations. While I could define them as variables in config.[env].yaml, I prefer to keep SQL-related variables separate from others in order to maintain clarity.

Hi Cognite Team,We are currently defining an integration pattern for a partner who will provide data from an OPC UA source into Cognite Data Fusion.Our preference is to use a managed MQTT broker in CDF and avoid deploying any self-hosted MQTT broker or additional bridge components on-prem.Based on the documentation, we understand that the OPC UA Extractor supports an mqtt: output mode, but it also references a requirement for the MQTT-CDF Bridge and a configured broker host/port.To ensure we define the correct and supported architecture, could you please clarify: Is it supported to configure the OPC UA Extractor to publish directly to the Cognite-hosted MQTT broker in CDF? If using the Cognite-managed MQTT broker, is the MQTT-CDF Bridge still required? Is a self-hosted MQTT broker mandatory when using the mqtt: configuration in the OPC UA Extractor? What is the recommended architecture today for a partner delivering OPC UA data into CDF using MQTT? Our intended target architectur

Hello Experts,I’m trying to use advanced filter on timeseries through SDK. I can do what I want except in the following situation: the asset_id property is null. Can anyone help me?I tried :from cognite.client.data_classes.time_series import TimeSeriesPropertyf = filters.And( filters.Equals("data_set_id", 123456), filters.Equals("name", "foobar"), filters.Equals(TimeSeriesProperty.asset_id, ""))[advancedFilter.[2].equals.value A string is not a valid value for the property "assetId"]from cognite.client.data_classes.time_series import TimeSeriesPropertyf = filters.And( filters.Equals("data_set_id", 123456), filters.Not(filters.Exists(TimeSeriesProperty.asset_id)),)-> 0I’m sure there is timeseries where timeseries.asset_id is None.Thanks experts !Regards,Pierre Rambourg