Product User Community

I have Cognite Hub Account configured in Authenticator App on my phone since some time. Recently I have installed the Authenticator also on an iPad. The CogniteHub account appears on the list , however it wants me to log in. And I can’t remember the password. Is there  a way to get the password from the Authenticator on my phone?

Hello, We have a very big CDF workflow with many parrallel steps. When a step fails, all the steps fails with “404: Session was not found.” (because the failing step deleted the session) all steps are market as failed. It makes debugging and finding the real failing step very tedious ! Any improvement planned for this? Any workaround? Thanks 

Lineage doesn’t work. You click on it and it shows for 1 second, before you get this screen. Multiple people gets the same error. 

In Charts, when choosing a core data model as your source, Events are disabled preventing the choosing of core model “CogniteActivity” as “events” to overlay timeseries data from CogniteTimeSeries.Is this functionality on the roadmap and if so, when can we expect?  

We are thrilled to announce the next evolution of your user experience. We’ve unified our ecosystem to help you master your industrial data without the friction.What’s new? One Identity: We’ve merged Cognite Academy and the Community. One login for everything. Unified Search: Find How-to guides, discussions, and courses in a single search. Guided Careers: New role-based learning paths, starting with the Data Engineer path! 👇 Read the full announcement here:  

Hi Team,I'm using the Cognite Toolkit to test the CDF download raw functionality. I’ve observed that the download process fails when the staging database name contains a special character such as :.For example, if the database name is state:odbc, the cdf download raw command fails because it attempts to create a temporary folder with the same name as the staging database. Since : is not a valid character for directory names, the folder creation fails.Additionally, when I try to download the raw data from another database (without special characters), three files are created in the temporary directory:.Table.yamlManifest.yamlRawRows.ndjson.gzCould you please confirm if the RawRows.ndjson.gz file is expected, or should the output be in .csv format?

I am going through the course Match entity python sdk ,  there one notebook in section “How to succeeded with this task ” .But the link is not accessible . Please can some one guide in it. https://cc.sj-cdn.net/instructor/hhi6o3bc6ve-cognite-academy/assets/1688105749/Entity_matching_notebook_copy.ipynb?response-content-disposition=attachment&Expires=1763969235&Signature=dX1a9jxYTvcqFQ7jWIhNing8AxVk9r~60Tm4ZFWuvUK232AuM7mIg4StnYkk~MwaarbraKyfxE0O4BFm8VQbKFsEzgjeIDufqRvP5XEDH4X9aTC0XY5YR1l8iJeNmU2ebqWArryM8296eFNOVtPsNdIPOGWIaDz2xdqQ2rkncm4Zo-awiI7u~VFf~83tTN10ZiLuoTBqa2hurinAD5W3Igo4HlRQG109N9PVBkL7fO2ZeyjHhbRqBtxXrU8um0u6l4fcHJRb1CL9sWHEjouuK55CTf4iIK~057XQM9yx6ylK8FWD9EuT3ayDmgdwXx~FPhKPNpdBaxWIcKj3VrEERw__&Key-Pair-Id=APKAI3B7HFD2VYJQK4MQ

 Issue SummaryCreating `CogniteFile` instances in Data Modeling Service (DMS) automatically creates unwanted "ghost files" - duplicate file resources that are incomplete and have no external_id, no dataset_id, and no content. This causes the Diagram Parser UI to display incorrect file IDs and show "Files not uploaded" errors.  Environment- Project: tridcognite-sanbox- Cluster: aw-was-gp-001.cognitedata.com- Dataset ID: 1242342447467142- Space: diagram_parser_space_1- Data Model: diagram_parser_data_model_1 (version 1)- Python SDK Version: 7.78.0---Methods Used:Method 1: SQL TransformationINSERT INTO cdf_cdm.CogniteFile(v1)SELECT   CAST(externalId AS STRING) as externalId,  CAST(name AS STRING) as name,  CAST('application/pdf' AS STRING) as mimeType,  CAST(TRUE AS BOOLEAN) as isUploadedFROM `Diag_Parser_Cement`.`parser_files_staging`WHERE externalId = 'cement-parser-file-crusher-pid' Transformation Configuration:- Destination: Instances- Space: diagram_parser_space_1- View: cdf_cdm.Cogn

Hello,When testing the new Diagram Parser tool on Vector PDFs, I found that it successfully finds some Assets ingested in the DMS and Symbols from a selected library. However, the Assets only show up as annotations even when verifying them. How can I get the suggested Assets to link to the Diagram in the DMS (i.e. open the diagram and see Assets directly linked to the diagram)? Thanks,Gabriel

During the calculation of adding two time series, if one of them has no value, will the calculated timesseries give me some interpolated value?In the example below, the ‘Error Intern’ series has a value of 2252, but in ‘Total Down Time’ I’m getting 2756.64, which is incorrect—it should also be 2252.Why is it still giving me a linear interpolation value even though I selected the interpolation type as ‘step’?  

Why are we making this change?The CDF CLI was originally designed to streamline application development on top of CDF, specifically focusing on Data Modeling workflows and the generation of type-safe SDKs.However, to ensure we are delivering the highest value to our developer ecosystem, we are refining our product strategy and priorities. The CDF CLI (beta) has not been under active development for some time, and maintaining it to our quality standards no longer aligns with our current roadmap. By discontinuing this tool, we can refocus our resources on the core tools and SDKs that best support your integration with Cognite Data Fusion.If your current CI/CD pipelines or local development environments rely on installing this package, you must take action before February 18th, 2026 to avoid build failures.Timeline of Events February 18th, 2026: The @cognite/cdf-cli package will be unpublished from NPM. You will no longer be able to install it via npm install or yarn add. What do you need

I’m searching for a specific document using Search. The doument ID is hidden in a metadata tag.If I search for this document ID in the search bar, the file is not discovered (obviously it doesn’t search in metadata)Plan B is to use the filter (which is inconvenient since it forces me to know that particular metadata key)The metadata key is document:abp_document_number, but it doesn’t show up when I search in the filters  If I use the Data explorer under Data management, it discovers the metadata key, allowing me to filter on it  But there is an even better way of doing this in Data explorer. There I can edit the search settings and instruct it to look in the metadata. Then it discovers it immediately. This allows me to find the document without having to know to filter on this very cryptic key. For classic data model, where a lot of info is dumped in metadata, with poor governance on the keys, the ability to search in metadata is VERY useful   

Can we visualize Visio files with.vsd extension in Industrial canvas?

Is there a way to annotate 3D models towards instances from a data model. In the API documentation I only see linking to assets as an option.  

We are using the cognite core data model and we extracted historian data which has been created as a bunch of time series. Now we have one asset that we would like to link these time series to. How do we do this?Please help, thanks.

Hi everyone! When importing data for a specific TAG, I notice multiple connecting lines or links between TAGs across the same P&ID. What steps can I take to minimize or reduce these connections?”

Is there a plan to support enumeration (or any way to restrict value of string properties) in the future?Enumerations are described here: https://graphql.org/learn/schema/, but seems like its currently not supported: 

Hello, In order to successfully engage with a potential industrial customer, we would like to showcase the New Diagram Parsing feature during a pre-sales meeting in December. However, we can’t utilize it without the Location Filter feature. Could you please help us to get access to the Manage Location Filter feature in the Admin workspace?  Project Name= tridcognite-sanbox Cluster=aw-was-gp-001 Thanks,Dinesh 

We are now planning to integrate a large amount of historical data. The context about historical data and our data organization is in this topic:We create an interface WellState, which contains 2 properties: WellId and Archived, in order to propagate the Archived flag to every object in our data models.After integrated large historical data in our development env, we found that DataFetchingException always occurred in Daily(550K lines) and Operation(2M lines) in using GraphQL. And if using cdf_nodes to load Daily and Operation, the transformation will be blocked util that your session is closed.Info: Daily has many relationships and Operation has only a relationship with Daily. But We have also other 3 objects like Operation(big dataset and a relationship with Daily), they can work well. GraphQL:query MyQuery { listOperation { edges { node { externalId space } } } listDaily { edges { node { externalId space } } }}the result is alwa

I am facing issue of jupyter notebook getting crashed frequently. It is observed for larger data and surprisingly ML model training on smallest data chunks as well.Sharing the image for reference.  

Hello,We are querying the view WfAlarms (structure as given below) which has direct reference to alarmCode. When we are querying it for say 3+ months, we get the CogniteApiError, but works fine when it is 1month. This error is seen in our application deployed in production. We expect almost 2 million of records in this view. Can you help to understand what can be done to fix this issue ? Any configuration that needs to be changed. Let me know if you need any other details.1. Properties being queried from view - WfAlarms ErrorView:type WfAlarms @view(version: "2") @container(indexes: [{identifier : "entityName", fields : ["entityName"], cursorable : true}, {identifier : "entityExternalId", fields : ["entityExternalId"], cursorable : true}, {identifier : "alarmFlags", fields : ["alarmFlags"], cursorable : true}, {identifier : "entityType", fields : ["entityType"], cursorable : true}, {identifier : "alarmCode", fields : ["alarmCode"], cursorable : true}, {identifier : "startDateTime", fie

As of January 1st, 2026, the Rule-Based Coloring feature in Industrial Canvas will transition from public beta to private beta. The change will not affect existing users who have previously enabled and are using the feature on a daily basis. The feature will be unavailable for new and existing users who are not in the private beta program.No action is needed. If you have any questions or concerns, please reach out.Thank you,Andreea. 

Can’t deselect tag when only 1 tag is selected. Pressing “Clear” will reset all filters and take me out of the menu.  Markus PettersenAker BP - Data Platform Architect

The “Ask a Question” feature inside the Canvas is very helpful—it allows users to quickly access specific documents and find information easily.However, I noticed that this feature is not available in the Industrial Tool – Search option or under the Data Management – Data Explorer tab. It would be very useful to have this functionality directly within the Data Explorer, as it would make the process more efficient and user-friendly without needing to switch to Canvas.What are your thoughts on this?Is it possible to consider adding this feature to Data Explorer in the future?

Hi team,I am looking for a way to trace data lineage in CDF. For example, if we have a table in RAW that feeds another table through a Transformation, is there a straightforward way to see where that data is being used downstream?Basically, I want to understand how to track which tables, transformations, or data products depend on a particular RAW table. Any native feature or recommended approach for this in CDF?Thanks!