Hi team,   I would like to sugget, if possible, adding the SAS Job Execution Web application as an option under the SAS Viya menu, similar to SAS Studio which is under Analytics Life Cycle menu. ... View more

Version SAS Viya 2025.03 LTS   Issue When there are a multitude of nodes present within the current schematic-style layout of the SAS Lineage tool - to the point that they don't even fit in the "Overview Window" - this presents a rather non-User friendly display of content to the Platform Admin.  Plenty of scrolling and clicking is involved in order to retrieve data on each node object -     Feature Request For usability, please can a "List View" be developed to provide easy + rapid consumption of Lineage content for the Platform Admin?   Example -     Additionally, please can a "Download" button be added? We would also be very happy if there was an option to update a cas table, such that this content could be viewed through a VA report. These 2 items (List View + Download button) could be added to the "View Settings" pane with Lineage -   ... View more

I am submitting this feature request on behalf of an active project we have with the Surrey Police Services.  As per tech support case CS0331348, our client would like to have the grouping behaviour in the advanced search UI of VI behave consistently when using AND or OR as a group operator.  Currently, OR maintains the group while AND removes the group and sets the search conditions linearly.  Our client would like to have the AND operator work in the same way as the OR operator.  Examples are in the tech support case. ... View more

in Environment Manager → Content, it is not visually clear that an item is a shortcut. The icon displayed is a standard folder icon, whereas in SAS Drive, shortcuts are marked with an arrow icon in the list view. We have encountered cases where users accidentally create shortcuts pointing to the original folder, which can result in users getting stuck in an endless loop. We can only identify such shortcuts by checking them in SAS Drive. Since SAS Drive will be deprecated soon, we suggest that this issue is addressed to prevent any data integrity problems in the future. regards   Karolina   ... View more

I would like to request a feature where there is a sound notification when the code finishes running. Currently, if there is code that takes a long time to run, it would be nice when I switch to another task to be notified by a sound that the code finished running. Possibly there can be different sounds for when there is an error, warning, or successful run. ... View more

Hello everyone,   at the moment it is not possible to use libname options (for example BULKUNLOAD ) for the Postgres engine in SAS 9.4. This is a major issue for the effective use of Postgres in a SAS 9.4environment. Enhancing the interface would be very helpful in enabling Postgres to be used in a way similar to DB2. Otherwise, we are missing the integration of a widely used Database Management System.   Thanks in advance.     ... View more

When using custom steps in flows, the name can be customized, but not the image. If a flow consists primarily of custom steps, images would be more helpful for clarity:   ... View more

Hi Team,   We understand from SAS release roadmap that SAS Drive will be decommissioned after December 2025.   We request the incorporation of a new feature or tab within the Explore and Visualize page to replicate the functionality of the former Reports tab in SAS Drive. This tab was essential to our end users' workflows, as it provided a centralized, flat list of all reports accessible to the user, eliminating the need to manually browse through multiple folder levels. The current Explore & Visualize interface does not offer a comparable experience, making it less efficient for users who frequently access a wide variety of reports. We kindly request that a dedicated section or tab be added to Explore and Visualize, offering: A flat, searchable list of all accessible reports Functionality to pin or favorite commonly used reports A user-friendly replacement for the navigation experience provided by the SAS Drive Reports tab   Thank you for your support. ... View more

I'd like to propose a necessary improvement to the user experience when fitting Generalized Estimating Equations (GEE) models in SAS procedures like PROC GENMOD or PROC GEE .   The Current Problem   When fitting a GEE model, Likelihood Ratio Confidence Intervals (LRCI) are fundamentally unavailable because the GEE approach only specifies a mean-variance relationship and not a full likelihood function.   Currently, if a user specifies the LRCI option in the MODEL statement for a GEE, SAS simply ignores the option without any notification. No warning or note is printed to the SAS log. No message is added to the results output.   This silent behavior can mislead users into assuming the intervals were calculated or even that their model setup might be wrong, wasting time and potentially leading to misinterpretation.   The Proposed Improvement   SAS should implement explicit and multiple notifications when the LRCI option is requested for a GEE model:   1. Immediate Log Warning (Highest Priority)   A clear WARNING should be issued to the SAS log explaining the technical reason the option is ignored. Suggested Log Message: WARNING: The LRCI (Likelihood Ratio Confidence Interval) option has been requested but is not supported for models fit using Generalized Estimating Equations (GEE). This is because GEE does not specify a full likelihood function. The option will be ignored, and default Wald confidence intervals will be reported.   2. Note in the Results Output   A brief note should be added to the output table (e.g., near the Parameter Estimates table) to confirm what was actually calculated. Suggested Note in Results: Note: Likelihood Ratio Confidence Intervals (LRCI) were requested but could not be computed for this GEE model. Default Wald confidence intervals are reported.   Why This Matters   This is a critical usability improvement. GEE is a common and powerful method, and providing clear, instantaneous feedback prevents confusion, saves user time, and ensures that new or experienced users fully understand the inferential methods being used in their analysis. ... View more

Hi all,   We are using SAS Viya and following the stable release process and currently looking into deploy SAS Viya 2025.10 in December. The whole setup is on Microsoft Azure and we are using Azure Database for PostgreSQL – Flexible Server as our main database for SAS Viya.   We would really like to request that the DataServer Custom Resource be enhanced to support Microsoft Entra ID (Azure AD) token-based authentication as a built-in alternative to using Kubernetes Secrets with static passwords.   Summary: We need the DataServer to be able to authenticate using Microsoft Entra access tokens obtained securely at runtime, ideally using Azure Workload Identity (federated OIDC), without relying on any long-lived credentials. The access token should be refreshed automatically before it expires and exposed to dependent workloads in the same way as the current mechanism (for example via environment variables or temporary files).   Problem: Today, the DataServer requires credentials to be provided via Kubernetes Secrets, which means a long-lived username and password must be stored in the cluster. This is no longer acceptable in environments that follow zero-trust or passwordless authentication standards. Azure Database for PostgreSQL Flexible Server supports Azure AD authentication using short-lived access tokens instead of static passwords, and our organization is required to use this method.   In practical terms, we are asking for a native input mode, such as “azureAdToken” or “workloadIdentity”, instead of only supporting static Secret-based credentials. This would eliminate the need for manual credential rotation, align with Azure’s passwordless identity approach, and meet our mandatory security and compliance requirements. This feature is important as the current static credential approach will soon become a blocker for production deployment in regulated enterprise environments.   I hope this makes sense and I do know that is a big ask, but this would be a really important feature for us and many other organizations that are moving towards zero-trust and passwordless authentication models.   Regards and thanks ... View more

SAS VIYA Workbench should be available on-prem and it should ideally be an integral part of the SAS Viya platform. ... View more

It would be nice if the logs from SAS Viya jobs could be viewed while the jobs are running. And not just after the job has ended. ... View more

Background: We've implemented, among others, two entities, accounts and transaction. Those are related to each other with a relationship, which is represented by a bridge table, and a respective traversal. On the account details page we've placed a related objects control, leveraging that traversal to display the transaction being related to an account. There are a few accounts with more than 2000 transactions related to it. Visual Investigator cannot display those transactions in the related objects grid: Changing the "Node expansion size limit (hard limit)", which is 2000 per default, to 5000 doesn't help, because there will not be more than 2000 transactions fetched and displayed, probably due to some hardcoded restriction. We investigated the API-Calls in the background and as the main player we identified the path-search-api (POST svi-sand/paths?limit=2000&expansionLimit=2000 with a filter in the Payload) According to the documentation (pathsSearch | SAS Developer Portal) a start and limit parameter is available suggesting that paging might be possible, but it is not used by the Related Objects control.   Our Change request: Our suggestion is to offer an option to activate paging for the Related Objects control to support displaying accounts with more than 2000 transactions related to it. Sorting, Filtering & Exporting might only adjustable in the control setup & not avalable to endusers;  ... View more

We would like to request an enhancement for SAS Visual Investigator to allow connections to a broader range of data sources beyond the currently listed relational databases (DB2, MySql, Oracle, PostgreSQL, SQL Server, Teradata). In some projects, we work with analytical or columnar databases such as ClickHouse, or with external files (e.g., Excel or CSV) that serve as key sources for entity modeling. It would be highly beneficial if Visual Investigator could either:   Support additional JDBC Connections for other databases or, Allow the creation of entities directly from external files. This capability would provide much greater flexibility for integrating diverse data sources into Visual Investigator and would expand its usability in modern data environments. ... View more

Currently when you write parquet files via the sas libname engine the parquet types are fully determined by the informat only.  That is to say if you have a numeric variable with a datetime20. format and no informat it will be written to the parquet file as a double.     The feature request is that if no informats are specified SAS should use sensible default types based on the format.  It is very intuitive that a datetime20. format column should be written as a timestamp thus users should not have to be burdened for defining informat types for what should be obvious conversions.  ... View more

Hello team, When attempting to export our SAS Visual Analytics report to PDF, we see: “This report has no printable content.” Our affected pages rely primarily on Data-Driven Content (DDC) objects. Please enable PDF export for the report and ensure sections containing Data-Driven Content can be exported. If native PDF rendering of DDC is not supported in our version/configuration, could we implement one of the following so stakeholders still get governed PDFs? Provide a printable fallback (e.g., mirrored KPI/table/chart in a print-friendly section). Allow the DDC app to expose a static image endpoint so VA can place a printable snapshot. Use a controlled image placeholder (updated on refresh) that represents the DDC state at export time. ... View more

Currently when SAS reads and writes parquet data files via the libname parquet engine labels are dropped and thus cannot be restored when the file is read back in.  Parquet though supports arbitrary metadata storage in the file header.     The suggestion here would be for SAS to store column labels in the parquet metadata and then retrieve and restore them when reading the file back in.    For reference the R arrow::write_parquet() function already does this by storing R specific attributes in the "R" field of the parquet files metadata.  ... View more

There should be an option to cancel SAS Viya jobs in the UI of the Environment Manager.  ... View more

When dates are viewed in the VI GUI they are automatically displayed in the user's local time zone BUT when you export to Excel they are in UTC? This is confusing our users as quite rightly they expect to see the same data in both formats, this should be a bug fix and not a feature request IMHO... ... View more