Skip to content

Release Notes

2025.1.1 (05-09-2025)

  • Fix - Third party dependency updates
  • Fix - Resolve overriding default ingress bind address with '::' (old default)
  • Fix - External certificate issues when external domain doesn't match server hostname

2025.1.0 (31-07-2025)

New defaults for Resource Requirements config group of 200 MHz for CPU and 300 MB for memory, which is a decrease from 1000 MHz (1 GHz) for CPU and 2048 MB (2 GB) for memory. The effect of this is that by default, Altair SLC Hub will run more workloads concurrently. To change the number of workloads that can be run concurrently, Resource Requirements Execution Profile Config Group can be configured.

If upgrading an installation of Altair SLC Hub to this version on Windows and using the internal database, you will need to install OpenSSL 1 on the Altair SLC Hub server. See instructions here.

If upgrading from any 2025.0 version of Altair SLC Hub then please follow these upgrade instructions.

  • New Feature - Store passwords for use when running workloads
  • New Feature - Replace Envoy Proxy with Traefik
  • Fix - Support for specifying a Vault Namespace in the case of using Vault Enterprise
  • Fix - Support for specifying a separate LDAP attribute defining the name of a users Unix username
  • Fix - Disable Vault integration with Authentication Domain Credentials by default.
  • Fix - Updgrade issue in RPMs
  • Fix - Renamed Facets to Config Groups
  • Fix - File ownership security improvements in Windows
  • Fix - Security improvements
  • Fix - Using credentials stored in Vault to authenticate object store

2025.0.2 (17-03-25)

  • Fix - Prevent slow license allocation on Workers (requires SLC 2023.10 or later, 2024.2 or later, 2025.0-ea.3 or later)
  • Fix - Increase idle timeout in proxy
  • Fix - Locking files on NFS drives
  • Fix - Prevent reading Authentication Domain passwords via the web portal
  • Fix - Update options for Library Definitions
  • Fix - Prevent users from erroneously running workloads in the SLC Hub with an additional membership to the root group

2025.0.1 (13-12-24)

Upgrading from pre 2025 versions of Hub, please follow these upgrade instructions.

This version introduces a new method of controlling the maximum number of concurrent workloads that can be run, including newly imposed defaults of 1000 MHz (1 GHz) for CPU and 2048 MB (2 GB) for memory which is an increase from 100 MHz for CPU and 300 MB for memory. The effect of this is that by default, Altair SLC Hub will run fewer workloads concurrently. To change the number of workloads that can be run concurrently, Resource Requirements Execution Profile Facets can be configured.

  • New Feature - Connection manager added to pipeline node edit dialog with improved bulk input/output management for links
  • New Feature - Connecting pipeline nodes has been improved with snapping and highlighting of valid node
  • New Feature - Ability to copy & paste nodes and links between different pipelines in web portal
  • New Feature - Expand hubcli to allow for user and group management
  • New Feature - Provide REST API for setting user password
  • New Feature - Make SLC execution log available through link session manager in web portal
  • New Feature - Include Content-Security-Policy and other similar security related headers
  • New Feature - Include (configurable) HTTP Strict-Transport-Security headers in web portal
  • New Feature - Ability to set configurable password validation policy
  • New Feature - Ability to track failed logon attempts and lock accounts if too many detected
  • New Feature - Ability to limit password duration and force passwords to be changed at next logon
  • New Feature - Resource Requirements Execution Profile Facets
  • New Feature - Password hash policy for passwords of internal SLC Hub users can be configured
  • New Feature - Support use of Argon2id hashing for passwords of internal SLC Hub users as per OWASP recommendations
  • New Feature - Ability to prevent modification of operating system user names associated with SLC Hub users
  • New Feature - Batch job command line interface (hubcli)
  • Fix - open_pam_session setting for the Nomad SLC Task driver is now the default
  • Fix - Clarify concept of "enabled" for SLC Hub services, make "autostart"
  • Fix - Server RPM can function as worker
  • Fix - Add OpenSSL 1 server RPM
  • Fix - LDAP sync now permits importing users with duplicate email addresses
  • Fix - LDAP sync now allows for deleted users in SLC Hub
  • Fix - LDAP sync now resurrects previously deleted users if LDAP query results change
  • Fix - Ensure that SLC reads source files in pipeline nodes in UTF8 (requires SLC 2024.1 or later)
  • Fix - Allow large file transfer from Altair Analytics Workbench file explorer
  • Fix - Improve database error handling in localised environments

2024.0.1 (15-04-24)

  • Fix - Internal object store running out of storage on Windows
  • Fix - Internal object store connection not secure on Linux

2024.0.0 (28-03-24)

This is a major release from the 2023 releases.

The upgrade involves permission changes on Linux, these will be updated by the upgrade and bootstrap.

The upgrade adds a new retention policy for pipline runs that defaults to 30 days. This should be amended if a different retention period is desired.

  • New Feature - Relocatable install on Linux
  • New Feature - Maintenance as non-root user on Linux
  • New Feature - Retention policy for pipeline runs
  • Fix - Move to OpenSSL 3.0
  • Fix - Improve behaviour with high umask
  • Fix - Don't bind to common ports like 8080/8181
  • Fix - Update third party dependencies

2023.4.3 (16-02-24)

  • Fix - Allow pipeline trigger 'run as' user to be changed to current user
  • Fix - Windows log file size limits
  • Fix - Pipeline runs layout improvements
  • Fix - Better error reporting in Portal
  • Fix - Portal UI improvements

2023.4.2 (12-02-24)

  • Fix - Ondemand CSV results not displaying in Portal
  • Fix - Ondemand choice parameter values not displaying

2023.4.1 (07-02-24)

  • Fix - RPM upgrade improvements
  • Fix - Object store configuration detection

2023.4.0 (23-11-23)

  • New Feature - Internal object store
  • New Feature - Internal database
  • Fix - Migrated Portal to Vue 3
  • Fix - Updated dependencies
  • Fix - Correctly constrain jobs based on SLC version specified
  • Fix - Hubctl output formatting improvements

2023.2-2.3.3.1 (24-5-23)

  • Fix - Timeout behaviour
  • Fix - Templating LDAP imports
  • Fix - Auth token not in local storage
  • Fix - Facets improvements

2023.2-2.3.1.45 (09-5-23)

  • New Feature - Execution Profile Facets

2023.1-2.3.0.462 (18-04-23)

Upgrading from 5EA-2.2.0.108 or earlier, please follow these upgrade instructions.

  • New Feature - Altair Licensing
  • New Feature - Hubctl Verify helpers
  • Documentation Improvements
  • Fix - Security Improvements

5EA-2.2.0.108 (15-12-22)

  • New Feature - Library Server

5EA-2.1.3.186 (22-11-22)

  • New Feature - Set IntermediatefilesDirectory per execution profile

5EA-2.0.22.488 (08-11-22)

  • MSI Fix - Missing files

5EA-2.0.22.473 (30-09-22)

Upgrading from 5EA-0.1.45 or earlier, please follow these upgrade instructions.

  • New Feature - Altair Naming
  • New Feature - Execution profiles
  • New Feature - Namespaces
  • New Feature - Library Server (experimental)
  • New Feature - Workload diagnostics
  • Fix - Access control Improvements
  • Fix - Improved how Operating System user used to run workload is selected (removal of "default user")
  • RPM Fix - Move to fewer RPMs, new name and locations
  • MSI Fix - New name and locations

5EA-0.1.45 (03-08-22)

  • MSI Fix - Issues installing example programs
  • Fix - Issue with LDAP when re-syncing

5EA-0.1.42 (06-07-22)

Upgrading from 5EA-0.1.33 or earlier, you must run dnf upgrade wpshub-* in order to update all RPMs correctly.

  • MSI Fix - issues starting link sessions
  • RPM Fix - upgrade RPMs incorrect
  • Fix - Ldap Sync improvements

5EA-0.1.33 (16-06-22)

Upgrading from 5EA-0.1.11 or earlier. If upgrading to this version, or any later version, from 5EA-0.1.11 or earlier it is necessary to run the following command:

hubctl scheduler --algorithm spread

This is regardless of whether the upgrade is done using a "dnf upgrade", or whether it is done by uninstalling and reinstalling the RPMs. Although a fix in this release changes the default scheduling algorithm to spread, this does not affect existing installations, and instead the above command must be run.

  • MSI Fix - Spaces in directory name
  • MSI Fix - Path fixes to allow demo deployment
  • Fix - Ensure hostname is lowercase for browser compatibility
  • Fix - Change the default workload scheduling algorithm from "binpack" to "spread"

5EA-0.1.11 (16-05-22)

  • Vault integration
  • Windows MSI installer
  • Allow single machine install of RPMs

5EA-0.0.24 (08-04-22)

  • Public availability of RPMs