This section lists the most important features from each release.
Qumulo Core 7.5.2
- Added the ability to upgrade NeuralCache on Azure Native Qumulo (ANQ) and Cloud Native Qumulo (CNQ) instances deployed on Qumulo Core versions lower than 7.4.1
- Added the ability to configure up to 32 hub portals and up to 32 spoke portals on the same cluster at the same time
- Made improvements to the S3 API:
- Extended the expiration of presigned S3 URLs
- Changed the
PutObjectS3 API action to accept all valid HTTP header keys - Resolved issues with the
CopyObjectS3 API action
- Added OpenMetrics for self-managed, on-premises clusters and CNQ instances
- Added support for latest Western Digital Ultrastar DC SN640 SSD firmware
Qumulo Core 7.5.1.2
Important
This release resolves an issue that previously caused an increase in logging output levels on CNQ and ANQ.
This release resolves an issue that previously caused an increase in logging output levels on CNQ and ANQ.
- For CNQ on Oracle Cloud Infrastructure (OCI), added support for Cold clusters with an object tier for infrequent access to files and directories
- Made improvements to the S3 API:
- Resolved an issue with S3 directory deletion
- Modified the XML response bodies for S3 API actions to return the canonical XML header
- Modified
filesREST APIs to return entity tag (ETag) values correctly - Added support for latest Western Digital Ultrastar DC SN640 SSD firmware
Qumulo Core 7.5.0.2 (Quarterly)
Important
- This release resolves a rare issue where previously, under certain conditions, concurrent quorum-related activity could result in file system data being overwritten.
- This release resolves an issue that previously occurred after a spoke portal was deleted.
-
This release resolves the following on-premises cluster issues:
- For clusters with certain erasure coding configurations deployed before July 4, 2020, this release resolves an issue that previously could occur during quorum formation, resulting in internal metadata being in an inconsistent state.
- This release makes Qumulo Core more resilient to unexpected drive controller behavior.
- Made improvements to Cloud Data Fabric:
- Enabled the creation of up to 32 spoke portals on a single Qumulo cluster
- Improved file creation latency and throughput for spoke portals
- Added the Priority Sync feature for read-write spoke portals
- Added the Qumulo Authoritative DNS (QDNS) to Qumulo Core
- Added support for latest Western Digital Ultrastar DC SN640 SSD firmware
- Resolved an issue with incorrect drive types for nodes in on-premises clusters
Qumulo Core 7.4.4
Important
Qumulo Core 7.4.4 is a platform upgrade for on-premises clusters which includes an update of the underlying Ubuntu operating system, from version 20.04 to 24.04.
- HPE hardware platforms users: We strongly recommend downloading and installing the latest HPE Service Pack (SPP) from the HPE Support Center. Lab testing of the upgrade process has shown that certain components (such as NICs and transceivers) can experience compatibility issues when running older firmware on Ubuntu 24.04.
- All hardware platform users: You must perform a rolling reboot (rebooting the nodes in your cluster one at a time) in order to avoid the interruption of connectivity between nodes in your cluster.
- Made improvements to Cloud Data Fabric:
- Added the ability to enable the S3 protocol for a spoke portal host cluster
- Removed previous restrictions from the placement of replication targets
- Enabled compression for data transferred between clusters in a portal relationship
- Made significant performance improvements for small-I/O workloads on the hub portal and for locally cached files on the spoke portal
- Added Web UI notifications about SSD endurance levels and a corresponding REST API endpoint
- Resolved an issue that previously affected read leases in the hub portal root directory
Qumulo Core 7.4.3
- Added the ability to use the NFSv4.1 protocol with spoke portals
- Added the ability to perform punch hole operations on spoke portals by using the
qqCLI and REST API - Added automatic functionality that cleans up portal relationships after synchronizing spoke portal changes
- Improved NeuralCache (formerly Predictive Read Cache) intelligent data ingestion and retention policy
- Added support for using EBS volumes for NeuralCache in AWS Local Zones
- Resolved an issue with NTLM authentication for certain image scanning devices and SMB shares
- Resolved an issue with the SMB server and showing previous versions for files and directories
Qumulo Core 7.4.2.1
Important
This release resolves an issue that can occur while following a non-standard upgrade path.
This release resolves an issue that can occur while following a non-standard upgrade path.
- Added support for Cloud Native Qumulo (CNQ) clusters with 1, 3, or more nodes
- Made improvements to Cloud Data Fabric:
- Moved the REST APIs out of preview
- Removed a
qqCLI command in favor of automatic clean-up of partially configured portal relationships - Added support for FTP communications for spoke portals
- Changed the names of portal states, which indicate the stages of the spoke portal creation process and the proposal or deletion of a portal relationship
- Added new portal statuses, which indicate the accessibility of a spoke portal or a hub portal
- Updated the default configuration for new NFS exports
- Added a REST API and
qqCLI command to facilitate deleting files from directories without constructing full paths
Qumulo Core 7.4.1.1
Important
- This release resolves an issue where previously Qumulo hub portal host clusters that communicated with read-write spoke portals crashed if the portal contained files with very large file IDs.
- This release resolves an issue that we identified in Qumulo Core releases 7.3.1.2 to 7.4.0.3, where making changes to networking configuration by using the API Tools page in the Web UI or by using the
qqCLI caused the default gateway to not populate. - Global Namespace is now a core component of Qumulo Cloud Data Fabric.
- To enable upgrades to this version of Qumulo Core, you must delete any existing spoke portals or hub portals from your cluster and then wait for any related tree-delete operations to complete before beginning the upgrade process.
- To enable the QFSD API to distinguish between cluster-local and spoke-cached data, from this release of Qumulo Core onward, 64-bit integers can no longer represent QFSD API file identifiers. API integrations must treat them as variable-length strings, in accordance with the OpenMetrics API schema. Although this change affects all REST API endpoints the Qumulo REST API continues to accept old-style identifiers for backwards compatibility.
- Made read-write spoke portals available by default
- Eliminated the REST APIs for manual removal of cached data from spoke portals (no changes to automatic cached data removal)
- Improved file identifiers to distinguish between cluster-local and spoke-cached data
- Implemented NeuralCache architecture for new cloud clusters
- Renamed the designation
preview(previously,beta) which marks REST APIs under development - Made public the REST API endpoints and
qqCLI commands that manage a cluster’s DNS configuration
Qumulo Core 7.4.0.3 (Quarterly)
The majority of functionality and stability improvements in this Qumulo Core release are internal.
- Added support for latest Western Digital Ultrastar DC SN640 SSD firmware
- Improved UID and GID sorting in the Web UI
- Resolved an issue with
PutObjectandCopyObjectS3 API actions
Important
- This release resolves a rare issue where previously, under certain conditions, concurrent quorum-related activity could result in file system data being overwritten.
- This release resolves an issue that previously occurred after a spoke portal was deleted.
-
This release resolves the following on-premises cluster issues:
- For clusters with certain erasure coding configurations deployed before July 4, 2020, this release resolves an issue that previously could occur during quorum formation, resulting in internal metadata being in an inconsistent state.
- This release makes Qumulo Core more resilient to unexpected drive controller behavior.
Qumulo Core 7.3.2.1
- Removed the limit from the number of floating IP addresses per cluster
- Added support for sharing S3 objects with presigned URLs to additional S3 API actions
Qumulo Core 7.3.1.2
Important
- For clusters that were created in 2016 (or earlier), this release resolves an issue identified in Qumulo Core 7.3.1.1 where, under certain conditions, the cluster was unable to start after being upgraded to 7.3.1.1, resulting in the cluster going offline.
- For clusters that have only tagged VLANs but no untagged networks, this release resolves an issue in Qumulo Core 7.3.1 where it was possible for the default gateways to be copied incorrectly during the upgrade process, resulting in a potential cluster outage.
- Made it easier to authorize hub portals from the hub portal host cluster by adding information to REST APIs and
qqCLI commands - Improved the behavior of NFSv3 and NFSv4.1
ACCESScommands - Increased the S3 connection limit
- Configured Qumulo Core to permit (but ignore) out-of-spec query parameters for S3 API actions
- Reverted floating IP assignment to Qumulo Core functionality prior to version 7.1.0
- Added a REST API and a
qqCLI command for querying the current fault tolerance of an object-backed cluster - Resolved an issue with NLM connections
- Resolved an issue with tree deletion operations
Qumulo Core 7.3.0.2 (Quarterly)
Important
This release resolves the following on-premises cluster issues:
This release resolves the following on-premises cluster issues:
- For clusters with certain erasure coding configurations deployed before July 4, 2020, this release resolves an issue that previously could occur during quorum formation, resulting in internal metadata being in an inconsistent state.
- This release makes Qumulo Core more resilient to unexpected drive controller behavior.
- Added the self-service permission model to Qumulo REST API access tokens and made changes to corresponding
qqCLI commands - Made it possible for a user to perform operations on their own access keys without having RBAC privileges
- Resolved issues concerning cached data on a spoke portal host cluster, directory metadata block updates made by S3 API actions, and the
GetObjectS3 API action with the Veeam client
Qumulo Core 7.2.3.2
Important
- This release resolves an issue with Cloud Data Fabric functionality in Qumulo Core 7.2.3.1 that could cause files to become unavailable temporarily.
- If your Qumulo cluster's nodes haven't been configured to use static routes, it isn't necessary to upgrade from Qumulo Core 7.2.3 to 7.2.3.1. To determine whether your cluster's nodes use static routes, SSH into the node and run the
ip route showcommand. Typically, static routes do not have tags such asproto dhcporproto kernel. For further help, contact the Qumulo Care team. - If your Qumulo cluster's nodes use static routes, this release resolves an issue in Qumulo Core 7.2.3 that, under certain conditions, caused the system to lose its static route configurations after a quorum event.
- Made CNQ on Azure available on Qumulo Nexus.
- Added support for read-only Cloud Data Fabric functionality between two or more Qumulo clusters.
- Added support for three-node clusters to CNQ on Azure and CNQ on AWS.
- Added support for CNQ clusters that adjust to multi-AZ configurations automatically.
- Improved S3 Object Lock functionality and made changes to corresponding S3 API actions.
- Made improvements in S3 Versioning API actions and the
CopyObjectS3 API action. - Allowed users without administrative role-based access control (RBAC) roles to modify S3 access keys and made changes to corresponding
qqCLI commands. - Updated some of the error types and HTTP status codes that appear during cluster membership operations.
- Added REST API resources and
qqCLI commands for creation and membership change of CNQ clusters. - Resolved an issue with an SSL configuration change while the S3 server is in plaintext mode.