July 2026
Continuous Data 2026.4.0.0 introduces V2P (Virtual-to-Physical) support for Virtualization SDK plugins, delivers extensive SQL Server and Oracle reliability improvements, enhances replication and offline storage capabilities through API updates, and includes security hardening for NTP and Tomcat.
New features
Virtual-to-Physical (V2P) support for Virtualization SDK (vSDK) plugins
Delphix Continuous Data Engine now supports the V2P workflow for datasets managed by vSDK plugins (supported only for Unix-based plugins), extending V2P beyond the built-in Oracle, SQL Server, and SAP ASE data sources.
For more information, read the Delphix Developer documentation.
Continuous Data user interfaces end of life notice
As the product experience transitions to Data Control Tower, Continuous Data management and setup interfaces are being sunset over the next three years.
For more information, read Continuous Data User Interfaces End of Life on this page.
NetBackup v11.x certification
Continuous Data is now certified for NetBackup v11.x on all compatible SQL Server/Windows versions.
For more information, read SQL Server matrix on Ecosystem Hub.
Fixed issues
Release 2026.4.0.0 (July 2026)
| Issue number | Description |
|---|---|
| DLPX-72502 | A stale host remained after a SQL Server cluster node's IP address changed while its FQDN stayed the same. The outdated host is now automatically removed when the node is updated; if the host cannot be removed because it is still in use, a warning is raised. |
| DLPX-78504 | Improved the error message shown when SQL Server VDB provisioning fails due to an invalid parent environment. |
| DLPX-78733 | Improved SQL Server VDB policy enforcement fault messages to show the actual underlying issue (for example connectivity or disk-space errors) instead of a generic unreachable-host message. |
| DLPX-80764 | When a Windows mount location is already in use by another volume, the engine now records that volume's label alongside the expected label in debug logs, making it easier to identify what is occupying the mount path during troubleshooting. |
| DLPX-81300 | Windows target environment discovery now reports a clear message instead of a generic internal error indicating the IQN is invalid when the host's iSCSI initiator name is malformed. |
| DLPX-86243 | SQL Server AG dSource snapshots now fail with a clear, actionable error if the database is removed from the AG. Restoring a database via an AG cluster environment requires it to be attached to the AG. |
| DLPX-86879 | Corrected SmartFailover behavior in MSSQL WSFC environments to use the Consolidate strategy instead of Rename Consolidations when appropriate. |
| DLPX-86898 | SQL Server FCI discovery silently fails when a cluster contains a duplicate "SQL Server" resource type. Discovery now auto-selects the correct cluster group and logs an engine-side warning about the misconfiguration. |
| DLPX-86999 | Hardened the NTP service configuration to prevent unauthenticated queries from disclosing system information (Qualys QID 38293). |
| DLPX-89903 | SQL Server VDB delete/shutdown/startup failures caused by an orphaned SPID holding a lock (error 5061) now show a suggested action recommending a SQL Server instance restart to release the lock. |
| DLPX-90574 | Cancelling a SQL Server VDB refresh or provision during the recovery phase left the database mounted and accessible on the target instance; the storage is now properly unmounted when the operation is cancelled after recovery begins. |
| DLPX-93933 | On GCP and OCI Elastic Data engines, debug log files may accumulate without being automatically cleaned up, significantly increasing support bundle sizes over time. |
| DLPX-94146 | Replication fails with a confusing internal error when the engine is left in a failed upgrade state; the engine now returns a clear message directing the user to resolve the upgrade before retrying. |
| DLPX-95255 | Multi-hop engine upgrades through a 23.x intermediate release to 2025.4+ incorrectly attempt to enable NTP even when NTP was disabled. |
| DLPX-96133 | SAP ASE virtual database point-in-time refresh and Virtual-to-Physical (V2P) export could fail when the source database had been shrunk. |
| DLPX-96708 | In rare cases, an incremental replication send can fail with an internal error when the storage bookmark marking the previous successful replication point is missing, halting further replication progress. |
| DLPX-96737 | Replication connection failures caused by SSL certificate or PKIX path validation errors may display a generic error message without sufficient detail to identify the root cause. |
| DLPX-96762 | If a VDB's cache priority is set to High after upgrading to 2026.2.0.0 without a reboot, this feature will not be enabled. An error is shown but the VDB's cache priority is not reset to Normal. |
| DLPX-96772 | Replication jobs could hang before sending data when temporary replication database services were not cleaned up after a shutdown retry failure. |
| DLPX-96952 | After an upgrade, storage disks on a Hyper-V Delphix engine may appear as unconfigured in the management interface even though data and ZFS pools remain fully operational. |
| DLPX-97025 | After an upgrade, the management service may fail to start due to the Nginx service failing to initialize, requiring a second reboot to recover. |
| DLPX-97032 | When provisioning or refreshing an Oracle VDB with the "Generate new DBID" option enabled, the redo log file size was incorrectly set to 4MB regardless of the configured redo log size. This has been resolved. |
| DLPX-97035 | Replication jobs may fail without surfacing a clear error message when the target engine runs out of disk space during replication. |
| DLPX-97040 | Oracle VDB validation for log_archive_dest_N could incorrectly accept invalid configurations, causing SnapSync failures when the archive path ended in a directory named "mandatory" or contained special characters. |
| DLPX-97041 | Replication jobs could hang or fail due to internal database processes not being properly cleaned up when the system service manager was unresponsive. |
| DLPX-97044 | Oracle archive logs required for VDB recovery are now correctly retained by log retention policies, preventing VDB snapshots from becoming unrecoverable. |
| DLPX-97117 | Environment refresh fails after NFS filesystem relocation. |
| DLPX-97171 | Oracle VDB provisioning and refresh now correctly handle cases where the source VDB uses the default log_archive_format and the target VDB is configured with a non-default log_archive_format value, preventing ORA-00308 errors. |
| DLPX-97215 | Improved handling of AppData plugin container state during engine upgrades and management stack restarts. |
| DLPX-97244 | Updated Tomcat as part of routine dependency maintenance, which includes a fix for CVE-2026-41293, CVE-2026-41284, CVE-2026-42498, CVE-2026-43513, CVE-2026-43515, CVE-2026-43514, and CVE-2026-43512. |
| DLPX-97295 | Audit log forwarding to external syslog systems stops after log rotation. |
| DLPX-97341 | Running the storage test operation may incorrectly target active pool vdevs for partition table erasure, potentially causing data loss after the next engine reboot. |
| DLPX-97362 | Provisioning or refreshing a vPDB could fail with ORA-17500 due to required archive logs being skipped during CDB recovery under some circumstances. |
| DLPX-97584 | Repeated replication and serialization operations could leave stale temporary service entries, slowing upgrade operations over time and potentially causing upgrade failures. |
| DLPX-97718 | Addressed CVE-2026-14254, where account lockout could be bypassed through a race condition. |
| DLPX-97803 | On VMware hosts using AMD CPUs, the Delphix Engine could unexpectedly restart with a kernel panic. |
| DLPX-97804 | After upgrading to 2026.2.0.0 or later, snapshot timezone changes for AIX hosts. |
| DLPX-97820 | After upgrading to 2026.2.0.0 or later, AIX hosts randomly generate core dumps. |
| DLPX-97896 | Updated JDK to 17.0.19_10 to address CVE-2026-34282. |
| DLPX-97930 | Increased the memory limit for the management stack to help prevent out-of-memory conditions. |
Known issues
| Issue number | Summary | Workaround |
|---|---|---|
| DLPX-97092 | Refreshing an Oracle virtual PDB may fail with ORA-01157 and ORA-01110 in ASM environments when the source Data Guard primary PDB contains offline datafiles that have not been offline-dropped on the standby. | No workaround is available. Customers affected by this issue should contact Delphix Support to obtain a hotfix or wait for the product fix which is planned for the 2026.5.0.0 engine release. |
API changes
In Delphix 2026.4.0.0, the new API version is 1.11.50. This section describes all API changes since API version 1.11.49 which was released with Continuous Data 2026.3.0.0. All URL paths are relative to /resources/json/delphix.
Updates
| API object | Path | Type | Name | Change |
|---|---|---|---|---|
OfflineReplicationReceiveSpec
|
NA | Property | nfsShare
|
Removed property related to NFS share for offline replication data transfer. |
OfflineReplicationReceiveSpec
|
NA | Property | offlineStorage
|
New property added. Offline storage location to read replicated data from. |
ReplicationOfflineTraffic
|
NA | Property | nfsShare
|
Removed property related to NFS share for offline replication data transfer. |
Additions
| API object | Path | Type | Name | Change |
|---|---|---|---|---|
EndEntityReplication
|
NA | API type | EndEntityReplication
|
New API type added. The end-entity certificate used for DCT replication TLS on the primary. |
NfsOfflineStorage
|
NA | API type | NfsOfflineStorage
|
New API type added. NFS share for offline replication data transfer (host:/export). |
ObjectStore
|
/storage/objectStorage
|
Root operation | testExistingConnection
|
Added root operation. Tests connectivity to the object store with the existing configuration. |
OfflineStorage
|
NA | API type | OfflineStorage
|
New API type added. Offline storage location to read replicated data from. |
PublicSystemInfo
|
/about
|
Property | engineType
|
Added enum value "UCF". Description updated to list UCF as a possible engine type. |
RepaveEngineSummary
|
NA | Property | engineType
|
Added enum value "UCF". Description updated to list UCF as a possible engine type. |
ReplicationOfflineTraffic
|
NA | Property | offlineStorage
|
New property added. Offline storage location to use for data transfer. Replaces nfsShare. |
SystemInfo
|
/system
|
Operation | provisionUcfOrganization
|
New operation added. Provisions UCF organization details during first-boot setup. |
UcfOrganizationParameters
|
NA | API type | UcfOrganizationParameters
|
New API type added. Organization details for first-boot UCF provisioning. |
Deprecated and EOL features
Continuous Data User Interfaces End of Life
As we transition to the Data Control Tower platform experience, we are sunsetting all Continuous Data management and setup interfaces. The transition happens in stages:
-
September 2026: Starting with 2026.5, new engine installations will no longer include the Management UI, while upgraded engines are unaffected.
-
July 2027: With 2027.4, all UI, CLI, and API interfaces are removed from new installs and upgrades, marking End of Maintenance; support will continue to triage issues and offer reasonable workarounds, with equivalent fixes delivered through Data Control Tower or interface-free engine upgrades.
-
January 2029: End of Maintenance and Support concludes.
As you migrate, your data, dSources, virtual databases, and configurations remain fully intact; only the interfaces change. All Delphix customers are entitled to Data Control Tower, and we recommend installing it and connecting your engines now to begin the transition.
For more information, read the full announcement and FAQ.
Last updated
Oct 9, 2026