> For the complete documentation index, see [llms.txt](https://mariadb.com/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://mariadb.com/docs/release-notes/enterprise-operator/26.06/26.06.3.md).

# MariaDB Enterprise Kubernetes Operator 26.06.3

MariaDB Enterprise Operator 26.06.3 is a maintenance release of MariaDB Enterprise Operator, released on 2026-09-09

**Release date**: 9 September 2026

MariaDB Enterprise Kubernetes Operator 26.06.3 is a maintenance release for the 26.06 series. It focuses on how semi-synchronous replication is applied across the nodes of a replication topology.

If you are updating from a previous version, **please follow the** [**UPDATE GUIDE**](https://mariadb.com/docs/tools/mariadb-enterprise-operator/updates/update-26.06.3) to ensure a safe transition.

### Changelog

* **K8SDKR-311**: Primary-side semi-synchronous replication is now managed per role instead of being left enabled in every node:
  * [`rpl_semi_sync_master_enabled`](https://mariadb.com/docs/server/ha-and-performance/standard-replication/semisynchronous-replication#rpl_semi_sync_master_enabled) is now converged to the role of each node on every reconciliation: enabled in the current primary, disabled in the replicas. It is still rendered as `ON` in the configuration file, so a node boots armed and the operator disables it in the replicas once it reconciles them. Previously it was left enabled in every node, so a replica binlogging the transactions it applied waited for an acknowledgement that could never arrive: with a high [`replication.semiSyncAckTimeout`](https://mariadb.com/docs/tools/mariadb-enterprise-operator/topologies/high-availability/replication#configuration) its apply thread parked indefinitely and replication stopped progressing.
  * [`replication.semiSyncWaitNoSlave`](https://mariadb.com/docs/tools/mariadb-enterprise-operator/topologies/high-availability/replication#configuration) now only takes effect where it is meaningful, in the primary, rather than acting as a cluster-wide policy. Keeping it `ON`, so that the primary never commits a transaction without a replica acknowledgement, no longer stalls the replicas.
  * New [`replication.semiSyncBootAsReplica`](https://mariadb.com/docs/tools/mariadb-enterprise-operator/topologies/high-availability/replication#configuration) field in the `MariaDB` resource. It renders `read_only=ON` and `rpl_semi_sync_master_enabled=OFF` in the configuration file, so that a node boots read-only and becomes writable only once the operator has enabled semi-synchronous replication on it. This closes the window in which a node that has just started is writable while unable to require a replica acknowledgement. It is disabled by default, and while it is enabled a restarted primary stays read-only until the operator reconciles it. Using this field requires the [data-plane](https://mariadb.com/docs/tools/mariadb-enterprise-operator/topologies/data-plane) to be updated to `26.6.3`.

### Platform and component versions

The current release has been tested with the following versions:

| Platform/Component        | Version  |
| ------------------------- | -------- |
| Kubernetes                | 1.36     |
| OpenShift                 | 4.20     |
| MariaDB Enterprise Server | 11.8.8-5 |
| MaxScale                  | 25.10.3  |

<sub>*This page is: Copyright © 2026 MariaDB. All rights reserved.*</sub>

{% @marketo/form formid="4316" formId="4316" %}