K8SPS-646 Documented repl type change in real-time

[nastena1606]

modified: docs/architecture.md
modified: docs/change-replication-type.md
modified: docs/operator.md

[Copilot]

Pull request overview

Updates the documentation to describe the new capability (starting in Operator 1.3.0) to switch a running cluster between Group Replication and asynchronous replication, including operational impact (downtime) and Orchestrator/proxy considerations.

Changes:

• Expanded the “Change replication type” guide with prerequisites, workflow explanation, and step-by-step procedures for both directions (GR ↔ async), plus legacy (<=1.2.0) instructions.
• Updated Operator reference docs to state mysql.clusterType can be changed on a running cluster and added Orchestrator enable/disable guidance for type switches.
• Updated architecture docs to reflect that replication type switching on a running cluster is supported starting in 1.3.0 (with downtime).

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.

File	Description
docs/operator.md	Adds notes about changing mysql.clusterType at runtime and ties Orchestrator enablement to replication-type switches.
docs/change-replication-type.md	Rewrites/expands the guide to cover proxy switching and real-time replication-type changes with prerequisites and workflows.
docs/architecture.md	Updates the replication-types section to reflect 1.3.0 runtime switching support and downtime impact.

Suppressed comments (4)

docs/operator.md:369

• mysql.clusterType is documented as having value type int, but the allowed values shown (async, group-replication) are strings. This mismatch can confuse readers and should be corrected.

This issue also appears on line 365 of the same file.

Starting with Operator version 1.3.0, you can change this value on a running cluster. The Operator converts the topology in place. The switch causes downtime. See [Change replication type](change-replication-type.md) to learn more.

| Value type  | Example    |
| ----------- | ---------- |
| :material-numeric-1-box: int     | `group-replication` |

docs/operator.md:365

• The description says the Operator "converts the topology in place", but this PR’s new change-replication guide describes deleting/recreating MySQL/proxy workloads. Rephrase here to match the actual workflow (reuse PVCs, but workloads are recreated) to avoid implying an in-place conversion without teardown.

Starting with Operator version 1.3.0, you can change this value on a running cluster. The Operator converts the topology in place. The switch causes downtime. See [Change replication type](change-replication-type.md) to learn more.

docs/change-replication-type.md:30

• This list item is split across lines without indentation, which can break Markdown list rendering. Keep it on one line (or indent the continuation) so it remains a single bullet.

* You can accept a maintenance window. Writes fail until the
cluster returns to `ready`.

docs/change-replication-type.md:33

• This list item is split across lines without indentation, which can break Markdown list rendering. Keep it on one line (or indent the continuation) so it remains a single bullet.

* Orchestrator is required for async replication. Enable it
when you switch to `async` replication type. Disable it when you
switch to `group-replication`.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.