High Availability mode for Designer

[mateuszkp96]

Describe your changes

Checklist before merge

• Related issue ID is placed at the beginning of PR title in [brackets] (can be GH issue or Nu Jira issue)
• Code is cleaned from temporary changes and commented out lines
• Parts of the code that are not easy to understand are documented in the code
• Changes are covered by automated tests
• Showcase in dev-application.conf added to demonstrate the feature
• Documentation added or updated
• Added entry in Changelog.md describing the change from the perspective of a public distribution user
• Added MigrationGuide.md entry in the appropriate subcategory if introducing a breaking change
• Verify that PR will be squashed during merge

[github-actions]

🟢 Cypress snapshots are up to date. No changes needed.

[Copilot]

Pull request overview

This PR introduces a High Availability (HA) mode for the Designer backend, enabling multiple instances to run concurrently with leader-elected, single-instance execution of coordinated/background operations (via DB-backed distributed locks). It also adds a public endpoint to expose leader status for load balancers and documents the feature.

Changes:

• Add DB-backed distributed locking + leader election services (PostgreSQL-only) and wire them into Designer startup and schedulers.
• Gate background/periodic work (deployments reconciliation/synchronization and periodic deployment/rescheduling actors) behind leadership/locks.
• Add GET /api/app/leader public endpoint, plus OpenAPI + migration guide/changelog updates and integration tests.

Reviewed changes

Copilot reviewed 31 out of 31 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
scalafix-rules/src/test/scala/utils/SchemaInSqlDetectorSpec.scala	Extends detector test coverage with SQL patterns used by distributed locks.
scalafix-rules/src/main/scala/utils/SchemaInSqlDetector.scala	Updates SQL token heuristics to reduce false positives for new lock SQL constructs.
docs/MigrationGuide.md	Documents the new optional ha { ... } configuration block and constraints.
docs/Changelog.md	Adds changelog entry describing HA mode and the new leader endpoint.
docs-internal/api/nu-designer-openapi.yaml	Documents /api/app/leader and its response schema in OpenAPI.
designer/server/src/test/scala/pl/touk/nussknacker/ui/process/periodic/flink/RescheduleFinishedActorTest.scala	Updates actor tests to include periodic locking behavior (skip when lock not acquired).
designer/server/src/test/scala/pl/touk/nussknacker/ui/process/periodic/flink/DeploymentActorTest.scala	Updates deployment actor tests for lock acquisition/renewal and in-flight behavior.
designer/server/src/test/scala/pl/touk/nussknacker/ui/ha/SlickDistributedLockSpec.scala	Adds Postgres-backed tests for acquire/renew/release semantics.
designer/server/src/test/scala/pl/touk/nussknacker/ui/ha/LeadershipServiceSpec.scala	Adds unit tests for leader election heartbeat, callbacks, and stop behavior.
designer/server/src/test/scala/pl/touk/nussknacker/ui/api/AppApiHttpServiceHaSpec.scala	Adds IT coverage for /api/app/leader in HA-enabled mode using Postgres DB.
designer/server/src/test/scala/pl/touk/nussknacker/ui/api/AppApiHttpServiceBusinessSpec.scala	Adds test for /api/app/leader when HA is disabled (no instanceId).
designer/server/src/test/scala/pl/touk/nussknacker/test/base/it/NuItTestWithPostgres.scala	Introduces an IT base trait that starts the app with Postgres for HA endpoint testing.
designer/server/src/main/scala/pl/touk/nussknacker/ui/server/TapirHttpServiceFactory.scala	Wires Leadership into AppApiHttpService.
designer/server/src/main/scala/pl/touk/nussknacker/ui/process/periodic/SchedulingDependencies.scala	Threads PeriodicLock through periodic scheduling dependencies.
designer/server/src/main/scala/pl/touk/nussknacker/ui/process/periodic/RescheduleFinishedActor.scala	Adds periodic lock acquisition before rescheduling/marking failures.
designer/server/src/main/scala/pl/touk/nussknacker/ui/process/periodic/PeriodicLock.scala	Introduces a shared periodic-work lock wrapper with HA-aware resource lifecycle.
designer/server/src/main/scala/pl/touk/nussknacker/ui/process/periodic/PeriodicDeploymentManager.scala	Uses PeriodicLock when creating periodic actors.
designer/server/src/main/scala/pl/touk/nussknacker/ui/process/periodic/DeploymentActor.scala	Adds lock acquisition/renewal logic to periodic deployments to avoid multi-instance duplication.
designer/server/src/main/scala/pl/touk/nussknacker/ui/process/newdeployment/synchronize/DeploymentsStatusesSynchronizationScheduler.scala	Runs deployment status synchronization only on the elected leader.
designer/server/src/main/scala/pl/touk/nussknacker/ui/process/deployment/reconciliation/FinishedDeploymentsStatusesSynchronizationScheduler.scala	Runs finished-deployments reconciliation only on the elected leader.
designer/server/src/main/scala/pl/touk/nussknacker/ui/ha/Leadership.scala	Adds leader election service with heartbeat + callbacks + shutdown handling.
designer/server/src/main/scala/pl/touk/nussknacker/ui/ha/LeaderLock.scala	Adds a dedicated leader lock wrapper (name + acquire/renew/release).
designer/server/src/main/scala/pl/touk/nussknacker/ui/ha/HaMode.scala	Adds HA config parsing + validation (durations/constraints).
designer/server/src/main/scala/pl/touk/nussknacker/ui/ha/DistributedLock.scala	Implements Slick/Postgres distributed lock operations and NoOp fallback.
designer/server/src/main/scala/pl/touk/nussknacker/ui/factory/NussknackerAppFactory.scala	Triggers recovery on leadership acquisition (not only at startup).
designer/server/src/main/scala/pl/touk/nussknacker/ui/factory/DomainServices.scala	Creates/wires DistributedLock, PeriodicLock, and Leadership into domain services and schedulers.
designer/server/src/main/scala/pl/touk/nussknacker/ui/config/DesignerConfig.scala	Adds haMode to resolved DesignerConfig.
designer/server/src/main/scala/pl/touk/nussknacker/ui/api/description/AppApiEndpoints.scala	Adds Tapir endpoint + DTO/encoder for leader status response.
designer/server/src/main/scala/pl/touk/nussknacker/ui/api/AppApiHttpService.scala	Exposes /api/app/leader and returns leader status + instanceId.
designer/server/src/main/resources/db/migration/postgres/V1_072__DistributedLocksTable.sql	Adds Postgres migration creating distributed_locks table.
designer/server/src/main/resources/db/migration/hsql/V1_072__DistributedLocksTable.sql	Adds HSQL no-op migration with explanation (HA requires Postgres).

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 32 out of 32 changed files in this pull request and generated 1 comment.

Comments suppressed due to low confidence (2)

designer/server/src/test/scala/pl/touk/nussknacker/ui/ha/SlickDistributedLockSpec.scala:81

• The Using(...) result is ignored here as well, which would hide failures inserting the expired lock and potentially make the test pass for the wrong reason. Append .get so a failed INSERT fails the test.

    Using(testDbRef.db.createSession()) { session =>
      session
        .prepareStatement(
          s"""INSERT INTO "${getSchemaName()}".distributed_locks (name, lock_until, locked_at, locked_by)
             |VALUES ('$lockName', CURRENT_TIMESTAMP - INTERVAL '1 second', CURRENT_TIMESTAMP - INTERVAL '60 seconds', '$holder')""".stripMargin
        )
        .execute()
    }

designer/server/src/test/scala/pl/touk/nussknacker/ui/ha/SlickDistributedLockSpec.scala:23

• The Using(...) result is ignored, so if the TRUNCATE fails the exception will be swallowed and the test may continue with dirty state. Call .get (or otherwise rethrow) to ensure DB cleanup failures fail the test.

This issue also appears on line 74 of the same file.

    Using(testDbRef.db.createSession()) { session =>
      session.prepareStatement(s"""TRUNCATE TABLE "${getSchemaName()}".distributed_locks""").execute()
    }

[piotrp]

HA mode's configuration should be described in documentation, along with new API endpoints