Fix #1711 - DDLWorker: don't recreate replica dirs on every main-loop tick

[CarlosFelipeOR]

Fix #1711

Test PR. The fix was authored with assistance from AI model Claude Opus 4.7.

After upstream PR ClickHouse#92339, DDLWorker::markReplicasActive() unconditionally re-creates /clickhouse/task_queue/replicas/<host_id> on every iteration of runMainThread, racing with external recursive cleanup of /clickhouse and deterministically breaking test_replication_without_zookeeper::test_startup_without_zookeeper on antalya-25.8 since 25.8.21 (PR #1600).

This change adds a should_create_dirs parameter to markReplicasActive and only calls createReplicaDirs from real events (initialization/reconnect, or when a new host ID was just observed via host_ids_updated), not on every tick. This restores the pre-ClickHouse#92339 invariant ("create host dirs once at startup/reconnect") while preserving the interserver-IO fix from ClickHouse#92339.

Changelog category (leave one):

• Bug Fix (user-visible misbehavior in an official stable release)

Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):

Fix race in DDLWorker that broke recursive deletion of the DDL queue path in ZooKeeper.

Documentation entry for user-facing changes

...

CI/CD Options

Exclude tests:

• Fast test
• Integration Tests
• Stateless tests
• Stateful tests
• Performance tests
• All with ASAN
• All with TSAN
• All with MSAN
• All with UBSAN
• All with Coverage
• All with Aarch64
• All Regression
• Disable CI Cache

Regression jobs to run:

• Fast suites (mostly <1h)
• Aggregate Functions (2h)
• Alter (1.5h)
• Benchmark (30m)
• ClickHouse Keeper (1h)
• Iceberg (2h)
• LDAP (1h)
• Parquet (1.5h)
• RBAC (1.5h)
• SSL Server (1h)
• S3 (2h)
• S3 Export (2h)
• Swarms (30m)
• Tiered Storage (2h)

Made with Cursor

[github-actions]

Workflow [PR], commit [98ddd66]