Merge remote-tracking branch 'origin/main' into problame/async-timeline-get/refactor-timeline-initialization-to-avoid-holding-tenants-timelines-lock

test_runner/fixtures/pageserver/utils.py

@@ -1,5 +1,5 @@
 import time
-from typing import Optional
+from typing import Any, Dict, Optional
 
 from fixtures.log_helper import log
 from fixtures.pageserver.http import PageserverHttpClient
@@ -72,7 +72,7 @@ def wait_until_tenant_state(
     expected_state: str,
     iterations: int,
     period: float = 1.0,
-) -> bool:
+) -> Dict[str, Any]:
     """
     Does not use `wait_until` for debugging purposes
     """
@@ -81,7 +81,7 @@ def wait_until_tenant_state(
             tenant = pageserver_http.tenant_status(tenant_id=tenant_id)
             log.debug(f"Tenant {tenant_id} data: {tenant}")
             if tenant["state"]["slug"] == expected_state:
-                return True
+                return tenant
         except Exception as e:
             log.debug(f"Tenant {tenant_id} state retrieval failure: {e}")
 

test_runner/regress/test_disk_usage_eviction.py

@@ -110,6 +110,12 @@ class EvictionEnv:
             overrides=(
                 "--pageserver-config-override=disk_usage_based_eviction="
                 + enc.dump_inline_table(disk_usage_config).replace("\n", " "),
+                # Disk usage based eviction runs as a background task.
+                # But pageserver startup delays launch of background tasks for some time, to prioritize initial logical size calculations during startup.
+                # But, initial logical size calculation may not be triggered if safekeepers don't publish new broker messages.
+                # But, we only have a 10-second-timeout in this test.
+                # So, disable the delay for this test.
+                "--pageserver-config-override=background_task_maximum_delay='0s'",
             ),
         )
 

test_runner/regress/test_remote_storage.py

@@ -146,7 +146,12 @@ def test_remote_storage_backup_and_restore(
     # listing the remote timelines will fail because of the failpoint,
     # and the tenant will be marked as Broken.
     client.tenant_attach(tenant_id)
-    wait_until_tenant_state(pageserver_http, tenant_id, "Broken", 15)
+
+    tenant_info = wait_until_tenant_state(pageserver_http, tenant_id, "Broken", 15)
+    assert tenant_info["attachment_status"] == {
+        "slug": "failed",
+        "data": {"reason": "storage-sync-list-remote-timelines"},
+    }
 
     # Ensure that even though the tenant is broken, we can't attach it again.
     with pytest.raises(Exception, match=f"tenant {tenant_id} already exists, state: Broken"):

test_runner/regress/test_tenant_detach.py

@@ -532,7 +532,7 @@ def test_ignored_tenant_reattach(
 ):
     neon_env_builder.enable_remote_storage(
         remote_storage_kind=remote_storage_kind,
-        test_name="test_remote_storage_backup_and_restore",
+        test_name="test_ignored_tenant_reattach",
     )
     env = neon_env_builder.init_start()
     pageserver_http = env.pageserver.http_client()