diff --git a/testsuite/replay-verify/archive_disk_utils.py b/testsuite/replay-verify/archive_disk_utils.py
index bb1e67a1d6060f..2ccf4424d0c8ca 100644
--- a/testsuite/replay-verify/archive_disk_utils.py
+++ b/testsuite/replay-verify/archive_disk_utils.py
@@ -22,6 +22,10 @@
 MAINNET_SNAPSHOT_NAME = "mainnet-archive"
 
 
+def get_region_from_zone(zone):
+    return zone.rsplit("-", 1)[0]
+
+
 def get_kubectl_credentials(project_id, region, cluster_name):
     try:
         # Command to get kubectl credentials for the cluster
@@ -141,6 +145,8 @@ def create_snapshot_with_gcloud(
         source_disk_link,
         "--project",
         target_project,
+        "--storage-location",
+        get_region_from_zone(source_zone),
     ]
 
     try:
@@ -199,7 +205,7 @@ def create_disk_pv_pvc_from_snapshot(
     wait_for_operation(project, zone, operation.name, compute_v1.ZoneOperationsClient())
     logger.info(f"Disk {disk_name} created from snapshot {og_snapshot_name}.")
 
-    region_name = zone.rsplit("-", 1)[0]
+    region_name = get_region_from_zone(zone)
     get_kubectl_credentials(project, region_name, cluster_name)
     # create_persistent_volume(disk_name, pv_name, pvc_name, namespace, True)
     # this is only for xfs replaying logs to repair the disk
@@ -228,6 +234,17 @@ def create_disk_pv_pvc_from_snapshot(
         time.sleep(10)
     logger.info(f"creating final snapshot")
     create_snapshot_with_gcloud(snapshot_name, project, disk_name, zone, project)
+    logger.info("deleting repair pvc and correpsonding pv and disks")
+    # delete the PVC used for repair
+    try:
+        client.CoreV1Api().delete_namespaced_persistent_volume_claim(
+            name=repair_pvc, namespace=namespace
+        )
+    except ApiException as e:
+        if e.status == 404:
+            print(f"PersistentVolumeClaim '{repair_pvc}' not found.")
+        else:
+            print(f"Exception when deleting PersistentVolumeClaim: {e}")
 
 
 def is_job_pod_cleanedup(namespace, job_name):
@@ -298,7 +315,7 @@ def create_persistent_volume(disk_name, pv_name, pvc_name, namespace, read_only)
                 fs_type="xfs",
                 read_only=read_only,
             ),
-            persistent_volume_reclaim_policy="Retain",
+            persistent_volume_reclaim_policy="Delete",
             storage_class_name="standard",
         ),
     )
@@ -427,7 +444,7 @@ def create_pvcs_from_snapshot(run_id, snapshot_name, namespace, pvc_num, label):
     return res
 
 
-def create_disk_pv_pvc(
+def create_repair_disk_and_its_snapshot(
     project, zone, cluster_name, og_snapshot_name, snapshot_name, prefix, namespace
 ):
     tasks = []
@@ -462,8 +479,6 @@ def create_disk_pv_pvc(
             except Exception as e:
                 logger.error(f"Task generated an exception: {e}")
 
-    # start a self deleteing job to mount the xfs disks for repairing
-
 
 def parse_args():
     parser = argparse.ArgumentParser(
@@ -506,7 +521,7 @@ def parse_args():
         source_namespace,
         project_id,
     )
-    create_disk_pv_pvc(
+    create_repair_disk_and_its_snapshot(
         project_id,
         zone,
         cluster_name,
diff --git a/testsuite/replay-verify/main.py b/testsuite/replay-verify/main.py
index e8b29e363b4ee5..e3cb4acc0b26d1 100644
--- a/testsuite/replay-verify/main.py
+++ b/testsuite/replay-verify/main.py
@@ -28,6 +28,7 @@
 
 REPLAY_CONCURRENCY_LEVEL = 1
 
+
 class Network(Enum):
     TESTNET = 1
     MAINNET = 2
@@ -241,6 +242,7 @@ def get_pod_status(self):
     def get_humio_log_link(self):
         return construct_humio_url(self.label, self.name, self.start_time, time.time())
 
+
 class ReplayConfig:
     def __init__(self, network):
         if network == Network.TESTNET:
@@ -253,9 +255,10 @@ def __init__(self, network):
             self.concurrent_replayer = 18
             self.pvc_number = 8
             self.min_range_size = 10_000
-            self.range_size = 2_000_000 
+            self.range_size = 2_000_000
             self.timeout_secs = 400
 
+
 class TaskStats:
     def __init__(self, name):
         self.name = name
@@ -308,7 +311,7 @@ def __init__(
         self.image = image
         self.pvcs = []
         self.config = replay_config
-        
+
     def __str__(self):
         return f"""ReplayScheduler:
             id: {self.id}
@@ -360,7 +363,11 @@ def create_pvc_from_snapshot(self):
             else MAINNET_SNAPSHOT_NAME
         )
         pvcs = create_pvcs_from_snapshot(
-            self.id, snapshot_name, self.namespace, self.config.pvc_number, self.get_label()
+            self.id,
+            snapshot_name,
+            self.namespace,
+            self.config.pvc_number,
+            self.get_label(),
         )
         assert len(pvcs) == self.config.pvc_number, "failed to create all pvcs"
         self.pvcs = pvcs
@@ -504,12 +511,16 @@ def get_image(image_tag=None):
     shell = forge.LocalShell()
     git = forge.Git(shell)
     image_name = "tools"
-    default_latest_image = forge.find_recent_images(
-        shell,
-        git,
-        1,
-        image_name=image_name,
-    )[0] if image_tag is None else image_tag
+    default_latest_image = (
+        forge.find_recent_images(
+            shell,
+            git,
+            1,
+            image_name=image_name,
+        )[0]
+        if image_tag is None
+        else image_tag
+    )
     full_image = f"{forge.GAR_REPO_NAME}/{image_name}:{default_latest_image}"
     return full_image
 
@@ -546,7 +557,7 @@ def print_logs(failed_workpod_logs, txn_mismatch_logs):
         range_size=range_size,
         image=image,
         replay_config=config,
-        network= network,
+        network=network,
         namespace=args.namespace,
     )
     logger.info(f"scheduler: {scheduler}")