Grow and shrink storage pool in maintenance

Author: sp16743

test/integration/plugins/ontap/README.md

@@ -332,13 +332,13 @@ self.assertEqual(result.state, "Maintenance")
 
 | Suite | File | Tests | What it covers |
 | ------- | ------ | ------- | --------------- |
-| NFS3 Pool Lifecycle | `nfs3/pool/test_pool_lifecycle.py` | 16 | Create, grow to 300 TiB, reject grow past 300 TiB, safe shrink, disable, resize while disabled, enable, maintenance, resize in maintenance, delete, then re-create with a volume, reject shrink below used capacity, deploy a VM and attach the volume, cleanup |
+| NFS3 Pool Lifecycle | `nfs3/pool/test_pool_lifecycle.py` | 16 | Create, grow to 300 TiB, reject grow past 300 TiB, safe shrink, disable, resize while disabled, enable, maintenance, grow and shrink in maintenance, delete, then re-create with a volume, reject shrink below used capacity, deploy a VM and attach the volume, cleanup |
 | NFS3 Pool with Volumes | `nfs3/pool/test_pool_with_volumes.py` | 7 | Same + live volume present; negative delete guard |
 | NFS3 Zone-Scoped Pool | `nfs3/pool/test_zone_scoped_pool.py` | 4 | Zone scope — all hosts connected via `attachZone` |
 | NFS3 Volume Lifecycle | `nfs3/volume/test_volume_lifecycle.py` | 5 | Volume is metadata-only; FlexVol unchanged on delete |
 | NFS3 VM + Volume Attach | `nfs3/instance/test_vm_volume_attach.py` | 13 | Full VM lifecycle with stopped-VM qcow2 grow; ROOT on tagged pool seeds/reuses template cache, which survives VM delete |
 | NFS3 Template Cache Negative | `nfs3/template/test_template_cache_negative.py` | 3 | Tag mismatch; undersized pool; out-of-band cache delete |
-| iSCSI Pool Lifecycle | `iscsi/pool/test_pool_lifecycle.py` | 16 | Create, grow to 300 TiB, reject grow past 300 TiB, safe shrink, disable, resize while disabled, enable, maintenance, resize in maintenance, delete + igroups, then re-create with a volume (LUN), reject shrink below used capacity, deploy a VM and attach the volume, cleanup |
+| iSCSI Pool Lifecycle | `iscsi/pool/test_pool_lifecycle.py` | 16 | Create, grow to 300 TiB, reject grow past 300 TiB, safe shrink, disable, resize while disabled, enable, maintenance, grow and shrink in maintenance, delete + igroups, then re-create with a volume (LUN), reject shrink below used capacity, deploy a VM and attach the volume, cleanup |
 | iSCSI Pool with Volumes | `iscsi/pool/test_pool_with_volumes.py` | 7 | Same + live LUN present; negative delete guard |
 | iSCSI Zone-Scoped Pool | `iscsi/pool/test_zone_scoped_pool.py` | 4 | Zone scope |
 | iSCSI Volume Lifecycle | `iscsi/volume/test_volume_lifecycle.py` | 5 | LUN created per CS volume; LUN removed on delete |

test/integration/plugins/ontap/TEST_CASES.md

@@ -51,16 +51,16 @@ Each suite is sequential — tests must run in numbered order; each step builds
 | 03 | `test_03_reject_grow_above_max_size` | Reject a grow past the 300 TiB ONTAP FlexVol maximum (overshoots by 1 GiB; asserts ONTAP's autosize-maximum error, not a generic failure such as aggregate space) | test_02 | `CloudstackAPIException`; `capacitybytes` stays at 300 TiB; pool stays `Up` | FlexVol `space.size` unchanged; export policy unchanged | negative |
 | 04 | `test_04_shrink_storage_pool` | Safely shrink the original empty pool to its initial capacity | test_02 | `capacitybytes` reaches the safe target; pool stays `Up` | FlexVol `space.size` reaches the safe target; remains `online`; export policy unchanged | positive |
 | 05 | `test_05_disable_storage_pool` | Disable the pool (admin operation) | test_04 | `pool.state == "Disabled"` | FlexVol still `online`; export policy still present | positive |
-| 06 | `test_06_resize_storage_pool_while_disabled` | Grow then shrink the pool back while it is Disabled | test_05 | Both `updateStoragePool` calls accepted; `capacitybytes` reaches each target; KVM hosts report each resized NFS capacity; `pool.state` stays `"Disabled"` throughout | FlexVol `space.size` reaches each target and stays `online`; export policy still present and still covers every host IP | positive |
+| 06 | `test_06_grow_and_shrink_storage_pool_while_disabled` | Grow then shrink the pool back while it is Disabled | test_05 | Both `updateStoragePool` calls accepted; `capacitybytes` reaches each target; KVM hosts report each resized NFS capacity; `pool.state` stays `"Disabled"` throughout | FlexVol `space.size` reaches each target and stays `online`; export policy still present and still covers every host IP | positive |
 | 07 | `test_07_enable_storage_pool` | Re-enable the pool | test_06 | `pool.state == "Up"` | FlexVol still `online`; export policy still present | positive |
 | 08 | `test_08_enter_maintenance_mode` | Put pool into maintenance (drains new volume allocations) | test_07 | `pool.state == "Maintenance"` | FlexVol still `online`; export policy still present (maintenance is CS-only state) | positive |
-| 09 | `test_09_resize_storage_pool_in_maintenance` | Grow the pool while it is in maintenance mode | test_08 | `updateStoragePool` accepted; `capacitybytes` reaches the requested size; `pool.state` stays `"Maintenance"` | FlexVol `space.size` reaches the requested size; remains `online`; export policy unchanged | positive |
+| 09 | `test_09_grow_and_shrink_storage_pool_in_maintenance` | Grow then shrink the pool back while it is in Maintenance | test_08 | Both `updateStoragePool` calls accepted; `capacitybytes` reaches each target; `pool.state` stays `"Maintenance"` throughout | FlexVol `space.size` reaches each target and stays `online`; export policy still present | positive |
 | 10 | `test_10_cancel_maintenance_mode` | Cancel maintenance, return pool to service | test_09 | `pool.state == "Up"` | FlexVol still `online`; export policy still present | positive |
 | 11 | `test_11_delete_pool_from_maintenance` | Enter maintenance then permanently delete the original pool | test_10 | Pool no longer returned by `listStoragePools` (CS 431 error expected on ID lookup) | FlexVol and export policy deleted | positive |
 | 12 | `test_12_create_volume_on_pool` | Create a fresh pool and allocate a CloudStack data volume | test_11 | New pool is `Up`; `createVolume` returns a volume | FlexVol `online`; export policy present | positive |
 | 13 | `test_13_create_vm_and_attach_volume` | Deploy a VM and attach the ONTAP data volume to it | test_12 | VM reaches `Running`; volume reports `virtualmachineid` equal to the VM id; VM still `Running` after attach; pool capacity unchanged | FlexVol still `online`; volume data file materialised in the FlexVol (NFS3 creates it lazily at attach) | positive |
 | 14 | `test_14_reject_shrink_below_used_capacity` | Reject shrink below the used space of the VM-attached volume, with the target held above the ONTAP FlexVol minimum. Writes incompressible data through the ONTAP files API when the VM-attached thin volume leaves used space under the minimum, and removes it before returning | test_13 (`pool`, `volume`, `vm`) | `CloudstackAPIException`; volume remains listed with unchanged `id`/`state`/`size`/`poolid`; pool capacity unchanged | FlexVol size and export policy unchanged; used space stays above the FlexVol minimum | negative |
-| 15 | `test_15_resize_pool_with_vm_attached` | Grow then shrink the pool back while the volume is attached to the running VM | test_13 (`vm`, `volume`) | Both resizes reach the requested capacity; pool stays `Up`; VM stays `Running` and keeps `virtualmachineid` on the volume | FlexVol reaches each size and stays `online`; export policy still covers every host IP | positive |
+| 15 | `test_15_grow_and_shrink_pool_with_vm_attached` | Grow then shrink the pool back while the volume is attached to the running VM | test_13 (`vm`, `volume`) | Both resizes reach the requested capacity; pool stays `Up`; VM stays `Running` and keeps `virtualmachineid` on the volume | FlexVol reaches each size and stays `online`; export policy still covers every host IP | positive |
 | 16 | `test_16_delete_volume_and_pool` | Detach and destroy the VM, delete the volume, then force-delete the pool | test_15 | VM destroyed; volume and pool no longer listed | FlexVol and export policy deleted | cleanup |
 
 ---
@@ -156,16 +156,16 @@ Each suite is sequential — tests must run in numbered order; each step builds
 | 03 | `test_03_reject_grow_above_max_size` | Reject a grow past the 300 TiB ONTAP FlexVol maximum (overshoots by 1 GiB; asserts ONTAP's autosize-maximum error, not a generic failure such as aggregate space) | test_02 | `CloudstackAPIException`; `capacitybytes` stays at 300 TiB; pool stays `Up` | FlexVol `space.size` unchanged; igroups unchanged | negative |
 | 04 | `test_04_shrink_storage_pool` | Safely shrink the original empty pool to its initial capacity | test_02 | `capacitybytes` reaches the safe target; pool stays `Up` | FlexVol reaches the safe target and remains `online`; igroups remain present | positive |
 | 05 | `test_05_disable_storage_pool` | Disable the pool | test_04 | `pool.state == "Disabled"` | FlexVol still `online` | positive |
-| 06 | `test_06_resize_storage_pool_while_disabled` | Grow then shrink the pool back while it is Disabled | test_05 | Both `updateStoragePool` calls accepted; `capacitybytes` reaches each target; `pool.state` stays `"Disabled"` throughout | FlexVol `space.size` reaches each target and stays `online`; per-host igroups still present | positive |
+| 06 | `test_06_grow_and_shrink_storage_pool_while_disabled` | Grow then shrink the pool back while it is Disabled | test_05 | Both `updateStoragePool` calls accepted; `capacitybytes` reaches each target; `pool.state` stays `"Disabled"` throughout | FlexVol `space.size` reaches each target and stays `online`; per-host igroups still present | positive |
 | 07 | `test_07_enable_storage_pool` | Re-enable the pool | test_06 | `pool.state == "Up"` | FlexVol still `online` | positive |
 | 08 | `test_08_enter_maintenance_mode` | Put pool into maintenance | test_07 | `pool.state == "Maintenance"` | FlexVol still `online`; igroups unchanged | positive |
-| 09 | `test_09_resize_storage_pool_in_maintenance` | Grow the pool while it is in maintenance mode | test_08 | `updateStoragePool` accepted; `capacitybytes` reaches the requested size; `pool.state` stays `"Maintenance"` | FlexVol reaches the requested size and remains `online`; igroups unchanged | positive |
+| 09 | `test_09_grow_and_shrink_storage_pool_in_maintenance` | Grow then shrink the pool back while it is in Maintenance | test_08 | Both `updateStoragePool` calls accepted; `capacitybytes` reaches each target; `pool.state` stays `"Maintenance"` throughout | FlexVol `space.size` reaches each target and stays `online`; per-host igroups still present | positive |
 | 10 | `test_10_cancel_maintenance_mode` | Cancel maintenance | test_09 | `pool.state == "Up"` | FlexVol still `online` | positive |
 | 11 | `test_11_enter_maintenance_and_delete_pool` | Enter maintenance then delete the original pool | test_10 | Pool no longer listed | FlexVol and all host igroups deleted | positive |
 | 12 | `test_12_create_volume_on_pool` | Create a fresh pool and allocate a CloudStack volume | test_11 | New pool is `Up`; `createVolume` returns a volume | FlexVol `online`; at least one LUN present | positive |
 | 13 | `test_13_create_vm_and_attach_volume` | Deploy a VM and attach the ONTAP data volume (LUN) to it | test_12 | VM reaches `Running`; volume reports `virtualmachineid` equal to the VM id; VM still `Running` after attach | FlexVol still `online`; LUN still present; per-host igroups unchanged | positive |
 | 14 | `test_14_reject_shrink_below_used_capacity` | Reject shrink below the used space of the VM-attached LUN, with the target held above the ONTAP FlexVol minimum. Writes incompressible data through the ONTAP files API when the VM-attached thin volume leaves used space under the minimum, and removes it before returning | test_13 (`pool`, `volume`, `vm`) | `CloudstackAPIException`; volume remains listed with unchanged `id`/`state`/`size`/`poolid`; pool capacity unchanged | FlexVol size and LUN uuid/name list unchanged; used space stays above the FlexVol minimum | negative |
-| 15 | `test_15_resize_pool_with_vm_attached` | Grow then shrink the pool back while the volume (LUN) is attached to the running VM | test_13 (`vm`, `volume`) | Both resizes reach the requested capacity; pool stays `Up`; VM stays `Running` and keeps `virtualmachineid` on the volume | FlexVol reaches each size and stays `online`; LUN still present | positive |
+| 15 | `test_15_grow_and_shrink_pool_with_vm_attached` | Grow then shrink the pool back while the volume (LUN) is attached to the running VM | test_13 (`vm`, `volume`) | Both resizes reach the requested capacity; pool stays `Up`; VM stays `Running` and keeps `virtualmachineid` on the volume | FlexVol reaches each size and stays `online`; LUN still present | positive |
 | 16 | `test_16_delete_volume_and_pool` | Detach and destroy the VM, delete the volume, then force-delete the pool | test_15 | VM destroyed; volume and pool no longer listed | LUN, FlexVol, and igroups deleted | cleanup |
 
 ---

test/integration/plugins/ontap/iscsi/pool/test_pool_lifecycle.py

@@ -31,7 +31,7 @@
   06  Grow and shrink the pool while it is disabled
   07  Enable storage pool
   08  Enter maintenance mode
-  09  Resize storage pool while in maintenance mode
+  09  Grow and shrink the pool while it is in maintenance mode
   10  Cancel maintenance mode
   11  Enter maintenance mode and delete the storage pool
   12  Create a new pool and allocate a CloudStack data volume (LUN created)
@@ -565,11 +565,11 @@ def test_05_disable_storage_pool(self):
         )
 
     # ------------------------------------------------------------------
-    # Step 06 - Resize storage pool while disabled
+    # Step 06 - Grow and shrink the pool while it is disabled
     # ------------------------------------------------------------------
 
     @attr(tags=["iscsi_workflow"], required_hardware=True)
-    def test_06_resize_storage_pool_while_disabled(self):
+    def test_06_grow_and_shrink_storage_pool_while_disabled(self):
         """
         Grow and shrink the disabled pool and verify:
           - Both capacity updates are accepted
@@ -695,19 +695,17 @@ def test_08_enter_maintenance_mode(self):
         )
 
     # ------------------------------------------------------------------
-    # Step 09 - Resize storage pool while in maintenance mode
+    # Step 09 - Grow and shrink the pool while it is in maintenance mode
     # ------------------------------------------------------------------
 
     @attr(tags=["iscsi_workflow"], required_hardware=True)
-    def test_09_resize_storage_pool_in_maintenance(self):
+    def test_09_grow_and_shrink_storage_pool_in_maintenance(self):
         """
-        Grow the pool while it is in maintenance mode and verify:
-          - updateStoragePool is accepted while the pool is in Maintenance
-          - CloudStack reports the new capacity
-          - ONTAP: FlexVol is resized and still online
-          - The pool stays in Maintenance throughout (resize must not
-            implicitly return it to Up)
-          - ONTAP: per-host igroups are unchanged
+        Grow and shrink the pool while it is in maintenance mode and verify:
+          - Both capacity updates are accepted
+          - CloudStack and ONTAP reach each requested size
+          - The pool remains in Maintenance throughout
+          - The FlexVol remains online and per-host igroups are unchanged
         """
         self.assertIsNotNone(
             self.__class__.pool, "Pool absent - test_01 must pass first"
@@ -719,44 +717,42 @@ def test_09_resize_storage_pool_in_maintenance(self):
         self.assertEqual(
             listed[0].state, "Maintenance",
             "Pool must be in Maintenance - test_08 must pass first, got '%s'"
-            % listed[0].state
+            % listed[0].state,
         )
 
         ontap_vol = self.ontap.get_volume(pool.name)
         self.assertIsNotNone(
             ontap_vol, "ONTAP FlexVol missing before maintenance resize"
         )
-        current_size = max(
+        original_size = max(
             int(getattr(listed[0], "capacitybytes", 0) or 0),
             int(ontap_vol.get("space", {}).get("size", 0) or 0),
         )
-        requested_size = current_size + TestData.ONTAP_MIN_VOLUME_SIZE
-
-        log_progress(
-            logger, "info",
-            "Resizing iSCSI pool '%s' while in Maintenance: current=%d B, "
-            "requested=%d B",
-            pool.name, current_size, requested_size,
-        )
-
-        cmd = updateStoragePoolAPI.updateStoragePoolCmd()
-        cmd.id = pool.id
-        cmd.capacitybytes = requested_size
-        self.apiClient.updateStoragePool(cmd)
+        grow_target = original_size + TestData.ONTAP_MIN_VOLUME_SIZE
 
-        resized_pool = self._poll_pool_capacity(
-            pool.id, requested_size, timeout=120
-        )
-        self.assertEqual(
-            resized_pool.state, "Maintenance",
-            "Pool should remain in Maintenance after resize, got '%s'"
-            % resized_pool.state
-        )
+        for target, operation in (
+                (grow_target, "grow"),
+                (original_size, "shrink")):
+            log_progress(
+                logger, "info",
+                "%s iSCSI pool '%s' while in Maintenance: target=%d B",
+                operation.capitalize(), pool.name, target,
+            )
+            cmd = updateStoragePoolAPI.updateStoragePoolCmd()
+            cmd.id = pool.id
+            cmd.capacitybytes = target
+            self.apiClient.updateStoragePool(cmd)
 
-        resized_ontap_vol = self._poll_ontap_volume_size(
-            pool.name, requested_size, timeout=120
-        )
-        self.assertEqual(resized_ontap_vol.get("state"), "online")
+            resized = self._poll_pool_capacity(pool.id, target, timeout=120)
+            self.assertEqual(
+                resized.state, "Maintenance",
+                "Pool should remain in Maintenance after %s, got '%s'"
+                % (operation, resized.state),
+            )
+            resized_ontap = self._poll_ontap_volume_size(
+                pool.name, target, timeout=120
+            )
+            self.assertEqual(resized_ontap.get("state"), "online")
 
         for host in self.cluster_hosts:
             iqn = (
@@ -767,7 +763,7 @@ def test_09_resize_storage_pool_in_maintenance(self):
                 igroup_name = _igroup_name(self.svm_name, host.name)
                 self.assertIsNotNone(
                     self.ontap.get_igroup(self.svm_name, igroup_name),
-                    "ONTAP igroup '%s' disappeared after maintenance resize"
+                    "ONTAP igroup '%s' disappeared during maintenance resize"
                     % igroup_name,
                 )
 
@@ -1130,11 +1126,11 @@ def test_14_reject_shrink_below_used_capacity(self):
             self._delete_filler_file(pool.name, filler_name)
 
     # ------------------------------------------------------------------
-    # Step 15 - Resize the pool while the VM holds the volume
+    # Step 15 - Grow and shrink the pool while the VM holds the volume
     # ------------------------------------------------------------------
 
     @attr(tags=["iscsi_workflow"], required_hardware=True)
-    def test_15_resize_pool_with_vm_attached(self):
+    def test_15_grow_and_shrink_pool_with_vm_attached(self):
         """
         Grow the pool and shrink it straight back while the volume is
         attached to the running VM from test_13, so resize is exercised