Skip to content

Commit 1fbec9d

Browse files
committed
F OpenNebula/one#6753: revisions and format fixes
Signed-off-by: Neal Hansen <[email protected]>
1 parent 16c54e8 commit 1fbec9d

File tree

1 file changed

+54
-7
lines changed

1 file changed

+54
-7
lines changed

source/open_cluster_deployment/storage_setup/netapp_ds.rst

Lines changed: 54 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -12,57 +12,89 @@ The `NetApp ONTAP documentation <https://docs.netapp.com/us-en/ontap/>`_ may be
1212

1313
NetApp ONTAP Setup
1414
================================================================================
15+
1516
The NetApp system requires specific configurations. This driver operates using a Storage VM that provides iSCSI connections, with volumes/LUNs mapped directly to each host after API creation. Configure and enable the iSCSI protocol according to your infrastructure requirements.
1617

1718
1. Define Aggregates/Local Tiers for your Storage VM:
19+
1820
- In ONTAP System Manager: **Storage > Storage VMs > Select your SVM > Edit > Limit volume creation to preferred local tiers**
1921
- Assign at least one aggregate/tier and note their UUID(s) for later use
2022

21-
2. For capacity monitoring:
23+
2. To enable capacity monitoring:
24+
2225
- Enable *Enable maximum capacity limit* on the same Edit Storage VM screen
23-
- If not configured, set ``DATASTORE_CAPACITY_CHECK=no`` in both datastore's attributes
26+
- If not configured, set ``DATASTORE_CAPACITY_CHECK=no`` in both of the OpenNebula datastores' attributes
2427

2528
3. No automated snapshot configuration is required - OpenNebula handles this.
2629

2730
Frontend Setup
2831
================================================================================
32+
2933
The frontend requires network access to the NetApp ONTAP API endpoint and proper NFS/iSCSI configuration:
3034

3135
1. API Access:
36+
3237
- Ensure network connectivity to the NetApp ONTAP API interface
3338

3439
2. NFS Exports:
40+
3541
- Add to ``/etc/exports`` on the frontend:
42+
3643
- Per-datastore: ``/var/lib/one/datastores/101``
3744
- Shared datastores: ``/var/lib/one/datastores``
3845

3946
.. note:: The frontend only needs to mount System Datastores, **not** Image Datastores.
4047

41-
3. iSCSI Configuration:
42-
- Maintain active iSCSI sessions for image transfers between files and NetApp storage
48+
3. iSCSI Initiators:
49+
50+
- Configure initiator security in NetApp Storage VM:
51+
52+
- **Storage VM > Settings > iSCSI Protocol > Initiator Security**
53+
- Add initiators from ``/etc/iscsi/initiatorname.conf`` (all nodes and frontend)
54+
55+
- Discover and login to the iSCSI targets
56+
57+
- ``iscsiadm -m discovery -t sendtargets -p <target_ip>`` for each iSCSI target
58+
- ``iscsiadm -m node -l`` to login to all discovered targets
59+
60+
4. Persistent iSCSI Configuration:
61+
62+
- Set ``node.startup = automatic`` in ``/etc/iscsi/iscsid.conf``
63+
4364

4465
Node Setup
4566
================================================================================
67+
4668
Configure nodes with persistent iSCSI connections and NFS mounts:
4769

4870
1. iSCSI Initiators:
71+
4972
- Configure initiator security in NetApp Storage VM:
73+
5074
- **Storage VM > Settings > iSCSI Protocol > Initiator Security**
5175
- Add initiators from ``/etc/iscsi/initiatorname.conf`` (all nodes and frontend)
5276

53-
2. Persistent Configuration:
77+
- Discover and login to the iSCSI targets
78+
79+
- ``iscsiadm -m discovery -t sendtargets -p <target_ip>`` for each iSCSI target
80+
- ``iscsiadm -m node -l`` to login to all discovered targets
81+
82+
2. Persistent iSCSI Configuration:
83+
5484
- Set ``node.startup = automatic`` in ``/etc/iscsi/iscsid.conf``
5585
- Add frontend NFS mounts to ``/etc/fstab``
5686

5787
OpenNebula Configuration
5888
================================================================================
89+
5990
Create both datastores for optimal performance (instant cloning/moving capabilities):
6091

6192
* System Datastore
6293
* Image Datastore
6394

6495
Create System Datastore
6596
--------------------------------------------------------------------------------
97+
6698
Template parameters:
6799

68100
+-----------------------+-------------------------------------------------+
@@ -115,6 +147,7 @@ Example template:
115147

116148
Create Image Datastore
117149
--------------------------------------------------------------------------------
150+
118151
Template parameters:
119152

120153
+-----------------------+-------------------------------------------------+
@@ -162,16 +195,22 @@ Example template:
162195
163196
Datastore Internals
164197
================================================================================
198+
165199
Storage architecture details:
166200

167-
- **Images**: Stored as volumes with single LUNs in NetApp
201+
- **Images**: Stored as a volume with single LUN in NetApp
168202
- **Naming Convention**:
203+
169204
- Image datastore: ``one_<datastore_id>_<image_id>`` (volume), ``one_<datastore_id>_<image_id>_lun`` (LUN)
170205
- System datastore: ``one_<vm_id>_disk_<disk_id>`` (volume), ``one_<datastore_id>_<vm_id>_disk_<disk_id>_lun`` (LUN)
206+
171207
- **Operations**:
208+
172209
- Non-persistent: FlexClone
173210
- Persistent: Rename
174211

212+
Symbolic links from the system datastore will be created for each virtual machine disk by the frontend and shared via NFS with the compute nodes.
213+
175214
.. important:: The system datastore requires a shared filesystem (e.g., NFS mount from frontend to nodes) for device link management and VM metadata distribution.
176215

177216

@@ -183,4 +222,12 @@ Additional Configuration
183222
+=======================+=================================================+
184223
| ``NETAPP_MULTIPATH`` | ``yes`` or ``no``, Default: ``yes`` |
185224
| | Set to ``no`` to disable multipath |
186-
+-----------------------+-------------------------------------------------+
225+
+-----------------------+-------------------------------------------------+
226+
227+
228+
System Considerations
229+
================================================================================
230+
231+
Occasionally, under network interruptions or if a volume is deleted directly from NetApp, the iSCSI connection may drop or fail. This can cause the system to hang on a ``sync`` command, which in turn may lead to OpenNebula operation failures on the affected host. Although the driver is designed to manage these issues automatically, it's important to be aware of these potential iSCSI connection challenges.
232+
233+
.. note:: This behavior stems from the inherent complexities of iSCSI connections and is not exclusive to OpenNebula or NetApp.

0 commit comments

Comments
 (0)