Multi-Host support for NFS3 and iSCSI

[sandeeplocharla]

Description

This PR...
Provides support to Multi-Host for NFS3 and iSCSI

Types of changes

• Breaking change (fix or feature that would cause existing functionality to change)
• New feature (non-breaking change which adds functionality)
• Bug fix (non-breaking change which fixes an issue)
• Enhancement (improves an existing feature and functionality)
• Cleanup (Code refactoring and cleanup, that may add test cases)
• Build/CI
• Test (unit or integration test code)

Feature/Enhancement Scale or Bug Severity

Feature/Enhancement Scale

• Major
• Minor

Bug Severity

• BLOCKER
• Critical
• Major
• Minor
• Trivial

Screenshots (if appropriate):

How Has This Been Tested?

Only 1 Host present in the Cluster

Created an NFS3 StoragePool with Cluster scope

ONTAP volume created for the SP

Export Policy rule and client

Added a new host

Updated client in the Export Policy

Removed the host from the cluster

The respective client has been removed from the Export Policy

Test Done	Result	Comments
Create an instance by specifying the host and disk offering pointed to NFS3 Primary Storage Pool (Cluster scoped with min of 2 hosts)	PASS
Create an instance specifying host and disk offering pointed to iSCSI Primary Storage Pool (Cluster scoped with min of 2 hosts)	PASS
Create an instance in a cluster with atleast 2 hosts in the cluster by not specifying the host and disk offering pointed to NFS3 Primary Storage Pool (Cluster scoped with min of 2 hosts)	PASS
Create an instance in a cluster with atleast 2 hosts in the cluster by not specifying the host and disk offering pointed to iSCSI Primary Storage Pool (Cluster scoped with min of 2 hosts)	PASS
Create an instance by specifying the host and disk offering pointed to 2 NFS3 Primary Storage Pools (Cluster scoped + Zone scoped)	FAIL	StoragePoolAllocator is coming as empty and the random strategy to allocate was failing.
Create an instance by specifying the host and disk offering pointed to 2 iSCSI Primary Storage Pools (Cluster scoped + Zone scoped)	FAIL	StoragePoolAllocator is coming as empty and the random strategy to allocate was failing.
Create an instance by specifying the host and disk offering pointed to NFS3 Primary Storage Pool (Zone scoped with min of 2 hosts)	FAIL	Zone scoped instance creation failing (CSTACKEX-188)
Create an instance specifying host and disk offering pointed to iSCSI Primary Storage Pool (Zone scoped with min of 2 hosts)	FAIL	Zone scoped instance creation failing (CSTACKEX-188)
Create an instance in a cluster with atleast 2 hosts in the cluster by not specifying the host and disk offering pointed to NFS3 Primary Storage Pool (Zone scoped with min of 2 hosts)	FAIL	Zone scoped instance creation failing (CSTACKEX-188)
Create an instance in a cluster with atleast 2 hosts in the cluster by not specifying the host and disk offering pointed to iSCSI Primary Storage Pool (Zone scoped with min of 2 hosts)	FAIL	Zone scoped instance creation failing (CSTACKEX-188)
Power off VM on Host-1 and start it on Host-2 (same cluster) with disk offering pointed to NFS3 primary storage pool (cluster scoped)	PASS
Power off and on a VM with 'last known host' selected. Create the instance in a cluster, without specifying a host.	PASS	Last known host was selected
Power off and on a VM with 'last known host' NOT selected. Create the instance in a cluster, without specifying a host.	PASS	VM got hosted on the chosen host.
Power off and on a VM with 'last known host' selected but choose a different host. Create the instance in a cluster, without specifying a host.	PASS	VM got hosted on the chosen host.
Insufficient resources in the storage pool. Instance creation should fail.	PASS	Generic error is being displayed instead of proper error regarding Insufficient resources.
One host in the cluster (min. 2 hosts) loaded, resulting in insufficient CPU. Disk Offering pointed to Cluster scoped primary storage pool.	PASS
Both host in the cluster (min. 2 hosts) loaded, resulting in insufficient CPU. Disk Offering pointed to Cluster scoped primary storage pool. Instance creation should fail but disks should first get created and then destroyed.

[github-actions]

This pull request has merge conflicts. Dear author, please fix the conflicts and sync your branch with the base branch.

[Copilot]

Pull request overview

Adds multi-host lifecycle handling for the ONTAP volume plugin by updating NFS export policy rules as hosts connect to / are removed from storage pools, and by extending the AccessGroup model/strategy API to support add/remove semantics.

Changes:

• Implemented UnifiedNASStrategy.updateAccessGroup to add/remove host client matches in an existing NFS export policy.
• Added HostRuleAction to AccessGroup and wired host-connect / host-removal flows to update NFS3 export policy rules.
• Tightened/adjusted model and API surface (e.g., StorageStrategy.updateAccessGroup visibility; ExportRule.ProtocolsEnum JSON annotations).

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 7 comments.

Show a summary per file

File	Description
plugins/storage/volume/ontap/src/test/java/org/apache/cloudstack/storage/service/StorageStrategyTest.java	Updates test scaffolding to match new abstract method visibility; import/method formatting adjustments.
plugins/storage/volume/ontap/src/main/java/org/apache/cloudstack/storage/service/UnifiedNASStrategy.java	Implements export-policy mutation logic for NFS host add/remove (multi-host enablement).
plugins/storage/volume/ontap/src/main/java/org/apache/cloudstack/storage/service/StorageStrategy.java	Makes updateAccessGroup explicitly public abstract for consistent overriding/call sites.
plugins/storage/volume/ontap/src/main/java/org/apache/cloudstack/storage/service/model/AccessGroup.java	Adds HostRuleAction (ADD/REMOVE) to drive update semantics.
plugins/storage/volume/ontap/src/main/java/org/apache/cloudstack/storage/listener/OntapHostListener.java	Updates host lifecycle handling to update NFS3 export policy rules on connect and before removal.
plugins/storage/volume/ontap/src/main/java/org/apache/cloudstack/storage/feign/model/ExportRule.java	Adjusts enum JSON serialization/deserialization behavior for export rule protocols.
plugins/storage/volume/ontap/src/main/java/org/apache/cloudstack/storage/driver/OntapPrimaryDatastoreDriver.java	Import reordering / formatting-only change.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

This pull request has merge conflicts. Dear author, please fix the conflicts and sync your branch with the base branch.

[Copilot]

Pull request overview

Copilot reviewed 17 out of 17 changed files in this pull request and generated 4 comments.

[Copilot]

Pull request overview

Copilot reviewed 6 out of 7 changed files in this pull request and generated 5 comments.

[Copilot]

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 1 comment.

[Copilot]

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 2 comments.

[rajiv-jain-netapp]

Thinking aloud, do we not have any usecase where this method returns false?

[sandeeplocharla]

If export policy rule hasn't been removed, we could catch the exception and return false

[rajiv-jain-netapp]

This is not true for export policy, right? we have one export policy per pool?

[sandeeplocharla]

As its an update method, the code is checking to see if atleast one rule exists. But, its not creating one policy per host, though its over doing the check currently. Will correct it as we plan to allow pool creation with empty rules set.

[rajiv-jain-netapp]

Okay, check the code in the remove block for the last client entry in that rule. If that is the case, we have to remove the rule itself rather than remove the client from the rule. Validate the values from ONTAP while testing.

[Copilot]

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 3 comments.

[Copilot]

Pull request overview

Copilot reviewed 7 out of 7 changed files in this pull request and generated 3 comments.

[rajiv-jain-netapp]

In the testing section, I need results for

1. User created an NFS pool for cluster scope; validate the export policy on the ONTAP for the rules and client added.
2. Now add a new host to this cluster and validate whether the export policy is getting updated for this new host or not.
3. Similar to #2, test for host removal as well.

Note: I noticed the term "Live migrate" in your current test case capture. Just to remind you, we do not support live migration yet; these details will lead to a wrong interpretation in audiences. Please correct them accordingly.

[sandeeplocharla]

In the testing section, I need results for

1. User created an NFS pool for cluster scope; validate the export policy on the ONTAP for the rules and client added.
2. Now add a new host to this cluster and validate whether the export policy is getting updated for this new host or not.
3. Similar to Feignconfiguration and volume feignClient along with desired POJOs #2, test for host removal as well.

Note: I noticed the term "Live migrate" in your current test case capture. Just to remind you, we do not support live migration yet; these details will lead to a wrong interpretation in audiences. Please correct them accordingly.

Updated the Testing done with the relevant screenshots.
Also, the test section mentions "Live migrate a VM between two hosts in the same cluster". As per my understanding, we currently don't support VM's storage migration i.e., between 2 storage pools, which the test case is not claiming to be. If the language could be made better, please suggest an alternative, will update it accordingly.