Displaying unnecessary notifications for creation of DigitalOceanConfig

[momesgin]

Setup

• Rancher version: 2.14 and above, but could be happening in older versions too
• Rancher UI Extensions: n/a
• Browser type & version: chrome

Describe the bug
When creating a DigitalOcean cluster we display a notification growl for each successfully created DigitalOceanConfig.

To Reproduce
Go to cluster creation page > Add required info > Click Create

Result
You see a notification growl for each successfully created DigitalOceanConfig.

Expected Result
No notification growl should be displayed for the creation of DigitalOceanConfig after creating a cluster.

Screenshots

issue-rec-do-config.mov

Additional context
We also display notification messages on the notification section which are not interactive.

[github-actions]

🎯 Agentic Issue Triage

Summary: When creating a DigitalOcean cluster, the UI displays a success notification growl for each DigitalOceanConfig resource created with an autogenerated name. These notifications are unnecessary since they relate to internal sub-resources, not the cluster itself.

🔍 Root Cause Analysis

The growl is triggered in shell/plugins/steve/steve-class.js → processSaveResponse(), which fires a growl/success notification whenever:

1. A resource is created (HTTP 201)
2. The resource uses metadata.generateName
3. opt.suppressSuccessToast is not set

During cluster provisioning (shell/edit/provisioning.cattle.io.cluster/rke2.vue), saveMachinePools() creates machine config resources (e.g. DigitalOceanConfig) with generateName = "nc-<prefix>-" and calls entry.config.save() without the suppressSuccessToast option — hence the unwanted growls.

Relevant code:

• shell/plugins/steve/steve-class.js L73–90: processSaveResponse logic
• shell/edit/provisioning.cattle.io.cluster/rke2.vue L1511–1524: saveMachinePools where configs are saved

🛠️ Suggested Fix

Pass { suppressSuccessToast: true } when saving machine config resources in saveMachinePools:

// rke2.vue ~L1515
const neu = await entry.config.save({ suppressSuccessToast: true });

// rke2.vue ~L1524
entry.config = await entry.config.save({ suppressSuccessToast: true });

This is the same approach already used in shell/edit/management.cattle.io.user.vue:151.

⚠️ Consider whether this fix should apply only to machine configs (per-provider config resources) or more broadly. The suppressSuccessToast option is the intended mechanism — see steve-class.js L77.

🔁 Reproduction Steps

1. Open a Rancher instance with DigitalOcean node driver enabled.
2. Navigate to Cluster Management → Create → RKE2 → DigitalOcean.
3. Fill in required node pool configuration.
4. Click Create.
5. Observe: a notification growl appears for each DigitalOceanConfig created (one per node pool).
6. Expected: no such growl should appear for internal sub-resources.

✅ Sub-tasks Checklist

• Pass suppressSuccessToast: true to entry.config.save() in saveMachinePools (both entry.create and entry.update branches in rke2.vue)
• Check if other providers (AWS, Azure, vSphere, etc.) follow the same code path and are affected
• Verify Notification Center entries are also handled appropriately
• Add/update unit tests for processSaveResponse to cover suppressSuccessToast behavior
• Manually test cluster creation with DigitalOcean to confirm no growl appears

📚 Relevant Files & Resources

• shell/plugins/steve/steve-class.js — processSaveResponse method (L67–91)
• shell/edit/provisioning.cattle.io.cluster/rke2.vue — saveMachinePools (L1445–1534)
• shell/edit/management.cattle.io.user.vue — existing usage of suppressSuccessToast: true (L151)
• shell/store/growl.js — growl store
• shell/components/GrowlManager.vue — growl UI component

Generated by Agentic Triage for issue #17399 · ◷

[github-actions]

🤖 Issue grooming questions

Thanks for the detailed bug report and the video evidence — the problem is clear. A couple of things are needed before this can move to triage:

1. Reproduction steps — The current steps say "Add required info," which is a bit vague. Could you expand on the preconditions and exact steps needed to reproduce? For example:

   • Are valid DigitalOcean cloud credentials already configured in Rancher?
   • Are there specific node pool settings or other options that need to be filled in?
   • Is this reproducible on every DigitalOcean cluster creation attempt, or only under certain configurations?

2. Technical scope — To help engineers scope the fix, could you identify any specific UI components or areas of the codebase related to this behavior? For example, is the notification growl triggered by the same component that handles other resource-creation notifications, or is there something specific to the DigitalOceanConfig resource type?

This comment was generated automatically to help gather the information needed to groom this issue.

Note

🔒 Integrity filter blocked 3 items

The following items were blocked because they don't meet the GitHub integrity level.

• diff() nulls parent objects when only nested keys exist #16414 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• UI Support for New Global Roles Fields #9728 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".
• Cannot create new cloud credentials for custom node drivers #8523 list_issues: has lower integrity than agent requires. The agent cannot read data with integrity below "approved".

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

Generated by Daily Issue Grooming · ● 405.6K · ◷

[rak-phillip]

@momesgin Are we sure that this is an issue - the behavior looks intentional:

• [16688] improve notification on generate name resource creation #16723
• Improve notification on generate-name resource creation #16688
• [12961] Added the AuditPolicy CRUD | Added growl for autogenerated name cases #16191

[momesgin]

@rak-phillip thanks for the info 🙏
cc @nwmac