Clean up primary-only snapshot DB records on volume expunge by Damans227 · Pull Request #12813 · apache/cloudstack

Damans227 · 2026-03-13T18:54:32Z

Description

When snapshot.backup.to.secondary=false (KVM + Ceph) and a VM is expunged, Ceph destroys the RBD snapshots along with the volume image, but the DB records (snapshots, snapshot_store_ref) are left behind as undeletable orphans.

Fix

In StorageManagerImpl.cleanupStorage(), clean up primary-only snapshot records before the volume is expunged from storage.

Fixes: #12002

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

Screenshots (if appropriate):

How Has This Been Tested?

Tested on KVM + Ceph (RBD) with snapshot.backup.to.secondary=false:

Deployed VM on Ceph, took volume snapshot, destroyed+expunged VM
Verified snapshot DB records cleaned up after storage scavenger cycle
Confirmed NFS snapshots (with secondary copies) are unaffected
Unit tests cover: primary-only cleanup, skip when secondary exists, skip destroyed snapshots

How did you try to break this feature and the system with this change?

Tested with snapshots having both primary and secondary refs, and with already-destroyed snapshots, both worked correctly skipped.

…revent orphaned entries

codecov · 2026-03-13T19:03:59Z

Codecov Report

❌ Patch coverage is 79.31034% with 6 lines in your changes missing coverage. Please review.
✅ Project coverage is 17.68%. Comparing base (7048944) to head (28548b2).
⚠️ Report is 284 commits behind head on 4.22.

Files with missing lines	Patch %	Lines
...ain/java/com/cloud/storage/StorageManagerImpl.java	79.31%	3 Missing and 3 partials ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##               4.22   #12813      +/-   ##
============================================
+ Coverage     17.61%   17.68%   +0.07%     
- Complexity    15662    15795     +133     
============================================
  Files          5917     5922       +5     
  Lines        531415   533149    +1734     
  Branches      64973    65208     +235     
============================================
+ Hits          93588    94283     +695     
- Misses       427271   428214     +943     
- Partials      10556    10652      +96

Flag	Coverage Δ
uitests	`3.69% <ø> (-0.01%)`	⬇️
unittests	`18.76% <79.31%> (+0.07%)`	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

DaanHoogland

clgtm, but a concern remains (might be mute): Are we sure, when marking the snapshots records, there are really no data files left on primary for those?
I do not see any code ensuring that.

GaOrtiga

lgtm

Copilot

Pull request overview

This PR addresses orphaned primary-only snapshot database records during volume expunge, especially for KVM + Ceph/RBD when snapshots are not backed up to secondary storage.

Changes:

Adds snapshot cleanup before volume expunge in StorageManagerImpl.cleanupStorage().
Introduces helper logic to expunge primary-only snapshot store refs and mark snapshots destroyed.
Adds unit tests for cleanup, secondary-copy skip, and destroyed-snapshot skip cases.

Reviewed changes

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

File	Description
`server/src/main/java/com/cloud/storage/StorageManagerImpl.java`	Adds primary-only snapshot record cleanup during storage garbage collection.
`server/src/test/java/com/cloud/storage/StorageManagerImplTest.java`	Adds unit tests covering the new cleanup helper behavior.

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

+                snapshot.setState(Snapshot.State.Destroyed);
+                _snapshotDao.update(snapshot.getId(), snapshot);


Damans227 · 2026-05-20T01:56:16Z

clgtm, but a concern remains (might be mute): Are we sure, when marking the snapshots records, there are really no data files left on primary for those? I do not see any code ensuring that.

Good catch @DaanHoogland fixed. Cleanup now routes through _snapshotService.deleteSnapshot(...) so the storage driver actually removes the data, and the parent snapshots row only flips to Destroyed after confirming no live store refs remain.

…nd improve error handling

… limits, and removing annotations

Damans227 · 2026-05-20T02:07:10Z

@blueorangutan package

blueorangutan · 2026-05-20T02:08:05Z

@Damans227 a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress.

blueorangutan · 2026-05-20T03:05:48Z

Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ el10 ✔️ debian ✔️ suse15. SL-JID 17917

DaanHoogland · 2026-05-20T14:30:48Z

@blueorangutan test

blueorangutan · 2026-05-20T14:32:04Z

@DaanHoogland a [SL] Trillian-Jenkins test job (ol8 mgmt + kvm-ol8) has been kicked to run smoke tests

blueorangutan · 2026-05-20T15:00:49Z

[SF] Trillian Build Failed (tid-16129)

vladimirpetrov · 2026-08-19T13:07:43Z

@blueorangutan test

blueorangutan · 2026-08-19T13:10:03Z

@vladimirpetrov a [SL] Trillian-Jenkins test job (ol8 mgmt + kvm-ol8) has been kicked to run smoke tests

blueorangutan · 2026-08-20T04:44:35Z

[SF] Trillian test result (tid-16783)
Environment: kvm-ol8 (x2), zone: Advanced Networking with Mgmt server ol8
Total time taken: 52714 seconds
Marvin logs: https://github.com/blueorangutan/acs-prs/releases/download/trillian/pr12813-t16783-kvm-ol8.zip
Smoke tests completed. 147 look OK, 2 have errors, 0 did not run
Only failed and skipped tests results shown below:

Test	Result	Time (s)	Test File
test_02_enableHumanReadableLogs	`Error`	0.35	test_human_readable_logs.py
test_01_ssl_offloading_isolated_network	`Failure`	716.09	test_ssl_offloading.py
test_02_ssl_offloading_project_vpc	`Failure`	692.53	test_ssl_offloading.py

weizhouapache · 2026-09-08T10:49:48Z

@Damans227

I confirmed that a primary-only snapshot (KVM+Ceph, snapshot.backup.to.secondary=false) is now cleaned up in-band. snapshot marked Destroyed, store-ref marked Destroyed, resource count decremented, RBD snapshot removed, before the volume itself is expunged. Before the fix, this same snapshot was permanently orphaned (BackedUp/Ready forever, deleteSnapshot failing).

However, the method sits in the generic StorageManagerImpl.cleanupStorage(), which runs for every volume being expunged regardless of storage type, it now runs unconditionally on every volume-expunge pass rather than being gated by pool type.

Could you please verify whether snapshots on primary storage are working for volumes on NFS, with snapshot.backup.to.secondary set to false? When the VM or volume has been expunged, can we still create a template, volume, or VM from its snapshot?

Damans227 added 4 commits March 13, 2026 13:43

Clean up primary-only snapshot DB records during volume deletion to p…

e580af8

…revent orphaned entries

Clean up primary-only snapshot in StorageManagerImpl scavenger instead

77515d1

cleanup

3935803

cleanup

79c00ba

boring-cyborg Bot added the component:storage label Mar 13, 2026

Damans227 added the component:ceph label Mar 13, 2026

sureshanaparti added this to the 4.22.1 milestone Mar 14, 2026

sureshanaparti added this to Apache CloudStack 4.22.1 Mar 14, 2026

github-project-automation Bot moved this to Todo in Apache CloudStack 4.22.1 Mar 14, 2026

sureshanaparti moved this from Todo to In Progress in Apache CloudStack 4.22.1 Mar 14, 2026

sureshanaparti linked an issue Mar 14, 2026 that may be closed by this pull request

snapshot as Abnormal residue on rbd #12002

Open

DaanHoogland reviewed Mar 16, 2026

View reviewed changes

GaOrtiga approved these changes Apr 16, 2026

View reviewed changes

sureshanaparti removed this from Apache CloudStack 4.22.1 Apr 24, 2026

sureshanaparti modified the milestones: 4.22.1, 4.22.2 Apr 24, 2026

vishesh92 requested a review from Copilot May 18, 2026 07:40

Copilot started reviewing on behalf of vishesh92 May 18, 2026 07:41 View session

Copilot AI reviewed May 18, 2026

View reviewed changes

Comment thread server/src/main/java/com/cloud/storage/StorageManagerImpl.java

Comment on lines +2318 to +2319

snapshot.setState(Snapshot.State.Destroyed);

_snapshotDao.update(snapshot.getId(), snapshot);

Comment thread server/src/main/java/com/cloud/storage/StorageManagerImpl.java

Damans227 added 2 commits May 19, 2026 21:56

Refactor snapshot cleanup logic to use SnapshotService for deletion a…

66f1e64

…nd improve error handling

Enhance snapshot cleanup process by updating state, managing resource…

28548b2

… limits, and removing annotations

DaanHoogland added this to CloudStack Testing Aug 31, 2026

github-project-automation Bot moved this to Backlog in CloudStack Testing Aug 31, 2026

DaanHoogland moved this from Backlog to Ready in CloudStack Testing Sep 1, 2026

weizhouapache self-assigned this Sep 7, 2026

		snapshot.setState(Snapshot.State.Destroyed);
		_snapshotDao.update(snapshot.getId(), snapshot);

Uh oh!

Conversation

Damans227 commented Mar 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Fix

Types of changes

Feature/Enhancement Scale or Bug Severity

Feature/Enhancement Scale

Bug Severity

Screenshots (if appropriate):

How Has This Been Tested?

How did you try to break this feature and the system with this change?

Uh oh!

codecov Bot commented Mar 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

DaanHoogland left a comment

Choose a reason for hiding this comment

Uh oh!

GaOrtiga left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Damans227 commented May 20, 2026

Uh oh!

Damans227 commented May 20, 2026

Uh oh!

blueorangutan commented May 20, 2026

Uh oh!

blueorangutan commented May 20, 2026

Uh oh!

DaanHoogland commented May 20, 2026

Uh oh!

blueorangutan commented May 20, 2026

Uh oh!

blueorangutan commented May 20, 2026

Uh oh!

vladimirpetrov commented Aug 19, 2026

Uh oh!

blueorangutan commented Aug 19, 2026

Uh oh!

blueorangutan commented Aug 20, 2026

Uh oh!

weizhouapache commented Sep 8, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

Damans227 commented Mar 13, 2026 •

edited

Loading

codecov Bot commented Mar 13, 2026 •

edited

Loading