ROX-33939: virtual machine v2 reconciliation improvements#19904
Open
dashrews78 wants to merge 2 commits intomasterfrom
Open
ROX-33939: virtual machine v2 reconciliation improvements#19904dashrews78 wants to merge 2 commits intomasterfrom
dashrews78 wants to merge 2 commits intomasterfrom
Conversation
…ciliation Reconciliation only needs IDs, not full VM objects. Use Search() + ResultsToIDSet() — the same pattern all other pipelines use — to avoid deserializing full rows and the manual ID-extraction loop. Partially generated by AI. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
Skipping CI for Draft Pull Request. |
Contributor
🚀 Build Images ReadyImages are ready for commit 1fd38ea. To use with deploy scripts: export MAIN_IMAGE_TAG=4.11.x-603-g1fd38ea3eb |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #19904 +/- ##
==========================================
- Coverage 49.60% 49.60% -0.01%
==========================================
Files 2766 2766
Lines 208567 208562 -5
==========================================
- Hits 103454 103448 -6
- Misses 97436 97437 +1
Partials 7677 7677
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Contributor
📝 WalkthroughSummary by CodeRabbit
WalkthroughThe reconciliation logic in the VM pipeline is refactored to use the Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes 🚥 Pre-merge checks | ✅ 1 | ❌ 2❌ Failed checks (1 warning, 1 inconclusive)
✅ Passed checks (1 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@central/sensor/service/pipeline/virtualmachines/pipeline.go`:
- Around line 89-90: reconcileV2 is calling p.virtualMachineV2Store.Search with
a query built by search.NewQueryBuilder().AddExactMatches(...).ProtoQuery()
which relies on the default 100-result page size and thus misses VMs past the
first page; modify reconcileV2 to set explicit pagination on the query (either a
sufficiently large PageSize or, preferably, implement paging by setting/reading
PageToken and looping calls to p.virtualMachineV2Store.Search until no more
results) so you collect the complete set of VMs for the cluster before
reconciling/cleaning up.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Central YAML (base), Organization UI (inherited)
Review profile: CHILL
Plan: Pro
Run ID: e7c9e062-aafa-4d2d-8080-83bc2c387d41
📒 Files selected for processing (2)
central/sensor/service/pipeline/virtualmachines/pipeline.gocentral/sensor/service/pipeline/virtualmachines/pipeline_test.go
Comment on lines
89
to
+90
| query := search.NewQueryBuilder().AddExactMatches(search.ClusterID, clusterID).ProtoQuery() | ||
| query.Pagination = &v1.QueryPagination{Limit: math.MaxInt32} | ||
| virtualMachines, err := p.virtualMachineV2Store.SearchRawVirtualMachines(ctx, query) | ||
| results, err := p.virtualMachineV2Store.Search(ctx, query) |
Contributor
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Locate v2 datastore interface/impl Search signatures.
rg -nP --type=go '\btype\s+DataStore\s+interface\b|\bSearch\s*\(\s*ctx\s+context\.Context,\s*q\s+\*v1\.Query\s*\)\s*\(\s*\[\]search\.Result,\s*error\s*\)' -C3
rg -nP --type=go '\bfunc\s+\(.*\)\s+Search\s*\(\s*ctx\s+context\.Context,\s*q\s+\*v1\.Query\s*\)\s*\(\s*\[\]search\.Result,\s*error\s*\)' -C8
# Find pagination normalization/defaulting logic in search/datastore layers.
rg -nP --type=go '\bQueryPagination\b|\bPagination\b|default.*limit|fill.*pagination|normalize.*pagination|MaxInt32' -C3
# Compare other reconciliation code paths using Search + ResultsToIDSet.
rg -nP --type=go '\bResultsToIDSet\s*\(' -C3Repository: stackrox/stackrox
Length of output: 50374
🏁 Script executed:
# Find the applyQueryDefaults implementation to understand pagination defaults
rg -nA 15 'func.*applyQueryDefaults' --type=go
# Check if there's a default pagination limit applied anywhere
rg -n 'QueryPagination|DefaultLimit|default.*limit' --type=go central/virtualmachine/v2/datastore/store/postgres/store.go
# Check the VM v2 store Search implementation in detail
rg -nA 20 'func (s \*storeImpl) Search\(ctx context.Context, q \*v1.Query\)' central/virtualmachine/v2/datastore/store/postgres/store.goRepository: stackrox/stackrox
Length of output: 883
🏁 Script executed:
# Search for pagination handling in the query execution layer
rg -n 'FillDefaultSortOption|RunQueryRequest|GetQueryPagination|default.*pagination|Limit.*0|Limit.*-1' --type=go pkg/search/postgres/ | head -40
# Check the Search implementation in VM store more carefully
cat central/virtualmachine/v2/datastore/store/postgres/store.go | sed -n '669,720p'
# Search for where Pagination limits might be applied by default
rg -n 'NewQueryRequestWithPagination|pagination.*limit|limit.*pagination' --type=go pkg/search/ -A 3 | head -50Repository: stackrox/stackrox
Length of output: 4415
🏁 Script executed:
# Check the reconcileV1 implementation to see the explicit pagination pattern
rg -nB 5 -A 10 'func.*reconcileV1' central/sensor/service/pipeline/virtualmachines/pipeline.go
# Check what reconcileV2 currently looks like around line 89-90
rg -nB 3 -A 10 'func.*reconcileV2' central/sensor/service/pipeline/virtualmachines/pipeline.goRepository: stackrox/stackrox
Length of output: 1438
Set explicit pagination for Search in reconciliation.
reconcileV2 needs a complete cluster VM ID snapshot for accurate reconciliation. Without explicit pagination, Search defaults to a 100-result limit, causing VMs beyond the first page to be missed and never reconciled or cleaned up.
Proposed fix
func (p *pipelineImpl) reconcileV2(ctx context.Context, clusterID string, storeMap *reconciliation.StoreMap) error {
query := search.NewQueryBuilder().AddExactMatches(search.ClusterID, clusterID).ProtoQuery()
+ query.Pagination = &v1.QueryPagination{Limit: math.MaxInt32}
results, err := p.virtualMachineV2Store.Search(ctx, query)
if err != nil {
return errors.Wrap(err, "retrieving v2 virtual machines for reconciliation")
}📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| query := search.NewQueryBuilder().AddExactMatches(search.ClusterID, clusterID).ProtoQuery() | |
| query.Pagination = &v1.QueryPagination{Limit: math.MaxInt32} | |
| virtualMachines, err := p.virtualMachineV2Store.SearchRawVirtualMachines(ctx, query) | |
| results, err := p.virtualMachineV2Store.Search(ctx, query) | |
| query := search.NewQueryBuilder().AddExactMatches(search.ClusterID, clusterID).ProtoQuery() | |
| query.Pagination = &v1.QueryPagination{Limit: math.MaxInt32} | |
| results, err := p.virtualMachineV2Store.Search(ctx, query) |
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@central/sensor/service/pipeline/virtualmachines/pipeline.go` around lines 89
- 90, reconcileV2 is calling p.virtualMachineV2Store.Search with a query built
by search.NewQueryBuilder().AddExactMatches(...).ProtoQuery() which relies on
the default 100-result page size and thus misses VMs past the first page; modify
reconcileV2 to set explicit pagination on the query (either a sufficiently large
PageSize or, preferably, implement paging by setting/reading PageToken and
looping calls to p.virtualMachineV2Store.Search until no more results) so you
collect the complete set of VMs for the cluster before reconciling/cleaning up.
ajheflin
approved these changes
Apr 9, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Reconciliation only needs IDs, not full VM objects. Use Search() +
ResultsToIDSet() — the same pattern all other pipelines use — to
avoid deserializing full rows and the manual ID-extraction loop.
Partially generated by AI.
User-facing documentation
Testing and quality
Automated testing
How I validated my change
change me!