redhat-performance / cloud-governance / 29923280304
73%

DEFAULT BRANCH: main
Ran
Jobs 0
Files 0
Run time –
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
29923280304

push

travis-pro

web-flow 
Fix zombie_cluster_resource false positive detection for CAPA clusters (#1013)

* Fix zombie_cluster_resource false positive detection for CAPA clusters

The policy incorrectly detected active OpenShift cluster security groups
as zombie resources when instances and security groups carried different
cluster tag prefixes (CCM vs CAPA).

Root cause: zombie detection compared full tag key strings, so
`kubernetes.io/cluster/X` and `sigs.k8s.io/cluster-api-provider-aws/cluster/X`
were treated as different clusters, causing CAPA-provisioned security groups
to be falsely classified as orphaned.

A secondary bug in Utils.is_cluster_resource caused non-deterministic
visibility of CAPA resources due to variable shadowing in the inner for-loop.

Fixes:
- F1: Extract cluster name from tag key before comparison so resources
  tagged with different cluster prefixes resolve to the same cluster.
  Applied to __get_zombie_resources and __get_all_zombie_resources.
- F2: Add VPC instance check as a tag-agnostic safety net. Resources in
  a VPC with running instances are excluded from zombie detection.
  Applied to all VPC-bound resource types via _filter_zombies_by_vpc.
- F3: Fix __get_vpc_tags to use prefix-aware name matching instead of
  substring match, preventing cross-cluster false positives.
- F11: Remove else clause in __get_cluster_resources that allowed
  resource_name to overwrite the cluster ID with an unrelated tag key.
- F12: Fix elastic IP zombie detection — merge both association and
  allocation zombies and use the correct resource_id_name (AssociationId).
- F13: Fix VPC endpoint loop variable (zombie_ids = [zombie] not [zombies]).
- is_cluster_resource: Fix variable shadowing where the for-loop variable
  shadowed the parameter, making CAPA resource detection non-deterministic
  based on AWS API tag ordering.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Move zombie_cluster_resource to dry_run=yes for perf accounts

While the detection f... (continued)
Source Files on build 29923280304
Detailed source file information is not available for this build.