Dependancy upgrades

[Miles-Garnsey]

What this PR does:

Upgrades controller-runtime

Which issue(s) this PR fixes:

Expanding this due to the presence of more bugs than we originally realised.

Fixes #1313 #1317 #1316

Checklist

• Changes manually tested
• Automated Tests added/updated
• Documentation added/updated
• CHANGELOG.md updated (not required for documentation PRs)
• CLA Signed: DataStax CLA

[codecov]

Codecov Report

Attention: Patch coverage is 76.66667% with 21 lines in your changes are missing coverage. Please review.

Project coverage is 58.53%. Comparing base (1864ddb) to head (edcfbf7).
Report is 27 commits behind head on main.

❗ Current head edcfbf7 differs from pull request most recent head d813636

Please upload reports for the commit d813636 to get more accurate results.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1314      +/-   ##
==========================================
+ Coverage   57.43%   58.53%   +1.09%     
==========================================
  Files         103      122      +19     
  Lines       10808    11837    +1029     
==========================================
+ Hits         6208     6929     +721     
- Misses       4063     4301     +238     
- Partials      537      607      +70     

Files	Coverage Δ
controllers/config/clientconfig_controller.go	86.86% <100.00%> (+3.64%)	⬆️
controllers/control/k8ssandratask_controller.go	65.62% <100.00%> (+1.62%)	⬆️
...ntrollers/k8ssandra/k8ssandracluster_controller.go	77.58% <100.00%> (+2.58%)	⬆️
pkg/medusa/reconcile.go	69.87% <100.00%> (+8.92%)	⬆️
pkg/test/test_objects.go	100.00% <100.00%> (ø)
pkg/test/testenv.go	58.27% <100.00%> (ø)
...trollers/medusa/medusabackupschedule_controller.go	9.90% <0.00%> (-61.10%)	⬇️
...is/medusa/v1alpha1/medusabackupschedule_webhook.go	87.50% <66.66%> (ø)
controllers/reaper/reaper_controller.go	52.68% <0.00%> (ø)
controllers/replication/secret_controller.go	67.73% <62.50%> (+2.71%)	⬆️
... and 2 more

... and 22 files with indirect coverage changes

[Miles-Garnsey]

Finally figured out the issues plaguing the e2e tests. I think I'm hitting this: kubernetes-sigs/controller-runtime#2628

EDIT: maybe not. There's an error on the K8ssandraCluster object unable to list: test-1 because of unknown namespace for the cache when I run (for example) make E2E_TEST="TestOperator/ClusterScoped/MultiDcMultiCluster" TEST_ARGS="-test.timeout 99999s" kind-multi-e2e-test. This is pretty deeply unhelpful, because there are no errors in the logs, and so I have no idea which call to .List() is actually failing, or - more importantly - which client doesn't have a namespace set.

This ticket says that leaving DefaultNamespaces unset on the manager.options.NewCache field should allow the operator to watch all namespaces. And my code here is identical to what is in cass-operator here.

So I'm thinking this must be some problem with one of the other clients, maybe on the data planes. @burmanm if you have any ideas I'd be grateful to hear them.

[Miles-Garnsey]

I've narrowed this down a bit more, I think I'm seeing this error from here and here at least.

What appears to happen is that when .List() is called, it strikes this line because the namespace specified in the ListOption does not exist within the list of namespaces being cached by this client. This is pretty wild, since I can't see any reason why this should actually be failing given that this cache should be watching all namespaces (based on the way the remote clients get constructed here).

So it looks to me as though the List() function in client-go might actually be broken, since a caching client which watches all namespaces doesn't seem to be able to List() resources across a single namespace due to it never being in the cache. We'd have to work around this by constructing our own List function that Lists() from all namespaces then filters based on the one we actually wanted.

Example of the type of error I'm seeing:

2024-05-20T05:46:56.079Z	ERROR	Failed to get seed pods	{"controller": "k8ssandracluster", "controllerGroup": "k8ssandra.io", "controllerKind": "K8ssandraCluster", "K8ssandraCluster": {"name":"test","namespace":"test-0"}, "namespace": "test-0", "name": "test", "reconcileID": "9002b761-90a2-4e37-bb7f-3c55a97e3ce1", "K8ssandraCluster": {"name":"test","namespace":"test-0"}, "K8sContext": "kind-k8ssandra-0", "DC": {"name":"dc1","namespace":"test-1"}, "error": "unable to list: test-1 because of unknown namespace for the cache"}
github.com/k8ssandra/k8ssandra-operator/controllers/k8ssandra.(*K8ssandraClusterReconciler).findSeeds
	/workspace/controllers/k8ssandra/seeds.go:45
github.com/k8ssandra/k8ssandra-operator/controllers/k8ssandra.(*K8ssandraClusterReconciler).reconcileDatacenters
	/workspace/controllers/k8ssandra/datacenters.go:55
github.com/k8ssandra/k8ssandra-operator/controllers/k8ssandra.(*K8ssandraClusterReconciler).reconcile
	/workspace/controllers/k8ssandra/k8ssandracluster_controller.go:145
github.com/k8ssandra/k8ssandra-operator/controllers/k8ssandra.(*K8ssandraClusterReconciler).Reconcile
	/workspace/controllers/k8ssandra/k8ssandracluster_controller.go:93
sigs.k8s.io/controller-runtime/pkg/internal/controller.(*Controller).Reconcile
	/go/pkg/mod/sigs.k8s.io/[email protected]/pkg/internal/controller/controller.go:119
sigs.k8s.io/controller-runtime/pkg/internal/controller.(*Controller).reconcileHandler
	/go/pkg/mod/sigs.k8s.io/[email protected]/pkg/internal/controller/controller.go:316
sigs.k8s.io/controller-runtime/pkg/internal/controller.(*Controller).processNextWorkItem
	/go/pkg/mod/sigs.k8s.io/[email protected]/pkg/internal/controller/controller.go:266
sigs.k8s.io/controller-runtime/pkg/internal/controller.(*Controller).Start.func2.2
	/go/pkg/mod/sigs.k8s.io/[email protected]/pkg/internal/controller/controller.go:227

[sonarqubecloud]

Quality Gate passed

Issues
28 New issues
0 Accepted issues

Measures
0 Security Hotspots
No data about Coverage
1.9% Duplication on New Code

See analysis details on SonarCloud