Skip to content

Create more test bucket to debug high cpu issues #31377

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
xueqzhan wants to merge 1 commit into
base: main
Choose a base branch
from
+57 −1
Open

Create more test bucket to debug high cpu issues #31377

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
58 changes: 57 additions & 1 deletion pkg/test/ginkgo/cmd_runsuite.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -583,6 +583,18 @@ func (o *GinkgoRunSuiteOptions) Run(suite *TestSuite, clusterConfig *clusterdisc
return strings.Contains(t.name, "OrderedNamespaceDeletion")
})

draTests, kubeTests := splitTests(kubeTests, func(t *testCase) bool {
return strings.Contains(t.name, "[DRA]")
})

inPlaceResizeTests, kubeTests := splitTests(kubeTests, func(t *testCase) bool {
return strings.Contains(t.name, "InPlace Resize")
})

probingTests, kubeTests := splitTests(kubeTests, func(t *testCase) bool {
return strings.Contains(t.name, "Probing container") || strings.Contains(t.name, "Probing restartable init container")
})

networkTests, openshiftTests := splitTests(openshiftTests, func(t *testCase) bool {
return strings.Contains(t.name, "[sig-network]")
})
Expand All @@ -600,6 +612,10 @@ func (o *GinkgoRunSuiteOptions) Run(suite *TestSuite, clusterConfig *clusterdisc
return strings.Contains(t.name, "[sig-cli] oc adm must-gather")
})

deploymentConfigTests, openshiftTests := splitTests(openshiftTests, func(t *testCase) bool {
return strings.Contains(t.name, "[Feature:DeploymentConfig]")
})

logrus.Infof("Found %d openshift tests", len(openshiftTests))
logrus.Infof("Found %d kubernetes tests", len(kubeTests))
logrus.Infof("Found %d storage tests", len(storageTests))
Expand All @@ -609,6 +625,10 @@ func (o *GinkgoRunSuiteOptions) Run(suite *TestSuite, clusterConfig *clusterdisc
logrus.Infof("Found %d netpol tests", len(netpolTests))
logrus.Infof("Found %d builds tests", len(buildsTests))
logrus.Infof("Found %d must-gather tests", len(mustGatherTests))
logrus.Infof("Found %d DeploymentConfig tests", len(deploymentConfigTests))
logrus.Infof("Found %d DRA tests", len(draTests))
logrus.Infof("Found %d InPlace Resize tests", len(inPlaceResizeTests))
logrus.Infof("Found %d Probing tests", len(probingTests))

// If user specifies a count, duplicate the kube and openshift tests that many times.
expectedTestCount := len(early) + len(late)
Expand All @@ -622,6 +642,10 @@ func (o *GinkgoRunSuiteOptions) Run(suite *TestSuite, clusterConfig *clusterdisc
originalNetpol := netpolTests
originalBuilds := buildsTests
originalMustGather := mustGatherTests
originalDRA := draTests
originalInPlaceResize := inPlaceResizeTests
originalProbing := probingTests
originalDeploymentConfig := deploymentConfigTests

for i := 1; i < count; i++ {
kubeTests = append(kubeTests, copyTests(originalKube)...)
Expand All @@ -633,9 +657,13 @@ func (o *GinkgoRunSuiteOptions) Run(suite *TestSuite, clusterConfig *clusterdisc
netpolTests = append(netpolTests, copyTests(originalNetpol)...)
buildsTests = append(buildsTests, copyTests(originalBuilds)...)
mustGatherTests = append(mustGatherTests, copyTests(originalMustGather)...)
draTests = append(draTests, copyTests(originalDRA)...)
inPlaceResizeTests = append(inPlaceResizeTests, copyTests(originalInPlaceResize)...)
probingTests = append(probingTests, copyTests(originalProbing)...)
deploymentConfigTests = append(deploymentConfigTests, copyTests(originalDeploymentConfig)...)
}
}
expectedTestCount += len(openshiftTests) + len(kubeTests) + len(storageTests) + len(networkK8sTests) + len(orderedNamespaceDeletionTests) + len(networkTests) + len(netpolTests) + len(buildsTests) + len(mustGatherTests)
expectedTestCount += len(openshiftTests) + len(kubeTests) + len(storageTests) + len(networkK8sTests) + len(orderedNamespaceDeletionTests) + len(networkTests) + len(netpolTests) + len(buildsTests) + len(mustGatherTests) + len(draTests) + len(inPlaceResizeTests) + len(probingTests) + len(deploymentConfigTests)

abortFn := neverAbort
testCtx := ctx
Expand Down Expand Up @@ -667,6 +695,27 @@ func (o *GinkgoRunSuiteOptions) Run(suite *TestSuite, clusterConfig *clusterdisc
logrus.Infof("Completed Kubernetes test bucket in %v", time.Since(kubeStartTime))
tests = append(tests, kubeTestsCopy...)

draTestsCopy := copyTests(draTests)
draIntervalID, draStartTime := recordTestBucketInterval(monitorEventRecorder, "DRA")
q.Execute(testCtx, draTestsCopy, parallelism, testOutputConfig, abortFn)
monitorEventRecorder.EndInterval(draIntervalID, time.Now())
logrus.Infof("Completed DRA test bucket in %v", time.Since(draStartTime))
tests = append(tests, draTestsCopy...)

inPlaceResizeTestsCopy := copyTests(inPlaceResizeTests)
inPlaceResizeIntervalID, inPlaceResizeStartTime := recordTestBucketInterval(monitorEventRecorder, "InPlaceResize")
q.Execute(testCtx, inPlaceResizeTestsCopy, parallelism, testOutputConfig, abortFn)
monitorEventRecorder.EndInterval(inPlaceResizeIntervalID, time.Now())
logrus.Infof("Completed InPlaceResize test bucket in %v", time.Since(inPlaceResizeStartTime))
tests = append(tests, inPlaceResizeTestsCopy...)

probingTestsCopy := copyTests(probingTests)
probingIntervalID, probingStartTime := recordTestBucketInterval(monitorEventRecorder, "Probing")
q.Execute(testCtx, probingTestsCopy, parallelism, testOutputConfig, abortFn)
monitorEventRecorder.EndInterval(probingIntervalID, time.Now())
logrus.Infof("Completed Probing test bucket in %v", time.Since(probingStartTime))
tests = append(tests, probingTestsCopy...)

// I thought about randomizing the order of the kube, storage, and openshift tests, but storage dominates our e2e runs, so it doesn't help much.
storageTestsCopy := copyTests(storageTests)
storageIntervalID, storageStartTime := recordTestBucketInterval(monitorEventRecorder, "Storage")
Expand Down Expand Up @@ -721,6 +770,13 @@ func (o *GinkgoRunSuiteOptions) Run(suite *TestSuite, clusterConfig *clusterdisc
logrus.Infof("Completed OpenShift test bucket in %v", time.Since(openshiftStartTime))
tests = append(tests, openshiftTestsCopy...)

deploymentConfigTestsCopy := copyTests(deploymentConfigTests)
deploymentConfigIntervalID, deploymentConfigStartTime := recordTestBucketInterval(monitorEventRecorder, "DeploymentConfig")
q.Execute(testCtx, deploymentConfigTestsCopy, parallelism, testOutputConfig, abortFn)
monitorEventRecorder.EndInterval(deploymentConfigIntervalID, time.Now())
logrus.Infof("Completed DeploymentConfig test bucket in %v", time.Since(deploymentConfigStartTime))
tests = append(tests, deploymentConfigTestsCopy...)

// run the must-gather tests after parallel tests to reduce resource contention
mustGatherTestsCopy := copyTests(mustGatherTests)
mustGatherIntervalID, mustGatherStartTime := recordTestBucketInterval(monitorEventRecorder, "MustGather")
Expand Down