Skip to content
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.

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

Log test runner instance config in e2e #8007

Merged
merged 1 commit into from
Apr 18, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
112 changes: 60 additions & 52 deletions internal/test/e2e/run.go
Original file line number Diff line number Diff line change
Expand Up @@ -125,7 +125,7 @@ func RunTestsInParallel(conf ParallelRunConf) error {
for c := range work {
r := instanceTestsResults{conf: c}

r.conf.instanceId, r.testCommandResult, err = RunTests(c, invCatalogue)
r.conf.InstanceID, r.testCommandResult, err = RunTests(c, invCatalogue)
if err != nil {
r.err = err
}
Expand All @@ -150,22 +150,22 @@ func RunTestsInParallel(conf ParallelRunConf) error {
// Once the tool is updated to support the unified message, remove them
if r.err != nil {
result = testResultError
conf.Logger.Error(r.err, "Failed running e2e tests for instance", "jobId", r.conf.jobId, "instanceId", r.conf.instanceId, "tests", r.conf.regex, "status", testResultFail)
conf.Logger.Error(r.err, "Failed running e2e tests for instance", "jobId", r.conf.JobID, "instanceId", r.conf.InstanceID, "tests", r.conf.Regex, "status", testResultFail)
failedInstances++
} else if !r.testCommandResult.Successful() {
result = testResultFail
conf.Logger.Info("An e2e instance run has failed", "jobId", r.conf.jobId, "instanceId", r.conf.instanceId, "commandId", r.testCommandResult.CommandId, "tests", r.conf.regex, "status", testResultFail)
conf.Logger.Info("An e2e instance run has failed", "jobId", r.conf.JobID, "instanceId", r.conf.InstanceID, "commandId", r.testCommandResult.CommandId, "tests", r.conf.Regex, "status", testResultFail)
failedInstances++
} else {
result = testResultPass
conf.Logger.Info("Instance tests completed successfully", "jobId", r.conf.jobId, "instanceId", r.conf.instanceId, "commandId", r.testCommandResult.CommandId, "tests", r.conf.regex, "status", testResultPass)
conf.Logger.Info("Instance tests completed successfully", "jobId", r.conf.JobID, "instanceId", r.conf.InstanceID, "commandId", r.testCommandResult.CommandId, "tests", r.conf.Regex, "status", testResultPass)
}
completedInstances++
conf.Logger.Info("Instance tests run finished",
"result", result,
"tests", r.conf.regex,
"jobId", r.conf.jobId,
"instanceId", r.conf.instanceId,
"tests", r.conf.Regex,
"jobId", r.conf.JobID,
"instanceId", r.conf.InstanceID,
"completedInstances", completedInstances,
"totalInstances", totalInstances,
)
Expand All @@ -179,29 +179,35 @@ func RunTestsInParallel(conf ParallelRunConf) error {
}

type instanceRunConf struct {
session *session.Session
instanceProfileName, storageBucket, jobId, parentJobId, regex, instanceId string
testReportFolder, branchName string
ipPool networkutils.IPPool
hardware []*api.Hardware
hardwareCount int
tinkerbellAirgappedTest bool
bundlesOverride bool
testRunnerType TestRunnerType
testRunnerConfig TestInfraConfig
cleanupVms bool
logger logr.Logger
InstanceProfileName string
StorageBucket string
JobID string
ParentJobID string
Regex string
InstanceID string
TestReportFolder string
BranchName string
IPPool networkutils.IPPool
Hardware []*api.Hardware
HardwareCount int
TinkerbellAirgappedTest bool
BundlesOverride bool
TestRunnerType TestRunnerType
TestRunnerConfig TestInfraConfig
CleanupVMs bool
Logger logr.Logger
Session *session.Session
}

//nolint:gocyclo, revive // RunTests responsible launching test runner to run tests is complex.
func RunTests(conf instanceRunConf, inventoryCatalogue map[string]*hardwareCatalogue) (testInstanceID string, testCommandResult *testCommandResult, err error) {
testRunner, err := newTestRunner(conf.testRunnerType, conf.testRunnerConfig)
testRunner, err := newTestRunner(conf.TestRunnerType, conf.TestRunnerConfig)
if err != nil {
return "", nil, err
}
if conf.hardwareCount > 0 {
if conf.HardwareCount > 0 {
var hardwareCatalogue *hardwareCatalogue
if conf.tinkerbellAirgappedTest {
if conf.TinkerbellAirgappedTest {
hardwareCatalogue = inventoryCatalogue[airgappedHardware]
} else {
hardwareCatalogue = inventoryCatalogue[nonAirgappedHardware]
Expand All @@ -214,16 +220,18 @@ func RunTests(conf instanceRunConf, inventoryCatalogue map[string]*hardwareCatal
defer releaseTinkerbellHardware(&conf, hardwareCatalogue)
}

conf.Logger.Info("Creating runner instance", "cfg", conf)
instanceId, err := testRunner.createInstance(conf)
if err != nil {
return "", nil, err
}
conf.logger.V(1).Info("TestRunner instance has been created", "instanceId", instanceId)
conf.Logger = conf.Logger.WithValues("instance_id", instanceId)
conf.Logger.Info("TestRunner instance has been created")

defer func() {
err := testRunner.decommInstance(conf)
if err != nil {
conf.logger.V(1).Info("WARN: Failed to decomm e2e test runner instance", "error", err)
conf.Logger.V(1).Info("WARN: Failed to decomm e2e test runner instance", "error", err)
}
}()

Expand All @@ -232,12 +240,12 @@ func RunTests(conf instanceRunConf, inventoryCatalogue map[string]*hardwareCatal
return "", nil, err
}

err = session.setup(conf.regex)
err = session.setup(conf.Regex)
if err != nil {
return session.instanceId, nil, err
}

testCommandResult, err = session.runTests(conf.regex)
testCommandResult, err = session.runTests(conf.Regex)
if err != nil {
return session.instanceId, nil, err
}
Expand Down Expand Up @@ -288,13 +296,13 @@ func (e *E2ESession) runTests(regex string) (testCommandResult *testCommandResul
}

func (c instanceRunConf) runPostTestsProcessing(e *E2ESession, testCommandResult *testCommandResult) error {
regex := strings.Trim(c.regex, "\"")
regex := strings.Trim(c.Regex, "\"")
tests := strings.Split(regex, "|")

for _, testName := range tests {
e.uploadJUnitReportFromInstance(testName)
if c.testReportFolder != "" {
e.downloadJUnitReportToLocalDisk(testName, c.testReportFolder)
if c.TestReportFolder != "" {
e.downloadJUnitReportToLocalDisk(testName, c.TestReportFolder)
}

if !testCommandResult.Successful() {
Expand Down Expand Up @@ -486,30 +494,30 @@ func getTinkerbellTestsWithCount(tinkerbellTests []string, conf ParallelRunConf)
func newInstanceRunConf(awsSession *session.Session, conf ParallelRunConf, jobNumber int, testRegex string, ipPool networkutils.IPPool, hardware []*api.Hardware, hardwareCount int, tinkerbellAirgappedTest bool, testRunnerType TestRunnerType, testRunnerConfig *TestInfraConfig) instanceRunConf {
jobID := fmt.Sprintf("%s-%d", conf.JobId, jobNumber)
return instanceRunConf{
session: awsSession,
instanceProfileName: conf.InstanceProfileName,
storageBucket: conf.StorageBucket,
jobId: jobID,
parentJobId: conf.JobId,
regex: testRegex,
ipPool: ipPool,
hardware: hardware,
hardwareCount: hardwareCount,
tinkerbellAirgappedTest: tinkerbellAirgappedTest,
bundlesOverride: conf.BundlesOverride,
testReportFolder: conf.TestReportFolder,
branchName: conf.BranchName,
cleanupVms: conf.CleanupVms,
testRunnerType: testRunnerType,
testRunnerConfig: *testRunnerConfig,
logger: conf.Logger.WithValues("jobID", jobID, "test", testRegex),
Session: awsSession,
InstanceProfileName: conf.InstanceProfileName,
StorageBucket: conf.StorageBucket,
JobID: jobID,
ParentJobID: conf.JobId,
Regex: testRegex,
IPPool: ipPool,
Hardware: hardware,
HardwareCount: hardwareCount,
TinkerbellAirgappedTest: tinkerbellAirgappedTest,
BundlesOverride: conf.BundlesOverride,
TestReportFolder: conf.TestReportFolder,
BranchName: conf.BranchName,
CleanupVMs: conf.CleanupVms,
TestRunnerType: testRunnerType,
TestRunnerConfig: *testRunnerConfig,
Logger: conf.Logger.WithValues("jobID", jobID, "test", testRegex),
}
}

func logTestGroups(logger logr.Logger, instancesConf []instanceRunConf) {
testGroups := make([]string, 0, len(instancesConf))
for _, i := range instancesConf {
testGroups = append(testGroups, i.regex)
testGroups = append(testGroups, i.Regex)
}
logger.V(1).Info("Running tests in parallel", "testsGroups", testGroups)
}
Expand Down Expand Up @@ -551,24 +559,24 @@ func getAirgappedHardwarePool(storageBucket string) ([]*api.Hardware, error) {
}

func reserveTinkerbellHardware(conf *instanceRunConf, invCatalogue *hardwareCatalogue) error {
reservedTinkerbellHardware, err := invCatalogue.reserveHardware(conf.hardwareCount)
reservedTinkerbellHardware, err := invCatalogue.reserveHardware(conf.HardwareCount)
if err != nil {
return fmt.Errorf("timed out waiting for hardware")
}
conf.hardware = reservedTinkerbellHardware
conf.Hardware = reservedTinkerbellHardware
logTinkerbellTestHardwareInfo(conf, "Reserved")
return nil
}

func releaseTinkerbellHardware(conf *instanceRunConf, invCatalogue *hardwareCatalogue) {
logTinkerbellTestHardwareInfo(conf, "Releasing")
invCatalogue.releaseHardware(conf.hardware)
invCatalogue.releaseHardware(conf.Hardware)
}

func logTinkerbellTestHardwareInfo(conf *instanceRunConf, action string) {
var hardwareInfo []string
for _, hardware := range conf.hardware {
for _, hardware := range conf.Hardware {
hardwareInfo = append(hardwareInfo, hardware.Hostname)
}
conf.logger.V(1).Info(action+" hardware for TestRunner", "hardwarePool", strings.Join(hardwareInfo, ", "))
conf.Logger.V(1).Info(action+" hardware for TestRunner", "hardwarePool", strings.Join(hardwareInfo, ", "))
}
21 changes: 11 additions & 10 deletions internal/test/e2e/setup.go
Original file line number Diff line number Diff line change
Expand Up @@ -49,19 +49,19 @@ type E2ESession struct {

func newE2ESession(instanceId string, conf instanceRunConf) (*E2ESession, error) {
e := &E2ESession{
session: conf.session,
session: conf.Session,
instanceId: instanceId,
instanceProfileName: conf.instanceProfileName,
storageBucket: conf.storageBucket,
jobId: conf.jobId,
ipPool: conf.ipPool,
instanceProfileName: conf.InstanceProfileName,
storageBucket: conf.StorageBucket,
jobId: conf.JobID,
ipPool: conf.IPPool,
testEnvVars: make(map[string]string),
bundlesOverride: conf.bundlesOverride,
cleanupVms: conf.cleanupVms,
bundlesOverride: conf.BundlesOverride,
cleanupVms: conf.CleanupVMs,
requiredFiles: requiredFiles,
branchName: conf.branchName,
hardware: conf.hardware,
logger: conf.logger,
branchName: conf.BranchName,
hardware: conf.Hardware,
logger: conf.Logger,
}

return e, nil
Expand Down Expand Up @@ -179,6 +179,7 @@ func (e *E2ESession) setup(regex string) error {
}

ipPool := e.ipPool.ToString()

if ipPool != "" {
e.testEnvVars[e2etests.ClusterIPPoolEnvVar] = ipPool
}
Expand Down
32 changes: 16 additions & 16 deletions internal/test/e2e/testRunner.go
Original file line number Diff line number Diff line change
Expand Up @@ -132,10 +132,10 @@ func (v *VSphereTestRunner) setEnvironment() (map[string]string, error) {
}

func (v *VSphereTestRunner) createInstance(c instanceRunConf) (string, error) {
name := getTestRunnerName(v.logger, c.jobId)
name := getTestRunnerName(v.logger, c.JobID)
v.logger.V(1).Info("Creating vSphere Test Runner instance", "name", name)

ssmActivationInfo, err := ssm.CreateActivation(c.session, name, c.instanceProfileName)
ssmActivationInfo, err := ssm.CreateActivation(c.Session, name, c.InstanceProfileName)
if err != nil {
return "", fmt.Errorf("unable to create ssm activation: %v", err)
}
Expand All @@ -152,7 +152,7 @@ func (v *VSphereTestRunner) createInstance(c instanceRunConf) (string, error) {
PropertyMapping: []vsphere.OVFProperty{
{Key: ssmActivationCodeKey, Value: ssmActivationInfo.ActivationCode},
{Key: ssmActivationIdKey, Value: ssmActivationInfo.ActivationID},
{Key: ssmActivationRegionKey, Value: *c.session.Config.Region},
{Key: ssmActivationRegionKey, Value: *c.Session.Config.Region},
},
}

Expand All @@ -163,7 +163,7 @@ func (v *VSphereTestRunner) createInstance(c instanceRunConf) (string, error) {

var ssmInstance *aws_ssm.InstanceInformation
err = retrier.Retry(10, 5*time.Second, func() error {
ssmInstance, err = ssm.GetInstanceByActivationId(c.session, ssmActivationInfo.ActivationID)
ssmInstance, err = ssm.GetInstanceByActivationId(c.Session, ssmActivationInfo.ActivationID)
if err != nil {
return fmt.Errorf("failed to get ssm instance info post ovf deployment: %v", err)
}
Expand All @@ -180,19 +180,19 @@ func (v *VSphereTestRunner) createInstance(c instanceRunConf) (string, error) {
}

func (e *Ec2TestRunner) createInstance(c instanceRunConf) (string, error) {
name := getTestRunnerName(e.logger, c.jobId)
name := getTestRunnerName(e.logger, c.JobID)
e.logger.V(1).Info("Creating ec2 Test Runner instance", "name", name)
instanceId, err := ec2.CreateInstance(c.session, e.AmiID, key, tag, c.instanceProfileName, e.SubnetID, name)
instanceID, err := ec2.CreateInstance(c.Session, e.AmiID, key, tag, c.InstanceProfileName, e.SubnetID, name)
if err != nil {
return "", fmt.Errorf("creating instance for e2e tests: %v", err)
}
e.logger.V(1).Info("Instance created", "instance-id", instanceId)
e.InstanceID = instanceId
return instanceId, nil
e.logger.V(1).Info("Instance created", "instance-id", instanceID)
e.InstanceID = instanceID
return instanceID, nil
}

func (v *VSphereTestRunner) tagInstance(c instanceRunConf, key, value string) error {
vmName := getTestRunnerName(v.logger, c.jobId)
vmName := getTestRunnerName(v.logger, c.JobID)
vmPath := fmt.Sprintf("/%s/vm/%s/%s", v.Datacenter, v.Folder, vmName)
tag := fmt.Sprintf("%s:%s", key, value)

Expand All @@ -203,17 +203,17 @@ func (v *VSphereTestRunner) tagInstance(c instanceRunConf, key, value string) er
}

func (e *Ec2TestRunner) tagInstance(c instanceRunConf, key, value string) error {
err := ec2.TagInstance(c.session, e.InstanceID, key, value)
err := ec2.TagInstance(c.Session, e.InstanceID, key, value)
if err != nil {
return fmt.Errorf("failed to tag Ec2 test runner: %v", err)
}
return nil
}

func (v *VSphereTestRunner) decommInstance(c instanceRunConf) error {
_, deregisterError := ssm.DeregisterInstance(c.session, v.InstanceID)
_, deactivateError := ssm.DeleteActivation(c.session, v.ActivationId)
deleteError := cleanup.VsphereRmVms(context.Background(), getTestRunnerName(v.logger, c.jobId), executables.WithGovcEnvMap(v.envMap))
_, deregisterError := ssm.DeregisterInstance(c.Session, v.InstanceID)
_, deactivateError := ssm.DeleteActivation(c.Session, v.ActivationId)
deleteError := cleanup.VsphereRmVms(context.Background(), getTestRunnerName(v.logger, c.JobID), executables.WithGovcEnvMap(v.envMap))

if deregisterError != nil {
return fmt.Errorf("failed to decommission vsphere test runner ssm instance: %v", deregisterError)
Expand All @@ -231,9 +231,9 @@ func (v *VSphereTestRunner) decommInstance(c instanceRunConf) error {
}

func (e *Ec2TestRunner) decommInstance(c instanceRunConf) error {
runnerName := getTestRunnerName(e.logger, c.jobId)
runnerName := getTestRunnerName(e.logger, c.JobID)
e.logger.V(1).Info("Terminating ec2 Test Runner instance", "instanceID", e.InstanceID, "runner", runnerName)
if err := ec2.TerminateEc2Instances(c.session, aws.StringSlice([]string{e.InstanceID})); err != nil {
if err := ec2.TerminateEc2Instances(c.Session, aws.StringSlice([]string{e.InstanceID})); err != nil {
return fmt.Errorf("terminating instance %s for runner %s: %w", e.InstanceID, runnerName, err)
}

Expand Down
1 change: 1 addition & 0 deletions test/framework/cluster.go
Original file line number Diff line number Diff line change
Expand Up @@ -59,6 +59,7 @@ const (
JobIdVar = "T_JOB_ID"
BundlesOverrideVar = "T_BUNDLES_OVERRIDE"
ClusterIPPoolEnvVar = "T_CLUSTER_IP_POOL"
ClusterIPEnvVar = "T_CLUSTER_IP"
CleanupVmsVar = "T_CLEANUP_VMS"
hardwareYamlPath = "hardware.yaml"
hardwareCsvPath = "hardware.csv"
Expand Down