analysis.go

package analysis

import (
	"context"
	"fmt"
	"strings"
	"sync"
	"time"

	log "github.com/sirupsen/logrus"
	corev1 "k8s.io/api/core/v1"
	k8serrors "k8s.io/apimachinery/pkg/api/errors"
	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
	"k8s.io/utils/pointer"

	"github.com/argoproj/argo-rollouts/pkg/apis/rollouts/v1alpha1"
	analysisutil "github.com/argoproj/argo-rollouts/utils/analysis"
	"github.com/argoproj/argo-rollouts/utils/defaults"
	logutil "github.com/argoproj/argo-rollouts/utils/log"
	"github.com/argoproj/argo-rollouts/utils/record"
	timeutil "github.com/argoproj/argo-rollouts/utils/time"
)

const (
	// DefaultMeasurementHistoryLimit is the default maximum number of measurements to retain per metric,
	// before trimming the list.
	DefaultMeasurementHistoryLimit = 10
	// DefaultErrorRetryInterval is the default interval to retry a measurement upon error, in the
	// event an interval was not specified
	DefaultErrorRetryInterval = 10 * time.Second
	// SuccessfulAssessmentRunTerminatedResult is used for logging purposes when the metrics evaluation
	// is successful and the run is terminated.
	SuccessfulAssessmentRunTerminatedResult = "Metric Assessment Result - Successful: Run Terminated"
)

// metricTask holds the metric which need to be measured during this reconciliation along with
// an in-progress measurement
type metricTask struct {
	metric                v1alpha1.Metric
	incompleteMeasurement *v1alpha1.Measurement
}

func (c *Controller) reconcileAnalysisRun(origRun *v1alpha1.AnalysisRun) *v1alpha1.AnalysisRun {
	logger := logutil.WithAnalysisRun(origRun)
	if origRun.Status.Phase.Completed() {
		err := c.maybeGarbageCollectAnalysisRun(origRun, logger)
		if err != nil {
			// TODO(jessesuen): surface errors to controller so they can be retried
			logger.Warnf("Failed to garbage collect analysis run: %v", err)
		}
		return origRun
	}
	run := origRun.DeepCopy()

	if run.Status.MetricResults == nil {
		run.Status.MetricResults = make([]v1alpha1.MetricResult, 0)
	}

	resolvedMetrics, err := getResolvedMetricsWithoutSecrets(run.Spec.Metrics, run.Spec.Args)
	if err != nil {
		message := fmt.Sprintf("Unable to resolve metric arguments: %v", err)
		logger.Warn(message)
		run.Status.Phase = v1alpha1.AnalysisPhaseError
		run.Status.Message = message
		c.recordAnalysisRunCompletionEvent(run)
		return run
	}

	err = analysisutil.ValidateMetrics(resolvedMetrics)
	if err != nil {
		message := fmt.Sprintf("Analysis spec invalid: %v", err)
		logger.Warn(message)
		run.Status.Phase = v1alpha1.AnalysisPhaseError
		run.Status.Message = message
		c.recordAnalysisRunCompletionEvent(run)
		return run
	}

	dryRunMetricsMap, err := analysisutil.GetDryRunMetrics(run.Spec.DryRun, resolvedMetrics)
	if err != nil {
		message := fmt.Sprintf("Analysis spec invalid: %v", err)
		logger.Warn(message)
		run.Status.Phase = v1alpha1.AnalysisPhaseError
		run.Status.Message = message
		c.recordAnalysisRunCompletionEvent(run)
		return run
	}

	measurementRetentionMetricsMap, err := analysisutil.GetMeasurementRetentionMetrics(run.Spec.MeasurementRetention, resolvedMetrics)
	if err != nil {
		message := fmt.Sprintf("Analysis spec invalid: %v", err)
		logger.Warn(message)
		run.Status.Phase = v1alpha1.AnalysisPhaseError
		run.Status.Message = message
		c.recordAnalysisRunCompletionEvent(run)
		return run
	}

	tasks := generateMetricTasks(run, resolvedMetrics)
	logger.Infof("Taking %d Measurement(s)...", len(tasks))
	err = c.runMeasurements(run, tasks, dryRunMetricsMap)
	if err != nil {
		message := fmt.Sprintf("Unable to resolve metric arguments: %v", err)
		logger.Warn(message)
		run.Status.Phase = v1alpha1.AnalysisPhaseError
		run.Status.Message = message
		c.recordAnalysisRunCompletionEvent(run)
		return run
	}

	newStatus, newMessage := c.assessRunStatus(run, resolvedMetrics, dryRunMetricsMap)
	if newStatus != run.Status.Phase {
		run.Status.Phase = newStatus
		run.Status.Message = newMessage
		if newStatus.Completed() {
			c.recordAnalysisRunCompletionEvent(run)
			if run.Status.CompletedAt == nil {
				now := timeutil.MetaNow()
				run.Status.CompletedAt = &now
			}
		}
	}

	err = c.garbageCollectMeasurements(run, measurementRetentionMetricsMap, DefaultMeasurementHistoryLimit)
	if err != nil {
		// TODO(jessesuen): surface errors to controller so they can be retried
		logger.Warnf("Failed to garbage collect measurements: %v", err)
	}

	nextReconcileTime := calculateNextReconcileTime(run, resolvedMetrics)
	if nextReconcileTime != nil {
		enqueueSeconds := nextReconcileTime.Sub(timeutil.Now())
		if enqueueSeconds < 0 {
			enqueueSeconds = 0
		}
		logger.Infof("Enqueueing analysis after %v", enqueueSeconds)
		c.enqueueAnalysisAfter(run, enqueueSeconds)
	}
	return run
}

func getResolvedMetricsWithoutSecrets(metrics []v1alpha1.Metric, args []v1alpha1.Argument) ([]v1alpha1.Metric, error) {
	newArgs := make([]v1alpha1.Argument, 0)
	for _, arg := range args {
		newArg := arg.DeepCopy()
		if newArg.ValueFrom != nil && newArg.ValueFrom.SecretKeyRef != nil {
			newArg.ValueFrom = nil
			newArg.Value = pointer.StringPtr("temp-for-secret")
		}
		newArgs = append(newArgs, *newArg)
	}
	resolvedMetrics := make([]v1alpha1.Metric, 0)
	for _, metric := range metrics {
		resolvedMetric, err := analysisutil.ResolveMetricArgs(metric, newArgs)
		if err != nil {
			return nil, err
		}
		resolvedMetrics = append(resolvedMetrics, *resolvedMetric)
	}
	return resolvedMetrics, nil
}

func (c *Controller) recordAnalysisRunCompletionEvent(run *v1alpha1.AnalysisRun) {
	eventType := corev1.EventTypeNormal
	switch run.Status.Phase {
	case v1alpha1.AnalysisPhaseError, v1alpha1.AnalysisPhaseFailed:
		eventType = corev1.EventTypeWarning
	}
	c.recorder.Eventf(run, record.EventOptions{EventType: eventType, EventReason: "AnalysisRun" + string(run.Status.Phase)}, "Analysis Completed. Result: %s", run.Status.Phase)
}

// generateMetricTasks generates a list of metrics tasks needed to be measured as part of this
// sync, based on the last completion times that metric was measured (if ever). If the run is
// terminating (e.g. due to manual termination or failing metric), will not schedule further
// measurements other than to resume any in-flight measurements.
func generateMetricTasks(run *v1alpha1.AnalysisRun, metrics []v1alpha1.Metric) []metricTask {
	logger := logutil.WithAnalysisRun(run)
	var tasks []metricTask
	terminating := analysisutil.IsTerminating(run)

	for i, metric := range metrics {
		if analysisutil.MetricCompleted(run, metric.Name) {
			continue
		}
		logCtx := logger.WithField("metric", metric.Name)
		lastMeasurement := analysisutil.LastMeasurement(run, metric.Name)
		if lastMeasurement != nil && lastMeasurement.FinishedAt == nil {
			now := timeutil.MetaNow()
			if lastMeasurement.ResumeAt != nil && lastMeasurement.ResumeAt.After(now.Time) {
				continue
			}
			// last measurement is still in-progress. need to complete it
			logCtx.Infof("Resuming in-progress measurement")
			tasks = append(tasks, metricTask{
				metric:                run.Spec.Metrics[i],
				incompleteMeasurement: lastMeasurement,
			})
			continue
		}
		if terminating {
			logCtx.Infof("Skipping measurement: run is terminating")
			continue
		}
		if lastMeasurement == nil {
			if metric.InitialDelay != "" {
				if run.Status.StartedAt == nil {
					continue
				}
				duration, err := metric.InitialDelay.Duration()
				if err != nil {
					logCtx.Warnf("failed to parse duration: %v", err)
					continue
				}
				if run.Status.StartedAt.Add(duration).After(timeutil.Now()) {
					logCtx.Infof("Waiting until start delay duration passes")
					continue
				}
			}
			// measurement never taken
			tasks = append(tasks, metricTask{metric: run.Spec.Metrics[i]})
			logCtx.Infof("Running initial measurement")
			continue
		}
		metricResult := analysisutil.GetResult(run, metric.Name)
		effectiveCount := metric.EffectiveCount()
		if effectiveCount != nil && metricResult.Count >= int32(effectiveCount.IntValue()) {
			// we have reached desired count
			continue
		}
		// if we get here, we know we need to take a measurement (eventually). check last measurement
		// to decide if it should be taken now. metric.Interval can be null because we may be
		// retrying a metric due to error.
		interval := DefaultErrorRetryInterval
		if lastMeasurement.Phase == v1alpha1.AnalysisPhaseError {
			interval = DefaultErrorRetryInterval
		} else if metric.Interval != "" {
			parsedInterval, err := parseMetricInterval(*logCtx, metric.Interval)
			if err != nil {
				continue
			}
			interval = parsedInterval
		}
		if timeutil.Now().After(lastMeasurement.FinishedAt.Add(interval)) {
			tasks = append(tasks, metricTask{metric: run.Spec.Metrics[i]})
			logCtx.Infof("Running overdue measurement")
			continue
		}
	}
	return tasks
}

// parseMetricInterval is a helper method to parse the given metric interval and return the
// parsed duration or error (if any)
func parseMetricInterval(logCtx log.Entry, metricDurationString v1alpha1.DurationString) (time.Duration, error) {
	metricInterval, err := metricDurationString.Duration()
	if err != nil {
		logCtx.Warnf("Failed to parse interval: %v", err)
		return -1, err
	}
	return metricInterval, nil
}

// resolveArgs resolves args for metricTasks, including secret references
// returns resolved metricTasks and secrets for log redaction
func (c *Controller) resolveArgs(tasks []metricTask, args []v1alpha1.Argument, namespace string) ([]metricTask, []string, error) {
	//create set of secret values for redaction
	secretSet := map[string]bool{}
	for i, arg := range args {
		//if secret specified in valueFrom, replace value with secret value
		//error if arg has both value and valueFrom
		if arg.ValueFrom != nil && arg.ValueFrom.SecretKeyRef != nil {
			name := arg.ValueFrom.SecretKeyRef.Name
			secret, err := c.kubeclientset.CoreV1().Secrets(namespace).Get(context.TODO(), name, metav1.GetOptions{})
			if err != nil {
				return nil, nil, err
			}

			secretContentBytes, ok := secret.Data[arg.ValueFrom.SecretKeyRef.Key]
			if !ok {
				err := fmt.Errorf("key '%s' does not exist in secret '%s'", arg.ValueFrom.SecretKeyRef.Key, arg.ValueFrom.SecretKeyRef.Name)
				return nil, nil, err
			}
			secretContent := string(secretContentBytes)
			secretSet[secretContent] = true
			resolvedArg := arg.DeepCopy()
			resolvedArg.Value = &secretContent
			args[i] = *resolvedArg
		} else {
			args[i] = arg
		}
	}

	// creates list of secret values from secretSet for RedactorFormatter
	secrets := make([]string, 0, len(secretSet))
	for k := range secretSet {
		secrets = append(secrets, k)
	}

	// resolves arguments in each metric task
	for i, task := range tasks {
		resolvedMetric, err := analysisutil.ResolveMetricArgs(task.metric, args)
		if err != nil {
			return nil, nil, err
		}
		tasks[i].metric = *resolvedMetric
	}

	return tasks, secrets, nil
}

// runMeasurements iterates a list of metric tasks, and runs, resumes, or terminates measurements
func (c *Controller) runMeasurements(run *v1alpha1.AnalysisRun, tasks []metricTask, dryRunMetricsMap map[string]bool) error {
	var wg sync.WaitGroup
	// resultsLock should be held whenever we are accessing or setting status.metricResults since
	// we are performing queries in parallel
	var resultsLock sync.Mutex
	terminating := analysisutil.IsTerminating(run)

	// resolve args for metric tasks
	// get list of secret values for log redaction
	tasks, secrets, err := c.resolveArgs(tasks, run.Spec.Args, run.Namespace)
	if err != nil {
		return err
	}

	for _, task := range tasks {
		wg.Add(1)

		go func(t metricTask) error {
			defer wg.Done()
			//redact secret values from logs
			logger := logutil.WithRedactor(*logutil.WithAnalysisRun(run).WithField("metric", t.metric.Name), secrets)

			var newMeasurement v1alpha1.Measurement
			provider, providerErr := c.newProvider(*logger, run.Namespace, t.metric)
			if providerErr != nil {
				log.Errorf("Error in getting metric provider :%v", providerErr)
				if t.incompleteMeasurement != nil {
					newMeasurement = *t.incompleteMeasurement
				} else {
					startedAt := timeutil.MetaNow()
					newMeasurement.StartedAt = &startedAt
				}
				newMeasurement.Phase = v1alpha1.AnalysisPhaseError
				newMeasurement.Message = providerErr.Error()
			} else {
				if t.incompleteMeasurement == nil {
					newMeasurement = provider.Run(run, t.metric)
				} else {
					// metric is incomplete. either terminate or resume it
					if terminating {
						logger.Infof("Terminating in-progress measurement")
						newMeasurement = provider.Terminate(run, t.metric, *t.incompleteMeasurement)
						if newMeasurement.Phase == v1alpha1.AnalysisPhaseSuccessful {
							newMeasurement.Message = "Metric Terminated"
						}
					} else {
						newMeasurement = provider.Resume(run, t.metric, *t.incompleteMeasurement)
					}
				}
			}

			resultsLock.Lock()
			metricResult := analysisutil.GetResult(run, t.metric.Name)
			resultsLock.Unlock()
			if metricResult == nil {
				metricResult = &v1alpha1.MetricResult{
					Name:   t.metric.Name,
					Phase:  v1alpha1.AnalysisPhaseRunning,
					DryRun: dryRunMetricsMap[t.metric.Name],
				}

				if provider != nil && providerErr == nil {
					metricResult.Metadata = provider.GetMetadata(t.metric)
				}
			}

			if newMeasurement.Phase.Completed() {
				logger.Infof("Measurement Completed. Result: %s", newMeasurement.Phase)
				if newMeasurement.FinishedAt == nil {
					finishedAt := timeutil.MetaNow()
					newMeasurement.FinishedAt = &finishedAt
				}

				switch newMeasurement.Phase {
				case v1alpha1.AnalysisPhaseSuccessful:
					metricResult.Successful++
					metricResult.Count++
					metricResult.ConsecutiveError = 0
					metricResult.ConsecutiveSuccess++
				case v1alpha1.AnalysisPhaseFailed:
					metricResult.Failed++
					metricResult.Count++
					metricResult.ConsecutiveError = 0
					metricResult.ConsecutiveSuccess = 0
				case v1alpha1.AnalysisPhaseInconclusive:
					metricResult.Inconclusive++
					metricResult.Count++
					metricResult.ConsecutiveError = 0
					metricResult.ConsecutiveSuccess = 0
				case v1alpha1.AnalysisPhaseError:
					metricResult.Error++
					metricResult.ConsecutiveError++
					metricResult.ConsecutiveSuccess = 0
					logger.Warnf("Measurement had error: %s", newMeasurement.Message)
				}
			}

			//redact secret values from measurement message
			for _, secret := range secrets {
				if secret != "" {
					newMeasurement.Message = strings.ReplaceAll(newMeasurement.Message, secret, "*****")
				}
			}

			if t.incompleteMeasurement == nil {
				metricResult.Measurements = append(metricResult.Measurements, newMeasurement)
			} else {
				metricResult.Measurements[len(metricResult.Measurements)-1] = newMeasurement
			}

			resultsLock.Lock()
			analysisutil.SetResult(run, *metricResult)
			resultsLock.Unlock()
			return nil
		}(task)
	}
	wg.Wait()

	return nil
}

// assessRunStatus assesses the overall status of this AnalysisRun
// If any metric is not yet completed, the AnalysisRun is still considered Running
// Once all metrics are complete, the worst status is used as the overall AnalysisRun status
func (c *Controller) assessRunStatus(run *v1alpha1.AnalysisRun, metrics []v1alpha1.Metric, dryRunMetricsMap map[string]bool) (v1alpha1.AnalysisPhase, string) {
	var worstStatus v1alpha1.AnalysisPhase
	var worstMessage string
	terminating := analysisutil.IsTerminating(run)
	everythingCompleted := true

	if run.Status.StartedAt == nil {
		now := timeutil.MetaNow()
		run.Status.StartedAt = &now
	}
	if run.Spec.Terminate {
		worstMessage = "Run Terminated"
	}

	// Initialize Run & Dry-Run summary object
	runSummary := v1alpha1.RunSummary{
		Count:        0,
		Successful:   0,
		Failed:       0,
		Inconclusive: 0,
		Error:        0,
	}
	dryRunSummary := v1alpha1.RunSummary{
		Count:        0,
		Successful:   0,
		Failed:       0,
		Inconclusive: 0,
		Error:        0,
	}

	// Iterate all metrics and update `MetricResult.Phase` fields based on latest measurement(s)
	for _, metric := range metrics {
		if dryRunMetricsMap[metric.Name] {
			log.Infof("Metric '%s' is running in the Dry-Run mode.", metric.Name)
			dryRunSummary.Count++
		} else {
			runSummary.Count++
		}
		if result := analysisutil.GetResult(run, metric.Name); result != nil {
			logger := logutil.WithAnalysisRun(run).WithField("metric", metric.Name)
			metricStatus := assessMetricStatus(metric, *result, terminating)
			if result.Phase != metricStatus {
				logger.Infof("Metric '%s' transitioned from %s -> %s", metric.Name, result.Phase, metricStatus)
				if metricStatus.Completed() {
					eventType := corev1.EventTypeNormal
					switch metricStatus {
					case v1alpha1.AnalysisPhaseError, v1alpha1.AnalysisPhaseFailed:
						eventType = corev1.EventTypeWarning
					}
					c.recorder.Eventf(run, record.EventOptions{EventType: eventType, EventReason: "Metric" + string(metricStatus)}, "Metric '%s' Completed. Result: %s", metric.Name, metricStatus)
				}
				if lastMeasurement := analysisutil.LastMeasurement(run, metric.Name); lastMeasurement != nil {
					result.Message = lastMeasurement.Message
				}
				result.Phase = metricStatus
				analysisutil.SetResult(run, *result)
			}
			if !metricStatus.Completed() {
				// if any metric is in-progress, then entire analysis run will be considered running
				everythingCompleted = false
			} else {
				phase, message := assessMetricFailureInconclusiveOrError(metric, *result)
				// NOTE: We don't care about the status if the metric is marked as a Dry-Run
				// otherwise, remember the worst status of all completed metric results
				if !dryRunMetricsMap[metric.Name] {
					if worstStatus == "" || analysisutil.IsWorse(worstStatus, metricStatus) {
						worstStatus = metricStatus
						if message != "" {
							worstMessage = fmt.Sprintf("Metric \"%s\" assessed %s due to %s", metric.Name, metricStatus, message)
							if result.Message != "" {
								worstMessage += fmt.Sprintf(": \"Error Message: %s\"", result.Message)
							}
						}
					}
					// Update Run Summary
					switch phase {
					case v1alpha1.AnalysisPhaseError:
						runSummary.Error++
					case v1alpha1.AnalysisPhaseFailed:
						runSummary.Failed++
					case v1alpha1.AnalysisPhaseInconclusive:
						runSummary.Inconclusive++
					case v1alpha1.AnalysisPhaseSuccessful:
						runSummary.Successful++
					default:
						// We'll mark the status as success by default if it doesn't match anything.
						runSummary.Successful++
					}
				} else {
					// We don't really care about the failures from dry-runs and hence, if there is no current status
					// found then we just set it to `AnalysisPhaseSuccessful`
					if worstStatus == "" {
						worstStatus = v1alpha1.AnalysisPhaseSuccessful
					}
					// Update metric result message
					if message != "" {
						result.Message = fmt.Sprintf("Metric assessed %s due to %s", metricStatus, message)
						analysisutil.SetResult(run, *result)
					}
					// Update DryRun Summary
					switch phase {
					case v1alpha1.AnalysisPhaseError:
						dryRunSummary.Error++
					case v1alpha1.AnalysisPhaseFailed:
						dryRunSummary.Failed++
					case v1alpha1.AnalysisPhaseInconclusive:
						dryRunSummary.Inconclusive++
					case v1alpha1.AnalysisPhaseSuccessful:
						dryRunSummary.Successful++
					default:
						// We'll mark the status as success by default if it doesn't match anything.
						dryRunSummary.Successful++
					}
				}
			}
		} else {
			// metric hasn't started running. possible cases where some metrics starts with delay
			everythingCompleted = false
		}
	}
	// Append Dry-Run metrics results if any.
	worstMessage = strings.TrimSpace(worstMessage)
	run.Status.RunSummary = runSummary
	run.Status.DryRunSummary = &dryRunSummary
	if terminating {
		if worstStatus == "" {
			// we have yet to take a single measurement, but have already been instructed to stop
			log.Infof(SuccessfulAssessmentRunTerminatedResult)
			return v1alpha1.AnalysisPhaseSuccessful, worstMessage
		}
		log.Infof("Metric Assessment Result - %s: Run Terminated", worstStatus)
		return worstStatus, worstMessage
	}
	if !everythingCompleted || worstStatus == "" {
		return v1alpha1.AnalysisPhaseRunning, ""
	}
	return worstStatus, worstMessage
}

// assessMetricStatus assesses the status of a single metric based on:
// * current or latest measurement status
// * parameters given by the metric (failureLimit, count, etc...)
// * whether we are terminating (e.g. due to failing run, or termination request)
func assessMetricStatus(metric v1alpha1.Metric, result v1alpha1.MetricResult, terminating bool) v1alpha1.AnalysisPhase {
	if result.Phase.Completed() {
		return result.Phase
	}
	logger := log.WithField("metric", metric.Name)
	if len(result.Measurements) == 0 {
		if terminating {
			// we have yet to take a single measurement, but have already been instructed to stop
			logger.Infof(SuccessfulAssessmentRunTerminatedResult)
			return v1alpha1.AnalysisPhaseSuccessful
		}
		return v1alpha1.AnalysisPhasePending
	}
	lastMeasurement := result.Measurements[len(result.Measurements)-1]
	if !lastMeasurement.Phase.Completed() {
		// we still have an in-flight measurement
		return v1alpha1.AnalysisPhaseRunning
	}

	// Check if metric was considered Failed, Inconclusive, or Error
	// If true, then return AnalysisRunPhase as Failed, Inconclusive, or Error respectively
	phaseFailureInconclusiveOrError, message := assessMetricFailureInconclusiveOrError(metric, result)
	if phaseFailureInconclusiveOrError != "" {
		logger.Infof("Metric Assessment Result - %s: %s", phaseFailureInconclusiveOrError, message)
		return phaseFailureInconclusiveOrError
	}

	// Check if consecutiveSuccessLimit is applicable and was reached.
	if metric.ConsecutiveSuccessLimit != nil && metric.ConsecutiveSuccessLimit.IntValue() > 0 && result.ConsecutiveSuccess >= int32(metric.ConsecutiveSuccessLimit.IntValue()) {
		logger.Infof("Metric Assessment Result - %s: ConsecutiveSuccessLimit (%s) Reached", v1alpha1.AnalysisPhaseSuccessful, metric.ConsecutiveSuccessLimit.String())
		return v1alpha1.AnalysisPhaseSuccessful
	}

	// If a count was specified, and we reached that count, then metric is considered Successful.
	// The Error, Failed, Inconclusive counters are ignored because those checks have already been
	// taken into consideration above, and we do not want to fail if failures < failureLimit.
	effectiveCount := metric.EffectiveCount()
	if effectiveCount != nil && result.Count >= int32(effectiveCount.IntValue()) {

		failureApplicable := (metric.FailureLimit != nil && metric.FailureLimit.IntValue() >= 0) || metric.FailureLimit == nil
		successApplicable := metric.ConsecutiveSuccessLimit != nil && metric.ConsecutiveSuccessLimit.IntValue() > 0

		if failureApplicable && successApplicable {

			// failureLimit was checked above and not reached.
			// consecutiveSuccessLimit was checked above and not reached.

			failureLimit := "0"
			if metric.FailureLimit != nil {
				failureLimit = metric.FailureLimit.String()
			}

			logger.Infof("Metric Assessment Result - %s: ConsecutiveSuccessLimit (%s) Not Reached and FailureLimit (%s) Not Violated", v1alpha1.AnalysisPhaseInconclusive, metric.ConsecutiveSuccessLimit.String(), failureLimit)
			return v1alpha1.AnalysisPhaseInconclusive

		} else if successApplicable {

			logger.Infof("Metric Assessment Result - %s: ConsecutiveSuccessLimit (%s) Not Reached", v1alpha1.AnalysisPhaseFailed, metric.ConsecutiveSuccessLimit.String())
			return v1alpha1.AnalysisPhaseFailed

		} else if failureApplicable {
			// failureLimit was not reached in assessMetricFailureInconclusiveOrError above.
			// AnalysisPhaseSuccessful below.
		} else {
			// This cannot happen, since one of failureLimit or consecutiveSuccessLimit will be applicable
			// We validate that failureLimit >= 0 when consecutiveSuccessLimit == 0
		}

		logger.Infof("Metric Assessment Result - %s: Count (%s) Reached", v1alpha1.AnalysisPhaseSuccessful, effectiveCount.String())
		return v1alpha1.AnalysisPhaseSuccessful
	}
	// if we get here, this metric runs indefinitely
	if terminating {
		logger.Infof(SuccessfulAssessmentRunTerminatedResult)
		return v1alpha1.AnalysisPhaseSuccessful
	}
	return v1alpha1.AnalysisPhaseRunning
}

func assessMetricFailureInconclusiveOrError(metric v1alpha1.Metric, result v1alpha1.MetricResult) (v1alpha1.AnalysisPhase, string) {
	var message string
	var phase v1alpha1.AnalysisPhase

	failureLimit := int32(0)
	if metric.FailureLimit != nil {
		failureLimit = int32(metric.FailureLimit.IntValue())
	}
	// If failureLimit is negative, that means it isn't applicable.
	if failureLimit >= 0 && result.Failed > failureLimit {
		phase = v1alpha1.AnalysisPhaseFailed
		message = fmt.Sprintf("failed (%d) > failureLimit (%d)", result.Failed, failureLimit)
	}

	inconclusiveLimit := int32(0)
	if metric.InconclusiveLimit != nil {
		inconclusiveLimit = int32(metric.InconclusiveLimit.IntValue())
	}
	if result.Inconclusive > inconclusiveLimit {
		phase = v1alpha1.AnalysisPhaseInconclusive
		message = fmt.Sprintf("inconclusive (%d) > inconclusiveLimit (%d)", result.Inconclusive, inconclusiveLimit)
	}

	consecutiveErrorLimit := defaults.GetConsecutiveErrorLimitOrDefault(&metric)
	if result.ConsecutiveError > consecutiveErrorLimit {
		phase = v1alpha1.AnalysisPhaseError
		message = fmt.Sprintf("consecutiveErrors (%d) > consecutiveErrorLimit (%d)", result.ConsecutiveError, consecutiveErrorLimit)
	}
	return phase, message
}

// calculateNextReconcileTime calculates the next time that this AnalysisRun should be reconciled,
// based on the earliest time of all metrics intervals, counts, and their finishedAt timestamps
func calculateNextReconcileTime(run *v1alpha1.AnalysisRun, metrics []v1alpha1.Metric) *time.Time {
	var reconcileTime *time.Time
	for _, metric := range metrics {
		if analysisutil.MetricCompleted(run, metric.Name) {
			// NOTE: this also covers the case where metric.Count is reached
			continue
		}
		logCtx := logutil.WithAnalysisRun(run).WithField("metric", metric.Name)
		lastMeasurement := analysisutil.LastMeasurement(run, metric.Name)
		if lastMeasurement == nil {
			if metric.InitialDelay != "" {
				startTime := timeutil.MetaNow()
				if run.Status.StartedAt != nil {
					startTime = *run.Status.StartedAt
				}
				parsedInterval, err := parseMetricInterval(*logCtx, metric.InitialDelay)
				if err != nil {
					continue
				}
				endInitialDelay := startTime.Add(parsedInterval)
				if reconcileTime == nil || reconcileTime.After(endInitialDelay) {
					reconcileTime = &endInitialDelay
				}
				continue
			}
			// no measurement was started . we should never get here
			logCtx.Warnf("Metric never started. Not factored into enqueue time.")
			continue
		}
		if lastMeasurement.FinishedAt == nil {
			// unfinished in-flight measurement.
			if lastMeasurement.ResumeAt != nil {
				if reconcileTime == nil || reconcileTime.After(lastMeasurement.ResumeAt.Time) {
					reconcileTime = &lastMeasurement.ResumeAt.Time
				}
			}
			continue
		}
		metricResult := analysisutil.GetResult(run, metric.Name)
		effectiveCount := metric.EffectiveCount()
		if effectiveCount != nil && metricResult.Count >= int32(effectiveCount.IntValue()) {
			// we have reached desired count
			continue
		}
		var interval time.Duration
		if lastMeasurement.Phase == v1alpha1.AnalysisPhaseError {
			interval = DefaultErrorRetryInterval
		} else if metric.Interval != "" {
			parsedInterval, err := parseMetricInterval(*logCtx, metric.Interval)
			if err != nil {
				continue
			}
			interval = parsedInterval
		} else {
			// if we get here, an interval was not set (meaning reoccurrence was not desired), and
			// there was no error (meaning we don't need to retry). no need to requeue this metric.
			// NOTE: we shouldn't ever get here since it means we are not doing proper bookkeeping
			// of count.
			logCtx.Warnf("Skipping requeue. No interval or error (count: %d, effectiveCount: %s)", metricResult.Count, metric.EffectiveCount().String())
			continue
		}
		// Take the earliest time of all metrics
		metricReconcileTime := lastMeasurement.FinishedAt.Add(interval)
		if reconcileTime == nil || reconcileTime.After(metricReconcileTime) {
			reconcileTime = &metricReconcileTime
		}
	}
	return reconcileTime
}

// garbageCollectMeasurements trims the measurement history to the specified limit and GCs old measurements
func (c *Controller) garbageCollectMeasurements(run *v1alpha1.AnalysisRun, measurementRetentionMetricNamesMap map[string]*v1alpha1.MeasurementRetention, limit int) error {
	var errors []error

	resolvedArgsMetric, err := getResolvedMetricsWithoutSecrets(run.Spec.Metrics, run.Spec.Args)
	if err != nil {
		return fmt.Errorf("failed to resolve args on metrics during garbage collection: %w", err)
	}

	metricsByName := make(map[string]v1alpha1.Metric)
	for _, metric := range resolvedArgsMetric {
		metricsByName[metric.Name] = metric
	}

	for i, result := range run.Status.MetricResults {
		length := len(result.Measurements)
		measurementRetentionObject := measurementRetentionMetricNamesMap[result.Name]
		measurementsLimit := limit
		if measurementRetentionObject != nil && measurementRetentionObject.Limit > 0 {
			measurementsLimit = int(measurementRetentionObject.Limit)
		}
		if length > measurementsLimit {
			metric, ok := metricsByName[result.Name]
			if !ok {
				continue
			}
			logger := logutil.WithAnalysisRun(run).WithField("metric", metric.Name)
			provider, err := c.newProvider(*logger, run.Namespace, metric)
			if err != nil {
				errors = append(errors, err)
				continue
			}
			err = provider.GarbageCollect(run, metric, measurementsLimit)
			if err != nil {
				return err
			}
			result.Measurements = result.Measurements[length-measurementsLimit : length]
		}
		run.Status.MetricResults[i] = result
	}
	if len(errors) > 0 {
		return errors[0]
	}
	return nil
}

func (c *Controller) maybeGarbageCollectAnalysisRun(run *v1alpha1.AnalysisRun, logger *log.Entry) error {
	ctx := context.TODO()
	if run.DeletionTimestamp != nil || !isAnalysisRunTtlExceeded(run) {
		return nil
	}
	logger.Infof("Trying to cleanup TTL exceeded analysis run")
	err := c.argoProjClientset.ArgoprojV1alpha1().AnalysisRuns(run.Namespace).Delete(ctx, run.Name, metav1.DeleteOptions{})
	if err != nil && !k8serrors.IsNotFound(err) {
		return err
	}
	return nil
}

func isAnalysisRunTtlExceeded(run *v1alpha1.AnalysisRun) bool {
	// TTL only counted for completed runs with TTL strategy.
	if !run.Status.Phase.Completed() || run.Spec.TTLStrategy == nil {
		return false
	}
	// Cannot determine TTL if run has no completion time.
	if run.Status.CompletedAt == nil {
		return false
	}
	secondsCompleted := timeutil.MetaNow().Sub(run.Status.CompletedAt.Time).Seconds()
	var ttlSeconds *int32
	if run.Status.Phase == v1alpha1.AnalysisPhaseSuccessful && run.Spec.TTLStrategy.SecondsAfterSuccess != nil {
		ttlSeconds = run.Spec.TTLStrategy.SecondsAfterSuccess
	} else if run.Status.Phase == v1alpha1.AnalysisPhaseFailed && run.Spec.TTLStrategy.SecondsAfterFailure != nil {
		ttlSeconds = run.Spec.TTLStrategy.SecondsAfterFailure
	} else if run.Spec.TTLStrategy.SecondsAfterCompletion != nil {
		ttlSeconds = run.Spec.TTLStrategy.SecondsAfterCompletion
	}
	if ttlSeconds == nil {
		return false
	}
	return int32(secondsCompleted) > *ttlSeconds
}