New Feature: provide failFast flag, allow a DAG to run all branches o…

…f the DAG (either success or failure)

api/openapi-spec/swagger.json

@@ -1160,6 +1160,10 @@
  "description": "Entrypoint is a template reference to the starting point of the workflow",
  "type": "string"
  },
+ "failFast": {
+ "description": "This flag is for DAG logic. The DAG logic has a built-in \"fail fast\" feature to stop scheduling new steps, as soon as it detects that one of the DAG nodes is failed. Then it waits until all DAG nodes are completed before failing the DAG itself. The FailFast flag default is true, if set to false, it will allow a DAG to run all branches of the DAG to completion (either success or failure), regardless of the failed outcomes of branches in the DAG. More info and example about this feature at https://github.com/argoproj/argo/issues/1442",
+ "type": "boolean"
+ },
  "hostAliases": {
  "description": "HostAliases is an optional list of hosts and IPs that will be injected into the pod spec",
  "type": "array",

pkg/apis/workflow/v1alpha1/types.go

@@ -167,6 +167,14 @@ type WorkflowSpec struct {
 
  // HostAliases is an optional list of hosts and IPs that will be injected into the pod spec
  HostAliases []apiv1.HostAlias `json:"hostAliases,omitempty"`
+
+ // This flag is for DAG logic. The DAG logic has a built-in "fail fast" feature to stop scheduling new steps,
+ // as soon as it detects that one of the DAG nodes is failed. Then it waits until all DAG nodes are completed
+ // before failing the DAG itself.
+ // The FailFast flag default is true, if set to false, it will allow a DAG to run all branches of the DAG to
+ // completion (either success or failure), regardless of the failed outcomes of branches in the DAG.
+ // More info and example about this feature at https://github.com/argoproj/argo/issues/1442
+ FailFast *bool `json:"failFast,omitempty"`
 }
 
 // Template is a reusable and composable unit of execution in a workflow

test/e2e/functional/dag-disable-failFast.yaml

@@ -0,0 +1,49 @@
+apiVersion: argoproj.io/v1alpha1
+kind: Workflow
+metadata:
+ generateName: dag-primay-branch-
+spec:
+ failFast: false
+ entrypoint: statis
+ templates:
+ - name: a
+ container:
+ image: docker/whalesay:latest
+ command: [cowsay]
+ args: ["hello world"]
+ - name: b
+ retryStrategy:
+ limit: 2
+ container:
+ image: alpine:latest
+ command: [sh, -c]
+ args: ["sleep 30; echo haha"]
+ - name: c
+ retryStrategy:
+ limit: 3
+ container:
+ image: alpine:latest
+ command: [sh, -c]
+ args: ["echo intentional failure; exit 2"]
+ - name: d
+ container:
+ image: docker/whalesay:latest
+ command: [cowsay]
+ args: ["hello world"]
+ - name: statis
+ dag:
+ tasks:
+ - name: A
+ template: a
+ - name: B
+ dependencies: [A]
+ template: b
+ - name: C
+ dependencies: [A]
+ template: c
+ - name: D
+ dependencies: [B]
+ template: d
+ - name: E
+ dependencies: [D]
+ template: d

workflow/controller/dag.go

@@ -87,7 +87,29 @@ func (d *dagContext) assessDAGPhase(targetTasks []string, nodes map[string]wfv1.
  retriesExhausted = false
  }
  }
+
  if unsuccessfulPhase != "" {
+ // If failFast set to false, we should return Running to continue this workflow for other DAG branch
+ if d.wf.Spec.FailFast != nil && !*d.wf.Spec.FailFast {
+ tmpOverAllFinished := true
+ // If all the nodes have finished, we should mark the failed node to finish overall workflow
+ // So we should check all the targetTasks have finished
+ for _, tmpDepName := range targetTasks {
+ tmpDepNode := d.getTaskNode(tmpDepName)
+ if tmpDepNode == nil {
+ tmpOverAllFinished = false
+ break
+ }
+ if tmpDepNode.Type == wfv1.NodeTypeRetry && hasMoreRetries(tmpDepNode, d.wf) {
+ tmpOverAllFinished = false
+ break
+ }
+ }
+ if !tmpOverAllFinished {
+ return wfv1.NodeRunning
+ }
+ }
+
  // if we were unsuccessful, we can return *only* if all retry nodes have ben exhausted.
  if retriesExhausted {
  return unsuccessfulPhase
@@ -250,25 +272,6 @@ func (woc *wfOperationCtx) executeDAGTask(dagCtx *dagContext, taskName string) {
  dependenciesSuccessful = false
  }
  continue
- } else if depNode.Type == wfv1.NodeTypeRetry {
- // For retry type node
- // Maybe one of children node already success, but the retry node hasn't sync to latest status
- // So skip these steps, the next process function will make the status to be right
- tmpContinueFlag := false
- for _, tmpChildName := range depNode.Children {
- tmpChild, tmpOK := woc.wf.Status.Nodes[tmpChildName]
- if !tmpOK {
- continue
- }
- if tmpChild.Successful() {
- tmpContinueFlag = true
- break
- }
- }
- if tmpContinueFlag {
- woc.markNodePhase(depNode.Name, wfv1.NodeSucceeded)
- continue
- }
  }
  }
  dependenciesCompleted = false