Files
temporal/service/history/api/verifyfirstworkflowtaskscheduled/api_test.go
Shivam 089fba53e1 Fix pinned workflow trampolining via revision-based signal suppression (#9895)
## Summary
- Matching now sends the real `targetDeploymentRevisionNumber` for
pinned workflows instead of hardcoded 0
- Added `revision_number` field to `LastNotifiedTargetVersion`
(server-internal) and `DeclinedTargetVersionUpgrade` (public API)
- Case 4 in the target version change switch now uses
`targetRevisionNumber <= declined.RevisionNumber` instead of version
string comparison, preventing trampolining when stale matching
partitions send outdated target versions

## Problem
When a pinned workflow CaNs and declines a target version upgrade, a
stale matching partition can send an older target version. The old case
4 compared version strings (`declined.buildId == target.buildId`), which
didn't match the stale version — causing the workflow to re-signal, CaN
again, and trampoline indefinitely between stale and up-to-date
partitions.

## Test plan
- [x] `TestStalePartition_RevisionSuppressesTrampolining` — integration
test that simulates a stale partition via `rollbackTaskQueueToVersion`
and verifies:
- Stale partition (revision 0) is suppressed after declining at a higher
revision
- Genuinely new version (v4 at higher revision) correctly fires the
signal
- [x] Verified test **fails** when matching + history changes are
reverted (matching sends 0, old string comparison)

## API repo dependency
- api: `temporalio/api@trampolining-rev-number`
- api-go: `temporalio/api-go@trampolining-rev-number`

🤖 Generated with [Claude Code](https://claude.com/claude-code)

<!-- CURSOR_SUMMARY -->
---

> [!NOTE]
> **Medium Risk**
> Touches pinned-workflow versioning logic across matching, history, and
persistence; a bug here could change when workflows are signaled to
continue-as-new or upgrade, affecting routing behavior.
> 
> **Overview**
> Prevents pinned workflows from repeatedly continue-as-new
“trampolining” when task queue partitions have stale routing data by
**tracking and comparing target routing-config revision numbers**.
> 
> Matching now propagates the real `targetDeploymentRevisionNumber` for
pinned workflow tasks, and history threads this through
`AddWorkflowTaskStartedEvent` to persist
`LastNotifiedTargetVersion.revision_number` and carry it into
`DeclinedTargetVersionUpgrade` on continue-as-new. The pinned
target-change decision in `workflow_task_state_machine` switches case-4
suppression from deployment version string equality to
`targetRevisionNumber <= declined.RevisionNumber`, and adds an
integration test (`TestStalePartition_RevisionSuppressesTrampolining`)
covering stale vs fresh partition behavior.
> 
> <sup>Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit
4114662b39. Bugbot is set up for automated
code reviews on this repo. Configure
[here](https://www.cursor.com/dashboard/bugbot).</sup>
<!-- /CURSOR_SUMMARY -->

---------

Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-21 20:15:49 -07:00

293 lines
11 KiB
Go

package verifyfirstworkflowtaskscheduled
import (
"context"
"testing"
"time"
"github.com/google/uuid"
"github.com/stretchr/testify/require"
"github.com/stretchr/testify/suite"
commonpb "go.temporal.io/api/common/v1"
enumspb "go.temporal.io/api/enums/v1"
historypb "go.temporal.io/api/history/v1"
"go.temporal.io/api/serviceerror"
taskqueuepb "go.temporal.io/api/taskqueue/v1"
"go.temporal.io/api/workflowservice/v1"
enumsspb "go.temporal.io/server/api/enums/v1"
"go.temporal.io/server/api/historyservice/v1"
persistencespb "go.temporal.io/server/api/persistence/v1"
workflowspb "go.temporal.io/server/api/workflow/v1"
"go.temporal.io/server/common"
"go.temporal.io/server/common/cluster"
"go.temporal.io/server/common/log"
"go.temporal.io/server/common/metrics"
"go.temporal.io/server/common/payloads"
"go.temporal.io/server/common/persistence"
"go.temporal.io/server/common/primitives"
"go.temporal.io/server/service/history/api"
"go.temporal.io/server/service/history/events"
"go.temporal.io/server/service/history/hsm"
historyi "go.temporal.io/server/service/history/interfaces"
"go.temporal.io/server/service/history/shard"
"go.temporal.io/server/service/history/tests"
"go.temporal.io/server/service/history/workflow"
wcache "go.temporal.io/server/service/history/workflow/cache"
"go.uber.org/mock/gomock"
"google.golang.org/protobuf/types/known/durationpb"
)
var defaultWorkflowTaskCompletionLimits = historyi.WorkflowTaskCompletionLimits{MaxResetPoints: primitives.DefaultHistoryMaxAutoResetPoints, MaxSearchAttributeValueSize: 2048}
type (
VerifyFirstWorkflowTaskScheduledSuite struct {
*require.Assertions
suite.Suite
controller *gomock.Controller
mockEventsCache *events.MockCache
mockExecutionMgr *persistence.MockExecutionManager
shardContext *shard.ContextTest
workflowConsistencyChecker api.WorkflowConsistencyChecker
logger log.Logger
}
)
func TestVerifyFirstWorkflowTaskScheduledSuite(t *testing.T) {
suite.Run(t, new(VerifyFirstWorkflowTaskScheduledSuite))
}
func (s *VerifyFirstWorkflowTaskScheduledSuite) SetupTest() {
s.Assertions = require.New(s.T())
s.controller = gomock.NewController(s.T())
config := tests.NewDynamicConfig()
s.shardContext = shard.NewTestContext(
s.controller,
&persistencespb.ShardInfo{
ShardId: 1,
RangeId: 1,
},
config,
)
reg := hsm.NewRegistry()
err := workflow.RegisterStateMachine(reg)
s.NoError(err)
s.shardContext.SetStateMachineRegistry(reg)
mockNamespaceCache := s.shardContext.Resource.NamespaceCache
mockNamespaceCache.EXPECT().GetNamespaceByID(tests.NamespaceID).Return(tests.LocalNamespaceEntry, nil).AnyTimes()
s.mockExecutionMgr = s.shardContext.Resource.ExecutionMgr
mockClusterMetadata := s.shardContext.Resource.ClusterMetadata
mockClusterMetadata.EXPECT().GetClusterID().Return(int64(1)).AnyTimes()
mockClusterMetadata.EXPECT().GetCurrentClusterName().Return(cluster.TestCurrentClusterName).AnyTimes()
mockClusterMetadata.EXPECT().ClusterNameForFailoverVersion(false, common.EmptyVersion).Return(cluster.TestCurrentClusterName).AnyTimes()
mockClusterMetadata.EXPECT().ClusterNameForFailoverVersion(true, tests.Version).Return(cluster.TestCurrentClusterName).AnyTimes()
s.workflowConsistencyChecker = api.NewWorkflowConsistencyChecker(
s.shardContext,
wcache.NewHostLevelCache(s.shardContext.GetConfig(), s.shardContext.GetLogger(), metrics.NoopMetricsHandler))
s.mockEventsCache = s.shardContext.MockEventsCache
s.mockEventsCache.EXPECT().PutEvent(gomock.Any(), gomock.Any()).AnyTimes()
s.logger = s.shardContext.GetLogger()
}
func (s *VerifyFirstWorkflowTaskScheduledSuite) TearDownTest() {
s.controller.Finish()
}
func (s *VerifyFirstWorkflowTaskScheduledSuite) TestVerifyFirstWorkflowTaskScheduled_WorkflowNotFound() {
request := &historyservice.VerifyFirstWorkflowTaskScheduledRequest{
NamespaceId: tests.NamespaceID.String(),
WorkflowExecution: &commonpb.WorkflowExecution{
WorkflowId: tests.WorkflowID,
RunId: tests.RunID,
},
}
s.mockExecutionMgr.EXPECT().GetWorkflowExecution(gomock.Any(), gomock.Any()).Return(nil, &serviceerror.NotFound{})
err := Invoke(context.Background(), request, s.workflowConsistencyChecker)
s.IsType(&serviceerror.NotFound{}, err)
}
func (s *VerifyFirstWorkflowTaskScheduledSuite) TestVerifyFirstWorkflowTaskScheduled_WorkflowCompleted() {
request := &historyservice.VerifyFirstWorkflowTaskScheduledRequest{
NamespaceId: tests.NamespaceID.String(),
WorkflowExecution: &commonpb.WorkflowExecution{
WorkflowId: tests.WorkflowID,
RunId: tests.RunID,
},
}
ms := workflow.TestGlobalMutableState(s.shardContext, s.mockEventsCache, s.logger, tests.Version, tests.WorkflowID, tests.RunID)
addWorkflowExecutionStartedEventWithParent(ms,
&commonpb.WorkflowExecution{
WorkflowId: tests.WorkflowID,
RunId: tests.RunID,
}, "wType", "testTaskQueue", payloads.EncodeString("input"),
25*time.Second, 20*time.Second, 200*time.Second, nil, "identity")
_, err := ms.AddTimeoutWorkflowEvent(
ms.GetNextEventID(),
enumspb.RETRY_STATE_RETRY_POLICY_NOT_SET,
uuid.NewString(),
)
s.NoError(err)
wfMs := workflow.TestCloneToProto(context.Background(), ms)
gwmsResponse := &persistence.GetWorkflowExecutionResponse{State: wfMs}
s.mockExecutionMgr.EXPECT().GetWorkflowExecution(gomock.Any(), gomock.Any()).Return(gwmsResponse, nil)
err = Invoke(context.Background(), request, s.workflowConsistencyChecker)
s.NoError(err)
}
func (s *VerifyFirstWorkflowTaskScheduledSuite) TestVerifyFirstWorkflowTaskScheduled_WorkflowZombie() {
request := &historyservice.VerifyFirstWorkflowTaskScheduledRequest{
NamespaceId: tests.NamespaceID.String(),
WorkflowExecution: &commonpb.WorkflowExecution{
WorkflowId: tests.WorkflowID,
RunId: tests.RunID,
},
}
ms := workflow.TestGlobalMutableState(s.shardContext, s.mockEventsCache, s.logger, tests.Version, tests.WorkflowID, tests.RunID)
addWorkflowExecutionStartedEventWithParent(ms,
&commonpb.WorkflowExecution{
WorkflowId: tests.WorkflowID,
RunId: tests.RunID,
}, "wType", "testTaskQueue", payloads.EncodeString("input"),
25*time.Second, 20*time.Second, 200*time.Second, nil, "identity")
// zombie state should be treated as open
_, err := ms.UpdateWorkflowStateStatus(
enumsspb.WORKFLOW_EXECUTION_STATE_ZOMBIE,
enumspb.WORKFLOW_EXECUTION_STATUS_RUNNING,
)
s.NoError(err)
wfMs := workflow.TestCloneToProto(context.Background(), ms)
gwmsResponse := &persistence.GetWorkflowExecutionResponse{State: wfMs}
s.mockExecutionMgr.EXPECT().GetWorkflowExecution(gomock.Any(), gomock.Any()).Return(gwmsResponse, nil)
err = Invoke(context.Background(), request, s.workflowConsistencyChecker)
s.IsType(&serviceerror.WorkflowNotReady{}, err)
}
func (s *VerifyFirstWorkflowTaskScheduledSuite) TestVerifyFirstWorkflowTaskScheduled_WorkflowRunning_TaskPending() {
request := &historyservice.VerifyFirstWorkflowTaskScheduledRequest{
NamespaceId: tests.NamespaceID.String(),
WorkflowExecution: &commonpb.WorkflowExecution{
WorkflowId: tests.WorkflowID,
RunId: tests.RunID,
},
}
ms := workflow.TestGlobalMutableState(s.shardContext, s.mockEventsCache, s.logger, tests.Version, tests.WorkflowID, tests.RunID)
addWorkflowExecutionStartedEventWithParent(ms,
&commonpb.WorkflowExecution{
WorkflowId: tests.WorkflowID,
RunId: tests.RunID,
}, "wType", "testTaskQueue", payloads.EncodeString("input"),
25*time.Second, 20*time.Second, 200*time.Second, nil, "identity")
_, _ = ms.AddWorkflowTaskScheduledEvent(false, enumsspb.WORKFLOW_TASK_TYPE_NORMAL)
wfMs := workflow.TestCloneToProto(context.Background(), ms)
gwmsResponse := &persistence.GetWorkflowExecutionResponse{State: wfMs}
s.mockExecutionMgr.EXPECT().GetWorkflowExecution(gomock.Any(), gomock.Any()).Return(gwmsResponse, nil)
err := Invoke(context.Background(), request, s.workflowConsistencyChecker)
s.NoError(err)
}
func (s *VerifyFirstWorkflowTaskScheduledSuite) TestVerifyFirstWorkflowTaskScheduled_WorkflowRunning_TaskProcessed() {
request := &historyservice.VerifyFirstWorkflowTaskScheduledRequest{
NamespaceId: tests.NamespaceID.String(),
WorkflowExecution: &commonpb.WorkflowExecution{
WorkflowId: tests.WorkflowID,
RunId: tests.RunID,
},
}
ms := workflow.TestGlobalMutableState(s.shardContext, s.mockEventsCache, s.logger, tests.Version, tests.WorkflowID, tests.RunID)
addWorkflowExecutionStartedEventWithParent(ms,
&commonpb.WorkflowExecution{
WorkflowId: tests.WorkflowID,
RunId: tests.RunID,
}, "wType", "testTaskQueue", payloads.EncodeString("input"),
25*time.Second, 20*time.Second, 200*time.Second, nil, "identity")
// Schedule WFT
wt, _ := ms.AddWorkflowTaskScheduledEvent(false, enumsspb.WORKFLOW_TASK_TYPE_NORMAL)
// Start WFT
workflowTasksStartEvent, _, _ := ms.AddWorkflowTaskStartedEvent(
wt.ScheduledEventID,
tests.RunID,
&taskqueuepb.TaskQueue{Name: "testTaskQueue"},
uuid.NewString(),
nil,
nil,
nil,
false,
nil,
0,
)
wt.StartedEventID = workflowTasksStartEvent.GetEventId()
// Complete WFT
workflowTask := ms.GetWorkflowTaskByID(wt.ScheduledEventID)
s.NotNil(workflowTask)
s.Equal(wt.StartedEventID, workflowTask.StartedEventID)
_, _ = ms.AddWorkflowTaskCompletedEvent(workflowTask,
&workflowservice.RespondWorkflowTaskCompletedRequest{Identity: "some random identity"}, defaultWorkflowTaskCompletionLimits)
ms.FlushBufferedEvents()
wfMs := workflow.TestCloneToProto(context.Background(), ms)
gwmsResponse := &persistence.GetWorkflowExecutionResponse{State: wfMs}
s.mockExecutionMgr.EXPECT().GetWorkflowExecution(gomock.Any(), gomock.Any()).Return(gwmsResponse, nil)
err := Invoke(context.Background(), request, s.workflowConsistencyChecker)
s.NoError(err)
}
func addWorkflowExecutionStartedEventWithParent(
ms historyi.MutableState,
workflowExecution *commonpb.WorkflowExecution,
workflowType, taskQueue string,
input *commonpb.Payloads,
executionTimeout, runTimeout, taskTimeout time.Duration,
parentInfo *workflowspb.ParentExecutionInfo,
identity string,
) *historypb.HistoryEvent {
startRequest := &workflowservice.StartWorkflowExecutionRequest{
WorkflowId: workflowExecution.WorkflowId,
WorkflowType: &commonpb.WorkflowType{Name: workflowType},
TaskQueue: &taskqueuepb.TaskQueue{Name: taskQueue},
Input: input,
WorkflowExecutionTimeout: durationpb.New(executionTimeout),
WorkflowRunTimeout: durationpb.New(runTimeout),
WorkflowTaskTimeout: durationpb.New(taskTimeout),
Identity: identity,
}
event, _ := ms.AddWorkflowExecutionStartedEvent(
workflowExecution,
&historyservice.StartWorkflowExecutionRequest{
Attempt: 1,
NamespaceId: tests.NamespaceID.String(),
StartRequest: startRequest,
ParentExecutionInfo: parentInfo,
},
)
return event
}