Skip to content

Eliminate TODO stored to error trace from job rescuer #1010

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

Merged
brandur merged 1 commit into from
Aug 14, 2025
Merged

Eliminate TODO stored to error trace from job rescuer #1010

Show file tree
Hide file tree
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
1 change: 1 addition & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,6 +36,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### Fixed

- Cleanly error on invalid schema names in `Config.Schema`. [PR #952](https://github.com/riverqueue/river/pull/952).
- Jobs rescued by `JobRescuer` no longer have their trace set to "TODO". This becomes an empty string instead. [PR #1010](https://github.com/riverqueue/river/pull/1010).

## [0.23.1] - 2025-06-04

Expand Down
4 changes: 2 additions & 2 deletions internal/maintenance/job_rescuer.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -187,8 +187,8 @@ func (s *JobRescuer) runOnce(ctx context.Context) (*rescuerRunOnceResult, error)
errorData, err := json.Marshal(rivertype.AttemptError{
At: now,
Attempt: max(job.Attempt, 0),
Error: "Stuck job rescued by Rescuer",
Trace: "TODO",
Error: "Stuck job rescued by JobRescuer",
Trace: "",
})
if err != nil {
return nil, fmt.Errorf("error marshaling error JSON: %w", err)
Expand Down
65 changes: 35 additions & 30 deletions internal/maintenance/job_rescuer_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -134,7 +134,6 @@ func TestJobRescuer(t *testing.T) {

t.Run("RescuesStuckJobs", func(t *testing.T) {
t.Parallel()
require := require.New(t)

rescuer, bundle := setup(t)

Expand Down Expand Up @@ -163,63 +162,69 @@ func TestJobRescuer(t *testing.T) {
longTimeOutJob1 := testfactory.Job(ctx, t, bundle.exec, &testfactory.JobOpts{Kind: ptrutil.Ptr(rescuerJobKindLongTimeout), State: ptrutil.Ptr(rivertype.JobStateRunning), AttemptedAt: ptrutil.Ptr(bundle.rescueHorizon.Add(-1 * time.Minute)), MaxAttempts: ptrutil.Ptr(5)})
longTimeOutJob2 := testfactory.Job(ctx, t, bundle.exec, &testfactory.JobOpts{Kind: ptrutil.Ptr(rescuerJobKindLongTimeout), State: ptrutil.Ptr(rivertype.JobStateRunning), AttemptedAt: ptrutil.Ptr(bundle.rescueHorizon.Add(-6 * time.Minute)), MaxAttempts: ptrutil.Ptr(5)})

require.NoError(rescuer.Start(ctx))
require.NoError(t, rescuer.Start(ctx))

rescuer.TestSignals.FetchedBatch.WaitOrTimeout()
rescuer.TestSignals.UpdatedBatch.WaitOrTimeout()

confirmRetried := func(jobBefore *rivertype.JobRow) {
jobAfter, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: jobBefore.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(rivertype.JobStateRetryable, jobAfter.State)
require.NoError(t, err)
require.Equal(t, rivertype.JobStateRetryable, jobAfter.State)

require.Len(t, jobAfter.Errors, 1)
attemptError := jobAfter.Errors[0]
require.Zero(t, attemptError.Attempt)
require.Equal(t, "Stuck job rescued by JobRescuer", attemptError.Error)
require.Empty(t, attemptError.Trace)
}

var err error
confirmRetried(stuckToRetryJob1)
confirmRetried(stuckToRetryJob2)

job3After, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: stuckToRetryJob3.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(stuckToRetryJob3.State, job3After.State) // not rescued
require.NoError(t, err)
require.Equal(t, stuckToRetryJob3.State, job3After.State) // not rescued

discardJob1After, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: stuckToDiscardJob1.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(rivertype.JobStateDiscarded, discardJob1After.State)
require.WithinDuration(time.Now(), *discardJob1After.FinalizedAt, 5*time.Second)
require.Len(discardJob1After.Errors, 1)
require.NoError(t, err)
require.Equal(t, rivertype.JobStateDiscarded, discardJob1After.State)
require.WithinDuration(t, time.Now(), *discardJob1After.FinalizedAt, 5*time.Second)
require.Len(t, discardJob1After.Errors, 1)

discardJob2After, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: stuckToDiscardJob2.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(rivertype.JobStateRunning, discardJob2After.State)
require.Nil(discardJob2After.FinalizedAt)
require.NoError(t, err)
require.Equal(t, rivertype.JobStateRunning, discardJob2After.State)
require.Nil(t, discardJob2After.FinalizedAt)

cancelJob1After, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: stuckToCancelJob1.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(rivertype.JobStateCancelled, cancelJob1After.State)
require.WithinDuration(time.Now(), *cancelJob1After.FinalizedAt, 5*time.Second)
require.Len(cancelJob1After.Errors, 1)
require.NoError(t, err)
require.Equal(t, rivertype.JobStateCancelled, cancelJob1After.State)
require.WithinDuration(t, time.Now(), *cancelJob1After.FinalizedAt, 5*time.Second)
require.Len(t, cancelJob1After.Errors, 1)

cancelJob2After, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: stuckToCancelJob2.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(rivertype.JobStateRunning, cancelJob2After.State)
require.Nil(cancelJob2After.FinalizedAt)
require.NoError(t, err)
require.Equal(t, rivertype.JobStateRunning, cancelJob2After.State)
require.Nil(t, cancelJob2After.FinalizedAt)

notRunningJob1After, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: notRunningJob1.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(notRunningJob1.State, notRunningJob1After.State)
require.NoError(t, err)
require.Equal(t, notRunningJob1.State, notRunningJob1After.State)
notRunningJob2After, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: notRunningJob2.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(notRunningJob2.State, notRunningJob2After.State)
require.NoError(t, err)
require.Equal(t, notRunningJob2.State, notRunningJob2After.State)
notRunningJob3After, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: notRunningJob3.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(notRunningJob3.State, notRunningJob3After.State)
require.NoError(t, err)
require.Equal(t, notRunningJob3.State, notRunningJob3After.State)

notTimedOutJob1After, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: longTimeOutJob1.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(rivertype.JobStateRunning, notTimedOutJob1After.State)
require.NoError(t, err)
require.Equal(t, rivertype.JobStateRunning, notTimedOutJob1After.State)
notTimedOutJob2After, err := bundle.exec.JobGetByID(ctx, &riverdriver.JobGetByIDParams{ID: longTimeOutJob2.ID, Schema: rescuer.Config.Schema})
require.NoError(err)
require.Equal(rivertype.JobStateRetryable, notTimedOutJob2After.State)
require.NoError(t, err)
require.Equal(t, rivertype.JobStateRetryable, notTimedOutJob2After.State)
})

t.Run("RescuesInBatches", func(t *testing.T) {
Expand Down
3 changes: 3 additions & 0 deletions rivertype/river_type.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -251,6 +251,9 @@ type AttemptError struct {

// Trace contains a stack trace from a job that panicked. The trace is
// produced by invoking `debug.Trace()`.
//
// In the case of a non-panic or an error produced as a stuck job was
// rescued, this value will be an empty string.
Trace string `json:"trace"`
}

Expand Down
Loading