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

CI: fix onlineddl_scheduler flakiness #13754

Merged
Merged
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
66 changes: 38 additions & 28 deletions go/test/endtoend/onlineddl/scheduler/onlineddl_scheduler_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -168,6 +168,38 @@ func TestParseTableName(t *testing.T) {
}
}

func waitForReadyToComplete(t *testing.T, uuid string, expected bool) {
ctx, cancel := context.WithTimeout(context.Background(), normalWaitTime)
defer cancel()

ticker := time.NewTicker(time.Second)
defer ticker.Stop()
for {

rs := onlineddl.ReadMigrations(t, &vtParams, uuid)
require.NotNil(t, rs)
for _, row := range rs.Named().Rows {
readyToComplete := row.AsInt64("ready_to_complete", 0)
if expected == (readyToComplete > 0) {
// all good. This is what we waited for
if expected {
// if migration is ready to complete, the nthe timestamp should be non-null
assert.False(t, row["ready_to_complete_timestamp"].IsNull())
} else {
assert.True(t, row["ready_to_complete_timestamp"].IsNull())
}

return
}
}
select {
case <-ticker.C:
case <-ctx.Done():
}
require.NoError(t, ctx.Err())
}
}

func TestMain(m *testing.M) {
defer cluster.PanicHandler(nil)
flag.Parse()
Expand Down Expand Up @@ -561,13 +593,7 @@ func testScheduler(t *testing.T) {

t.Run("check ready to complete (before)", func(t *testing.T) {
for _, uuid := range []string{t1uuid, t2uuid} {
rs := onlineddl.ReadMigrations(t, &vtParams, uuid)
require.NotNil(t, rs)
for _, row := range rs.Named().Rows {
readyToComplete := row.AsInt64("ready_to_complete", 0)
assert.Equal(t, int64(0), readyToComplete)
assert.True(t, row["ready_to_complete_timestamp"].IsNull())
}
waitForReadyToComplete(t, uuid, false)
}
})
t.Run("unthrottle, expect t2 running", func(t *testing.T) {
Expand Down Expand Up @@ -599,13 +625,7 @@ func testScheduler(t *testing.T) {
})
t.Run("check ready to complete (after)", func(t *testing.T) {
for _, uuid := range []string{t1uuid, t2uuid} {
rs := onlineddl.ReadMigrations(t, &vtParams, uuid)
require.NotNil(t, rs)
for _, row := range rs.Named().Rows {
readyToComplete := row.AsInt64("ready_to_complete", 0)
assert.Equal(t, int64(1), readyToComplete)
assert.False(t, row["ready_to_complete_timestamp"].IsNull())
}
waitForReadyToComplete(t, uuid, true)
}
})

Expand All @@ -614,21 +634,16 @@ func testScheduler(t *testing.T) {
onlineddl.CheckThrottledApps(t, &vtParams, throttlerapp.OnlineDDLName, false)

t.Run("REVERT both tables concurrent, postponed", func(t *testing.T) {
t1uuid = testRevertMigration(t, createRevertParams(t1uuid, ddlStrategy+" -allow-concurrent -postpone-completion", "vtgate", "", true))
t2uuid = testRevertMigration(t, createRevertParams(t2uuid, ddlStrategy+" -allow-concurrent -postpone-completion", "vtgate", "", true))
t1uuid = testRevertMigration(t, createRevertParams(t1uuid, ddlStrategy+" --allow-concurrent --postpone-completion", "vtgate", "", true))
t2uuid = testRevertMigration(t, createRevertParams(t2uuid, ddlStrategy+" --allow-concurrent --postpone-completion", "vtgate", "", true))

testAllowConcurrent(t, "t1", t1uuid, 1)
t.Run("expect both migrations to run", func(t *testing.T) {
onlineddl.WaitForMigrationStatus(t, &vtParams, shards, t1uuid, normalWaitTime, schema.OnlineDDLStatusRunning)
onlineddl.WaitForMigrationStatus(t, &vtParams, shards, t2uuid, normalWaitTime, schema.OnlineDDLStatusRunning)
})
t.Run("test ready-to-complete", func(t *testing.T) {
rs := onlineddl.ReadMigrations(t, &vtParams, t1uuid)
require.NotNil(t, rs)
for _, row := range rs.Named().Rows {
readyToComplete := row.AsInt64("ready_to_complete", 0)
assert.Equal(t, int64(1), readyToComplete)
}
waitForReadyToComplete(t, t1uuid, true)
})
t.Run("complete t2", func(t *testing.T) {
// now that both are running, let's unblock t2. We expect it to complete.
Expand Down Expand Up @@ -766,12 +781,7 @@ func testScheduler(t *testing.T) {
onlineddl.CheckMigrationStatus(t, &vtParams, shards, drop1uuid, schema.OnlineDDLStatusReady)
})
t.Run("t3 ready to complete", func(t *testing.T) {
rs := onlineddl.ReadMigrations(t, &vtParams, drop1uuid)
require.NotNil(t, rs)
for _, row := range rs.Named().Rows {
readyToComplete := row.AsInt64("ready_to_complete", 0)
assert.Equal(t, int64(1), readyToComplete)
}
waitForReadyToComplete(t, drop1uuid, true)
})
t.Run("t3drop complete", func(t *testing.T) {
// drop3 migration should not block. It can run concurrently to t1, and does not conflict
Expand Down