zitadel/internal/eventstore/spooler/spooler.go
Silvan a6c4702b8e
fix: lock again (#1132)
* start sub

* start implement subsciptions

* start subscription

* implementation for member done

* admin done

* fix: tests

* extend handlers

* prepary notification

* no errors in adminapi

* changed current sequence in all packages

* ignore mocks

* works

* subscriptions as singleton

* tests

* refactor: rename function scope var

* fix: process ALL previous sequences

* fix: spooler and pubsub

* handler check

* fix: process events until all done

* fix break on query err

* fix: handler

* fix: process sequence or return error

* check aggregate id

* fix: log only in error case

* fix tests

* fix: handlers

* fix: spooler

* fix: spooler

* fix: tests

* fix: continue

* fix: locker duration

* fix: variable lock duration

* fix: test

* fix: test

* fix: test min max time

Co-authored-by: Livio Amstutz <livio.a@gmail.com>
2020-12-22 12:27:55 +01:00

207 lines
5.2 KiB
Go

package spooler
import (
"context"
"strconv"
"sync"
"time"
"github.com/caos/logging"
"github.com/caos/zitadel/internal/eventstore"
"github.com/caos/zitadel/internal/eventstore/models"
"github.com/caos/zitadel/internal/eventstore/query"
"github.com/caos/zitadel/internal/telemetry/tracing"
"github.com/caos/zitadel/internal/view/repository"
)
type Spooler struct {
handlers []query.Handler
locker Locker
lockID string
eventstore eventstore.Eventstore
workers int
queue chan *spooledHandler
}
type Locker interface {
Renew(lockerID, viewModel string, waitTime time.Duration) error
}
type spooledHandler struct {
query.Handler
locker Locker
queuedAt time.Time
eventstore eventstore.Eventstore
}
func (s *Spooler) Start() {
defer logging.LogWithFields("SPOOL-N0V1g", "lockerID", s.lockID, "workers", s.workers).Info("spooler started")
if s.workers < 1 {
return
}
for i := 0; i < s.workers; i++ {
go func(workerIdx int) {
workerID := s.lockID + "--" + strconv.Itoa(workerIdx)
for task := range s.queue {
go requeueTask(task, s.queue)
task.load(workerID)
}
}(i)
}
go func() {
for _, handler := range s.handlers {
s.queue <- &spooledHandler{Handler: handler, locker: s.locker, queuedAt: time.Now(), eventstore: s.eventstore}
}
}()
}
func requeueTask(task *spooledHandler, queue chan<- *spooledHandler) {
time.Sleep(task.MinimumCycleDuration() - time.Since(task.queuedAt))
task.queuedAt = time.Now()
queue <- task
}
func (s *spooledHandler) load(workerID string) {
errs := make(chan error)
defer close(errs)
ctx, cancel := context.WithCancel(context.Background())
go s.awaitError(cancel, errs, workerID)
hasLocked := s.lock(ctx, errs, workerID)
if <-hasLocked {
for {
events, err := s.query(ctx)
if err != nil {
errs <- err
break
}
err = s.process(ctx, events, workerID)
if err != nil {
errs <- err
break
}
if uint64(len(events)) < s.QueryLimit() {
// no more events to process
// stop chan
if ctx.Err() == nil {
errs <- nil
}
break
}
}
}
<-ctx.Done()
}
func (s *spooledHandler) awaitError(cancel func(), errs chan error, workerID string) {
select {
case err := <-errs:
cancel()
logging.Log("SPOOL-OT8di").OnError(err).WithField("view", s.ViewModel()).WithField("worker", workerID).Debug("load canceled")
}
}
func (s *spooledHandler) process(ctx context.Context, events []*models.Event, workerID string) error {
for i, event := range events {
select {
case <-ctx.Done():
logging.LogWithFields("SPOOL-FTKwH", "view", s.ViewModel(), "worker", workerID, "traceID", tracing.TraceIDFromCtx(ctx)).Debug("context canceled")
return nil
default:
if err := s.Reduce(event); err != nil {
err = s.OnError(event, err)
if err == nil {
continue
}
time.Sleep(100 * time.Millisecond)
return s.process(ctx, events[i:], workerID)
}
}
}
err := s.OnSuccess()
logging.LogWithFields("SPOOL-49ods", "view", s.ViewModel(), "worker", workerID, "traceID", tracing.TraceIDFromCtx(ctx)).OnError(err).Warn("could not process on success func")
return err
}
func (s *spooledHandler) query(ctx context.Context) ([]*models.Event, error) {
query, err := s.EventQuery()
if err != nil {
return nil, err
}
factory := models.FactoryFromSearchQuery(query)
sequence, err := s.eventstore.LatestSequence(ctx, factory)
logging.Log("SPOOL-7SciK").OnError(err).WithField("traceID", tracing.TraceIDFromCtx(ctx)).Debug("unable to query latest sequence")
var processedSequence uint64
for _, filter := range query.Filters {
if filter.GetField() == models.Field_LatestSequence {
processedSequence = filter.GetValue().(uint64)
}
}
if sequence != 0 && processedSequence == sequence {
return nil, nil
}
query.Limit = s.QueryLimit()
return s.eventstore.FilterEvents(ctx, query)
}
//lock ensures the lock on the database.
// the returned channel will be closed if ctx is done or an error occured durring lock
func (s *spooledHandler) lock(ctx context.Context, errs chan<- error, workerID string) chan bool {
renewTimer := time.After(0)
locked := make(chan bool)
go func(locked chan bool) {
var firstLock sync.Once
defer close(locked)
for {
select {
case <-ctx.Done():
return
case <-renewTimer:
err := s.locker.Renew(workerID, s.ViewModel(), s.LockDuration())
firstLock.Do(func() {
locked <- err == nil
})
if err == nil {
renewTimer = time.After(s.LockDuration())
continue
}
if ctx.Err() == nil {
errs <- err
}
return
}
}
}(locked)
return locked
}
func HandleError(event *models.Event, failedErr error,
latestFailedEvent func(sequence uint64) (*repository.FailedEvent, error),
processFailedEvent func(*repository.FailedEvent) error,
processSequence func(*models.Event) error,
errorCountUntilSkip uint64) error {
failedEvent, err := latestFailedEvent(event.Sequence)
if err != nil {
return err
}
failedEvent.FailureCount++
failedEvent.ErrMsg = failedErr.Error()
err = processFailedEvent(failedEvent)
if err != nil {
return err
}
if errorCountUntilSkip <= failedEvent.FailureCount {
return processSequence(event)
}
return failedErr
}
func HandleSuccess(updateSpoolerRunTimestamp func() error) error {
return updateSpoolerRunTimestamp()
}