mirror of
https://github.com/hanzoai/tasks.git
synced 2026-10-30 14:56:19 +00:00
## What Changed
The replication progress cache was reusing
`MutableStateCacheTypeTagValue` as its `cache_type` metrics tag, causing
its `cache_size` gauge to overwrite the mutable state cache's gauge via
Prometheus last-write-wins semantics.
This made `cache_size{cache_type="mutablestate"}` always report `128000`
(the replication progress cache's default size) instead of the actual
mutable state cache capacity — particularly misleading when
`cacheSizeBasedLimit: true` is configured.
## Fix
- Add a new `ReplicationProgressCacheTypeTagValue =
"replication_progress"` constant in `common/metrics/metric_defs.go`
- Use it in `service/history/replication/progress_cache.go` instead of
`MutableStateCacheTypeTagValue`
- Add a unit test verifying the progress cache uses a distinct
`cache_type` tag value
Both caches now report their `cache_size` independently via distinct
`cache_type` label values.
## Why
This is a one-line behavioral change (plus the new constant and test).
It only affects metric label values — no logic, API surface, or
persistence changes.
Fixes #9600
Co-authored-by: Jiechen Zhong <[email protected]>
202 lines
6.5 KiB
Go
202 lines
6.5 KiB
Go
package replication
|
|
|
|
import (
|
|
"math/rand"
|
|
"testing"
|
|
|
|
"github.com/google/uuid"
|
|
"github.com/stretchr/testify/require"
|
|
"github.com/stretchr/testify/suite"
|
|
historyspb "go.temporal.io/server/api/history/v1"
|
|
persistencespb "go.temporal.io/server/api/persistence/v1"
|
|
"go.temporal.io/server/common/metrics"
|
|
"go.temporal.io/server/common/metrics/metricstest"
|
|
"go.temporal.io/server/common/persistence/versionhistory"
|
|
"go.temporal.io/server/common/testing/protorequire"
|
|
"go.temporal.io/server/service/history/shard"
|
|
"go.temporal.io/server/service/history/tests"
|
|
"go.uber.org/mock/gomock"
|
|
)
|
|
|
|
type (
|
|
progressCacheSuite struct {
|
|
suite.Suite
|
|
*require.Assertions
|
|
protorequire.ProtoAssertions
|
|
|
|
controller *gomock.Controller
|
|
mockShard *shard.ContextTest
|
|
|
|
shardContext *shard.ContextTest
|
|
progressCache ProgressCache
|
|
namespaceID string
|
|
workflowID string
|
|
runID string
|
|
}
|
|
)
|
|
|
|
func TestWorkflowCacheSuite(t *testing.T) {
|
|
s := new(progressCacheSuite)
|
|
suite.Run(t, s)
|
|
}
|
|
|
|
func (s *progressCacheSuite) SetupSuite() {
|
|
}
|
|
|
|
func (s *progressCacheSuite) TearDownSuite() {
|
|
}
|
|
|
|
func (s *progressCacheSuite) SetupTest() {
|
|
s.Assertions = require.New(s.T())
|
|
|
|
s.controller = gomock.NewController(s.T())
|
|
s.mockShard = shard.NewTestContext(
|
|
s.controller,
|
|
&persistencespb.ShardInfo{
|
|
ShardId: 0,
|
|
RangeId: 1,
|
|
},
|
|
tests.NewDynamicConfig(),
|
|
)
|
|
|
|
s.mockShard.Resource.ClusterMetadata.EXPECT().IsGlobalNamespaceEnabled().Return(false).AnyTimes()
|
|
|
|
s.shardContext = shard.NewTestContext(
|
|
s.controller,
|
|
&persistencespb.ShardInfo{
|
|
ShardId: 0,
|
|
RangeId: 1,
|
|
Owner: "test-shard-owner",
|
|
},
|
|
tests.NewDynamicConfig(),
|
|
)
|
|
s.progressCache = NewProgressCache(s.shardContext.GetConfig(), s.mockShard.GetLogger(), metrics.NoopMetricsHandler)
|
|
s.namespaceID = tests.NamespaceID.String()
|
|
s.workflowID = uuid.NewString()
|
|
s.runID = uuid.NewString()
|
|
|
|
}
|
|
|
|
func (s *progressCacheSuite) TearDownTest() {
|
|
s.controller.Finish()
|
|
s.mockShard.StopForTest()
|
|
}
|
|
|
|
func (s *progressCacheSuite) TestProgressCache() {
|
|
targetClusterID := rand.Int31()
|
|
firstEventID := int64(999)
|
|
versionedTransition := &persistencespb.VersionedTransition{
|
|
NamespaceFailoverVersion: 80,
|
|
TransitionCount: 10,
|
|
}
|
|
versionedTransitions := []*persistencespb.VersionedTransition{versionedTransition}
|
|
versionHistoryItems := []*historyspb.VersionHistoryItem{
|
|
versionhistory.NewVersionHistoryItem(firstEventID, versionedTransition.NamespaceFailoverVersion),
|
|
}
|
|
expected := &ReplicationProgress{
|
|
versionedTransitions: [][]*persistencespb.VersionedTransition{versionedTransitions},
|
|
eventVersionHistoryItems: [][]*historyspb.VersionHistoryItem{versionHistoryItems},
|
|
lastVersionTransitionIndex: 0,
|
|
}
|
|
|
|
// get non-existing progress
|
|
cachedProgress := s.progressCache.Get(s.runID, targetClusterID)
|
|
s.Nil(cachedProgress)
|
|
|
|
err := s.progressCache.Update(s.runID, targetClusterID, versionedTransitions, versionHistoryItems)
|
|
s.Nil(err)
|
|
|
|
// get existing progress
|
|
cachedProgress = s.progressCache.Get(s.runID, targetClusterID)
|
|
s.DeepEqual(expected, cachedProgress)
|
|
|
|
// update existing versioned transition and version history
|
|
versionedTransitions2 := []*persistencespb.VersionedTransition{
|
|
{
|
|
NamespaceFailoverVersion: 80,
|
|
TransitionCount: 20,
|
|
},
|
|
}
|
|
versionHistoryItems2 := []*historyspb.VersionHistoryItem{
|
|
versionhistory.NewVersionHistoryItem(firstEventID+1, versionedTransition.NamespaceFailoverVersion),
|
|
}
|
|
err = s.progressCache.Update(s.runID, targetClusterID, versionedTransitions2, versionHistoryItems2)
|
|
s.Nil(err)
|
|
|
|
expected2 := &ReplicationProgress{
|
|
versionedTransitions: [][]*persistencespb.VersionedTransition{versionedTransitions2},
|
|
eventVersionHistoryItems: [][]*historyspb.VersionHistoryItem{versionHistoryItems2},
|
|
lastVersionTransitionIndex: 0,
|
|
lastEventVersionHistoryIndex: 0,
|
|
}
|
|
cachedProgress = s.progressCache.Get(s.runID, targetClusterID)
|
|
s.DeepEqual(expected2, cachedProgress)
|
|
|
|
// add new versioned transition and version history
|
|
versionedTransitions3 := []*persistencespb.VersionedTransition{
|
|
{
|
|
NamespaceFailoverVersion: 90,
|
|
TransitionCount: 15,
|
|
},
|
|
}
|
|
versionHistoryItems3 := []*historyspb.VersionHistoryItem{
|
|
versionhistory.NewVersionHistoryItem(firstEventID, versionedTransition.NamespaceFailoverVersion),
|
|
versionhistory.NewVersionHistoryItem(firstEventID+1, versionedTransition.NamespaceFailoverVersion+1),
|
|
}
|
|
err = s.progressCache.Update(s.runID, targetClusterID, versionedTransitions3, versionHistoryItems3)
|
|
s.Nil(err)
|
|
|
|
expected3 := &ReplicationProgress{
|
|
versionedTransitions: [][]*persistencespb.VersionedTransition{versionedTransitions2, versionedTransitions3},
|
|
eventVersionHistoryItems: [][]*historyspb.VersionHistoryItem{versionHistoryItems2, versionHistoryItems3},
|
|
lastVersionTransitionIndex: 1,
|
|
lastEventVersionHistoryIndex: 1,
|
|
}
|
|
cachedProgress = s.progressCache.Get(s.runID, targetClusterID)
|
|
s.DeepEqual(expected3, cachedProgress)
|
|
|
|
// noop update: versioned transition and version history are already included in the existing progress
|
|
err = s.progressCache.Update(s.runID, targetClusterID, versionedTransitions, versionHistoryItems)
|
|
s.Nil(err)
|
|
|
|
cachedProgress = s.progressCache.Get(s.runID, targetClusterID)
|
|
s.DeepEqual(expected3, cachedProgress)
|
|
}
|
|
|
|
func TestProgressCacheUsesDistinctMetricsTag(t *testing.T) {
|
|
// Verify that the progress cache uses its own cache_type tag value
|
|
// ("replication_progress") rather than the mutable state cache's tag value
|
|
// ("mutablestate"), preventing Prometheus gauge collision. See #9600.
|
|
ctrl := gomock.NewController(t)
|
|
mockShard := shard.NewTestContext(
|
|
ctrl,
|
|
&persistencespb.ShardInfo{
|
|
ShardId: 0,
|
|
RangeId: 1,
|
|
},
|
|
tests.NewDynamicConfig(),
|
|
)
|
|
defer ctrl.Finish()
|
|
defer mockShard.StopForTest()
|
|
|
|
metricsHandler := metricstest.NewCaptureHandler()
|
|
capture := metricsHandler.StartCapture()
|
|
|
|
NewProgressCache(mockShard.GetConfig(), mockShard.GetLogger(), metricsHandler)
|
|
|
|
snapshot := capture.Snapshot()
|
|
cacheSizeRecordings := snapshot[metrics.CacheSize.Name()]
|
|
require.NotEmpty(t, cacheSizeRecordings, "expected cache_size metric to be recorded")
|
|
|
|
// Verify the cache_type tag is "replication_progress", not "mutablestate"
|
|
found := false
|
|
for _, recording := range cacheSizeRecordings {
|
|
if tag, ok := recording.Tags[metrics.CacheTypeTagName]; ok {
|
|
require.Equal(t, metrics.ReplicationProgressCacheTypeTagValue, tag,
|
|
"progress cache should use ReplicationProgressCacheTypeTagValue, not MutableStateCacheTypeTagValue")
|
|
found = true
|
|
}
|
|
}
|
|
require.True(t, found, "expected cache_size metric with cache_type tag")
|
|
}
|