mirror of
https://github.com/milvus-io/milvus.git
synced 2026-07-21 02:05:41 +00:00
issue: #47420 ## What this PR does Project-wide migration of raw `fmt.Errorf` / `errors.New` in function bodies onto the `merr` framework, plus the Sys-vs-Input error classification and the machinery it drives (retriability, fine-grained metrics, segcore unification), plus the convention docs and a linter that keeps it from regressing. Scope: storage, proxy, coordinators (root/data/query), query node, data node, `pkg/util` & `internal/util`, expression parser, message queue, streaming, and misc packages. Bare raw-error usages went from ~3000 to a ~340 allowlist (package-level sentinels / build-tag / test sites). --- ## How to review this PR It is large but the vast majority is mechanical. Changes fall into three tiers; spend review budget on Part 2 and Part 3. ### Part 1 — Mechanical standardization (low risk, verify by rule) Each converted call follows one of a small fixed set of rules. To review, check that each site obeys the matching rule rather than reading every line: | Pattern | Rule | |---|---| | `fmt.Errorf("...")` originating a new error | → `merr.WrapErrXxxMsg("...")` with a code matching the failure's meaning | | Adding context to an existing typed error | → `merr.Wrap(err, "...")` / `merr.Wrapf(...)` — **preserves** the inner code (never `WrapErr*Err`, which overwrites it) | | Errors inside the streaming subsystem | → `status.New*` factories (StreamingError), **not** merr — this is the component-internal dialect (see `docs/dev/error_handling_guide.md`) | | Low-level / control-flow signal caught by `errors.Is` | → kept as a package-level `errors.New` sentinel (lowercase, same-package) | Conventions are documented in `docs/dev/error_handling_guide.md` (how-to) and `docs/dev/error_sentinel_convention.md` (rules + audit). A `gocritic`/`ruleguard` rule (`rawmerrerror`, in `rules.go`) enforces "no raw `return errors.New/fmt.Errorf`" under `make verifiers`. ### Part 2 — Behavior changes (review these closely) These are the sites where the wire contract or runtime behavior changes, not just the source text. Listed by category; representative locations given, full set in the diff. **A. gRPC wire-code shifts: `UnexpectedError(1)/Code 65535` → typed code.** Where a handler previously returned a raw error (collapsed to `Code=65535` on the wire), it now returns a typed merr, so the client sees a real code. The most common shift is to `IllegalArgument(5)/Code 1100` (ParameterInvalid). Touch points include datanode task handlers (CreateTask/Query/Drop), proxy Upsert, querynode GetMetrics, datacoord CreateIndex, httpserver query-response builder, and typeutil schema validation. One code refinement: an index-param validation moved `1100` → `1101` (ParameterMissing). **Client/SDK assertions and any code that switched on `Code=65535` for these paths must be re-checked** (the go_client e2e assertions were already aligned in this PR). **B. Prometheus `status` label contract change (externally visible).** The proxy metric's coarse `fail` / `rejected` values are split into `fail_input` / `fail_system` and `rejected_user` / `rejected_system` (in `requestutil.ParseMetricLabel`; auth/privilege rejections count as `rejected_user`), so dashboards can attribute a failure to caller vs operator. **Dashboards/alerts querying `status="fail"` must migrate to `status=~"fail_.*"`, and `status="rejected"` to `status=~"rejected_.*"`.** The in-repo Grafana dashboard is already migrated; external dashboards built on the old values silently go empty after upgrade. This is the one change that requires an ops-side migration. **C. Retriability semantics.** - C1: `merr.Status(err)` now forces `Retriable=false` when the error is an `InputError` — a malformed request can never succeed on blind retry, so clients never get the self-contradictory "your input is wrong but you may retry". - C2: `retry.Do` short-circuits an `InputError` (non-retriable) — **but only when the caller did not pass a `RetryErr` predicate**. The check is an `if c.isRetryErr != nil { ... } else if InputError { ... }` *mutually exclusive* branch (`pkg/util/retry/retry.go`): an explicit `RetryErr` takes precedence and bypasses the InputError abort. `retry.Handle` deliberately does **not** apply the InputError abort (its callers signal abort via `shouldRetry=false`). Four flusher startup callsites that must retry through transient "not ready" errors were given explicit `RetryErr` escape hatches. **D. segcore (C++→Go) error classification.** A single shared Go-side table (`pkg/util/merr/segcore.go`) maps each segcore code to a merr sentinel + InputError/signal category, replacing scattered hand-written `if errorCode == ...` switches in the cgo wrappers. **Wire `Code` values change for every segcore pass-through error, not just the remapped ones.** Named sentinels remap (C++ `2003` → merr `2001`, `2033` → `2002`, Folly/Knowhere codes likewise); **all remaining pass-through codes (`2004`–`2043`, previously surfaced to clients as raw C++ enum values) now serialize as `2000`** (`ErrSegcore`), with the original C++ code preserved in the `Reason` text (`segcoreCode=...`); unknown/future codes collapse to `2000` as well (pinned by the `wire_code_projection` test). Transient segcore classes (object storage / file IO / OOM / mmap / FieldNotLoaded — 11 codes) now report `Retriable=true`. **Any client switching on raw segcore codes in the `2004`–`2043` range must be re-checked**; the in-Reason code remains available for diagnostics. Signal codes (PretendFinished / FollyCancel) are recognized centrally. `errors.Is`-based control flow on these (e.g. scheduler skip/retry) is preserved. **E. InputError classification (25 sentinels + dynamic marks).** 25 sentinels in `errors.go` carry `WithErrorType(InputError)` (the Collection / ResourceGroup / Database families, `ErrIndexDuplicate`, `ErrParameterInvalid`, `ErrPrivilegeNotAuthenticated`, `ErrImportFailed`, `ErrQueryPlan`, ...), plus dynamic marks for the 8 segcore input codes (ExprInvalid, DimNotMatch, MetricTypeInvalid, FieldIDInvalid, ...) and `WrapErrAsInputError`. The widest blast radius is `ErrParameterInvalid` (1100): ~2335 `WrapErrParameterInvalid*` callsites now classify as input / non-retriable. Because of C1/C2 this changes retriability for any path that returns these. **The audit to confirm no transient path was mis-marked is the single most important review item** (see Part 3). One reverse correction: storage field-stats parsing moved from `ErrParameterInvalid` (input) to `ErrDataIntegrity` — a corrupted stored stat is data corruption, not user input. ### Part 3 — Known risks & traps (called out proactively) 1. **`merr.Wrap` vs `WrapErr*Err` (code-masking).** `WrapErr*Err` builds a `wrappedMilvusError{sentinel: ErrServiceInternal}` whose `code()` returns the *outer* sentinel — it overwrites the inner typed code and hides the `errors.Is` chain. This is intentional (use it to *deliberately* downgrade), but it was a recurring conversion defect; the rule "add context with `merr.Wrap`, downgrade with `WrapErr*Err`" is enforced by convention and reviewed across the diff. 2. **InputError × `retry.Do` blast radius.** Marking a sentinel `InputError` makes any `retry.Do(...)` without a `RetryErr` predicate stop retrying it. Reviewers should sanity-check that no transient use of the 19 newly-marked sentinels (especially `ErrParameterInvalid`) sits inside a retry loop that needed to keep spinning. The known flusher cases were handled (see C2). 3. **The ~340 raw-error allowlist.** What remains as bare `errors.New` is, by design: package-level sentinels (caught by `errors.Is`), `//go:build test` sites, and out-of-band trees (`cmd/`, `tests/`, codegen, walimpls). The linter only bans the *direct-return* form; assignment-then-return escapes and the full no-exceptions ban are deferred to an AST-based linter (Tier 2, documented). 4. **segcore C++ second step deferred.** This PR unifies classification on the Go side; splitting the dual-semantic C++ codes at the source is a follow-up. --- ## Validation - `make verifiers`: Go side clean (gofmt + static-check across modules, including the new `rawmerrerror` rule with a 0-hit baseline repo-wide). - `make test-go`: passing; the one real regression introduced (a datanode `invalid_task_type` assertion shifting `1` → `5` from a ParameterInvalid conversion) was fixed in-tree. - go_client e2e CreateIndex assertions aligned to the new merr messages. --------- Signed-off-by: zhenshan.cao <zhenshan.cao@zilliz.com> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
453 lines
17 KiB
Go
453 lines
17 KiB
Go
package datacoord
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"time"
|
|
|
|
"github.com/cockroachdb/errors"
|
|
"github.com/samber/lo"
|
|
"go.uber.org/atomic"
|
|
"go.uber.org/zap"
|
|
"google.golang.org/protobuf/proto"
|
|
|
|
"github.com/milvus-io/milvus/internal/compaction"
|
|
"github.com/milvus-io/milvus/internal/datacoord/allocator"
|
|
"github.com/milvus-io/milvus/internal/datacoord/session"
|
|
"github.com/milvus-io/milvus/internal/metastore/kv/binlog"
|
|
"github.com/milvus-io/milvus/internal/util/fileresource"
|
|
"github.com/milvus-io/milvus/pkg/v3/log"
|
|
"github.com/milvus-io/milvus/pkg/v3/metrics"
|
|
"github.com/milvus-io/milvus/pkg/v3/proto/datapb"
|
|
"github.com/milvus-io/milvus/pkg/v3/taskcommon"
|
|
"github.com/milvus-io/milvus/pkg/v3/util/merr"
|
|
"github.com/milvus-io/milvus/pkg/v3/util/paramtable"
|
|
)
|
|
|
|
var _ CompactionTask = (*mixCompactionTask)(nil)
|
|
|
|
type mixCompactionTask struct {
|
|
taskProto atomic.Value // *datapb.CompactionTask
|
|
|
|
allocator allocator.Allocator
|
|
meta CompactionMeta
|
|
|
|
ievm IndexEngineVersionManager
|
|
|
|
times *taskcommon.Times
|
|
|
|
slotUsage atomic.Int64
|
|
}
|
|
|
|
func (t *mixCompactionTask) GetTaskID() int64 {
|
|
return t.GetTaskProto().GetPlanID()
|
|
}
|
|
|
|
func (t *mixCompactionTask) GetTaskType() taskcommon.Type {
|
|
return taskcommon.Compaction
|
|
}
|
|
|
|
func (t *mixCompactionTask) GetTaskState() taskcommon.State {
|
|
return taskcommon.FromCompactionState(t.GetTaskProto().GetState())
|
|
}
|
|
|
|
func (t *mixCompactionTask) GetTaskSlot() int64 {
|
|
slotUsage := t.slotUsage.Load()
|
|
if slotUsage == 0 {
|
|
slotUsage = paramtable.Get().DataCoordCfg.MixCompactionSlotUsage.GetAsInt64()
|
|
if t.GetTaskProto().GetType() == datapb.CompactionType_SortCompaction {
|
|
segment := t.meta.GetHealthySegment(context.Background(), t.GetTaskProto().GetInputSegments()[0])
|
|
if segment != nil {
|
|
segSize := segment.getSegmentSize()
|
|
slotUsage = calculateStatsTaskSlot(segSize)
|
|
log.Info("mixCompactionTask get task slot",
|
|
zap.Int64("segment size", segSize), zap.Int64("task slot", slotUsage))
|
|
}
|
|
}
|
|
t.slotUsage.Store(slotUsage)
|
|
}
|
|
return slotUsage
|
|
}
|
|
|
|
func (t *mixCompactionTask) SetTaskTime(timeType taskcommon.TimeType, time time.Time) {
|
|
t.times.SetTaskTime(timeType, time)
|
|
}
|
|
|
|
func (t *mixCompactionTask) GetTaskTime(timeType taskcommon.TimeType) time.Time {
|
|
return timeType.GetTaskTime(t.times)
|
|
}
|
|
|
|
func (t *mixCompactionTask) GetTaskVersion() int64 {
|
|
return int64(t.GetTaskProto().GetRetryTimes())
|
|
}
|
|
|
|
func (t *mixCompactionTask) CreateTaskOnWorker(nodeID int64, cluster session.Cluster) {
|
|
log := log.With(zap.Int64("triggerID", t.GetTaskProto().GetTriggerID()),
|
|
zap.Int64("PlanID", t.GetTaskProto().GetPlanID()),
|
|
zap.Int64("collectionID", t.GetTaskProto().GetCollectionID()),
|
|
zap.Int64("nodeID", nodeID))
|
|
|
|
plan, err := t.BuildCompactionRequest()
|
|
if err != nil {
|
|
log.Warn("mixCompactionTask failed to build compaction request", zap.Error(err))
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_failed), setFailReason(err.Error()))
|
|
if err != nil {
|
|
log.Warn("mixCompactionTask failed to updateAndSaveTaskMeta", zap.Error(err))
|
|
}
|
|
return
|
|
}
|
|
|
|
err = cluster.CreateCompaction(nodeID, plan, t.GetTaskProto().GetCollectionID())
|
|
if err != nil {
|
|
// Compaction tasks may be refused by DataNode because of slot limit. In this case, the node id is reset
|
|
// to enable a retry in compaction.checkCompaction().
|
|
// This is tricky, we should remove the reassignment here.
|
|
originNodeID := t.GetTaskProto().GetNodeID()
|
|
log.Warn("mixCompactionTask failed to notify compaction tasks to DataNode",
|
|
zap.Int64("planID", t.GetTaskProto().GetPlanID()),
|
|
zap.Int64("nodeID", originNodeID),
|
|
zap.Error(err))
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_pipelining), setNodeID(NullNodeID))
|
|
if err != nil {
|
|
log.Warn("mixCompactionTask failed to updateAndSaveTaskMeta", zap.Error(err))
|
|
}
|
|
metrics.DataCoordCompactionTaskNum.WithLabelValues(fmt.Sprintf("%d", originNodeID), t.GetTaskProto().GetType().String(), metrics.Executing).Dec()
|
|
metrics.DataCoordCompactionTaskNum.WithLabelValues(fmt.Sprintf("%d", NullNodeID), t.GetTaskProto().GetType().String(), metrics.Pending).Inc()
|
|
return
|
|
}
|
|
log.Info("mixCompactionTask notify compaction tasks to DataNode")
|
|
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_executing), setNodeID(nodeID))
|
|
if err != nil {
|
|
log.Warn("mixCompactionTask failed to updateAndSaveTaskMeta", zap.Error(err))
|
|
}
|
|
}
|
|
|
|
func (t *mixCompactionTask) QueryTaskOnWorker(cluster session.Cluster) {
|
|
log := log.With(zap.Int64("triggerID", t.GetTaskProto().GetTriggerID()),
|
|
zap.Int64("PlanID", t.GetTaskProto().GetPlanID()),
|
|
zap.Int64("collectionID", t.GetTaskProto().GetCollectionID()))
|
|
result, err := cluster.QueryCompaction(t.GetTaskProto().GetNodeID(), &datapb.CompactionStateRequest{
|
|
PlanID: t.GetTaskProto().GetPlanID(),
|
|
})
|
|
if err != nil || result == nil {
|
|
log.Warn("mixCompactionTask failed to get compaction result", zap.Error(err))
|
|
if err := t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_pipelining), setNodeID(NullNodeID)); err != nil {
|
|
log.Warn("mixCompactionTask failed to updateAndSaveTaskMeta", zap.Error(err))
|
|
}
|
|
return
|
|
}
|
|
switch result.GetState() {
|
|
case datapb.CompactionTaskState_completed:
|
|
if len(result.GetSegments()) == 0 {
|
|
log.Info("compaction result is empty, all data may have been deleted")
|
|
}
|
|
err = t.meta.ValidateSegmentStateBeforeCompleteCompactionMutation(t.GetTaskProto())
|
|
if err != nil {
|
|
t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_failed), setFailReason(err.Error()))
|
|
return
|
|
}
|
|
if err := t.saveSegmentMeta(result); err != nil {
|
|
log.Warn("mixCompactionTask failed to save segment meta", zap.Error(err))
|
|
if errors.Is(err, merr.ErrIllegalCompactionPlan) {
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_failed))
|
|
if err != nil {
|
|
log.Warn("mixCompactionTask failed to setState failed", zap.Error(err))
|
|
}
|
|
}
|
|
return
|
|
}
|
|
UpdateCompactionSegmentSizeMetrics(result.GetSegments())
|
|
t.processMetaSaved()
|
|
case datapb.CompactionTaskState_pipelining, datapb.CompactionTaskState_executing:
|
|
return
|
|
case datapb.CompactionTaskState_timeout:
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_timeout))
|
|
if err != nil {
|
|
log.Warn("update clustering compaction task meta failed", zap.Error(err))
|
|
return
|
|
}
|
|
case datapb.CompactionTaskState_failed:
|
|
log.Info("mixCompactionTask fail in datanode")
|
|
err := t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_failed))
|
|
if err != nil {
|
|
log.Warn("fail to updateAndSaveTaskMeta")
|
|
}
|
|
default:
|
|
log.Error("not support compaction task state", zap.String("state", result.GetState().String()))
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_failed))
|
|
if err != nil {
|
|
log.Warn("update clustering compaction task meta failed", zap.Error(err))
|
|
return
|
|
}
|
|
}
|
|
}
|
|
|
|
func (t *mixCompactionTask) DropTaskOnWorker(cluster session.Cluster) {
|
|
log := log.With(zap.Int64("triggerID", t.GetTaskProto().GetTriggerID()),
|
|
zap.Int64("PlanID", t.GetTaskProto().GetPlanID()),
|
|
zap.Int64("collectionID", t.GetTaskProto().GetCollectionID()))
|
|
if err := cluster.DropCompaction(t.GetTaskProto().GetNodeID(), t.GetTaskProto().GetPlanID()); err != nil {
|
|
log.Warn("mixCompactionTask processCompleted unable to drop compaction plan")
|
|
}
|
|
}
|
|
|
|
func (t *mixCompactionTask) GetTaskProto() *datapb.CompactionTask {
|
|
task := t.taskProto.Load()
|
|
if task == nil {
|
|
return nil
|
|
}
|
|
return task.(*datapb.CompactionTask)
|
|
}
|
|
|
|
func newMixCompactionTask(t *datapb.CompactionTask,
|
|
allocator allocator.Allocator,
|
|
meta CompactionMeta,
|
|
ievm IndexEngineVersionManager,
|
|
) *mixCompactionTask {
|
|
task := &mixCompactionTask{
|
|
allocator: allocator,
|
|
meta: meta,
|
|
ievm: ievm,
|
|
times: taskcommon.NewTimes(),
|
|
}
|
|
task.taskProto.Store(t)
|
|
return task
|
|
}
|
|
|
|
func (t *mixCompactionTask) processMetaSaved() bool {
|
|
log := log.With(zap.Int64("triggerID", t.GetTaskProto().GetTriggerID()), zap.Int64("PlanID", t.GetTaskProto().GetPlanID()), zap.Int64("collectionID", t.GetTaskProto().GetCollectionID()))
|
|
if err := t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_completed)); err != nil {
|
|
log.Warn("mixCompactionTask failed to proccessMetaSaved", zap.Error(err))
|
|
return false
|
|
}
|
|
|
|
return t.processCompleted()
|
|
}
|
|
|
|
func (t *mixCompactionTask) saveTaskMeta(task *datapb.CompactionTask) error {
|
|
return t.meta.SaveCompactionTask(context.TODO(), task)
|
|
}
|
|
|
|
func (t *mixCompactionTask) SaveTaskMeta() error {
|
|
return t.saveTaskMeta(t.GetTaskProto())
|
|
}
|
|
|
|
func (t *mixCompactionTask) saveSegmentMeta(result *datapb.CompactionPlanResult) error {
|
|
log := log.With(zap.Int64("triggerID", t.GetTaskProto().GetTriggerID()), zap.Int64("PlanID", t.GetTaskProto().GetPlanID()), zap.Int64("collectionID", t.GetTaskProto().GetCollectionID()))
|
|
if err := binlog.CompressCompactionBinlogs(result.GetSegments()); err != nil {
|
|
return err
|
|
}
|
|
// Also prepare metric updates.
|
|
newSegments, metricMutation, err := t.meta.CompleteCompactionMutation(context.TODO(), t.taskProto.Load().(*datapb.CompactionTask), result)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
// Apply metrics after successful meta update.
|
|
newSegmentIDs := lo.Map(newSegments, func(s *SegmentInfo, _ int) UniqueID { return s.GetID() })
|
|
metricMutation.commit()
|
|
for _, newSegID := range newSegmentIDs {
|
|
select {
|
|
case getBuildIndexChSingleton() <- newSegID:
|
|
default:
|
|
}
|
|
}
|
|
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_meta_saved), setResultSegments(newSegmentIDs))
|
|
if err != nil {
|
|
log.Warn("mixCompaction failed to setState meta saved", zap.Error(err))
|
|
return err
|
|
}
|
|
log.Info("mixCompactionTask success to save segment meta")
|
|
return nil
|
|
}
|
|
|
|
// Note: return True means exit this state machine.
|
|
// ONLY return True for Completed, Failed or Timeout
|
|
func (t *mixCompactionTask) Process() bool {
|
|
log := log.With(zap.Int64("triggerID",
|
|
t.GetTaskProto().GetTriggerID()),
|
|
zap.Int64("PlanID", t.GetTaskProto().GetPlanID()),
|
|
zap.Int64("collectionID", t.GetTaskProto().GetCollectionID()))
|
|
lastState := t.GetTaskProto().GetState().String()
|
|
processResult := false
|
|
switch t.GetTaskProto().GetState() {
|
|
case datapb.CompactionTaskState_meta_saved:
|
|
processResult = t.processMetaSaved()
|
|
case datapb.CompactionTaskState_completed:
|
|
processResult = t.processCompleted()
|
|
case datapb.CompactionTaskState_failed:
|
|
processResult = t.processFailed()
|
|
case datapb.CompactionTaskState_timeout:
|
|
processResult = true
|
|
}
|
|
currentState := t.GetTaskProto().GetState().String()
|
|
if currentState != lastState {
|
|
log.Info("mix compaction task state changed", zap.String("lastState", lastState), zap.String("currentState", currentState))
|
|
}
|
|
return processResult
|
|
}
|
|
|
|
func (t *mixCompactionTask) GetLabel() string {
|
|
return fmt.Sprintf("%d-%s", t.taskProto.Load().(*datapb.CompactionTask).PartitionID, t.GetTaskProto().GetChannel())
|
|
}
|
|
|
|
func (t *mixCompactionTask) NeedReAssignNodeID() bool {
|
|
return t.GetTaskProto().GetState() == datapb.CompactionTaskState_pipelining && (t.GetTaskProto().GetNodeID() == 0 || t.GetTaskProto().GetNodeID() == NullNodeID)
|
|
}
|
|
|
|
func (t *mixCompactionTask) processCompleted() bool {
|
|
log := log.With(zap.Int64("triggerID", t.GetTaskProto().GetTriggerID()),
|
|
zap.Int64("PlanID", t.GetTaskProto().GetPlanID()),
|
|
zap.Int64("collectionID", t.GetTaskProto().GetCollectionID()))
|
|
t.resetSegmentCompacting()
|
|
log.Info("mixCompactionTask processCompleted done")
|
|
return true
|
|
}
|
|
|
|
func (t *mixCompactionTask) resetSegmentCompacting() {
|
|
t.meta.SetSegmentsCompacting(context.TODO(), t.taskProto.Load().(*datapb.CompactionTask).GetInputSegments(), false)
|
|
}
|
|
|
|
func (t *mixCompactionTask) ShadowClone(opts ...compactionTaskOpt) *datapb.CompactionTask {
|
|
taskClone := proto.Clone(t.GetTaskProto()).(*datapb.CompactionTask)
|
|
for _, opt := range opts {
|
|
opt(taskClone)
|
|
}
|
|
return taskClone
|
|
}
|
|
|
|
func (t *mixCompactionTask) processFailed() bool {
|
|
return true
|
|
}
|
|
|
|
func (t *mixCompactionTask) Clean() bool {
|
|
return t.doClean() == nil
|
|
}
|
|
|
|
func (t *mixCompactionTask) doClean() error {
|
|
log := log.With(zap.Int64("triggerID", t.GetTaskProto().GetTriggerID()), zap.Int64("PlanID", t.GetTaskProto().GetPlanID()), zap.Int64("collectionID", t.GetTaskProto().GetCollectionID()))
|
|
err := t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_cleaned))
|
|
if err != nil {
|
|
log.Warn("mixCompactionTask fail to updateAndSaveTaskMeta", zap.Error(err))
|
|
return err
|
|
}
|
|
// resetSegmentCompacting must be the last step of Clean, to make sure resetSegmentCompacting only called once
|
|
// otherwise, it may unlock segments locked by other compaction tasks
|
|
t.resetSegmentCompacting()
|
|
log.Info("mixCompactionTask clean done")
|
|
return nil
|
|
}
|
|
|
|
func (t *mixCompactionTask) updateAndSaveTaskMeta(opts ...compactionTaskOpt) error {
|
|
// if task state is completed, cleaned, failed, timeout, then do append end time and save
|
|
if t.GetTaskProto().State == datapb.CompactionTaskState_completed ||
|
|
t.GetTaskProto().State == datapb.CompactionTaskState_cleaned ||
|
|
t.GetTaskProto().State == datapb.CompactionTaskState_failed ||
|
|
t.GetTaskProto().State == datapb.CompactionTaskState_timeout {
|
|
ts := time.Now().Unix()
|
|
opts = append(opts, setEndTime(ts))
|
|
}
|
|
|
|
task := t.ShadowClone(opts...)
|
|
err := t.saveTaskMeta(task)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
t.SetTask(task)
|
|
return nil
|
|
}
|
|
|
|
func (t *mixCompactionTask) SetNodeID(id UniqueID) error {
|
|
return t.updateAndSaveTaskMeta(setNodeID(id))
|
|
}
|
|
|
|
func (t *mixCompactionTask) SetTask(task *datapb.CompactionTask) {
|
|
t.taskProto.Store(task)
|
|
}
|
|
|
|
func (t *mixCompactionTask) BuildCompactionRequest() (*datapb.CompactionPlan, error) {
|
|
log := log.With(zap.Int64("triggerID", t.GetTaskProto().GetTriggerID()), zap.Int64("PlanID", t.GetTaskProto().GetPlanID()), zap.Int64("collectionID", t.GetTaskProto().GetCollectionID()))
|
|
taskProto := t.taskProto.Load().(*datapb.CompactionTask)
|
|
taskSchema := taskProto.GetSchema()
|
|
if taskSchema == nil {
|
|
return nil, merr.WrapErrIllegalCompactionPlan("compaction task schema is nil")
|
|
}
|
|
taskSchemaVersion := taskSchema.GetVersion()
|
|
compactionParams, err := compaction.GenerateJSONParams(taskSchema)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
plan := &datapb.CompactionPlan{
|
|
PlanID: taskProto.GetPlanID(),
|
|
StartTime: taskProto.GetStartTime(),
|
|
Type: taskProto.GetType(),
|
|
Channel: taskProto.GetChannel(),
|
|
CollectionTtl: taskProto.GetCollectionTtl(),
|
|
TotalRows: taskProto.GetTotalRows(),
|
|
Schema: taskSchema,
|
|
PreAllocatedSegmentIDs: taskProto.GetPreAllocatedSegmentIDs(),
|
|
SlotUsage: t.GetSlotUsage(),
|
|
MaxSize: taskProto.GetMaxSize(),
|
|
JsonParams: compactionParams,
|
|
CurrentScalarIndexVersion: t.ievm.ResolveScalarIndexVersion(),
|
|
}
|
|
|
|
// set analyzer resource for text match index if use ref mode
|
|
if fileresource.IsRefMode(paramtable.Get().CommonCfg.DNFileResourceMode.GetValue()) && taskProto.GetType() == datapb.CompactionType_SortCompaction && len(taskSchema.GetFileResourceIds()) > 0 {
|
|
resources, err := t.meta.GetFileResources(context.Background(), taskSchema.GetFileResourceIds()...)
|
|
if err != nil {
|
|
log.Warn("get file resources for collection failed", zap.Int64("collectionID", taskProto.GetCollectionID()), zap.Error(err))
|
|
return nil, merr.Wrap(err, "get file resources for sort compaction failed")
|
|
}
|
|
plan.FileResources = resources
|
|
}
|
|
|
|
segIDMap := make(map[int64][]*datapb.FieldBinlog, len(plan.SegmentBinlogs))
|
|
segments := make([]*SegmentInfo, 0, len(taskProto.GetInputSegments()))
|
|
for _, segID := range taskProto.GetInputSegments() {
|
|
segInfo := t.meta.GetHealthySegment(context.TODO(), segID)
|
|
if segInfo == nil {
|
|
return nil, merr.WrapErrSegmentNotFound(segID)
|
|
}
|
|
if taskSchemaVersion < segInfo.GetSchemaVersion() {
|
|
return nil, merr.WrapErrIllegalCompactionPlanMsg("compaction task schema version %d is older than input segment %d schema version %d", taskSchemaVersion, segInfo.GetID(), segInfo.GetSchemaVersion())
|
|
}
|
|
plan.SegmentBinlogs = append(plan.SegmentBinlogs, &datapb.CompactionSegmentBinlogs{
|
|
SegmentID: segID,
|
|
CollectionID: segInfo.GetCollectionID(),
|
|
PartitionID: segInfo.GetPartitionID(),
|
|
Level: segInfo.GetLevel(),
|
|
InsertChannel: segInfo.GetInsertChannel(),
|
|
FieldBinlogs: segInfo.GetBinlogs(),
|
|
Field2StatslogPaths: segInfo.GetStatslogs(),
|
|
Deltalogs: segInfo.GetDeltalogs(),
|
|
IsSorted: segInfo.GetIsSorted(),
|
|
IsSortedByNamespace: segInfo.GetIsSortedByNamespace(),
|
|
StorageVersion: segInfo.GetStorageVersion(),
|
|
Manifest: segInfo.GetManifestPath(),
|
|
CommitTimestamp: segInfo.GetCommitTimestamp(),
|
|
})
|
|
segIDMap[segID] = segInfo.GetDeltalogs()
|
|
segments = append(segments, segInfo)
|
|
}
|
|
|
|
logIDRange, err := PreAllocateBinlogIDs(t.allocator, segments, taskSchema)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
plan.PreAllocatedLogIDs = logIDRange
|
|
// BeginLogID is deprecated, but still assign it for compatibility.
|
|
plan.BeginLogID = logIDRange.Begin
|
|
|
|
WrapPluginContext(taskProto.GetCollectionID(), taskSchema.GetProperties(), plan)
|
|
|
|
log.Info("Compaction handler refreshed mix compaction plan", zap.Int64("maxSize", plan.GetMaxSize()),
|
|
zap.Any("PreAllocatedLogIDs", logIDRange), zap.Any("segID2DeltaLogs", segIDMap))
|
|
return plan, nil
|
|
}
|
|
|
|
func (t *mixCompactionTask) GetSlotUsage() int64 {
|
|
return t.GetTaskSlot()
|
|
}
|