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>
497 lines
18 KiB
Go
497 lines
18 KiB
Go
// Licensed to the LF AI & Data foundation under one
|
|
// or more contributor license agreements. See the NOTICE file
|
|
// distributed with this work for additional information
|
|
// regarding copyright ownership. The ASF licenses this file
|
|
// to you under the Apache License, Version 2.0 (the
|
|
// "License"); you may not use this file except in compliance
|
|
// with the License. You may obtain a copy of the License at
|
|
//
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
//
|
|
// Unless required by applicable law or agreed to in writing, software
|
|
// distributed under the License is distributed on an "AS IS" BASIS,
|
|
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
// See the License for the specific language governing permissions and
|
|
// limitations under the License.
|
|
|
|
package datacoord
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"time"
|
|
|
|
"go.uber.org/atomic"
|
|
"go.uber.org/zap"
|
|
"google.golang.org/protobuf/proto"
|
|
|
|
"github.com/milvus-io/milvus-proto/go-api/v3/commonpb"
|
|
"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/storagev2/packed"
|
|
"github.com/milvus-io/milvus/pkg/v3/common"
|
|
"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 = (*l0CompactionTask)(nil)
|
|
|
|
type l0CompactionTask struct {
|
|
taskProto atomic.Value // *datapb.CompactionTask
|
|
|
|
allocator allocator.Allocator
|
|
meta CompactionMeta
|
|
|
|
times *taskcommon.Times
|
|
committedV3Manifests map[int64]string
|
|
}
|
|
|
|
func (t *l0CompactionTask) GetTaskID() int64 {
|
|
return t.GetTaskProto().GetPlanID()
|
|
}
|
|
|
|
func (t *l0CompactionTask) GetTaskType() taskcommon.Type {
|
|
return taskcommon.Compaction
|
|
}
|
|
|
|
func (t *l0CompactionTask) GetTaskState() taskcommon.State {
|
|
return taskcommon.FromCompactionState(t.GetTaskProto().GetState())
|
|
}
|
|
|
|
func (t *l0CompactionTask) GetTaskSlot() int64 {
|
|
batchSize := paramtable.Get().CommonCfg.BloomFilterApplyBatchSize.GetAsInt()
|
|
factor := paramtable.Get().DataCoordCfg.L0DeleteCompactionSlotUsage.GetAsInt64()
|
|
slot := factor * t.GetTaskProto().GetTotalRows() / int64(batchSize)
|
|
if slot < 1 {
|
|
return 1
|
|
}
|
|
return slot
|
|
}
|
|
|
|
func (t *l0CompactionTask) SetTaskTime(timeType taskcommon.TimeType, time time.Time) {
|
|
t.times.SetTaskTime(timeType, time)
|
|
}
|
|
|
|
func (t *l0CompactionTask) GetTaskTime(timeType taskcommon.TimeType) time.Time {
|
|
return timeType.GetTaskTime(t.times)
|
|
}
|
|
|
|
func (t *l0CompactionTask) GetTaskVersion() int64 {
|
|
return int64(t.GetTaskProto().GetRetryTimes())
|
|
}
|
|
|
|
func (t *l0CompactionTask) CreateTaskOnWorker(nodeID int64, cluster session.Cluster) {
|
|
log := log.With(zap.Int64("triggerID", t.GetTaskProto().GetTriggerID()), zap.Int64("nodeID", t.GetTaskProto().GetNodeID()))
|
|
plan, err := t.BuildCompactionRequest()
|
|
if err != nil {
|
|
log.Warn("l0CompactionTask failed to build compaction request", zap.Error(err))
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_failed), setFailReason(err.Error()))
|
|
if err != nil {
|
|
log.Warn("l0CompactionTask failed to updateAndSaveTaskMeta", zap.Error(err))
|
|
}
|
|
return
|
|
}
|
|
|
|
// Check if this is a fast finish case (no target segments to compact with)
|
|
// Fast finish plan only contains L0 input segments, no target L1/L2 segments
|
|
if len(plan.SegmentBinlogs) == len(t.GetTaskProto().GetInputSegments()) {
|
|
log.Info("l0CompactionTask fast finish: no target segments, directly marking L0 segments as dropped",
|
|
zap.Int64("planID", t.GetTaskProto().GetPlanID()))
|
|
|
|
// Save segment meta with empty output segments (marks L0 input segments as dropped)
|
|
if err = t.saveSegmentMeta([]*datapb.CompactionSegment{}); err != nil {
|
|
log.Warn("l0CompactionTask fast finish failed to save segment meta", zap.Error(err))
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_failed), setFailReason(err.Error()))
|
|
if err != nil {
|
|
log.Warn("l0CompactionTask failed to updateAndSaveTaskMeta", zap.Error(err))
|
|
}
|
|
return
|
|
}
|
|
|
|
// Transition to meta_saved state
|
|
if err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_meta_saved)); err != nil {
|
|
log.Warn("l0CompactionTask fast finish failed to save task meta_saved state", zap.Error(err))
|
|
return
|
|
}
|
|
|
|
log.Info("l0CompactionTask fast finish completed", zap.Int64("planID", t.GetTaskProto().GetPlanID()))
|
|
return
|
|
}
|
|
|
|
err = cluster.CreateCompaction(nodeID, plan, t.GetTaskProto().GetCollectionID())
|
|
if err != nil {
|
|
originNodeID := t.GetTaskProto().GetNodeID()
|
|
log.Warn("l0CompactionTask 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("l0CompactionTask failed to updateAndSaveTaskMeta", zap.Int64("planID", t.GetTaskProto().GetPlanID()), zap.Error(err))
|
|
return
|
|
}
|
|
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
|
|
}
|
|
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_executing), setNodeID(nodeID))
|
|
if err != nil {
|
|
log.Warn("l0CompactionTask failed to updateAndSaveTaskMeta", zap.Error(err))
|
|
}
|
|
}
|
|
|
|
func (t *l0CompactionTask) QueryTaskOnWorker(cluster session.Cluster) {
|
|
log := log.With(zap.Int64("planID", t.GetTaskProto().GetPlanID()), zap.Int64("nodeID", t.GetTaskProto().GetNodeID()))
|
|
result, err := cluster.QueryCompaction(t.GetTaskProto().GetNodeID(), &datapb.CompactionStateRequest{
|
|
PlanID: t.GetTaskProto().GetPlanID(),
|
|
})
|
|
if err != nil || result == nil {
|
|
log.Warn("l0CompactionTask failed to get compaction result", zap.Error(err))
|
|
err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_pipelining), setNodeID(NullNodeID))
|
|
if err != nil {
|
|
log.Warn("update l0 compaction task meta failed", zap.Error(err))
|
|
}
|
|
return
|
|
}
|
|
switch result.GetState() {
|
|
case datapb.CompactionTaskState_completed:
|
|
err = t.meta.ValidateSegmentStateBeforeCompleteCompactionMutation(t.GetTaskProto())
|
|
if err != nil {
|
|
t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_failed), setFailReason(err.Error()))
|
|
return
|
|
}
|
|
|
|
if err = t.saveSegmentMeta(result.GetSegments()); err != nil {
|
|
log.Warn("l0CompactionTask failed to save segment meta", zap.Error(err))
|
|
return
|
|
}
|
|
|
|
if err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_meta_saved)); err != nil {
|
|
log.Warn("l0CompactionTask failed to save task meta_saved state", 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:
|
|
if err = t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_failed)); err != nil {
|
|
log.Warn("l0CompactionTask failed to set task failed state", zap.Error(err))
|
|
return
|
|
}
|
|
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 *l0CompactionTask) DropTaskOnWorker(cluster session.Cluster) {
|
|
if t.hasAssignedWorker() {
|
|
err := cluster.DropCompaction(t.GetTaskProto().GetNodeID(), t.GetTaskProto().GetPlanID())
|
|
if err != nil {
|
|
log.Warn("l0CompactionTask unable to drop compaction plan", zap.Int64("planID", t.GetTaskProto().GetPlanID()), zap.Error(err))
|
|
}
|
|
}
|
|
}
|
|
|
|
func (t *l0CompactionTask) GetTaskProto() *datapb.CompactionTask {
|
|
task := t.taskProto.Load()
|
|
if task == nil {
|
|
return nil
|
|
}
|
|
return task.(*datapb.CompactionTask)
|
|
}
|
|
|
|
func newL0CompactionTask(t *datapb.CompactionTask, allocator allocator.Allocator, meta CompactionMeta) *l0CompactionTask {
|
|
task := &l0CompactionTask{
|
|
allocator: allocator,
|
|
meta: meta,
|
|
times: taskcommon.NewTimes(),
|
|
committedV3Manifests: make(map[int64]string),
|
|
}
|
|
task.taskProto.Store(t)
|
|
return task
|
|
}
|
|
|
|
// Note: return True means exit this state machine.
|
|
// ONLY return True for Completed, Failed
|
|
func (t *l0CompactionTask) Process() bool {
|
|
switch t.GetTaskProto().GetState() {
|
|
case datapb.CompactionTaskState_meta_saved:
|
|
return t.processMetaSaved()
|
|
case datapb.CompactionTaskState_completed:
|
|
return t.processCompleted()
|
|
case datapb.CompactionTaskState_failed:
|
|
return true
|
|
case datapb.CompactionTaskState_timeout:
|
|
return true
|
|
default:
|
|
return false
|
|
}
|
|
}
|
|
|
|
func (t *l0CompactionTask) processMetaSaved() bool {
|
|
err := t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_completed))
|
|
if err != nil {
|
|
log.Warn("l0CompactionTask unable to processMetaSaved", zap.Int64("planID", t.GetTaskProto().GetPlanID()), zap.Error(err))
|
|
return false
|
|
}
|
|
return t.processCompleted()
|
|
}
|
|
|
|
func (t *l0CompactionTask) processCompleted() bool {
|
|
t.resetSegmentCompacting()
|
|
task := t.taskProto.Load().(*datapb.CompactionTask)
|
|
log.Info("l0CompactionTask processCompleted done", zap.Int64("planID", task.GetPlanID()),
|
|
zap.Duration("costs", time.Duration(task.GetEndTime()-task.GetStartTime())*time.Second))
|
|
return true
|
|
}
|
|
|
|
func (t *l0CompactionTask) doClean() error {
|
|
log := log.With(zap.Int64("planID", t.GetTaskProto().GetPlanID()))
|
|
err := t.updateAndSaveTaskMeta(setState(datapb.CompactionTaskState_cleaned))
|
|
if err != nil {
|
|
log.Warn("l0CompactionTask failed 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("l0CompactionTask clean done")
|
|
return nil
|
|
}
|
|
|
|
func (t *l0CompactionTask) Clean() bool {
|
|
return t.doClean() == nil
|
|
}
|
|
|
|
func (t *l0CompactionTask) SetTask(task *datapb.CompactionTask) {
|
|
t.taskProto.Store(task)
|
|
}
|
|
|
|
func (t *l0CompactionTask) GetLabel() string {
|
|
return fmt.Sprintf("%d-%s", t.GetTaskProto().PartitionID, t.GetTaskProto().GetChannel())
|
|
}
|
|
|
|
func (t *l0CompactionTask) NeedReAssignNodeID() bool {
|
|
return t.GetTaskProto().GetState() == datapb.CompactionTaskState_pipelining && (!t.hasAssignedWorker())
|
|
}
|
|
|
|
func (t *l0CompactionTask) ShadowClone(opts ...compactionTaskOpt) *datapb.CompactionTask {
|
|
taskClone := proto.Clone(t.GetTaskProto()).(*datapb.CompactionTask)
|
|
for _, opt := range opts {
|
|
opt(taskClone)
|
|
}
|
|
return taskClone
|
|
}
|
|
|
|
func (t *l0CompactionTask) selectFlushedSegment() ([]*SegmentInfo, []*datapb.CompactionSegmentBinlogs, error) {
|
|
taskProto := t.taskProto.Load().(*datapb.CompactionTask)
|
|
// Select flushed L1/L2 segments for LevelZero compaction that meets the condition:
|
|
// dmlPos < triggerInfo.pos
|
|
flushedSegments := t.meta.SelectSegments(context.TODO(), WithCollection(taskProto.GetCollectionID()), SegmentFilterFunc(func(info *SegmentInfo) bool {
|
|
return (taskProto.GetPartitionID() == common.AllPartitionsID || info.GetPartitionID() == taskProto.GetPartitionID()) &&
|
|
info.GetInsertChannel() == taskProto.GetChannel() &&
|
|
(info.GetState() == commonpb.SegmentState_Sealed || isFlushState(info.GetState())) &&
|
|
!info.GetIsImporting() &&
|
|
info.GetLevel() != datapb.SegmentLevel_L0 &&
|
|
segmentEffectiveTs(info.SegmentInfo) < taskProto.GetPos().GetTimestamp()
|
|
}))
|
|
|
|
sealedSegBinlogs := []*datapb.CompactionSegmentBinlogs{}
|
|
for _, info := range flushedSegments {
|
|
// Sealed is unexpected, fail fast
|
|
if info.GetState() == commonpb.SegmentState_Sealed {
|
|
return nil, nil, merr.WrapErrServiceInternalMsg("L0 compaction selected invalid sealed segment %d", info.GetID())
|
|
}
|
|
|
|
sealedSegBinlogs = append(sealedSegBinlogs, &datapb.CompactionSegmentBinlogs{
|
|
SegmentID: info.GetID(),
|
|
Field2StatslogPaths: info.GetStatslogs(),
|
|
InsertChannel: info.GetInsertChannel(),
|
|
Level: info.GetLevel(),
|
|
CollectionID: info.GetCollectionID(),
|
|
PartitionID: info.GetPartitionID(),
|
|
IsSorted: info.GetIsSorted(),
|
|
IsSortedByNamespace: info.GetIsSortedByNamespace(),
|
|
Manifest: info.GetManifestPath(),
|
|
CommitTimestamp: info.GetCommitTimestamp(),
|
|
})
|
|
}
|
|
|
|
return flushedSegments, sealedSegBinlogs, nil
|
|
}
|
|
|
|
func (t *l0CompactionTask) BuildCompactionRequest() (*datapb.CompactionPlan, error) {
|
|
taskProto := t.taskProto.Load().(*datapb.CompactionTask)
|
|
compactionParams, err := compaction.GenerateJSONParams(taskProto.GetSchema())
|
|
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: taskProto.GetSchema(),
|
|
SlotUsage: t.GetSlotUsage(),
|
|
JsonParams: compactionParams,
|
|
}
|
|
|
|
log := log.With(zap.Int64("taskID", taskProto.GetTriggerID()), zap.Int64("planID", plan.GetPlanID()))
|
|
segments := make([]*SegmentInfo, 0)
|
|
for _, segID := range taskProto.GetInputSegments() {
|
|
segInfo := t.meta.GetHealthySegment(context.TODO(), segID)
|
|
if segInfo == nil {
|
|
return nil, merr.WrapErrSegmentNotFound(segID)
|
|
}
|
|
plan.SegmentBinlogs = append(plan.SegmentBinlogs, &datapb.CompactionSegmentBinlogs{
|
|
SegmentID: segID,
|
|
CollectionID: segInfo.GetCollectionID(),
|
|
PartitionID: segInfo.GetPartitionID(),
|
|
Level: segInfo.GetLevel(),
|
|
InsertChannel: segInfo.GetInsertChannel(),
|
|
Deltalogs: segInfo.GetDeltalogs(),
|
|
IsSorted: segInfo.GetIsSorted(),
|
|
IsSortedByNamespace: segInfo.GetIsSortedByNamespace(),
|
|
Manifest: segInfo.GetManifestPath(),
|
|
CommitTimestamp: segInfo.GetCommitTimestamp(),
|
|
})
|
|
segments = append(segments, segInfo)
|
|
}
|
|
|
|
flushedSegments, flushedSegBinlogs, err := t.selectFlushedSegment()
|
|
if err != nil {
|
|
log.Warn("invalid L0 compaction plan, unable to select flushed segments", zap.Error(err))
|
|
return nil, err
|
|
}
|
|
if len(flushedSegments) == 0 {
|
|
// Fast finish: no target segments to compact with, return plan with only L0 segments
|
|
log.Info("l0Compaction available non-L0 Segments is empty, will fast finish",
|
|
zap.Any("target position", taskProto.GetPos()))
|
|
return plan, nil
|
|
}
|
|
|
|
segments = append(segments, flushedSegments...)
|
|
logIDRange, err := PreAllocateBinlogIDs(t.allocator, segments, nil)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
plan.PreAllocatedLogIDs = logIDRange
|
|
// BeginLogID is deprecated, but still assign it for compatibility.
|
|
plan.BeginLogID = logIDRange.Begin
|
|
|
|
plan.SegmentBinlogs = append(plan.SegmentBinlogs, flushedSegBinlogs...)
|
|
log.Info("l0CompactionTask refreshed level zero compaction plan",
|
|
zap.Any("target position", taskProto.GetPos()),
|
|
zap.Any("target segments count", len(flushedSegBinlogs)),
|
|
zap.Any("PreAllocatedLogIDs", logIDRange))
|
|
|
|
WrapPluginContext(taskProto.GetCollectionID(), taskProto.GetSchema().GetProperties(), plan)
|
|
return plan, nil
|
|
}
|
|
|
|
func (t *l0CompactionTask) resetSegmentCompacting() {
|
|
t.meta.SetSegmentsCompacting(context.TODO(), t.GetTaskProto().GetInputSegments(), false)
|
|
}
|
|
|
|
func (t *l0CompactionTask) hasAssignedWorker() bool {
|
|
return t.GetTaskProto().GetNodeID() != 0 && t.GetTaskProto().GetNodeID() != NullNodeID
|
|
}
|
|
|
|
func (t *l0CompactionTask) SetNodeID(id UniqueID) error {
|
|
return t.updateAndSaveTaskMeta(setNodeID(id))
|
|
}
|
|
|
|
func (t *l0CompactionTask) SaveTaskMeta() error {
|
|
return t.saveTaskMeta(t.GetTaskProto())
|
|
}
|
|
|
|
func (t *l0CompactionTask) 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 *l0CompactionTask) saveTaskMeta(task *datapb.CompactionTask) error {
|
|
return t.meta.SaveCompactionTask(context.TODO(), task)
|
|
}
|
|
|
|
func buildL0V3DeltaLogEntries(segmentID int64, deltalogs []*datapb.FieldBinlog) ([]packed.DeltaLogEntry, error) {
|
|
entries := make([]packed.DeltaLogEntry, 0)
|
|
for _, fieldBinlog := range deltalogs {
|
|
for _, binlog := range fieldBinlog.GetBinlogs() {
|
|
path := binlog.GetLogPath()
|
|
if path == "" {
|
|
return nil, merr.WrapErrServiceInternalMsg("L0 V3 compaction result missing deltalog path for segment %d, logID %d", segmentID, binlog.GetLogID())
|
|
}
|
|
entries = append(entries, packed.DeltaLogEntry{
|
|
Path: path,
|
|
NumEntries: binlog.GetEntriesNum(),
|
|
})
|
|
}
|
|
}
|
|
return entries, nil
|
|
}
|
|
|
|
func (t *l0CompactionTask) saveSegmentMeta(outputSegs []*datapb.CompactionSegment) error {
|
|
var operators []UpdateOperator
|
|
storageConfig := compaction.CreateStorageConfig()
|
|
for _, seg := range outputSegs {
|
|
if len(seg.GetDeltalogs()) > 0 {
|
|
operators = append(operators, AddL0DeltalogsAndUpdateManifestOperator(
|
|
seg.GetSegmentID(),
|
|
seg.GetDeltalogs(),
|
|
storageConfig,
|
|
t.committedV3Manifests,
|
|
))
|
|
}
|
|
}
|
|
|
|
for _, segID := range t.GetTaskProto().InputSegments {
|
|
operators = append(operators, UpdateStatusOperator(segID, commonpb.SegmentState_Dropped), UpdateCompactedOperator(segID))
|
|
}
|
|
|
|
log.Info("meta update: update segments info for level zero compaction",
|
|
zap.Int64("planID", t.GetTaskProto().GetPlanID()),
|
|
)
|
|
|
|
return t.meta.UpdateSegmentsInfo(context.TODO(), operators...)
|
|
}
|
|
|
|
func (t *l0CompactionTask) GetSlotUsage() int64 {
|
|
return t.GetTaskSlot()
|
|
}
|