Files
seaweedFS/weed/plugin/worker/worker.go
Chris Lu 8ec9ff4a12 Refactor plugin system and migrate worker runtime (#8369)
* admin: add plugin runtime UI page and route wiring

* pb: add plugin gRPC contract and generated bindings

* admin/plugin: implement worker registry, runtime, monitoring, and config store

* admin/dash: wire plugin runtime and expose plugin workflow APIs

* command: add flags to enable plugin runtime

* admin: rename remaining plugin v2 wording to plugin

* admin/plugin: add detectable job type registry helper

* admin/plugin: add scheduled detection and dispatch orchestration

* admin/plugin: prefetch job type descriptors when workers connect

* admin/plugin: add known job type discovery API and UI

* admin/plugin: refresh design doc to match current implementation

* admin/plugin: enforce per-worker scheduler concurrency limits

* admin/plugin: use descriptor runtime defaults for scheduler policy

* admin/ui: auto-load first known plugin job type on page open

* admin/plugin: bootstrap persisted config from descriptor defaults

* admin/plugin: dedupe scheduled proposals by dedupe key

* admin/ui: add job type and state filters for plugin monitoring

* admin/ui: add per-job-type plugin activity summary

* admin/plugin: split descriptor read API from schema refresh

* admin/ui: keep plugin summary metrics global while tables are filtered

* admin/plugin: retry executor reservation before timing out

* admin/plugin: expose scheduler states for monitoring

* admin/ui: show per-job-type scheduler states in plugin monitor

* pb/plugin: rename protobuf package to plugin

* admin/plugin: rename pluginRuntime wiring to plugin

* admin/plugin: remove runtime naming from plugin APIs and UI

* admin/plugin: rename runtime files to plugin naming

* admin/plugin: persist jobs and activities for monitor recovery

* admin/plugin: lease one detector worker per job type

* admin/ui: show worker load from plugin heartbeats

* admin/plugin: skip stale workers for detector and executor picks

* plugin/worker: add plugin worker command and stream runtime scaffold

* plugin/worker: implement vacuum detect and execute handlers

* admin/plugin: document external vacuum plugin worker starter

* command: update plugin.worker help to reflect implemented flow

* command/admin: drop legacy Plugin V2 label

* plugin/worker: validate vacuum job type and respect min interval

* plugin/worker: test no-op detect when min interval not elapsed

* command/admin: document plugin.worker external process

* plugin/worker: advertise configured concurrency in hello

* command/plugin.worker: add jobType handler selection

* command/plugin.worker: test handler selection by job type

* command/plugin.worker: persist worker id in workingDir

* admin/plugin: document plugin.worker jobType and workingDir flags

* plugin/worker: support cancel request for in-flight work

* plugin/worker: test cancel request acknowledgements

* command/plugin.worker: document workingDir and jobType behavior

* plugin/worker: emit executor activity events for monitor

* plugin/worker: test executor activity builder

* admin/plugin: send last successful run in detection request

* admin/plugin: send cancel request when detect or execute context ends

* admin/plugin: document worker cancel request responsibility

* admin/handlers: expose plugin scheduler states API in no-auth mode

* admin/handlers: test plugin scheduler states route registration

* admin/plugin: keep worker id on worker-generated activity records

* admin/plugin: test worker id propagation in monitor activities

* admin/dash: always initialize plugin service

* command/admin: remove plugin enable flags and default to enabled

* admin/dash: drop pluginEnabled constructor parameter

* admin/plugin UI: stop checking plugin enabled state

* admin/plugin: remove docs for plugin enable flags

* admin/dash: remove unused plugin enabled check method

* admin/dash: fallback to in-memory plugin init when dataDir fails

* admin/plugin API: expose worker gRPC port in status

* command/plugin.worker: resolve admin gRPC port via plugin status

* split plugin UI into overview/configuration/monitoring pages

* Update layout_templ.go

* add volume_balance plugin worker handler

* wire plugin.worker CLI for volume_balance job type

* add erasure_coding plugin worker handler

* wire plugin.worker CLI for erasure_coding job type

* support multi-job handlers in plugin worker runtime

* allow plugin.worker jobType as comma-separated list

* admin/plugin UI: rename to Workers and simplify config view

* plugin worker: queue detection requests instead of capacity reject

* Update plugin_worker.go

* plugin volume_balance: remove force_move/timeout from worker config UI

* plugin erasure_coding: enforce local working dir and cleanup

* admin/plugin UI: rename admin settings to job scheduling

* admin/plugin UI: persist and robustly render detection results

* admin/plugin: record and return detection trace metadata

* admin/plugin UI: show detection process and decision trace

* plugin: surface detector decision trace as activities

* mini: start a plugin worker by default

* admin/plugin UI: split monitoring into detection and execution tabs

* plugin worker: emit detection decision trace for EC and balance

* admin workers UI: split monitoring into detection and execution pages

* plugin scheduler: skip proposals for active assigned/running jobs

* admin workers UI: add job queue tab

* plugin worker: add dummy stress detector and executor job type

* admin workers UI: reorder tabs to detection queue execution

* admin workers UI: regenerate plugin template

* plugin defaults: include dummy stress and add stress tests

* plugin dummy stress: rotate detection selections across runs

* plugin scheduler: remove cross-run proposal dedupe

* plugin queue: track pending scheduled jobs

* plugin scheduler: wait for executor capacity before dispatch

* plugin scheduler: skip detection when waiting backlog is high

* plugin: add disk-backed job detail API and persistence

* admin ui: show plugin job detail modal from job id links

* plugin: generate unique job ids instead of reusing proposal ids

* plugin worker: emit heartbeats on work state changes

* plugin registry: round-robin tied executor and detector picks

* add temporary EC overnight stress runner

* plugin job details: persist and render EC execution plans

* ec volume details: color data and parity shard badges

* shard labels: keep parity ids numeric and color-only distinction

* admin: remove legacy maintenance UI routes and templates

* admin: remove dead maintenance endpoint helpers

* Update layout_templ.go

* remove dummy_stress worker and command support

* refactor plugin UI to job-type top tabs and sub-tabs

* migrate weed worker command to plugin runtime

* remove plugin.worker command and keep worker runtime with metrics

* update helm worker args for jobType and execution flags

* set plugin scheduling defaults to global 16 and per-worker 4

* stress: fix RPC context reuse and remove redundant variables in ec_stress_runner

* admin/plugin: fix lifecycle races, safe channel operations, and terminal state constants

* admin/dash: randomize job IDs and fix priority zero-value overwrite in plugin API

* admin/handlers: implement buffered rendering to prevent response corruption

* admin/plugin: implement debounced persistence flusher and optimize BuildJobDetail memory lookups

* admin/plugin: fix priority overwrite and implement bounded wait in scheduler reserve

* admin/plugin: implement atomic file writes and fix run record side effects

* admin/plugin: use P prefix for parity shard labels in execution plans

* admin/plugin: enable parallel execution for cancellation tests

* admin: refactor time.Time fields to pointers for better JSON omitempty support

* admin/plugin: implement pointer-safe time assignments and comparisons in plugin core

* admin/plugin: fix time assignment and sorting logic in plugin monitor after pointer refactor

* admin/plugin: update scheduler activity tracking to use time pointers

* admin/plugin: fix time-based run history trimming after pointer refactor

* admin/dash: fix JobSpec struct literal in plugin API after pointer refactor

* admin/view: add D/P prefixes to EC shard badges for UI consistency

* admin/plugin: use lifecycle-aware context for schema prefetching

* Update ec_volume_details_templ.go

* admin/stress: fix proposal sorting and log volume cleanup errors

* stress: refine ec stress runner with math/rand and collection name

- Added Collection field to VolumeEcShardsDeleteRequest for correct filename construction.
- Replaced crypto/rand with seeded math/rand PRNG for bulk payloads.
- Added documentation for EcMinAge zero-value behavior.
- Added logging for ignored errors in volume/shard deletion.

* admin: return internal server error for plugin store failures

Changed error status code from 400 Bad Request to 500 Internal Server Error for failures in GetPluginJobDetail to correctly reflect server-side errors.

* admin: implement safe channel sends and graceful shutdown sync

- Added sync.WaitGroup to Plugin struct to manage background goroutines.
- Implemented safeSendCh helper using recover() to prevent panics on closed channels.
- Ensured Shutdown() waits for all background operations to complete.

* admin: robustify plugin monitor with nil-safe time and record init

- Standardized nil-safe assignment for *time.Time pointers (CreatedAt, UpdatedAt, CompletedAt).
- Ensured persistJobDetailSnapshot initializes new records correctly if they don't exist on disk.
- Fixed debounced persistence to trigger immediate write on job completion.

* admin: improve scheduler shutdown behavior and logic guards

- Replaced brittle error string matching with explicit r.shutdownCh selection for shutdown detection.
- Removed redundant nil guard in buildScheduledJobSpec.
- Standardized WaitGroup usage for schedulerLoop.

* admin: implement deep copy for job parameters and atomic write fixes

- Implemented deepCopyGenericValue and used it in cloneTrackedJob to prevent shared state.
- Ensured atomicWriteFile creates parent directories before writing.

* admin: remove unreachable branch in shard classification

Removed an unreachable 'totalShards <= 0' check in classifyShardID as dataShards and parityShards are already guarded.

* admin: secure UI links and use canonical shard constants

- Added rel="noopener noreferrer" to external links for security.
- Replaced magic number 14 with erasure_coding.TotalShardsCount.
- Used renderEcShardBadge for missing shard list consistency.

* admin: stabilize plugin tests and fix regressions

- Composed a robust plugin_monitor_test.go to handle asynchronous persistence.
- Updated all time.Time literals to use timeToPtr helper.
- Added explicit Shutdown() calls in tests to synchronize with debounced writes.
- Fixed syntax errors and orphaned struct literals in tests.

* Potential fix for code scanning alert no. 278: Slice memory allocation with excessive size value

Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>

* Potential fix for code scanning alert no. 283: Uncontrolled data used in path expression

Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>

* admin: finalize refinements for error handling, scheduler, and race fixes

- Standardized HTTP 500 status codes for store failures in plugin_api.go.
- Tracked scheduled detection goroutines with sync.WaitGroup for safe shutdown.
- Fixed race condition in safeSendDetectionComplete by extracting channel under lock.
- Implemented deep copy for JobActivity details.
- Used defaultDirPerm constant in atomicWriteFile.

* test(ec): migrate admin dockertest to plugin APIs

* admin/plugin_api: fix RunPluginJobTypeAPI to return 500 for server-side detection/filter errors

* admin/plugin_api: fix ExecutePluginJobAPI to return 500 for job execution failures

* admin/plugin_api: limit parseProtoJSONBody request body to 1MB to prevent unbounded memory usage

* admin/plugin: consolidate regex to package-level validJobTypePattern; add char validation to sanitizeJobID

* admin/plugin: fix racy Shutdown channel close with sync.Once

* admin/plugin: track sendLoop and recv goroutines in WorkerStream with r.wg

* admin/plugin: document writeProtoFiles atomicity — .pb is source of truth, .json is human-readable only

* admin/plugin: extract activityLess helper to deduplicate nil-safe OccurredAt sort comparators

* test/ec: check http.NewRequest errors to prevent nil req panics

* test/ec: replace deprecated ioutil/math/rand, fix stale step comment 5.1→3.1

* plugin(ec): raise default detection and scheduling throughput limits

* topology: include empty disks in volume list and EC capacity fallback

* topology: remove hard 10-task cap for detection planning

* Update ec_volume_details_templ.go

* adjust default

* fix tests

---------

Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
2026-02-18 13:42:41 -08:00

940 lines
24 KiB
Go

package pluginworker
import (
"context"
"crypto/rand"
"encoding/hex"
"fmt"
"os"
"sort"
"strings"
"sync"
"time"
"github.com/seaweedfs/seaweedfs/weed/glog"
"github.com/seaweedfs/seaweedfs/weed/pb"
"github.com/seaweedfs/seaweedfs/weed/pb/plugin_pb"
"google.golang.org/grpc"
"google.golang.org/protobuf/proto"
"google.golang.org/protobuf/types/known/timestamppb"
)
const (
defaultHeartbeatInterval = 15 * time.Second
defaultReconnectDelay = 5 * time.Second
defaultSendBufferSize = 256
)
// DetectionSender sends detection responses for one request.
type DetectionSender interface {
SendProposals(*plugin_pb.DetectionProposals) error
SendComplete(*plugin_pb.DetectionComplete) error
SendActivity(*plugin_pb.ActivityEvent) error
}
// ExecutionSender sends execution progress/completion responses for one request.
type ExecutionSender interface {
SendProgress(*plugin_pb.JobProgressUpdate) error
SendCompleted(*plugin_pb.JobCompleted) error
}
// JobHandler implements one plugin job type on the worker side.
type JobHandler interface {
Capability() *plugin_pb.JobTypeCapability
Descriptor() *plugin_pb.JobTypeDescriptor
Detect(context.Context, *plugin_pb.RunDetectionRequest, DetectionSender) error
Execute(context.Context, *plugin_pb.ExecuteJobRequest, ExecutionSender) error
}
// WorkerOptions configures one plugin worker process.
type WorkerOptions struct {
AdminServer string
WorkerID string
WorkerVersion string
WorkerAddress string
HeartbeatInterval time.Duration
ReconnectDelay time.Duration
MaxDetectionConcurrency int
MaxExecutionConcurrency int
GrpcDialOption grpc.DialOption
Handlers []JobHandler
Handler JobHandler
}
// Worker runs one plugin job handler over plugin.proto stream.
type Worker struct {
opts WorkerOptions
detectSlots chan struct{}
execSlots chan struct{}
handlers map[string]JobHandler
runningMu sync.RWMutex
runningWork map[string]*plugin_pb.RunningWork
workCancelMu sync.Mutex
workCancel map[string]context.CancelFunc
workerID string
connectionMu sync.RWMutex
connected bool
}
// NewWorker creates a plugin worker instance.
func NewWorker(options WorkerOptions) (*Worker, error) {
if strings.TrimSpace(options.AdminServer) == "" {
return nil, fmt.Errorf("admin server is required")
}
if options.GrpcDialOption == nil {
return nil, fmt.Errorf("grpc dial option is required")
}
if options.HeartbeatInterval <= 0 {
options.HeartbeatInterval = defaultHeartbeatInterval
}
if options.ReconnectDelay <= 0 {
options.ReconnectDelay = defaultReconnectDelay
}
if options.MaxDetectionConcurrency <= 0 {
options.MaxDetectionConcurrency = 1
}
if options.MaxExecutionConcurrency <= 0 {
options.MaxExecutionConcurrency = 1
}
if strings.TrimSpace(options.WorkerVersion) == "" {
options.WorkerVersion = "dev"
}
workerID := strings.TrimSpace(options.WorkerID)
if workerID == "" {
workerID = generateWorkerID()
}
workerAddress := strings.TrimSpace(options.WorkerAddress)
if workerAddress == "" {
hostname, _ := os.Hostname()
workerAddress = hostname
}
opts := options
opts.WorkerAddress = workerAddress
allHandlers := make([]JobHandler, 0, len(opts.Handlers)+1)
if opts.Handler != nil {
allHandlers = append(allHandlers, opts.Handler)
}
allHandlers = append(allHandlers, opts.Handlers...)
if len(allHandlers) == 0 {
return nil, fmt.Errorf("at least one job handler is required")
}
handlers := make(map[string]JobHandler, len(allHandlers))
for i, handler := range allHandlers {
if handler == nil {
return nil, fmt.Errorf("job handler at index %d is nil", i)
}
handlerJobType, err := resolveHandlerJobType(handler)
if err != nil {
return nil, fmt.Errorf("resolve job handler at index %d: %w", i, err)
}
key := normalizeJobTypeKey(handlerJobType)
if key == "" {
return nil, fmt.Errorf("job handler at index %d has empty job type", i)
}
if _, found := handlers[key]; found {
return nil, fmt.Errorf("duplicate job handler for job type %q", handlerJobType)
}
handlers[key] = handler
}
if opts.Handler == nil {
opts.Handler = allHandlers[0]
}
w := &Worker{
opts: opts,
detectSlots: make(chan struct{}, opts.MaxDetectionConcurrency),
execSlots: make(chan struct{}, opts.MaxExecutionConcurrency),
handlers: handlers,
runningWork: make(map[string]*plugin_pb.RunningWork),
workCancel: make(map[string]context.CancelFunc),
workerID: workerID,
}
return w, nil
}
// Run keeps the plugin worker connected and reconnects on stream failures.
func (w *Worker) Run(ctx context.Context) error {
adminAddress := pb.ServerToGrpcAddress(w.opts.AdminServer)
for {
select {
case <-ctx.Done():
return nil
default:
}
if err := w.runOnce(ctx, adminAddress); err != nil {
if ctx.Err() != nil {
return nil
}
glog.Warningf("Plugin worker %s stream ended: %v", w.workerID, err)
}
select {
case <-ctx.Done():
return nil
case <-time.After(w.opts.ReconnectDelay):
}
}
}
func (w *Worker) runOnce(ctx context.Context, adminAddress string) error {
defer w.setConnected(false)
dialCtx, cancelDial := context.WithTimeout(ctx, 5*time.Second)
defer cancelDial()
conn, err := pb.GrpcDial(dialCtx, adminAddress, false, w.opts.GrpcDialOption)
if err != nil {
return fmt.Errorf("dial admin %s: %w", adminAddress, err)
}
defer conn.Close()
client := plugin_pb.NewPluginControlServiceClient(conn)
connCtx, cancelConn := context.WithCancel(ctx)
defer cancelConn()
stream, err := client.WorkerStream(connCtx)
if err != nil {
return fmt.Errorf("open worker stream: %w", err)
}
w.setConnected(true)
sendCh := make(chan *plugin_pb.WorkerToAdminMessage, defaultSendBufferSize)
sendErrCh := make(chan error, 1)
send := func(msg *plugin_pb.WorkerToAdminMessage) bool {
if msg == nil {
return false
}
msg.WorkerId = w.workerID
if msg.SentAt == nil {
msg.SentAt = timestamppb.Now()
}
select {
case <-connCtx.Done():
return false
case sendCh <- msg:
return true
}
}
go func() {
for {
select {
case <-connCtx.Done():
return
case msg := <-sendCh:
if msg == nil {
continue
}
if err := stream.Send(msg); err != nil {
select {
case sendErrCh <- err:
default:
}
cancelConn()
return
}
}
}
}()
if !send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_Hello{Hello: w.buildHello()},
}) {
return fmt.Errorf("send worker hello: stream closed")
}
heartbeatTicker := time.NewTicker(w.opts.HeartbeatInterval)
defer heartbeatTicker.Stop()
go func() {
for {
select {
case <-connCtx.Done():
return
case <-heartbeatTicker.C:
w.sendHeartbeat(send)
}
}
}()
for {
select {
case <-connCtx.Done():
return connCtx.Err()
case err := <-sendErrCh:
return fmt.Errorf("send to admin stream: %w", err)
default:
}
message, err := stream.Recv()
if err != nil {
return fmt.Errorf("recv admin message: %w", err)
}
w.handleAdminMessage(connCtx, message, send)
}
}
// IsConnected reports whether the worker currently has an active stream to admin.
func (w *Worker) IsConnected() bool {
w.connectionMu.RLock()
defer w.connectionMu.RUnlock()
return w.connected
}
func (w *Worker) setConnected(connected bool) {
w.connectionMu.Lock()
w.connected = connected
w.connectionMu.Unlock()
}
func (w *Worker) handleAdminMessage(
ctx context.Context,
message *plugin_pb.AdminToWorkerMessage,
send func(*plugin_pb.WorkerToAdminMessage) bool,
) {
if message == nil {
return
}
switch body := message.Body.(type) {
case *plugin_pb.AdminToWorkerMessage_Hello:
_ = body
case *plugin_pb.AdminToWorkerMessage_RequestConfigSchema:
w.handleSchemaRequest(message.GetRequestId(), body.RequestConfigSchema, send)
case *plugin_pb.AdminToWorkerMessage_RunDetectionRequest:
w.handleDetectionRequest(ctx, message.GetRequestId(), body.RunDetectionRequest, send)
case *plugin_pb.AdminToWorkerMessage_ExecuteJobRequest:
w.handleExecuteRequest(ctx, message.GetRequestId(), body.ExecuteJobRequest, send)
case *plugin_pb.AdminToWorkerMessage_CancelRequest:
cancel := body.CancelRequest
targetID := ""
if cancel != nil {
targetID = strings.TrimSpace(cancel.TargetId)
}
accepted := false
ackMessage := "cancel target is required"
if targetID != "" {
if w.cancelWork(targetID) {
accepted = true
ackMessage = "cancel request accepted"
} else {
ackMessage = "cancel target not found"
}
}
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_Acknowledge{Acknowledge: &plugin_pb.WorkerAcknowledge{
RequestId: message.GetRequestId(),
Accepted: accepted,
Message: ackMessage,
}},
})
case *plugin_pb.AdminToWorkerMessage_Shutdown:
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_Acknowledge{Acknowledge: &plugin_pb.WorkerAcknowledge{
RequestId: message.GetRequestId(),
Accepted: true,
Message: "shutdown acknowledged",
}},
})
default:
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_Acknowledge{Acknowledge: &plugin_pb.WorkerAcknowledge{
RequestId: message.GetRequestId(),
Accepted: false,
Message: "unsupported request body",
}},
})
}
}
func (w *Worker) handleSchemaRequest(requestID string, request *plugin_pb.RequestConfigSchema, send func(*plugin_pb.WorkerToAdminMessage) bool) {
jobType := ""
if request != nil {
jobType = strings.TrimSpace(request.JobType)
}
handler, resolvedJobType, err := w.findHandler(jobType)
if err != nil {
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_ConfigSchemaResponse{ConfigSchemaResponse: &plugin_pb.ConfigSchemaResponse{
RequestId: requestID,
JobType: jobType,
Success: false,
ErrorMessage: err.Error(),
}},
})
return
}
descriptor := handler.Descriptor()
if descriptor == nil || descriptor.JobType == "" {
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_ConfigSchemaResponse{ConfigSchemaResponse: &plugin_pb.ConfigSchemaResponse{
RequestId: requestID,
JobType: resolvedJobType,
Success: false,
ErrorMessage: "handler descriptor is not configured",
}},
})
return
}
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_ConfigSchemaResponse{ConfigSchemaResponse: &plugin_pb.ConfigSchemaResponse{
RequestId: requestID,
JobType: descriptor.JobType,
Success: true,
JobTypeDescriptor: descriptor,
}},
})
}
func (w *Worker) handleDetectionRequest(
ctx context.Context,
requestID string,
request *plugin_pb.RunDetectionRequest,
send func(*plugin_pb.WorkerToAdminMessage) bool,
) {
if request == nil {
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_DetectionComplete{DetectionComplete: &plugin_pb.DetectionComplete{
RequestId: requestID,
Success: false,
ErrorMessage: "run detection request is nil",
}},
})
return
}
handler, resolvedJobType, err := w.findHandler(request.JobType)
if err != nil {
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_DetectionComplete{DetectionComplete: &plugin_pb.DetectionComplete{
RequestId: requestID,
JobType: request.JobType,
Success: false,
ErrorMessage: err.Error(),
}},
})
return
}
workKey := "detect:" + requestID
w.setRunningWork(workKey, &plugin_pb.RunningWork{
WorkId: requestID,
Kind: plugin_pb.WorkKind_WORK_KIND_DETECTION,
JobType: resolvedJobType,
State: plugin_pb.JobState_JOB_STATE_ASSIGNED,
ProgressPercent: 0,
Stage: "queued",
})
w.sendHeartbeat(send)
requestCtx, cancelRequest := context.WithCancel(ctx)
w.setWorkCancel(cancelRequest, requestID)
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_Acknowledge{Acknowledge: &plugin_pb.WorkerAcknowledge{
RequestId: requestID,
Accepted: true,
Message: "detection request accepted",
}},
})
go func() {
detectionSender := &detectionSender{
requestID: requestID,
jobType: resolvedJobType,
send: send,
}
defer func() {
w.clearWorkCancel(requestID)
cancelRequest()
w.clearRunningWork(workKey)
w.sendHeartbeat(send)
}()
select {
case <-requestCtx.Done():
detectionSender.SendComplete(&plugin_pb.DetectionComplete{
Success: false,
ErrorMessage: requestCtx.Err().Error(),
})
return
case w.detectSlots <- struct{}{}:
}
defer func() {
<-w.detectSlots
w.sendHeartbeat(send)
}()
w.setRunningWork(workKey, &plugin_pb.RunningWork{
WorkId: requestID,
Kind: plugin_pb.WorkKind_WORK_KIND_DETECTION,
JobType: resolvedJobType,
State: plugin_pb.JobState_JOB_STATE_RUNNING,
ProgressPercent: 0,
Stage: "detecting",
})
w.sendHeartbeat(send)
if err := handler.Detect(requestCtx, request, detectionSender); err != nil {
detectionSender.SendComplete(&plugin_pb.DetectionComplete{
Success: false,
ErrorMessage: err.Error(),
})
}
}()
}
func (w *Worker) handleExecuteRequest(
ctx context.Context,
requestID string,
request *plugin_pb.ExecuteJobRequest,
send func(*plugin_pb.WorkerToAdminMessage) bool,
) {
if request == nil || request.Job == nil {
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_JobCompleted{JobCompleted: &plugin_pb.JobCompleted{
RequestId: requestID,
Success: false,
ErrorMessage: "execute request/job is nil",
}},
})
return
}
handler, resolvedJobType, err := w.findHandler(request.Job.JobType)
if err != nil {
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_JobCompleted{JobCompleted: &plugin_pb.JobCompleted{
RequestId: requestID,
JobId: request.Job.JobId,
JobType: request.Job.JobType,
Success: false,
ErrorMessage: err.Error(),
}},
})
return
}
select {
case w.execSlots <- struct{}{}:
default:
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_JobCompleted{JobCompleted: &plugin_pb.JobCompleted{
RequestId: requestID,
JobId: request.Job.JobId,
JobType: resolvedJobType,
Success: false,
ErrorMessage: "executor is at capacity",
}},
})
return
}
w.sendHeartbeat(send)
workKey := "exec:" + requestID
w.setRunningWork(workKey, &plugin_pb.RunningWork{
WorkId: request.Job.JobId,
Kind: plugin_pb.WorkKind_WORK_KIND_EXECUTION,
JobType: resolvedJobType,
State: plugin_pb.JobState_JOB_STATE_RUNNING,
ProgressPercent: 0,
Stage: "starting",
})
w.sendHeartbeat(send)
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_Acknowledge{Acknowledge: &plugin_pb.WorkerAcknowledge{
RequestId: requestID,
Accepted: true,
Message: "execute request accepted",
}},
})
go func() {
requestCtx, cancelRequest := context.WithCancel(ctx)
w.setWorkCancel(cancelRequest, requestID, request.Job.JobId)
defer func() {
w.clearWorkCancel(requestID, request.Job.JobId)
cancelRequest()
<-w.execSlots
w.clearRunningWork(workKey)
w.sendHeartbeat(send)
}()
executionSender := &executionSender{
requestID: requestID,
jobID: request.Job.JobId,
jobType: resolvedJobType,
send: send,
onProgress: func(progress float64, stage string) {
w.updateRunningExecution(workKey, progress, stage)
},
}
if err := handler.Execute(requestCtx, request, executionSender); err != nil {
executionSender.SendCompleted(&plugin_pb.JobCompleted{
Success: false,
ErrorMessage: err.Error(),
})
}
}()
}
func (w *Worker) buildHello() *plugin_pb.WorkerHello {
jobTypeKeys := make([]string, 0, len(w.handlers))
for key := range w.handlers {
jobTypeKeys = append(jobTypeKeys, key)
}
sort.Strings(jobTypeKeys)
capabilities := make([]*plugin_pb.JobTypeCapability, 0, len(jobTypeKeys))
jobTypes := make([]string, 0, len(jobTypeKeys))
for _, key := range jobTypeKeys {
handler := w.handlers[key]
if handler == nil {
continue
}
jobType, _ := resolveHandlerJobType(handler)
capability := handler.Capability()
if capability == nil {
capability = &plugin_pb.JobTypeCapability{}
} else {
capability = proto.Clone(capability).(*plugin_pb.JobTypeCapability)
}
if strings.TrimSpace(capability.JobType) == "" {
capability.JobType = jobType
}
capability.MaxDetectionConcurrency = int32(cap(w.detectSlots))
capability.MaxExecutionConcurrency = int32(cap(w.execSlots))
capabilities = append(capabilities, capability)
if capability.JobType != "" {
jobTypes = append(jobTypes, capability.JobType)
}
}
instanceID := generateWorkerID()
return &plugin_pb.WorkerHello{
WorkerId: w.workerID,
WorkerInstanceId: "inst-" + instanceID,
Address: w.opts.WorkerAddress,
WorkerVersion: w.opts.WorkerVersion,
ProtocolVersion: "plugin.v1",
Capabilities: capabilities,
Metadata: map[string]string{
"runtime": "plugin",
"job_types": strings.Join(jobTypes, ","),
},
}
}
func (w *Worker) buildHeartbeat() *plugin_pb.WorkerHeartbeat {
w.runningMu.RLock()
running := make([]*plugin_pb.RunningWork, 0, len(w.runningWork))
for _, work := range w.runningWork {
if work == nil {
continue
}
cloned := *work
running = append(running, &cloned)
}
w.runningMu.RUnlock()
detectUsed := len(w.detectSlots)
execUsed := len(w.execSlots)
return &plugin_pb.WorkerHeartbeat{
WorkerId: w.workerID,
RunningWork: running,
DetectionSlotsUsed: int32(detectUsed),
DetectionSlotsTotal: int32(cap(w.detectSlots)),
ExecutionSlotsUsed: int32(execUsed),
ExecutionSlotsTotal: int32(cap(w.execSlots)),
QueuedJobsByType: map[string]int32{},
Metadata: map[string]string{
"runtime": "plugin",
},
}
}
func (w *Worker) sendHeartbeat(send func(*plugin_pb.WorkerToAdminMessage) bool) {
if send == nil {
return
}
send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_Heartbeat{
Heartbeat: w.buildHeartbeat(),
},
})
}
func (w *Worker) setRunningWork(key string, work *plugin_pb.RunningWork) {
if strings.TrimSpace(key) == "" || work == nil {
return
}
w.runningMu.Lock()
w.runningWork[key] = work
w.runningMu.Unlock()
}
func (w *Worker) clearRunningWork(key string) {
w.runningMu.Lock()
delete(w.runningWork, key)
w.runningMu.Unlock()
}
func (w *Worker) updateRunningExecution(key string, progress float64, stage string) {
w.runningMu.Lock()
if running := w.runningWork[key]; running != nil {
running.ProgressPercent = progress
if strings.TrimSpace(stage) != "" {
running.Stage = stage
}
running.State = plugin_pb.JobState_JOB_STATE_RUNNING
}
w.runningMu.Unlock()
}
type detectionSender struct {
requestID string
jobType string
send func(*plugin_pb.WorkerToAdminMessage) bool
}
func (s *detectionSender) SendProposals(proposals *plugin_pb.DetectionProposals) error {
if proposals == nil {
return fmt.Errorf("detection proposals are nil")
}
if proposals.RequestId == "" {
proposals.RequestId = s.requestID
}
if proposals.JobType == "" {
proposals.JobType = s.jobType
}
if !s.send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_DetectionProposals{DetectionProposals: proposals},
}) {
return fmt.Errorf("stream closed")
}
return nil
}
func (s *detectionSender) SendComplete(complete *plugin_pb.DetectionComplete) error {
if complete == nil {
return fmt.Errorf("detection complete is nil")
}
if complete.RequestId == "" {
complete.RequestId = s.requestID
}
if complete.JobType == "" {
complete.JobType = s.jobType
}
if !s.send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_DetectionComplete{DetectionComplete: complete},
}) {
return fmt.Errorf("stream closed")
}
return nil
}
func (s *detectionSender) SendActivity(activity *plugin_pb.ActivityEvent) error {
if activity == nil {
return fmt.Errorf("detection activity is nil")
}
if activity.CreatedAt == nil {
activity.CreatedAt = timestamppb.Now()
}
if activity.Source == plugin_pb.ActivitySource_ACTIVITY_SOURCE_UNSPECIFIED {
activity.Source = plugin_pb.ActivitySource_ACTIVITY_SOURCE_DETECTOR
}
update := &plugin_pb.JobProgressUpdate{
RequestId: s.requestID,
JobType: s.jobType,
State: plugin_pb.JobState_JOB_STATE_RUNNING,
ProgressPercent: 0,
Stage: activity.Stage,
Message: activity.Message,
Activities: []*plugin_pb.ActivityEvent{activity},
UpdatedAt: timestamppb.Now(),
}
if !s.send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_JobProgressUpdate{JobProgressUpdate: update},
}) {
return fmt.Errorf("stream closed")
}
return nil
}
type executionSender struct {
requestID string
jobID string
jobType string
send func(*plugin_pb.WorkerToAdminMessage) bool
onProgress func(progress float64, stage string)
}
func (s *executionSender) SendProgress(progress *plugin_pb.JobProgressUpdate) error {
if progress == nil {
return fmt.Errorf("job progress is nil")
}
if progress.RequestId == "" {
progress.RequestId = s.requestID
}
if progress.JobId == "" {
progress.JobId = s.jobID
}
if progress.JobType == "" {
progress.JobType = s.jobType
}
if progress.UpdatedAt == nil {
progress.UpdatedAt = timestamppb.Now()
}
if s.onProgress != nil {
s.onProgress(progress.ProgressPercent, progress.Stage)
}
if !s.send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_JobProgressUpdate{JobProgressUpdate: progress},
}) {
return fmt.Errorf("stream closed")
}
return nil
}
func (s *executionSender) SendCompleted(completed *plugin_pb.JobCompleted) error {
if completed == nil {
return fmt.Errorf("job completed is nil")
}
if completed.RequestId == "" {
completed.RequestId = s.requestID
}
if completed.JobId == "" {
completed.JobId = s.jobID
}
if completed.JobType == "" {
completed.JobType = s.jobType
}
if completed.CompletedAt == nil {
completed.CompletedAt = timestamppb.Now()
}
if !s.send(&plugin_pb.WorkerToAdminMessage{
Body: &plugin_pb.WorkerToAdminMessage_JobCompleted{JobCompleted: completed},
}) {
return fmt.Errorf("stream closed")
}
return nil
}
func generateWorkerID() string {
random := make([]byte, 3)
if _, err := rand.Read(random); err != nil {
return fmt.Sprintf("plugin-%d", time.Now().UnixNano())
}
return "plugin-" + hex.EncodeToString(random)
}
func (w *Worker) setWorkCancel(cancel context.CancelFunc, keys ...string) {
if cancel == nil {
return
}
w.workCancelMu.Lock()
defer w.workCancelMu.Unlock()
for _, key := range keys {
key = strings.TrimSpace(key)
if key == "" {
continue
}
w.workCancel[key] = cancel
}
}
func (w *Worker) clearWorkCancel(keys ...string) {
w.workCancelMu.Lock()
defer w.workCancelMu.Unlock()
for _, key := range keys {
key = strings.TrimSpace(key)
if key == "" {
continue
}
delete(w.workCancel, key)
}
}
func (w *Worker) cancelWork(targetID string) bool {
targetID = strings.TrimSpace(targetID)
if targetID == "" {
return false
}
w.workCancelMu.Lock()
cancel := w.workCancel[targetID]
w.workCancelMu.Unlock()
if cancel == nil {
return false
}
cancel()
return true
}
func (w *Worker) findHandler(jobType string) (JobHandler, string, error) {
trimmed := strings.TrimSpace(jobType)
if trimmed == "" {
if len(w.handlers) == 1 {
for _, handler := range w.handlers {
resolvedJobType, err := resolveHandlerJobType(handler)
return handler, resolvedJobType, err
}
}
return nil, "", fmt.Errorf("job type is required when worker serves multiple job types")
}
key := normalizeJobTypeKey(trimmed)
handler := w.handlers[key]
if handler == nil {
return nil, "", fmt.Errorf("job type %q is not handled by this worker", trimmed)
}
resolvedJobType, err := resolveHandlerJobType(handler)
if err != nil {
return nil, "", err
}
return handler, resolvedJobType, nil
}
func resolveHandlerJobType(handler JobHandler) (string, error) {
if handler == nil {
return "", fmt.Errorf("job handler is nil")
}
if descriptor := handler.Descriptor(); descriptor != nil {
if jobType := strings.TrimSpace(descriptor.JobType); jobType != "" {
return jobType, nil
}
}
if capability := handler.Capability(); capability != nil {
if jobType := strings.TrimSpace(capability.JobType); jobType != "" {
return jobType, nil
}
}
return "", fmt.Errorf("handler job type is not configured")
}
func normalizeJobTypeKey(jobType string) string {
return strings.ToLower(strings.TrimSpace(jobType))
}