Files
Byron.wang ca3d00229e Feat/Implement structured logging and Trace ID propagation (#552)
* use slog instead of log package and format to new log schema

* update the environment name to LOG_OUTPUT_FORMAT

* add the env to .env.example

* fix log reference error

* change the order of milldlewares

* delete unused code

* fix the concurrently session potential race condition

* fix the log format in tests

* update the duplicate code

* refactor: convert log functions to slog structured format

- Change log.Error/Info/Warn/Debug/Panic to accept msg + key-value pairs
- Remove printf-style formatting from log functions
- Update log calls in internal/cluster, internal/db, internal/core/session_manager
- Remove unused 'initialized' variable from log package
- Remaining files will be updated in follow-up commits

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* refactor: update all log call sites to use slog structured format

Convert all log.Error, log.Info, log.Warn, log.Debug, and log.Panic
calls from printf-style formatting to slog key-value pairs.

Before: log.Error("failed to do something: %s", err.Error())
After:  log.Error("failed to do something", "error", err)

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* refactor: update cmd/ log calls to use slog structured format

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* feat: implement GnetLogger for structured logging in gnet

* refactor: remove deprecated log visibility functions and related calls

* feat: enhance session management with trace and identity context propagation

* feat: implement serverless transaction handler and writer for plugin runtime

* refactor: rename context field to traceCtx in RealBackwardsInvocation

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: Yeuoly <admin@srmxy.cn>
2025-12-30 11:00:48 +08:00

128 lines
3.4 KiB
Go

package tasks
import (
"errors"
"time"
"github.com/langgenius/dify-plugin-daemon/internal/cluster"
"github.com/langgenius/dify-plugin-daemon/internal/db"
"github.com/langgenius/dify-plugin-daemon/internal/types/app"
"github.com/langgenius/dify-plugin-daemon/internal/types/models"
"github.com/langgenius/dify-plugin-daemon/pkg/utils/log"
"github.com/langgenius/dify-plugin-daemon/pkg/utils/mapping"
)
var (
// reference to running installation tasks
installingTasks mapping.Map[string, bool]
)
func startTasks(taskIDs []string) {
for _, taskID := range taskIDs {
log.Info("start new install task", "task_id", taskID)
installingTasks.Store(taskID, true)
}
}
func endTasks(taskIDs []string) {
for _, taskID := range taskIDs {
log.Info("install task finished", "task_id", taskID)
installingTasks.Delete(taskID)
}
}
// RecycleTasks is a finalizer to update the status of all running installation tasks to failed
// when the daemon is shutting down
func RecycleTasks() error {
var errs []error
installingTasks.Range(func(taskId string, _ bool) bool {
log.Info("updating task status to failed", "task_id", taskId)
// update task status to failed
task, err := db.GetOne[models.InstallTask](
db.Equal("id", taskId),
db.InArray("status", []any{
string(models.InstallTaskStatusRunning),
string(models.InstallTaskStatusPending)},
),
)
if err != nil {
errs = append(errs, err)
return true
}
task.Status = models.InstallTaskStatusFailed
for i := range task.Plugins {
plugin := &task.Plugins[i]
plugin.Status = models.InstallTaskStatusFailed
plugin.Message = "An unexpected daemon shutdown occurred"
}
err = db.Update(task)
if err != nil {
errs = append(errs, err)
}
return true
})
return errors.Join(errs...)
}
func markTasksAsTimeout(tasks []*models.InstallTask) {
if len(tasks) == 0 {
return
}
for _, task := range tasks {
task.Status = models.InstallTaskStatusFailed
for i := range task.Plugins {
plugin := &task.Plugins[i]
plugin.Status = models.InstallTaskStatusFailed
plugin.Message = "Task timed out but not properly terminated"
}
}
err := db.Update(tasks)
if err != nil {
log.Error("failed to update tasks", "error", err)
}
}
// Just in case some tasks may stuck for some reason that we don't know.
func MonitorTimeoutTasks(cluster *cluster.Cluster, config *app.Config) {
go func() {
var timeout time.Duration
if config.Platform == app.PLATFORM_SERVERLESS {
timeout = time.Duration(config.DifyPluginServerlessConnectorLaunchTimeout) * time.Second
} else {
timeout = time.Duration(config.PythonEnvInitTimeout) * time.Second
}
// add some tolerance to timeout to avoid race condition
timeout = timeout + time.Minute
ticker := time.NewTicker(time.Minute)
defer ticker.Stop()
for range ticker.C {
if !cluster.IsMaster() {
continue
}
tasksToProcess := []*models.InstallTask{}
// get all tasks that are pending or running
tasks, err := db.GetAll[models.InstallTask](
db.InArray("status", []any{
string(models.InstallTaskStatusPending),
string(models.InstallTaskStatusRunning),
}),
)
if err != nil {
log.Error("failed to get all tasks", "error", err)
continue
}
for i := range tasks {
task := &tasks[i]
if time.Since(task.CreatedAt) > timeout {
log.Info("task timed out", "task_id", task.ID)
tasksToProcess = append(tasksToProcess, task)
}
}
markTasksAsTimeout(tasksToProcess)
}
}()
}