在這種情境中，你大概會希望「每個容器只有一個（Uvicorn）行程」，因為你已在叢集層級處理了複本。

所以這種情況下，你「不會」想在容器中使用多個 workers（例如用 `--workers` 命令列選項）。你會想每個容器只執行「一個 Uvicorn 行程」（但可能有多個容器）。

在容器內再放一個行程管理器（如同多 workers 的情況）只會增加「不必要的複雜度」，而你很可能已用叢集系統處理好了。

### 多行程容器與特殊情境 { #containers-with-multiple-processes-and-special-cases }

當然，有些「特殊情況」你可能會想在「一個容器內」執行數個「Uvicorn worker 行程」。

在這些情況中，你可以用 `--workers` 命令列選項來設定要啟動的 workers 數量：

```{ .dockerfile .annotate }
FROM python:3.14

        val gradleStep = CompileAll(buildModel, buildModel.stages[0]).steps.getGradleStep(GRADLE_RUNNER_STEP_NAME)
        assertEquals(
            listOf(
                "-Dorg.gradle.workers.max=%maxParallelForks%",
                "-PmaxParallelForks=%maxParallelForks%",
                "-Dorg.gradle.internal.plugins.portal.url.override=%gradle.plugins.portal.url%",
                "-Dscan.value.tcPipeline=master",

                "\"-Porg.gradle.performance.db.username=%performance.db.username%\"",
                "-DenableTestDistribution=%enableTestDistribution%",
                "-Dorg.gradle.workers.max=%maxParallelForks%",
                "-PmaxParallelForks=%maxParallelForks%",
                PLUGINS_PORTAL_URL_OVERRIDE,
                "-Dscan.value.tcPipeline=master",
                "-s",

type NotificationGroup struct {
	workers    *workers.Workers
	errs       []NotificationPeerErr
	retryCount int
}

// WithNPeers returns a new NotificationGroup with length of errs slice upto nerrs,
// upon Wait() errors are returned collected from all tasks.
func WithNPeers(nerrs int) *NotificationGroup {
	if nerrs <= 0 {
		nerrs = 1
	}
	wk, _ := workers.New(nerrs)

	recordsRead int                       // number of records read in current slice
	input       chan *queueItem           // input for workers
	queue       chan *queueItem           // output from workers in order
	err         error                     // global error state, only touched by Reader.Read
	bufferPool  bpool.Pool[[]byte]        // pool of []byte objects for input

		config.HelpKV{
			Key:         apiReplicationMaxWorkers,
			Description: `set the maximum number of replication workers` + defaultHelpPostfix(apiReplicationMaxWorkers),
			Optional:    true,
			Type:        "number",
		},
		config.HelpKV{
			Key:         apiTransitionWorkers,
			Description: `set the number of transition workers` + defaultHelpPostfix(apiTransitionWorkers),
			Optional:    true,
			Type:        "number",
		},

Pero en muchos casos, querrás ejecutar varios worker processes al mismo tiempo.

### Múltiples Procesos - Workers { #multiple-processes-workers }

import org.gradle.api.tasks.PathSensitivity
import org.gradle.api.tasks.TaskAction
// Using star import to workaround https://youtrack.jetbrains.com/issue/KTIJ-24390
import org.gradle.kotlin.dsl.*
import org.gradle.workers.WorkerExecutor
import javax.inject.Inject


@CacheableTask
abstract class IncubatingApiReportTask : DefaultTask() {

    private val additionalClasspath = project.objects.fileCollection().apply {

</component>...

          del /f /s /q %USERPROFILE%\.gradle\init.d\*.*
          mkdir %USERPROFILE%\.gradle\init.d
          copy .ci\init.gradle %USERPROFILE%\.gradle\init.d\
          (
            echo call %GRADLEW_BAT% --max-workers=4 -Dbwc.checkout.align=true %GRADLE_TASK% ^|^| exit /b 1