CoroutineWorker的doWork方法在Job取消后自动重启问题求助

执行14次服务器API调用时，首次执行到第14次API时触发以下错误：

kotlinx.coroutines.JobCancellationException: Job was cancelled; job=JobImpl{Cancelling}@c8489c9
kotlinx.coroutines.JobCancellationException: Job was cancelled; job=JobImpl{Cancelling}@c8489c9
Job was cancelled
Work [ id=690a0f6f-e2e4-4ea2-baaf-86650b02608b, tags={ xxx.DownloadWorker,download_entity_service_tag } ] was cancelled
androidx.work.impl.WorkerStoppedException
at androidx.work.impl.WorkerWrapper.interrupt(WorkerWrapper.kt:348)
at androidx.work.impl.Processor.interrupt(Processor.java:439)
at androidx.work.impl.Processor.stopWork(Processor.java:259)
at androidx.work.impl.utils.StopWorkRunnable.run(StopWorkRunnable.kt:49)
at androidx.work.impl.utils.SerialExecutorImpl$Task.run(SerialExecutorImpl.java:96)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:644)
at java.lang.Thread.run(Thread.java:1012)

DownloadWorker核心代码如下：

class DownloadWorker(context: Context, parameters: WorkerParameters ) :
    CoroutineWorker(context, parameters) {
    private val tag by lazy { this::class.java.simpleName }
    private lateinit var wifiLock: WifiManager.WifiLock


    override suspend fun doWork(): Result {
        val syncType = getSyncType()
        val progress = applicationContext.getString(R.string.download_worker_start_download)
        displayServiceNotification(createForegroundInfo(progress))
        acquireWifiLock()
        runCatching {
            download(getSyncMethod(),DataToBeSyncedExtractor.generate(syncType))
            finishDownloadProcedure(syncType)
        }.onFailure {
            w(it)
            releaseWifiLock()
            return Result.failure()
        }
        releaseWifiLock()
        return Result.success()
    }

      ...其他代码
}

断点显示代码正确进入onFailure分支并返回Result.failure()，但doWork方法却自动重启，重新下载全部14个API的数据，最终无错误完成。预期返回Result.failure()后不会从头重启任务，寻求原因及解决办法。

原因分析 #

• 任务被强制终止而非正常失败：错误栈中的WorkerStoppedException和JobCancellationException表明，任务是先被系统/WorkManager强制取消，触发协程Job取消后才进入onFailure分支。此时WorkManager会判定任务为意外终止，而非用户主动返回失败，因此会触发默认重试逻辑。
• 默认重试策略生效：WorkManager的WorkRequest默认配置了重试策略（如线性回退重试），当任务被意外取消或抛出未捕获异常时，会自动重启任务。

解决办法 #

1. 区分取消场景和业务失败 #

在doWork中单独捕获WorkerStoppedException，明确区分任务被系统取消和业务逻辑错误的场景，避免触发重试：

override suspend fun doWork(): Result {
    val syncType = getSyncType()
    val progress = applicationContext.getString(R.string.download_worker_start_download)
    displayServiceNotification(createForegroundInfo(progress))
    acquireWifiLock()

    return try {
        download(getSyncMethod(), DataToBeSyncedExtractor.generate(syncType))
        finishDownloadProcedure(syncType)
        releaseWifiLock()
        Result.success()
    } catch (e: WorkerStoppedException) {
        // 任务被系统强制取消，返回failure但不触发重试
        releaseWifiLock()
        Result.failure()
    } catch (e: Exception) {
        // 业务逻辑错误，根据需求决定是否重试
        w(e)
        releaseWifiLock()
        Result.failure()
    }
}

2. 显式配置重试策略 #

创建WorkRequest时，自定义重试规则，比如禁用重试或仅在特定错误时重试：

val downloadWork = OneTimeWorkRequestBuilder<DownloadWorker>()
    // 禁用重试：自定义RetryPolicy始终返回false
    .setRetryPolicy(object : RetryPolicy {
        override fun shouldRetry(result: Result): Boolean {
            return false
        }
    })
    .build()

如果需要针对特定错误重试，可以在shouldRetry中判断result的错误类型。

3. 排查任务被取消的根源 #

• 检查任务执行时长是否超过WorkManager后台Worker的默认超时限制；
• 确认前台服务通知配置正确，避免系统将Worker判定为后台任务而终止；
• 排查是否存在系统资源紧张导致的任务终止。

内容的提问来源于stack exchange，提问作者ricP87