// the worker has yet to start) then we don't need to start the worker.
      if (workerRunningState == RUNNING || workerRunningState == QUEUED) {
        queue.add(task);
        return;
      }

      oldRunCount = workerRunCount;

      // If the worker is not yet running, the delegate Executor might reject our attempt to start

we don't consider this as a vulnerability, given that models can exhaust
resources in many different ways and solutions exist to prevent this from
happening (e.g., rate limits, ACLs, monitors to restart broken servers).

### Model sharing

If the multitenant design allows sharing models, make sure that tenants and
users are aware of the security risks detailed here and that they are going to

        this.maxResourceAge = maxAge;
        this.cleanupInterval = cleanupInterval;
        this.leakDetectionEnabled = enableLeakDetection;
        this.autoCleanupEnabled = enableAutoCleanup;

        // Restart cleanup task with new interval
        startCleanupTask();

        log.info("ResourceManager configured: maxAge={} ms, cleanupInterval={} ms, leakDetection={}, autoCleanup={}", maxAge,

	for _, file := range files {
		go removeAll(file)
	}

	// Remove the entire folder in case there are leftovers that didn't get cleaned up before restart.
	go removeAll(pathJoin(diskPath, minioMetaTmpBucket+"-old"))

	// Renames and schedules for purging all bucket metacache.
	go renameAllBucketMetacache(diskPath)
}

storage by _(newly_added_servers\*m)_ more drives, taking the total count to _(existing_servers\*m)+(newly_added_servers\*m)_ drives. New object upload requests automatically start using the least used cluster. This expansion strategy works endlessly, so you can perpetually expand your clusters as needed.  When you restart, it is immediate and non-disruptive to the applications. Each group of servers in the command-line is called a pool. There are 2 server pools in this example. New objects are placed...

### 更嚴重的錯誤 - 當機 { #bigger-errors-crashes }

然而，仍可能有一些情況，我們寫的程式碼「讓整個應用當機」，使 Uvicorn 與 Python 都崩潰。💥

即便如此，你大概也不會希望應用因為某處錯誤就一直處於死亡狀態，你可能會希望它「繼續運作」，至少讓沒有壞掉的「路徑操作（path operations）」能持續服務。

### 當機後重新啟動 { #restart-after-crash }

在這些會讓「執行中行程」整個崩潰的嚴重錯誤案例裡，你會希望有個外部元件負責「重新啟動」該行程，至少嘗試幾次...

/// tip

...不過，如果整個應用「一啟動就立刻」崩潰，那持續無止境地重啟大概沒有意義。但在這類情況下，你很可能會在開發過程中就發現，或至少在部署後馬上注意到。

./mc mb minio2/newbucket2
# delete bucket2 on minio2. This should replicate to minio1 after it comes online.
./mc rb minio2/bucket2

# Restart minio1 instance
minio server --address ":9001" --console-address ":10000" /tmp/minio1/{1...4} >/tmp/minio1_1.log 2>&1 &
sleep 200

# Test whether most recent tag update on minio2 is replicated to minio1

* Bump to container-vm-v20160321 ([#23313](https://github.com/kubernetes/kubernetes/pull/23313), [@zmerlynn](https://github.com/zmerlynn))
* Remove the restart-kube-proxy and restart-apiserver functions ([#23180](https://github.com/kubernetes/kubernetes/pull/23180), [@roberthbailey](https://github.com/roberthbailey))

- Fixed a bug where, after a kubelet restart, regular containers in a pod with a
  sidecar (initContainer with restartPolicy: Always) and a startupProbe failed
  to restart after crashing. Affected pods remained stuck with RestartCount: 0

## Running on Startup and Restarts { #running-on-startup-and-restarts }

There is normally another tool in charge of **starting and running** your container.

It could be **Docker** directly, **Docker Compose**, **Kubernetes**, a **cloud service**, etc.