[enhancement] PR #2319 ctrd/container_lock.go TrylockWithRetry is not FIFO

[xiaoxubeii]

Ⅰ. Issue Description

In #2319 ctrd/container_lock.go:

func (l *containerLock) TrylockWithRetry(ctx context.Context, id string) bool {
	var retry = 32


	for {
		ok := l.Trylock(id)
		if ok {
			return true
		}


		// sleep random duration by retry
		select {
		case <-time.After(time.Millisecond * time.Duration(rand.Intn(retry))):
			if retry < 2048 {
				retry = retry << 1
			}
			continue
		case <-ctx.Done():
			return false
		}
	}
}

The retry mechanism is not FIFO, so the probability of getting a lock is dependent on the retry intervals. Is this acceptable?

Ⅱ. Describe what happened

Ⅲ. Describe what you expected to happen

Ⅳ. How to reproduce it (as minimally and precisely as possible)

Ⅴ. Anything else we need to know?

Ⅵ. Environment:

• pouch version (use pouch version):
• OS (e.g. from /etc/os-release):
• Kernel (e.g. uname -a):
• Install tools:
• Others:

[fuweid]

basically, we don't consider it as normal mode, as you mentioned. It's like starvation mode. I think it's ok here.

[xiaoxubeii]

LGTM