Just want the final solution for some previous similar issues(*nodes marked as dead)

[yudidi]

Please use the following questions as a guideline to help me answer
your issue/question without further inquiry. Thank you.

Which version of Elastic are you using?

elastic.v5 (for Elasticsearch 5.x)

Please describe the expected behavior

Just want the final solution for some previous similar issues:

1. When the context is cancelled the node is marked dead #484
2. AWS endpoint with varying IP address #1091

Please describe the actual behavior

I saw this log sometimes in my app. But I don't know what causes this and I don't find the final solution after I view some similar issues. Could you please tell me the final solution?

ELASTIC 2020/12/14 10:53:38 elastic: http://xxxdomainxxx.elasticsearch.aliyuncs.com:9200 is dead
ELASTIC 2020/12/14 10:53:38 elastic: all 1 nodes marked as dead; resurrecting them to prevent deadlock

Any steps to reproduce the behavior?

It's hard to describe. The information I can provide is as following, are they enough?

My current code is:

elastic.SetSniff(false)

Should I solve it by closing the health check?

elastic.SetSniff(false),
elastic.SetHealthcheck(false),

[yudidi]

I found the source code:

elastic/client.go

Lines 1164 to 1173 in 423089d

	// We have a deadlock here: All nodes are marked as dead.
	// If sniffing is disabled, connections will never be marked alive again.
	// So we are marking them as alive--if sniffing is disabled.
	// They'll then be picked up in the next call to PerformRequest.
	if !c.snifferEnabled {
	c.errorf("elastic: all %d nodes marked as dead; resurrecting them to prevent deadlock", len(c.conns))
	for _, conn := range c.conns {
	conn.MarkAsAlive()
	}
	}

If you want to know what is the purpose of the code, click the red rectangle for more detail.