Jiri Danek created ARTEMIS-1177:
-----------------------------------
Summary: clientFailureCheckPeriod option does not do what Doc says
it does
Key: ARTEMIS-1177
URL: https://issues.apache.org/jira/browse/ARTEMIS-1177
Project: ActiveMQ Artemis
Issue Type: Bug
Components: Broker
Affects Versions: 2.1.0
Reporter: Jiri Danek
Priority: Minor
If I run the following
{noformat}
@Test fun `clientFailureCheckPeriod option`() {
val f =
org.apache.activemq.artemis.jms.client.ActiveMQConnectionFactory("tcp://127.0.0.1:61616?clientFailureCheckPeriod=1000&connectionTTL=13000")
val c = f.createConnection()
Thread.sleep(5 * 1000)
Runtime.getRuntime().exec("kill -SIGSTOP 27314")
println(java.time.LocalDateTime.now())
Thread.sleep(120 * 1000)
}
{noformat}
(where 27314 is PID of Artemis), I get the following output
{noformat}
2017-05-22T14:18:07.027
14:18:28,023 WARN AMQ212037: Connection failure has been detected: AMQ119011:
Did not receive data from server for
org.apache.activemq.artemis.core.remoting.impl.netty.NettyConnection@1d90b3b6[local=
/127.0.0.1:54584, remote=/127.0.0.1:61616] [code=CONNECTION_TIMEDOUT]
{noformat}
There is approx 13 second delay between broker going down and client noticing
it. Given the documentation for clientFailureCheckPeriod, I'd expect the client
notices it after just 1 second.
"If the client does not receive any packets for a configurable number of
milliseconds then it will consider the connection failed and will either
initiate failover, or call any FailureListener instances (or ExceptionListener
instances if you are using JMS) depending on how it has been configured."
https://github.com/apache/activemq-artemis/blob/master/docs/user-manual/en/connection-ttl.md
I think that the documentation is wrong.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
