Network bridge stopped after network outage

[johndoe <do...@gmail.com>]

Good Afternoon,

We're currently performing a trial of a network connector between two sites
in a production environment to push data from one site to another (just two
brokers, Site A and Site B).   We experienced a network outage (in the order
of minutes), and noticed that the network bridge was stopped, and did not
re-establish the connection when the network came back up again.  It was
only when we restarted Site A's broker that the network bridge was
re-established.

Our network connector (on Site A) is configured as follows:

 <networkConnector uri="static://(tcp://10.123.4.5:61616)"
	name="Site A Publisher"
	duplex="false"
	conduitSubscriptions="false"
	decreaseNetworkConsumerPriority="false">

The following is an excerpt from the activemq broker's log file (Site A):

2015-09-26 12:44:15,149 | WARN  | Caught an exception processing local
command | org.apache.activemq.network.DemandForwardingBridgeSupport |
ActiveMQ Connection Dispatcher: vm://SiteABroker#8594
java.net.SocketException: Connection reset by peer: socket write error
	at java.net.SocketOutputStream.socketWrite0(Native Method)[:1.7.0_51]
	at java.net.SocketOutputStream.socketWrite(Unknown Source)[:1.7.0_51]
...
...
2015-09-26 12:44:15,150 | INFO  | Network connection between
vm://SiteABroker#8594 and tcp:///10.123.4.5:61616@49214 shutdown due to a
local error: java.net.SocketException: Connection reset by peer: socket
write error | org.apache.activemq.network.DemandForwardingBridgeSupport |
ActiveMQ Connection Dispatcher: vm://SiteABroker#8594
2015-09-26 12:44:15,149 | INFO  | Network connection between
vm://SiteABroker#8594 and tcp:///10.123.4.5:61616@49214 shutdown due to a
local error: java.net.SocketException: Connection reset |
org.apache.activemq.network.DemandForwardingBridgeSupport | ActiveMQ
Transport: tcp:///10.123.4.5:61616@49214
2015-09-26 12:44:15,151 | INFO  | Establishing network connection from
vm://SiteABroker?async=false&network=true to tcp://10.123.4.5:61616 |
org.apache.activemq.network.DiscoveryNetworkConnector | Simple Discovery
Agent-4298
2015-09-26 12:44:15,151 | WARN  | Caught an exception processing local
command | org.apache.activemq.network.DemandForwardingBridgeSupport |
ActiveMQ Connection Dispatcher: vm://SiteABroker#8594
java.net.SocketException: Connection reset
	at java.net.SocketInputStream.read(Unknown Source)[:1.7.0_51]
	at java.net.SocketInputStream.read(Unknown Source)[:1.7.0_51]
	at
org.apache.activemq.transport.tcp.TcpBufferedInputStream.fill(TcpBufferedInputStream.java:50)[activemq-client-5.11.1.jar:5.11.1]
..
...
*2015-09-26 12:44:15,161 | INFO  | SiteABroker bridge to SiteB stopped |
org.apache.activemq.network.DemandForwardingBridgeSupport | ActiveMQ
BrokerService[SiteABroker] Task-12980*
2015-09-26 13:14:07,239 | INFO  | Apache ActiveMQ 5.11.1 (SiteABroker,
ID:SiteA-52538-1443025978152-0:1) is shutting down |
org.apache.activemq.broker.BrokerService | ActiveMQ ShutdownHook
..

The broker was restarted at 13:14:07 as can be seen.  I've since noticed
that the static URI contains redundant '//' characters, but I can't imagine
this would cause an issue.

When attempting to reproduce the issue, the network bridge is always
re-established, but I've never seen the exceptions listed in the log file
excerpt above.  

Has anyone seen this before?

Regards,
John



--
View this message in context: http://activemq.2283324.n4.nabble.com/Network-bridge-stopped-after-network-outage-tp4702764.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.