On 07/25/2014 12:02 PM, anow789 wrote:
My ActiveMQ setup is 2 brokers using kahadb on a shared ntfs file system. The
URI I am using for a .NET client connections is:
,
<value>failover:(tcp://host1:61616,tcp://host2:61616)?transport.randomize=false</value>
Everything works fine when connecting to host1, but when host2 is the active
broker there is a 15-20 sec delay for every message sent/received due to the
connection failure to host1. Randomize is set to false so a connection to
host1 is tried first as it's the primary broker.
What can I do to remove the connection delay when the secondary broker is
active? I've tried a few different timeout options with no luck.
Thank you,
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Failover-Delay-tp4683666.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Are you creating a new connection for each send / receive ? If so
that's a bit of an anti-pattern for a connection oriented API like NMS /
JMS.
You can configure the connection timeout option for the tcp transport
via the transport.connectTimeout value, default is 30 secs
--
Tim Bish
Sr Software Engineer | RedHat Inc.
[email protected] | www.fusesource.com | www.redhat.com
skype: tabish121 | twitter: @tabish121
blog: http://timbish.blogspot.com/
