Failover- use multicast, then static
Is it possible to use MultiCast for failover, but in the event that all those brokers become unavailable, you would use a static ip address? failover://(discovery:multicast://default,tcp://test.you.com:61616) -- View this message in context: http://www.nabble.com/Failover--use-multicast%2C-then-static-tp15358576s2354p15358576.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Re: Constantly adding and removing consumer
I have seen this same behaviour an am interested in any info on this! david.la wrote: > > Hi, > > I'm using activemq with camel and when i turn the debug logs, I'm seeing > activity every second or so. It's adding and removing consumer. > > I'd like to know what is that for may be i'm misconfiguring my activemq > because even if there is no external activity, it's still happening. > > I recently found that the advisory module was driving ActiveMQ to use more > and more memory and then crash. I disabled it. I want to make sure the > constant activity i'm seeing the logs will not result in the same kind of > issues. > > The debug messages i'm seeing look like: > > DEBUG AbstractRegion - Removing consumer: > ID:test.com-38805-1205442501944-3:2:1016:1 > DEBUG AbstractRegion - Removing consumer: > ID:test.com-38805-1205442501944-3:1:1016:1 > DEBUG AbstractRegion - Adding consumer: > ID:test.com-38805-1205442501944-3:1:1017:1 > DEBUG AbstractRegion - Adding consumer: > ID:test.com-38805-1205442501944-3:2:1017:1 > DEBUG AbstractRegion - Removing consumer: > ID:test.com-38805-1205442501944-3:2:1017:1 > DEBUG AbstractRegion - Removing consumer: > ID:test.com-38805-1205442501944-3:1:1017:1 > DEBUG AbstractRegion - Adding consumer: > ID:test.com-38805-1205442501944-3:2:1018:1 > DEBUG AbstractRegion - Adding consumer: > ID:test.com-38805-1205442501944-3:1:1018:1 > DEBUG AbstractRegion - Removing consumer: > ID:test.com-38805-1205442501944-3:1:1018:1 > DEBUG AbstractRegion - Removing consumer: > ID:test.com-38805-1205442501944-3:2:1018:1 > DEBUG AbstractRegion - Adding consumer: > ID:test.com-38805-1205442501944-3:1:1019:1 > DEBUG AbstractRegion - Adding consumer: > ID:test.com-38805-1205442501944-3:2:1019:1 > DEBUG AbstractRegion - Removing consumer: > ID:test.com-38805-1205442501944-3:1:1019:1 > DEBUG AbstractRegion - Removing consumer: > ID:test.com-38805-1205442501944-3:2:1019:1 > DEBUG AbstractRegion - Adding consumer: > ID:test.com-38805-1205442501944-3:2:1020:1 > DEBUG AbstractRegion - Adding consumer: > ID:test.com-38805-1205442501944-3:1:1020:1 > DEBUG AbstractRegion - Removing consumer: > ID:test.com-38805-1205442501944-3:1:1020:1 > DEBUG AbstractRegion - Removing consumer: > ID:test.com-38805-1205442501944-3:2:1020:1 > DEBUG AbstractRegion - Adding consumer: > ID:test.com-38805-1205442501944-3:2:1021:1 > DEBUG AbstractRegion - Adding consumer: > ID:test.com-38805-1205442501944-3:1:1021:1 > > I have declared my camel routing (inside activemq.xml) as > > ---8ß- > xmlns="http://activemq.apache.org/camel/schema/spring";> > > Where bean “myProcessor” class extends > “org.apache.camel.processor.DelegateProcessor” > > > -- View this message in context: http://www.nabble.com/Constantly-adding-and-removing-consumer-tp16038772s2354p16040946.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
timeStampingBrokerPlugin and Time to live
I have set the producer time to live value to expire messages after 10 sec. However, if the sys clock on the producer is different than the brokers, chances are high that this 10 sec window will never be realized and ALL messages will be considered expired. I saw the timeStampingBrokerPlugin is the solution to this problem, however, when I use it, the message timestamp = brokerInTime (which it should) but the expiration time is wrong (timestamp + 10 sec). Do I need to do anything additional to get this to work? I added this xml snippet to my xbean: -- View this message in context: http://www.nabble.com/timeStampingBrokerPlugin-and-Time-to-live-tp16041093s2354p16041093.html Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Re: timeStampingBrokerPlugin and Time to live
More info on my issue:
Setup:
Running a producer locally, with TTL set to 3000ms (on my laptop)
Connecting to remote broker via mulitcast
I set my laptop clock back 5 min and try to hit the broker with the correct
time. I get this...
DEBUG RegionBroker - Message expired ActiveMQObjectMessage
{commandId = 6, responseRequired = true, transactionId = null, expiration =
1205525689525, timestamp = 1205525686525, arrival = 0, brokerInTime =
1205526013325, brokerOutTime = 0, correlationId = null}
I added the timeStampingBrokerPlugin so timeouts can be calculated using
Broker clock. I get this msg:
DEBUG RegionBroker - Message expired ActiveMQObjectMessage
{commandId = 6, responseRequired = true, transactionId = null, expiration =
1205526055718, timestamp = 1205526379498, arrival = 0, brokerInTime =
1205526379498, brokerOutTime = 0, correlationId = null}
The timestamp and brokerInTime are in sync, but expiration is wrong. It
should be timestamp + 3 seconds
Then I set my laptop time 5 minutes faster than my broker. The only thing
that appears in my brokers logs are:
DEBUG AMQMessageStore- Journalled message add for:
ID:patrick-32776-1205523020988-0:0:1:30:1, at: offset = 91661, file = 1,
size = 746, type = 1
The message gets to my service. In my service logs (which runs on the same
server as broker), the message output is:
2008-03-14 16:37:40,423 DEBUG [ActiveMQ Session Task] ActiveMQSession (1624)
- ID:patrick-32776-1205523020988-0:0:1 sending message:
ActiveMQObjectMessage {commandId = 0, responseRequ
ired = false, expiration = 0, timestamp = 1205527060423, arrival = 0,
brokerInTime = 0, brokerOutTime = 0, correlationId = null}
I stripped out contents of the message that would have made this post long,
but let me know if you want to see the entire thing. In none of these
scenarios did the expiration time work as explained in docs. I am running
v5.0.0 and here are my broker configs. Please help!
--
View this message in context:
http://www.nabble.com/timeStampingBrokerPlugin-and-Time-to-live-tp16041093s2354p16055598.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
How to get Broker server info when using multicast://default
The clients currently find our brokers by using the multicast://default discovery, however, I am tasked to log what servers are used to perform the messaging, once a connection is made. When I retrieve the brokerURL on the session, it is always multicast://default. There does not seem to be any methods on the producer/consumer objects that would give you this information either. The end goal is to get the ip address:port of the server currently performing the work. Any tips would be muchly appreciated! -- View this message in context: http://www.nabble.com/How-to-get-Broker-server-info-when-using-multicast%3A--default-tf4715723s2354.html#a13480472 Sent from the ActiveMQ - User mailing list archive at Nabble.com.
