uima-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jörn Kottmann <kottm...@gmail.com>
Subject Re: UIMA AS Client with failover broker url which contains randomize option
Date Tue, 10 Nov 2009 15:25:19 GMT
Jörn Kottmann wrote:
> Jörn Kottmann wrote:
>> Jörn Kottmann wrote:
>>> By Broker: tcp://XXX2:61616)?randomize=false
>>>
>>> The brokerUrl is:
>>> failover:(tcp://XXX1:61616,tcp://XXX2:61616)?randomize=false
>>>
>> It looks like that JmsMessageContext.chooseServerUri transform it
>> from
>> failover:(tcp://XXX1:61616,tcp://XXX2:61616)?randomize=false
>> into
>> tcp://XXX2:61616)?randomize=false
>>
>> which is then later rejected.
>>
>> Why does uima tries to select a server uri instead of directly using
>> the broker uri ?
> I just removed the call to chooseServerUri, but I guess that breaks
> the http case ? Not sure what that is about, maybe some who knows the 
> code
> could have a look ?

I did some fail over testing, the setup we have is a pure master slave 
broker.
At the beginning everything is talking to the master until it fails 
(kill -9), then everyone
should talk to the slave instead.

Here is the log from my worker node:

INFO  FailoverTransport              - Successfully connected to 
tcp://XXX1:61616

##### Here I stopped the master broker process with kill -9 #####

WARN  FailoverTransport              - Transport failed to 
tcp://XXX1:61616 , attempting to automatically reconnect due to: 
java.io.EOFException
WARN  FailoverTransport              - Transport failed to 
tcp://XXX1:61616 , attempting to automatically reconnect due to: 
java.io.EOFException
WARN  FailoverTransport              - Transport failed to 
tcp://XXX1:61616 , attempting to automatically reconnect due to: 
java.io.EOFException
INFO  FailoverTransport              - Successfully reconnected to 
tcp://XXX2:61616
INFO  FailoverTransport              - Successfully reconnected to 
tcp://XXX2:61616
INFO  FailoverTransport              - Successfully reconnected to 
tcp://XXX2:61616

The client is reading messages ids from a queue and is then sending them 
in a CAS
to the worker nodes, well through a configuration error the reading 
failed after the master failed.
So I fixed the configuration error and tried to restart the client, but 
it always gets a time out
during initialize. But it sends a message to the uima as queue, but the 
message is not retrieved
by the worker node.

I used the activemq web interface to get some details about the message 
which is never
retrieved:

Command 	2001
MessageFrom 	ID:XXXX-51032-1257865414664-0:1:1
ServerURI 	failover:(tcp://XXX1:61616,tcp://XXX2:61616)?randomize=false
MessageType 	3000


Why does it reconnects three times ? The queue only has 2 consumers on the
activemq web interface.

The log output from activemq is only written to the console not to the 
uima.log file,
has the disadvantage that the time stamp is missing.

Jörn

Mime
View raw message