"Emergency Response Requried"Handling stream of events simultaneouls in CEP - Complex Event Processing

Hi,
We have some 60 sample data coming per second
I have an JMS adapter to read messages from the queue.
In my First-CQL i have a requirement to do a pattern matching for some duration set to 5 seconds.
After 5 seconds the First-CQL will send out an response as "sucess" or "failure" accordingly after
which i need to send the message to a queue and receive the response and process.
Till all these process is complete( i.e. sending and receiving message from queue) which is taking atleast 20 seconds.
Non of my other events are processed.
It is supposed to process rest of the incoming messages parallel.
We even tried with a simple program an adapter -> channel -> eventLogBean.
The code after "event.send()" code in the adapter will be executed only after completing the execution of eventLogBean.
Is there any configuration required ??
If any of you have come accross such a scenario any help would be greatly appreciated.
Please provide a solution.
Edited by: user11083081 on Mar 23, 2010 3:12 AM 

Hi,
It's not really clear from your post what the problem you are having is. Could you perhaps explain what you would like to happen?
andy 

Hi,
The issue was regarding creating new threads.
Untill the execution of one event completed another event was not even read.
We came out of the issue by adding a channel with configuration of
<max-threads>10</max-threads> .
Thanks,
Edited by: c devi on Apr 13, 2010 12:11 AM

Related

BPEL Recieve activity waits for message in an asynchronous process

Hi,
I am facing a strange problem ...
I have one Oracle JMS queue in which messages are posted from a BPEL process(Producer).
I have another BPEL porcess to conusme those...
When I write an empty BPEL process with the starting node as a Recieve, it works fine(it is able to consume the message)..
But if the same recieve is placed within another process(Asynchronous), it waits for the message indefinitely.
When I monitored the queue for pending messages, it shows nothing.(which means the message has been consumed).
Can someone help me please...
saptarishi
Edited by: saptarishi on Jun 3, 2010 5:06 PM 
Still stuck in this...
Can someone help me please ...
saptarishi 
Thanks to pavan found out the problem.
Please refer to these links as well.... Both are related.
Re: Problem in implementing multiple callbacks in BPEL
http://forums.oracle.com/forums/message.jspa?messageID=4473579#4473579
Need to put correlation into it..
saptarishi

Transaction Rollback in SOA 11g

Hi,
I have created a BPLE process which is doing some DB transactions and after that it is posting XML to Output JMS Queue.
During load testing when i post more that 70 xmls per sec in input then i get double xmls in output Queue.
its Flow trace is showing that after completing whole process due to some error transaction rolls back and it again initiates the whole process. so while rolling back it rolls back all the DB transactions but messages poseted in output JMS Queue are still there. so i am getting double messages in output Queue.
Even m not able to get the reason of rolling back the transactions.
Please help to solve the issue.
thanks in Advance. :)
Edited by: user13641647 on Jan 19, 2011 4:26 PM 
Hi,
you should make the push jms message operation participate in a XA transaction.
I am not so expert, so I cant tell you how to do it in bpel, it's just a tip for you to check.
Edited by: GheParDo on Jan 19, 2011 4:54 PM

Unified Timeout Issue in BPEL

Hi all,
We have two BPEL processess one ReqCS which is the "calling BPEL" and PS which is the "called BPEL". The called BPEL PS has a number of calls to database and wrappers. The problem is that the ReqCS BPEL which is calling the PS BPEL gets faulted. But the PS BPEL gets completed. We r getting the error Description as "Waiting for response has timedout. We tried changing the sycnMaxTime of the BPEL but it didnt help. PS has quiet a number of calls which makes the output XML as a huge one. We can see the output XML in the instance of the PS. But we couldnt see the instance of the ReqCS BPEL.
Regards
Prabhu
Edited by: Prabhu on Mar 23, 2011 6:42 AM 
And Also this is working fine when the output xml is less sized.
Regards
Prabhu

BEA-Error Scenarios

Dear all,
What are all the scenarios we get the BEA-382000 error code in OSB. can i do Re try when i get this error code every time as i do it for time out scenario.
as per the error code doc it says General runtime error, but what are all General runtime error's.
thanks
pr
Edited by: 947858 on May 10, 2013 2:09 AM 
This code is generally use for those errors which cannot be identified or classified at run time so you may consider putting a retry for this error code but make sure that it is not an infinite retry loop. If even after few number of retries, you are continuously getting the same error then issue may be serious and must be looked into by admins.
Regards,
Anuj 
thanks Anuj.
What all other error codes that may need re try other than above mentioned and 380000.
Edited by: 947858 on May 10, 2013 1:47 AM

DB Adapter polling issue

Hi All,
I have one question on DB Polling ?when i am using DElete polling Strategy option .
for Example DB Adater is Reading 100 Records,70 suuceeful but some 30 records are failed what happened to thses 30 records? where it stores? please any body help on it........any soliution
Thank you,
Ram.
Edited by: 969014 on Jun 5, 2013 5:30 AM 
Hi Ram,
DbAdapter is transactional and it should not delete failed records...
Ref docs here...
http://docs.oracle.com/cd/E28280_01/integration.1111/e10231/adptr_db.htm#BABCIICF
Cheers,
Vlad

Categories

Resources