Hi To analyse this issue further can you create a ticket regarding this in our issue tracking system[1] and attach zip of the following directories from wso2cep-3.1.0/repository/deployment/server
inputeventadaptors outputeventadaptors executionplans eventbuilders eventformatters Thanks Suho [1]https://wso2.org/jira/browse/CEP On Thu, Dec 18, 2014 at 3:23 PM, Kacper Pietrasik < [email protected]> wrote: > Hi, > > thanks for your replay. Exactly, I have this situation with queue, but I > see strange behaviour. > > For example: > 1) Sometimes two CEP instances work in active-active mode (request > load-balancer works fine) > 2) Sometimes when CEP instances work in active-active I get duplicated > messages > 3) Sometimes this instances start to work in failover mode, and one is > Passive, second is Active > > This two instances have the same flow (Streams, Execution Plan, > distributed Execution Plan, etc). > > I don't know why this situations happen. Maybe you could know why I get > this 3 situtaions and why active-active doesn't work all the time? > If I was able somehow to run this instances in failovermode, so is this > possible to run this instances (in my situation, two CEP connect to one > queue) all the time? > > > 2014-12-16 13:29 GMT+01:00 Sriskandarajah Suhothayan <[email protected]>: >> >> Hi Kacper >> >> You are correct, in all cases the server level clustering configuration >> is the same, what changes it how we wire the events. >> >> In failover mode, its expected that client will only send messages to one >> CEP and when it fails it will send to the other. But this is not always >> possible, if you use JMS topics the message will be delivered to both CEP >> nodes (Hence here you have to use the full-active-active ) and if you use >> JMS queue then the message will be delivered to either of the CEP nodes but >> not for both (Hence here you have to use the active-active). >> >> Therefore how we wire the events should depend on how the events are send >> to the streams, in all 3 cases the ultimate goal is the send the same event >> to the identical execution plans in both servers and in all 3 cases one of >> them will become Active and the other will become Passive. >> Hence the output will be published only from the Execution plan that has >> become Active. >> >> Its not necessary to setup distributed database, but you have to share >> the database so in that case you can create a mySQL database [1] and point >> both the CEPs to that. >> Then you have to start CEP with the -Dsetup command as[2] and this will >> create the appropriate tables. >> >> Hope this clarifies your doubts >> Let me know if you need further clarifications. >> >> Thanks >> Suho >> >> [1]https://docs.wso2.com/display/CLUSTER420/Setting+up+the+Database >> [2] >> https://docs.wso2.com/display/CLUSTER420/Configuring+the+Manager+Node#ConfiguringtheManagerNode-StartingtheASserver >> >> On Tue, Dec 16, 2014 at 7:45 AM, Srinath Perera <[email protected]> wrote: >> >>> Hi Kacper, >>> >>> Adding Suho and team, someone will answer. >>> >>> --Srinath >>> >>> On Mon, Dec 15, 2014 at 3:19 PM, Kacper Pietrasik < >>> [email protected]> wrote: >>>> >>>> Hi, >>>> >>>> I work with WSO2 CEP and I have a little problem with HA. I was looking >>>> for somebody who might help me with this component, and I noticed in your >>>> website that you have been working with CEP, so I decided write to you a >>>> message. >>>> >>>> In CEP documentation about HA, WSO2 members show how to set HA in one >>>> server/PC and recommend to set HA in two servers/PC. I want to to that in >>>> failover mode. My problem is that when I turn on CEP instances it works in >>>> active-active mode instead of failover. I noticed that sometimes one >>>> instance is Passive member and second is active. In documentation >>>> active-active and failover mode are set up in the same way. >>>> >>>> My question are. Is there any properties that I can change to have all >>>> the time failover mode (when I turn on 2 instance of CEP, one become >>>> Passive, second become active)? Is it necessary to set up distributed >>>> database when I have 2 CEP on 2 different servers/PC (I noticed that when I >>>> changed in second CEP database URL and when I don't change it, it works the >>>> same)? >>>> If you don't know how to help me, maybe you know someone in WSO2 how >>>> can help me with this problem? >>>> >>>> -- >>>> Best regards, >>>> Kacper Pietrasik >>>> >>>> >>> >>> -- >>> ============================ >>> Blog: http://srinathsview.blogspot.com twitter:@srinath_perera >>> Site: http://people.apache.org/~hemapani/ >>> Photos: http://www.flickr.com/photos/hemapani/ >>> Phone: 0772360902 >>> >> >> >> >> -- >> >> *S. Suhothayan* >> Technical Lead & Team Lead of WSO2 Complex Event Processor >> *WSO2 Inc. *http://wso2.com >> * <http://wso2.com/>* >> lean . enterprise . middleware >> >> >> *cell: (+94) 779 756 757 <%28%2B94%29%20779%20756%20757> | blog: >> http://suhothayan.blogspot.com/ <http://suhothayan.blogspot.com/>twitter: >> http://twitter.com/suhothayan <http://twitter.com/suhothayan> | linked-in: >> http://lk.linkedin.com/in/suhothayan <http://lk.linkedin.com/in/suhothayan>* >> > > > -- > Best regards, > Kacper Pietrasik > > > -- *S. Suhothayan* Technical Lead & Team Lead of WSO2 Complex Event Processor *WSO2 Inc. *http://wso2.com * <http://wso2.com/>* lean . enterprise . middleware *cell: (+94) 779 756 757 | blog: http://suhothayan.blogspot.com/ <http://suhothayan.blogspot.com/>twitter: http://twitter.com/suhothayan <http://twitter.com/suhothayan> | linked-in: http://lk.linkedin.com/in/suhothayan <http://lk.linkedin.com/in/suhothayan>*
_______________________________________________ Dev mailing list [email protected] http://wso2.org/cgi-bin/mailman/listinfo/dev
