Hi Sunder, I would recommend you have a look at the TimerService as decribed in the EJB 2.1 spec. I suspect that it is what you are looking for.
If so then I would recommend you check out the various commercial/open source Timer Services around (as noted Weblogic has one in its product) and then crate (if necessary) a TImerService API wrapper around your product of choice. That way you get the TimerService functionality now but can delegate to the AppServer's implementation in the future (when they all become 2.1 compliant). William -----Original Message----- From: Sunder Rajan [mailto:[EMAIL PROTECTED]] Sent: Friday, September 20, 2002 6:48 AM To: [EMAIL PROTECTED] Subject: Re: J2EE Scheduling Architecture I Think the Weblogic Time API has been deprecated. Anyway , We now are also thinking of using JMS instead of spawning of threads in the Appserver. The alternate architecture is Create a JMS queue and add report request objects to that queue, Schedule the object to run in 30 mins(Weblogic JMS has a scheduling API for 6.1). After 30 mins, the JMS creates a Message driven bean instance which reads the external DB and checks to see if data ready, if it is , it generates the report, else adds the same object to the queue again to run in 30 mins. Any ccomments? Sounds simple , but not sure if I missed something. ----- Original Message ----- From: Albert Pi <[EMAIL PROTECTED]> Date: Thursday, September 19, 2002 1:24 pm Subject: Re: J2EE Scheduling Architecture > Sunder: > I use the same way like you do (using WebLogic Time > API ). so far mine is working fine. > > Albert Pi > Corp IS System Delivery > 516-803-3762 > > > >>> Sunder Rajan <[EMAIL PROTECTED]> 09/19/02 01:25PM >>> > Hi Everyone, > I have a general question on J2EE scheduling. We run an application > that processes financial reports. We need to automate the report > generation. The idea is , The user will schedule certain reports > from a > web page. The reports will run based on whether certain data is > existent on a different domain. So lets say a user schedules Job A to > run as an automated report. Job A will and only run when certain data > is available in database X. So the current architecture is to > store the > different jobs in a table B, a startup class in weblogic will > check the > database X periodically(30 mins) to check and see if data is available > and update the job in table B(set data to true). A different startup > class will periodically check table B and see if data flag is true and > kick of a report by starting a new thread(A new thread is required > since the reports may run for 30-45 mins). Do you guys see any > pitfalls/drawbacks with this approach. Is there a better way of > solvingthis? > > Thanks, > Sunder > > ======================================================================== === > To unsubscribe, send email to [EMAIL PROTECTED] and include in > the body > of the message "signoff EJB-INTEREST". For general help, send > email to > [EMAIL PROTECTED] and include in the body of the message "help". > > fffffffffffffffffffffffff > To unsubscribe, send email to [EMAIL PROTECTED] and include in > the body > of the message "signoff EJB-INTEREST". For general help, send > email to > [EMAIL PROTECTED] and include in the body of the message "help". > > =========================================================================== To unsubscribe, send email to [EMAIL PROTECTED] and include in the body of the message "signoff EJB-INTEREST". For general help, send email to [EMAIL PROTECTED] and include in the body of the message "help". =========================================================================== To unsubscribe, send email to [EMAIL PROTECTED] and include in the body of the message "signoff EJB-INTEREST". For general help, send email to [EMAIL PROTECTED] and include in the body of the message "help".
