Skip to main content
Announcements
Introducing Qlik Answers: A plug-and-play, Generative AI powered RAG solution. READ ALL ABOUT IT!
cancel
Showing results for 
Search instead for 
Did you mean: 
Burgo
Contributor
Contributor

Talend scheduled job restart without killing Java Zulu istance

Hi all,

I have a Talend scheduled job via Window Task Scheduler that starts every day at 4 AM.

At every start it runs a new Java Zulu instance without ending the older one.

Am I missing some set up?

0695b00000UzKHxAAN.png

Labels (3)
4 Replies
Anonymous
Not applicable

Hello,

Could you please provide the following information for us to be better evaluate the issue? 

- Product version (such as Talend Cloud Real-Time Big Data Platform (7.3.120200219_1130)

- Operating System (such as Windows 10, 64 bit)

- Java version (such as Oracle JDK build 11.0.8+10-LTS)

Best regards

Sabrina

Anonymous
Not applicable

Are you sure that the job will finish if you run it manually from the command line? This looks like the jobs are still doing something.

Burgo
Contributor
Contributor
Author

Hi Sabrina,

Talend Open Studio for ESB Version: 8.0.1

Windows Server 2019 Datacenter, 64 bit, 32GB RAM

Azul Zulu JDK 11.58.15, 64 bit

 

 

Hi rhall,

The job read from an event queue with tMicrosoftAzureHubInput component and keep listening for getting new messages so it's always doin something I guess:

 

 

0695b00000UzMAFAA3.png 

, but I've set the task to stop and restart ad 4 AM:

 

0695b00000UzM7zAAF.png 

Thanks

Anonymous
Not applicable

OK, I doubt this is a Talend problem to be honest. How long does the job last when run in the Studio? Does it ever stop? I doubt it will if it is reading from a queue unless you have a process to stop it within the job. Note: Talend jobs are not intended to run continually. A better way of doing this is to use a Route. Routes are written to run continuously.

 

The scheduling tool will be kicking off the batch file for the job I imagine. This will start a JVM to run the job. I suspect that what is happening is that the JVM is not being stopped by the scheduling tool and that the batch file is being terminated. This will not necessarily kill the JVM. I suggest that you add a new scheduled rule to kill the previous process.