CODEWITHSUNDEEP
mulebatch-processing
Hans
Hans

Reputation: 23

mulesoft batch job is not executed

I work in Runtime 4.1.5 and batch job undertakes the work of synchronizing data

If the batch job is completed normally, the log should look like this:

Created instance 'dc97a040-009e-11ec-a7bf-00155d801499' for batch job 'sendFlow_Job'
splitAndLoad: Starting loading phase for instance 'dc97a040-009e-11ec-a7bf-00155d801499' of job 'sendFlow_Job'
Finished loading phase for instance dc97a040-009e-11ec-a7bf-00155d801499 of job sendFlow_Job. 1 records were loaded

Started execution of instance 'dc97a040-009e-11ec-a7bf-00155d801499' of job 'sendFlow_Job'

batch step customer log ....

Finished execution for instance 'dc97a040-009e-11ec-a7bf-00155d801499'                              of job 'sendFlow_Job'. Total Records processed: 1. Successful records: 1. Failed Records: 0


=================end=======================

The log in question is as follows:

Created instance 'dc97a040-009e-11ec-a7bf-00155d801499' for batch job 'sendFlow_Job'
splitAndLoad: Starting loading phase for instance 'dc97a040-009e-11ec-a7bf-00155d801499' of job 'sendFlow_Job'
Finished loading phase for instance dc97a040-009e-11ec-a7bf-00155d801499 of job sendFlow_Job. 1 records were loaded

Started execution of instance 'dc97a040-009e-11ec-a7bf-00155d801499' of job 'sendFlow_Job'

=================end===================

It can be clearly seen that the log shows that the batch job has only completed the first stage of work. After that, the batch job is as if it has never existed, there is no log output, and no errors are thrown.And from the target database, the data is indeed not synchronized

I tested it in the local environment and the problem was reproduced. Use kill -9 to kill the process while the batch step is executing, then the process will restart, and then all batch jobs will have problems

I found the queue file used by batch job in the .mule folder. It is similar to BSQ-batch-job-flow-name-dc97a040-009e-11ec-a7bf-00155d801499-XXX Under normal circumstances, each batch job will create three BSQ file and delete it at the mplete. In my question, the BSQ file will be created but not deleted

I looked up some posts and they suggested deleting the .mule folder and restarting. In the actual environment, I don’t know when there will be a problem and deleting the .mule folder does not completely solve the problem of batch job not being executed. Is anyone proficient in mule batch job? Can you give me some suggestions, thanks

Upvotes: 1

Views: 1193

Answers (1)

aled
aled

Reputation: 25837

You should not delete the .mule directory. There is other information in there unrelated to batch that would be lost, like clustering configurations, persistent object stores, other applications batches and queues. It may be ok to delete it inside the Studio embedded runtime because that just your development environment and you probably are not losing production data, but in any case is not a solution just to delete information.

There are too many possible causes to identify the right one, and you should provide a lot more information. My first recommendation is to ensure your Mule 4.1.5 has the latest cumulative patch to ensure all known issues are resolved. Note that Mule 4.1.5 has been released almost 3 years ago. If possible at all migrate to the latest Mule 4.3.0 with the latest cumulative patching. It should be more stable and performant than 4.1.5.

Upvotes: 1

Related Questions