Search:

Hello,

I'm trying to migrate a J2EE application to Spring. We have a special MDB configured to handle the transaction itself (BMT). The idea behind that is to by-pass the container transaction...

hello,

I'm searching a way to configure resources used by Spring JMS when it's deployed on Jboss. Indeed, JMSTemplate creates a session for each message sent, or received. In ActiveMQ, the Spring...

Thanks for your help :)

I've just tried with/without time interval on H2 (in memory):


without any interval: first attempt lasts 15ms, the second one 16ms => then crash
with an interval of 50ms: 63ms, 62ms and 63ms...

public boolean shouldSkip(Throwable t, int skipCount) throws SkipLimitExceededException {
final int maximum = manager.getMaxNumberIgnoredExceptions();
if (skipCount == maximum) {
...

Mostly: H2, sometimes Postgresql and Oracle (target platform)

These executions are executed sequentially without any delay, so I guess they are always executed within a second :)

Hello :)

I don't if it's a bug in the framework, or if my test is not correct, but I got an exception when I start a job for the third times (since exceptions have occurred in its executions):

...

Moreover, when I've looked at the log, I've seen that our SkipPolicy receives a negative skipCount for that test (i < 0):



public boolean shouldSkip(final Throwable throwable, final int i)...

The test case is: commit-interval of 2, max-retries = 1, exception thrown every time the item a is written.

This is the behaviour we have seen:


Chunk: a, b
- process of a, b
- write of a, b...

About recovering from an exception thrown by a writer: we have just made new accurate tests which have the following behaviour:

Chunk: a, b
- process of a, b
- write of a, b => exception
-...

Yes it's thrown by the writer.

"If it generates that exception again, then the step should stop. " But if we have put an attempt counter greater than 1 (the counter starts when it starts to...

Hello,

We are facing a behaviour that we cannot understand. In the code bellow, that's the parent step our test step. We've created 3 kinds of exceptions:

Fatal Exceptions: if a reader,...

We have found why the onSkipInProcess is not called. The only item processed by the partition was skipped, so the writer was called with an empty set of items leading to the crash of our writer (it...

Hello,

I've defined the following partition step:


<step id="partitionStep" parent="abstractPartitionerStagedStep" xmlns="http://www.springframework.org/schema/batch">
...

yes because we are using a database table keeping track of items that are completed, in error and skipped. So we want a allow-start-if-complete step with an item reader able to know which items must...

Ok I've understood the behaviour:

SimpleStepExecutionSplitter checks if the partion is restating, if it's restarting (the previous execution of the partition handler was not completed) then the...

so, only partitions that cause a batch failure will be reprocessed? and thus, the Partitionner must be called only one times even if the batch is restarted several times?

Hello,

I prefer to post in the forum before creating a JIRA ticket, I've the following problem:

I've 4 partitions (gridSize = 4), the batch crashed (it's wanted). Item 3 and 4 have not been...

I don't think it's possible to make a real "pause". But, the stop/restart functionnality is built-in.



public boolean stopJob(JobExecution execution)
throws...

Thanks, it works ;)



<step id="partitionStep" parent="abstractPartitionerStagedStep" xmlns="http://www.springframework.org/schema/batch">
<partition step="mainStep"...

Is it possible to improve that check ? It could be interesting to be able to customize that check especially when you know that your code works correctly. In our case, we're using an ItemStreamReader...

thanks for that suggestion, I'm also interested :)

hello,

Tasklet step supports listeners, unfortunately partition step doesn't support listeners. So, if I want to know the current step execution before executing inner partitions, I'm obliged to...

thank you :)

finally we have followed your suggestion:



/**
* {@link StagingItemsTracking} keeps track of items read and written by every batch process. Each
* of them are executed in...

There is no listener called after all unsucessfull retries with the associated items (that cause the batch failure) ?