Search:

Sorry for the false alarm. It turned out to be due to this bug: [URL="https://jira.springsource.org/browse/INT-1820"]. So I didn't have to muck with the ordering to fix it.

Hi Gary,
Since I am performing other database operations within the flow, I do need the failed flow to rollback. This is why I need a retry of the flow in which only the failed message is consumed...

I want to do a stateful retry of the flow (actually, zero retry), so I need access to the message. The Callable.call() does not expose a message. I would get the failed message from the messaging...

Hi,
As I tried to configure the retry, I just realized that there isn't any out-of-the-box retry advice for a polling consumer. It wasn't that obvious from the documentation, but this post...

Hi,
When a message fails (transaction rolls back) to process on a queue backed by a transactional channel message store, like the JdbcChannelMessageStore, the message would remain in the store and...

Hi,
I have used the JdbcChannelMessageStore and it does use the database as a queue.
What I really want is to back a queue channel with a table only for the sake of disaster recovery. What I mean...

JIRA created: https://jira.springsource.org/browse/BATCH-1875

Hi,
When the reader is non-transactional, and the exception is skippable, Spring Batch does a good job of isolating the failing item(s) in a chunk write by rescoping the transaction around each item...

Hi,
I would like to use the highly concurrent tomcat jdbc pool instead of DBCP. Since I need to use Oracle LOBs, it is not clear to me what NativeJdbcExtractor to use with this pool. Will...

I see that this has been fixed in 2.1.8 :D

Hi,
I have a batch job (version 2.1.7) with a partitioned step. A skippable exception was thrown but the partition failed with the stack trace below:

at...

Hi,
I have existing code which is more pipe-lined i.e. read buffer from driving table->read supplemental data->transform/validate->write->read (some of the previously written...

Hi,
Is it possible to pass the job instance id to a processor using late-binding? I need this for auditing in the processor business logic.
Thanks,
Hari.

I believe I can do pessimistic locking as long as I add a processing indicator column. So the reader locks and buffers unprocessed rows using the query (with a filter on processing indicator), sets...

@dma_k
That is exactly what I am trying to do
@Dave
The need for pessimistic locking comes because I need to run multiple instances of the same batch (for increased throughput). It is simpler for...

So I should do the pessimistic query in beforeChunk and put the items in the step execution context? The reader then reads from the context?
When individual item processing happens as part of the...

Sure. I'd like to contribute. How do I do that?

Use-case:
I run the same batch on multiple machines:
reader: reads a page of rows with pessimistic lock (query for unprocessed rows with sorting...

I guess I really need to understand how the fault tolerant processing works. This isn't documented enough and I came to know about this only through questions in the forum and looking at the source...

Thanks. I presume the afterChunk method is the right place to check for the updated rollback counts? The reference guide says that afterChunk is invoked before the commit while the javadoc says that...

Hi,
I would like to pessimistically lock a page of rows from a table in the reader and have the locks released when the chunk commits. Is this reader a transactional reader? I will be configuring...

Hi,
There is specific processing I need to do if the chunk transaction rolls back. I could even do this processing at the start of the next chunk. How do I determine the transaction status of the...

Hi,
I have been through the documentation but for some reason could not figure out where the transaction begins for a chunk. I can see that it ends when the after the writer writes out a chunk....

Thanks Dave. I wrapped the CommandLineJobRunner in my main implementation. I have a CLI where I either load just the context in case of polling jobs or delegate to the CommandLineJobRunner to launch...

Thanks Dave. It's hard to control such a job since the poller creates multiple job instances. There is no single point from where to stop/restart these instances. It would have been ideal to have a...

Hi,
I have the spring batch admin/integration approach to running a file polling job implemented. I could test it easily enough by running a method that just loads the context and sleeps. But I now...