simpleflow | simple TensorFlow-like graph computation framework | Machine Learning library

Your RetryableItemReader only retries the read operation, but the timeout is happening while initializing the reader, ie in the open method. This can be seen here:

Spring Batch does not allow alternative branches in the flow to be implicit. In other words, you need an on(...) for each case.

Assuming decider() yields a proxied bean, it should work fine with

The ScriptItemProcessor delegates item processing to a org.springframework.scripting.ScriptEvaluator. As of v5.3, Spring Framework provides three implementations of the ScriptEvaluator interface: BshScriptEvaluator, GroovyScriptEvaluator and StandardScriptEvaluator.

By default, if you don't specify which ScriptEvaluator to use, the ScriptItemProcessor will use a StandardScriptEvaluator, see Javadoc of setScriptEvaluator. Now this StandardScriptEvaluator is based on JSR-223 javax.script, so you need to make sure the language you want to use is supported by the default ScriptEngine. In your case, python is not a supported language by default, and hence the error.

If you want to use python through JSR-223, you can use Jython (adding the jar of Jython to the classpath should be enough, see Calling Python from Java through scripting engine (jython)?)

Use the exception methods documented here to troubleshoot the root cause for the exception in the stack trace of yours.

https://docs.oracle.com/cd/E17802_01/products/products/persistence/javadoc-1_0-fr/javax/persistence/PersistenceException.html

1. Does the spring batch process query and fetch all the 350000 rows at once and give the records to processor in chunks of 50 (chunk size) ?

No, it will read only pageSize items at once, not all of them (that's the point of a paging reader). ChunkSize and PageSize are different parameters. For example, you can read a page of 100 items and have 10 chunks of 10 items for each page. Matching the pageSize and chunkSize usually gives better performance. Please refer to the Javadoc.

1. I have this line in the code JpaPagingItemReader reader = new JpaPagingItemReader<>();. Should the Person class be a Entity object for this to work or any will a POJO work fine?

Yes, otherwise I don't see the purpose of using a JPA reader (a JDBC one would be enough).

1. What could be causing the above exception?

You have a NullPointerException here which means the entity manager is null. You need to call afterPropertiesSet on your reader to make sure it is correctly configured.