Hi to all of the members of this forum, this is my first post here. 
I'm developing a java backed webscript, on an Alfresco 3.3.4, which has to read some numerical properties from nodes, aggregate them using a certain set of rules, then outputs the results in a csv file.
When I run the first tests of this webscript on a few tens of nodes I had no problems whatsoever, leaving aside bugs related to the inner logic of the webscript itself, which I had long fixed anyway.
The problems started getting quite serious when I ran the tests on a realistical scenario, with thousands of nodes to be computed, tens of thousands in the future most likely.
The program just shuts down with no errors and exits, NOT DOING ANY ROLLBACK on the modifications it does on the nodes it visits (it puts a timestamp in them to understand when and if it visited a specific node).
In the logs there are no errors, no exceptions, just the execution stops at a certain point of a class, always the same point, which does nothing in particular: it simply prints a log message.
The only particular things this webscript does is it prints tens of thousands of log messages in a very short amount of time: I used them to check if it makes the right aggregations. They're not essential and I will disable them in the future, most likely. Can this be related to the problem I'm having?
I encoutered the 1000 nodes limit "feature" of lucene too, but I had already handled that, this problem verifies long after I've already loaded all of the nodes I need in memory in fact.
Thanks for any help.
I'm developing a java backed webscript, on an Alfresco 3.3.4, which has to read some numerical properties from nodes, aggregate them using a certain set of rules, then outputs the results in a csv file.
When I run the first tests of this webscript on a few tens of nodes I had no problems whatsoever, leaving aside bugs related to the inner logic of the webscript itself, which I had long fixed anyway.
The problems started getting quite serious when I ran the tests on a realistical scenario, with thousands of nodes to be computed, tens of thousands in the future most likely.
The program just shuts down with no errors and exits, NOT DOING ANY ROLLBACK on the modifications it does on the nodes it visits (it puts a timestamp in them to understand when and if it visited a specific node).
In the logs there are no errors, no exceptions, just the execution stops at a certain point of a class, always the same point, which does nothing in particular: it simply prints a log message.
The only particular things this webscript does is it prints tens of thousands of log messages in a very short amount of time: I used them to check if it makes the right aggregations. They're not essential and I will disable them in the future, most likely. Can this be related to the problem I'm having?
I encoutered the 1000 nodes limit "feature" of lucene too, but I had already handled that, this problem verifies long after I've already loaded all of the nodes I need in memory in fact.
Thanks for any help.
