[ https://issues.apache.org/jira/browse/OFBIZ-13282 ]
Jacques Le Roux deleted comment on OFBIZ-13282:
-----------------------------------------
was (Author: jacques.le.roux):
This follows this thread :
https://lists.apache.org/thread/12l6mqzgc7xb7b49b45j3qnfpsgy27yz
> JobTracker object
> -----------------
>
> Key: OFBIZ-13282
> URL: https://issues.apache.org/jira/browse/OFBIZ-13282
> Project: OFBiz
> Issue Type: Improvement
> Components: framework/service
> Affects Versions: Upcoming Branch
> Reporter: Gaetan Chaboussie
> Priority: Major
> Labels: pull-request-available
>
> Hello community,
> We recently got a client with an interesting need, and began reflecting on it
> (Nicolas and I).
> The need is the following : be able to easily follow the advancement of a
> very large process.
> For example, when an operator launches the generation of three thousands
> invoices, they want to be able to see how many have been treated, have an
> estimate on how long it would take, rerun the failed generations, how many
> failed so far, etc.
> To this end, we started the development of what we called the JobTracker.
> The idea is the following:
> * I launch the top service that plans all the unit generation jobs. This top
> service uses a dedicated engine (TrackedServiceEngine) that inits and injects
> a JobTracker object in the context.
> ** You can check the `testTrackedServiceAreTracked` unit test for an example
> implementation.
> * The JobTracker object has its corresponding persisting entity, that allows
> it to be retrieved from anywhere in the framework, associated with the
> JobTrackerFactory and its cache.
> * Each of the generating services is watched by the JobTracker object. At
> the end of each unit job, there is a check for the status of the whole list
> of tracked jobs to check completion.
> * When all the jobs have been run, the JobTracker status switches to
> complete.
> * Any time during the process of the list of unit jobs, the tracker can be
> accessed, and gathers all of the data of the jobs linked to it. It can (for
> example) give the completion percentage of the generation process.
> * At the end of each job, the result of the job can be stored in a dedicated
> new entity TrackedJobResult. This entity stores the Job final status,
> potential error messages, and can be extended for storing business related
> data, useful in a lot of ways. We didn’t use JobSandbox to store the result
> because they are automatically purged after some time.
> If you want to test it for yourself, you can run the
> ‘ServiceForTestingTracker’ service that uses the description above.
> You can then go to the view of the JobTrackers[1], and follow the real-time
> advancement of the tracker once it’s been selected.
> [1] [https://localhost:8443/webtools/control/FindJobTracker]
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
