Openbravo Issue Tracking System - Openbravo ERP
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0006648Openbravo ERP09. Financial managementpublic2008-12-18 16:062009-05-21 16:43
Reportershuehner 
Assigned Toanthony_wolski 
PriorityhighSeveritymajorReproducibilityhave not tried
StatusclosedResolutionfixed 
PlatformOS5OS Version
Product Version2.40 
Target VersionFixed in Versionpi 
Merge Request Status
Review Assigned To
OBNetwork customer
Web browser
ModulesCore
Support ticket
Regression level
Regression date
Regression introduced in release
Regression introduced by commit
Triggers an Emergency PackNo
Summary0006648: When background accounting is disabled and you use posting by db tables, after the process background accounting will be running
DescriptionWhen you disable background accounting (i.e. by update ad_process set isactive='N' where ad_process_id = 800064).

When you then run posting by db-tables and this accounting run is finished. The background process is running again and keeps running until you stop tomcat. (After the restart the background accounting is not running, until you use posting by db tables again).
Steps To ReproduceDisable background accounting (see description on how)
Check that is is not running (by i.e. putting debug code into AcctServer.java)
Use Posting by db-tables
Check with your debug-code that every 10s the background accounting is checking for documents to account)
Proposed Solution
Additional Information
TagsNo tags attached.
Relationships
blocks defect 0006589 closed anthony_wolski When background accounting is disabled and you use posting by db tables, after the process background accounting will be running 
Attached Files
Issue History
Date ModifiedUsernameFieldChange
2008-12-18 16:06rafarodaNew Issue
2008-12-18 16:06rafarodaAssigned To => anthony_wolski
2008-12-18 16:06rafarodaStatusnew => scheduled
2009-03-24 09:22anthony_wolskiNote Added: 0014882
2009-03-24 09:22anthony_wolskiStatusscheduled => feedback
2009-04-14 11:59anthony_wolskiStatusfeedback => scheduled
2009-04-14 11:59anthony_wolskifix_in_branch2.40 => pi
2009-04-14 12:01anthony_wolskiStatusscheduled => resolved
2009-04-14 12:01anthony_wolskiFixed in Version => pi
2009-04-14 12:01anthony_wolskiFixed in SCM revision => N/A
2009-04-14 12:01anthony_wolskiResolutionopen => fixed
2009-04-14 12:01anthony_wolskiNote Added: 0015354
2009-05-21 16:43psarobeStatusresolved => closed

Notes
(0014882)
anthony_wolski   
2009-03-24 09:22   
Cannot reproduce. I've followed the steps to reproduce after a clean install.source. Debug code is in the first line of PeriodicAcctServer:

System.out.println(new Date() + "" + this.getClass().getName() + " running...");

Note: the background process is always running, but nothing happens (it is not executed) if it is set to inactive. Setting the process to inactive (as perthe issue description -> i.e. by update ad_process set isactive='N' where ad_process_id = 800064 requires a server restart for the change to take effect.
(0015354)
anthony_wolski   
2009-04-14 12:01   
Hi,

i just did retest...

and with the sample data i could not reproduce the problem. But it guess what my problem is/was... Using direct launch (posting by db-tables) does run PeriodicAcctServer.processPL once.. inside is a loop until acct.checkDocuments == false. Checking if there are more documents to process. I think when we found the isse 6648 this condition was always true so the one processPL call did never finish but loop endlessly.

Local change was a hardcoded limitation (datefilter) inside AcctServer so that AcctServer did not find more documents to post but the checkDocuments method was not modified and tried to endlessly repost these skipped documents.

@Tony, @Ivan, @Eduardo:
Could you recheck that this explanation makes sense for you? If yes i think we can close the issue without changes needed.

Makes sense... closing.