This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.

Bug 143406 - All exceptions with ConcurrentModificationException message are closed as duplicate of 125
Summary: All exceptions with ConcurrentModificationException message are closed as dup...
Status: VERIFIED FIXED
Alias: None
Product: ide
Classification: Unclassified
Component: Exceptions Reporter (show other bugs)
Version: 6.x
Hardware: All All
: P1 blocker (vote)
Assignee: Jindrich Sedek
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2008-08-11 08:05 UTC by Marian Mirilovic
Modified: 2008-10-08 15:57 UTC (History)
0 users

See Also:
Issue Type: DEFECT
Exception Reporter:

Attachments
changelog (423.59 KB, application/x-bzip2)
2008-09-03 19:07 UTC, Jindrich Sedek 		Details
report 138 needed to be split into separate issue - fixed by 144297 (6.80 KB, application/x-gzip)
2008-09-09 13:19 UTC, Jindrich Sedek 		Details
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Marian Mirilovic 2008-08-11 08:05:41 UTC 
Look at :
http://statistics.netbeans.org/analytics/detail.do?id=125
there are 156 duplicates .. I recently moved 13 of them to another issue :
http://statistics.netbeans.org/analytics/detail.do?id=38706
and it looks like that another tens of them are totally different ...
http://statistics.netbeans.org/analytics/detail.do?id=88727
Comment 1 Marian Mirilovic 2008-08-11 09:23:35 UTC 
~ 60 of duplicates were moved to 88727 ... this is serious, hence we've been getting a lot of exceptions from one place
and we didn't know about that ;(
Comment 2 Jindrich Sedek 2008-08-15 15:38:13 UTC 
I'm trying to find the root of this problem. I've writen a duplicates algorithm test and it marks issues as different 
so it's correct.

I've submitted the logs of these issues to the actual data on statistics.netbeans.org (influenced by Marian manual 
changes on duplicates marking) and all issues were marked correctly as separate issues and as duplicates of appropriate 
issues (id 38706 or 88727).

I've submitted the logs of these issues to the one month old data on qa-man-sol17 (non influenced by manual changes) 
and all issues were marked together as duplicates of 125.

All this meens that algorithm works correcly, but there are alredy wrongly marked duplicates - as 38706 was - and these 
wrongly marked duplicates are causing wrong decision of newly incomming issues. I have alredy faced similar problem in 
#136675 and I'm affraid there might be many similar cases.

I'm affraid we have to fix this problem properly and "recalcucate" whole database where there are no changes done by 
users manualy. We should also incorporate some support for different NB versions while looking for duplicates - I mean 
issue from 6.5 should not be blindly marked as duplicate of the report from 6.0, but it should be double checked for 
new issue and issue #125029 should be also fixed earlier before getting new bug tracking system (see latest comments 
there).
Comment 3 Jindrich Sedek 2008-09-03 19:05:05 UTC 
The database data recalculation has just finished.
I'll attach change log done be recalculation application
Comment 4 Jindrich Sedek 2008-09-03 19:07:20 UTC 
Created attachment 68979 [details]
changelog
Comment 5 Marian Mirilovic 2008-10-08 15:57:23 UTC 
verified