|
The expected value for total issues on April 12 was 61.
The expected value for total issues on May 10 was 110. The filter I'm using captures all issues in the database (all projects, all issues attributes, any status, user, etc.) The current values do accurately reflect what I expect to see. Please let me know what information I can provide that would be helpful. Thanks, Asaf. Would it be possible to get a backup of the data?The ordinary .zip backups that Jira makes would be fine. If that's not possible, then perhaps a copy of the logs/catalina.out logfile after the debugging has been enabled? Details of how to enable this are in the Troubleshooting section of http://confluence.atlassian.com/display/CODEGEIST/Timecharts
I will be able to send the logs/catalina.out file, but won't be able to send a backup (at least for now).
I have updated the properties file, but I have to wait to shutdown the server over the weekend, for the changes to take affect. I will send the logs by Monday. I updated the properties file with the following two lines (at the end):
log4j.category.com.packetmotion.jira.plugin.report.timecharts = DEBUG, filelog I can't find any output in the logs/catalina.out file with "timecharts." What output should I be looking for? BTW, as an example of how the data changes, consider the following data from the same database a few days later. You will notice that the numbers changed for Apr 12 (as well as the other dates) when compared to the data provided in the description of the issue. Thu Apr 12 00:00:00 PDT 2007 4 2 87 1 20 114 Attached is the atlassian-jira.log with the debug output.
Matt,
I'm just checking in. Is there any other information I can provide that can be helpful? Is the log file that I attached OK? Thanks, Asaf We are experiencing the same behavior.
In fact we saw a clear difference between the results provided by the pending stock chart The pending stock chart is simply showing the number of unresolved issues over the time for a given filter. Several tests later and we had to confirm that the Pending stock chart was right, not the time chart. Test made on Jira 3.10. Vince We seem to have the same problem. We started our JIRA system on May 17th, 2007 and on that day we had just about 3 issues. Now we have about 402 issues in the system. But when the Time Chart is displayed, it shows the system to have about 361 issues on May 17th. And the total number of issues always remains the same even if new issues are added every day.
We use this system extensively and it would really help if you could tell us if there is a workaround for this problem, perhaps the section in the source code where things are going wrong. We use JIRA 3.9. I've also encountered this problem and I analyzed what's wrong. The reason is that while the plugin collects data, and progresses backwards in time it ignores the fact, that issues may not existed at a certain point in time.
I.e. at tuesday I've created an issue with state open. Now i generate timechart report from monday till wednesday. The report shows:
I've managed to fix the code, but it is an ugly workaround, so i won't attach the patch. Is this likely to be resolved in the near future. This is a feature of the JIRA that we would like to use as part of our reporting procedure, but at the moment it is not possible. I am probably interested in the 'ugly workaround' if there is no official fix due soon.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||
And is your filter catching all issue types? Please would you add the Filter Summary details.
Does the current value (today) accurately reflect what you expect to see?