New and Improved 3.13 Beta. Highlights: Shareable filters and dashboards and lots of other goodies. Any feedback can be raised as JIRA issues in the JIRA project.
Issue Details (XML | Word | Printable)

Key: JRA-8448
Type: Bug Bug
Status: Resolved Resolved
Resolution: Fixed
Priority: Critical Critical
Assignee: Unassigned
Reporter: Richard THIBAULT
Votes: 3
Watchers: 6
Operations

If you were logged in you would be able to see more operations.
JIRA

Excel Export memory usage is not constant

Created: 08/Nov/05 10:23 AM   Updated: 11/Aug/07 03:03 PM
Component/s: Bulk Operations, Import / Export, Issue navigator, Performance
Affects Version/s: 3.3.2
Fix Version/s: 3.7

Time Tracking:
Not Specified

Environment:
ystem Date mardi, 08 nov. 2005
System Time 17:16:51
Java Version 1.4.2_09
Java Vendor Sun Microsystems Inc.
JVM Version 1.0
JVM Vendor Sun Microsystems Inc.
JVM Implementation Version 1.4.2_09-b05
Java Runtime Java(TM) 2 Runtime Environment, Standard Edition
Java VM Java HotSpot(TM) Client VM
User Name adminweb
User Timezone Europe/Paris
User Locale français (France)
System Encoding Cp1252
Operating System Windows 2000 5.0
OS Architecture x86
Application Server Container Apache Tomcat/5.0.19
Database type oracle
Database JNDI address java:comp/env/jdbc/JiraDS
Issue Links:
Cause
 
Reference

Participants: Akshaya Kumar Naik, Akshaya Kumar Naik, Brian Nguyen, Igor Sereda, Jeff Turner [Atlassian], Mark Chaimungkalanont [Atlassian], Richard THIBAULT and Scott Farquhar [Atlassian]
Since last comment: 2 years, 8 weeks ago
Resolution Date: 05/Jul/06 10:34 AM
Labels:


 Description  « Hide
When we start exporting a large number of issues to Excel (we tried to export all fields of 14000 issues), JIRA throws a java.lang.OutOfMemory.
On less than 1000 issues, the process finishes.
It means that the memory usage of the Excel export is not constant and depends on the export volume.

It's a very big problem since a lot of users are interesting in using the Excel export every day, and the database is growing a lot.

Could you solve this problem quickly ?



 All   Comments   Work Log   Change History      Sort Order: Ascending order - Click to sort in descending order
Jeff Turner [Atlassian] added a comment - 13/Nov/05 04:23 PM
Richard,

How much memory are you starting JIRA with? After extended use this can be seen in Admin -> System -> System Info.

Cheers,
Jeff


Richard THIBAULT added a comment - 14/Nov/05 04:44 AM
We are starting JIRA with 1 GB of memory :

Java VM Memory Statistics
Total Memory 1016 MB
Free Memory 317 MB
Used Memory 699 MB
Memory Graph
[Used Memory ( 69 %)] [Free Memory ( 31 %)]
31 % Free (Force garbage collection)

There are 16300 issues in our database.

If we try to export more than 2000 issues, JIRA is getting out of memory. We cannot upgrade our server memory at that time and I'm not sure it would be sufficient since the memory used depends on the number of exported issues.


Richard THIBAULT added a comment - 14/Nov/05 04:53 AM
For your information, our database is growing a lot : 2000 issues are created every month.
We are anxious about the capability of JIRA to support such volumes when we see that some processes are running out of memory with a 1 GB tomcat instance.

Akshaya Kumar Naik added a comment - 17/Nov/05 02:07 PM
We have the same problem with 13000 issues. We have problem with filters too. After running few filters, memory spikes up and never gets garbage collected, eventually causing the instance to freeze. We are rebooting our instance twice a day. Can you please fix this in 3.4.2 as our production instance is affected.

Mark Chaimungkalanont [Atlassian] added a comment - 17/Nov/05 11:35 PM
Akshaya,

What JIRA version are you using? In 3.3, you should turn off issue caching. If that happens, just executing filters should not cause major spikes in the memory (as all issues are not added to memory)

Cheers

Mark C


Akshaya Kumar Naik added a comment - 20/Dec/05 12:16 AM
Sorry about the delay in resonding. Our issue was caused by a leak in IBM type 2 jdbc driver. We are using BEA Type 4 JDBC driver for db2 now. It's working fine.

Brian Nguyen added a comment - 20/Dec/05 01:58 AM
Hi Akshay,

Thanks for the update. Great to hear that you have resolved the issue. If you have any further questions do not hesitate to contact us.

Thanks,
Brian


Igor Sereda added a comment - 26/Jan/06 05:55 AM
I just tried to export all issues into Excel from jira.atlassian.com. I got "502 Proxy Error" and jira instance seemed to be unavailable for some time.

I also have tried to download all bugs via RSS format, and got the same error.

This seems critical! Please advise.

Here's the exact response from your proxy:
======== cut =============
Proxy Error

The proxy server received an invalid response from an upstream server.
The proxy server could not handle the request GET /secure/IssueNavigator.jspa.

Reason: Error reading from remote server

Apache/2.0.46 (Red Hat) Server at jira.atlassian.com Port 80


Jeff Turner [Atlassian] added a comment - 29/Jan/06 10:10 PM
Igor,

Why do you want all 15k issues from our system in Excel?

Or is this just to illustrate the problem?

Cheers,
Jeff


Igor Sereda added a comment - 30/Jan/06 03:01 AM
Hi Jeff,

Of course it was a demonstration of the problem, I don't need your issues, I have a lot of my own But don't you think it's possible that someone who has 15k issues in JIRA would want them exported for, say, graphical analysis?

As for my personal concern, I'm working on a desktop client for JIRA (similar to what we already have for Bugzilla, you can take a look at http://deskzilla.com). We'd like to offer an ability for JIRA users to work offline and do other "desktop" stuff. Naturally, I need to get issues from a JIRA instance - not necessary all of them, but it could be a lot.

Best regards,
Igor


Scott Farquhar [Atlassian] added a comment - 05/Jul/06 10:34 AM
As of JIRA 3.7, the excel export exports all issues, and the memory usage is constant for the export.

Export away!