csv import file size

[Shinjani Gaur]

I am trying to import Bugzilla data into Jira via CSV import. I created some test files that were less than 1 MB and was able to import the data. I am now trying to import via a larger csv file (4 MB) and the import gets stuck at the "Begin Import" step. Are there any config changes that I need to make so that the import goes through?

[Shinjani Gaur · Accepted solution]

I am using Jira 6.1.3. I have tried the import with FireFox as well as Chrome. The version of Firefox is 8.0. Version of Chrome is 32.0.1700.107 m. The dialog box about a running script only pops up on FireFox. While doing the import via Chrome after I select “Begin Import” the page just gets reloaded and I am back at the top of the “Map Values” page.

Also, once I am on the "Begin Import" page the Back button doesn't work. I am stuck at that page.

[pborkowski]

The problem seems to be much more complex than I've thought. The situation that you are experiencing is not expected and I belive that you may have found a new bug in our importing system. Please open a ticket under JIM Project ( https://ecosystem.atlassian.net/browse/JIM) and share with us your CSV file if possible ( you can set Security level to "Reporter and Atlassian Staff if you don't want the data to be public ).

[pborkowski]

Hi,

I don't think that's a memory problem, we are loading whole CSV file into the memory after first step in the import wizard, so if you have completed almost whole import then memory probably isn't the issue here.

It looks like there is some JavaScript error in the last import step. Could you try to perform an import using different browser in the newest version (e.g. Chrome/Firefox)? Regardless of the result please share with us what's your current JIRA Version and what browser are you using - this will help us to better understand your issue.

Regards,

Przemek

[Shinjani Gaur]

I did use the Bugzilla importer plugin first but the Bugzilla Bug Ids get lost during the import. So I decided to use the csv import plugin via which the IDs can be somewhat similar. I checked the log files and there are no errors. A minute or so after I select "Begin Import", a dialog box pops up about a script running for too long and the option to stop the script or to let it continue. Once I hit continue, nothing happens for sometime and then the script pops up again.

I increased the memory allocated to JVM but that didn't solve the problem either.

[Wojciech Seliga]

Hi,

Have you considered using dedicated Bugzilla importer (bundled with JIRA self-hosted)? If you happen to use JIRA OnDemand (where Bugzilla importer is not available as it requires direct DB access to your Bugzilla instance) it's still advisable to set up a temporary self-hosted JIRA instance, do the import via Bugzilla importer, export such data to XML and then import it to OnDemand.

However if you really cannot use Bugzilla importer and what to proceed with CSV importer, then I can tell you that importing 4MB file should be an issue at all. We know customers who have successfully imported 1GB+ CSV files.

Do you mean that nothing happens after you click Begin Import?

Can you take a look at JIRA log files and let us know if there is anything suspicious there?

Cheers,

Wojtek