export to csv file from on prem jira (for purpose of import to cloud jira)

[Neelam Manhas]

field: issue description

The export to the csv file is cutting off a lot of issue descriptions in various issues. This is the key field for us, it has the scenario/acceptance criteria.

I can see the issue description on the screen after the search no problem.

The descriptions with straight text seem to be fine, even some special characters seem to be fine.

But there are many issues I have where there is a 20 line issue description however only 2 lines get loaded into the csv file, what is the limitation here, are there know things like perhaps a tab or certain characters that won't be loaded?

Is there a better way to do this?  The import on the cloud only appears to except a csv file.

The migration assistant tool is not an option at this  time due to our old version of on prem.

[Dave Theodore [Coyote Creek Consulting] · Accepted solution]

How are you reading the CSV file? I find that Excel and even OpenOffice have trouble with complex CSV files.  They sometimes split cells and add line feeds even on a legal CSV file. You are also going to need to split some columns.  Any multi-select field will need to have only one value per column in order to import in to Jira. You will also need to deal with attachments somehow, and host them somewhere that Cloud can access.  Unless you have a small amount of data to migrate, it might be easier to upgrade your on-prem version and use JCMA.

[Neelam Manhas]

thanks @Dave Theodore [Coyote Creek Consulting]  I am actually not doing anything to the file, its a straight export from on prem jira, when I open the csv file some of the descriptions are cut off already.  Yes we would have liked to use JCMA but our on prem version is very old and not being supported with this tool.

[Dave Theodore [Coyote Creek Consulting]]

My question was, what tool are you using to read the file? I you use a generic text tool that just reads what is there (such as cat, more, notepad, etc) you will see the text as it exported. If the data is truncated when you read it that way, you can conclude that the issue is in Jira.  You need to confirm that it's not the tool you are using to read the CSV that is munging the text. We use command line CSV editors to avoid this issue because all of the UI CSV tools have issues of one form or fashion.

With regard to JCMA, it might be worth the effort to upgrade your Jira to a compatible version, then use JCMA.  If you don't want your users to feel the pain of different versions twice, you could make a clone of your production instance, upgrade it, then use JCMA.  This has the added benefit that if something goes wrong with the migration, you can just bring the production instance back online and you are right back where you started.

[Neelam Manhas]

@Dave Theodore [Coyote Creek Consulting]  thank you, yep it was the csv file!  If i don't open and save the csv and import it its fine, once i open and save it messes up the description and cause issues.  Was able to see the differences by using notepad.