Can't import xml backup from 2.6.4 version

[ION Information Systems]

Hi Jarek,

we have set up a test environment where we upgraded Jira to version 6.3.10 and The Scheduler to 3.2.2.

In the production instance we keep the Scheduler still disable due to [this issue|https://bitbucket.org/apdc/the-scheduler/issue/41/issues-created-with-the-scheduler-plugin], but before disabling it I took an xml backup of the scheduled issues.

Now I'm trying to re-import them in the new environment but I get the attached error even if I tick the "I am importing a 2.6.4 data". The file Am I doing something wrong? Is this a known issue?

Can you please advise?

Thanks Graziella

[jost]

Hi,

Please to create folder "plugin" in "... /jira-home/data/".

Then place in to this folder attached files (attached PreviousGenSchema.xsd and CurrentGenSchema.xsd).

That should resolve this problem for now. We working on this bug and it will be fixed in near future.

[ION Information Systems]

Hi,

I copied those files, and tried again. It allowed me to chose among a list of projects (which it is ok), but then I got another error (attached for the logs). I also restarted, but same error.

Please advice.

Cheers Graziella

[jost]

Hi,

That means the scheduler is trying to import a scheduled issue that cannot be created. The error is caused by change or miss some configuration. For example lack of custom field or Project.

Cheers Jacek

[ION Information Systems]

The environment we set up it's a copy of production and settings on that project have not been changed. The error seems to be related to fields "Required" in the field configuration scheme that have not a tag in the xml backup, so the scheduled issues can't be created due to missing field/s. Am I correct?

Cheers Graziella

[jost]

HI

Exactly, as i described above this error is caused by change or miss some configuration (including fields and projects). For example missing custom field or Project.

Cheers Jacek

[ION Information Systems]

Please find attached an extract (anonymized) of the backup I'm using.

I checked the Field Configuration Scheme of this project, and I set to optional the fields that are not in the file. So at the moment the only required fields are: * Description * Type * Priority * Summary and "Epic Name" which is an Agile field, that I can' set to optional as it is required in every field configuration, but I removed it from any screen this project is using (I guess during the import the screen used is only the Create one, right?). So removing it from the screen would do the trick.

Also I noticed that in the xml backup there is the Jira version, I also change it to be 6.3.10, but again it didn't work.

If you check the xml, you will see that the only empty tag is the "customfield_10100" which is not required in the field configuration scheme. All the field and this project exist in the environment we have set up.

I hope it helps to clarify.

Cheers Graziella

[jost]

Hi Graziella , the error is caused missing or wrong configuration of Custom Field/s, not just Field. That means there are no required Custom Field in your system. You can find configuration of Custom Fields on Admin>Issues > "Custom FIelds"That would be great if you can compare fields in instance of export to your current instance.

Please see:

Additing a Custom Field

Configuring a Custom Field

[ION Information Systems]

Hi Jarek,

I found out what was the issue, again it is the interaction with another plugin the nFeed in this case. In few scheduled issues (93 in total that's why it was quite difficult to spot), there were some tags related to an nFeed, removing it made the trick. * The tag was correctly populated * The custom field does exist in Jira * The nFeed is defined correctly its side

The environment I'm using is a copy of production, and I'm sure those custom fields have not been changed; of course the version changed as we are testing the upgrade of Jira from 5.2 to 6.3, and both plugins have been upgraded accordingly. But given that I'm ticking the "I am importing a 2.6.4 data" checkbox while importing the xml, I expect the Schedule knows how to map any field previously backuped. Are nFeeds supposed to work?

In this case that nFeed custom field is not strictly important, so it is not a problem for us to remove it from the import and add it manually after those issues have been created.

Cheers Graziella

[ION Information Systems]

Additionally, there was another tag that didn't allow the import. It is related to a multi user picker, the value itself was empty in the xml. Thus I'm not sure it is related to the empty value or the type of the custom field (native Jira custom field).

Removing both type of custom fields did the trick.

Cheers Graziella

[jost]

Hi Graziella,

Does that mean that you were able finally to do import?

Cheers Jacek

[ION Information Systems]

yes