Why are my projects not showing after successful import from JIRA Server 7.3.1?

Alexander Ackerman

JIRA Server Version: 7.3.1

I created a backup of my JIRA Server using the JIRA Backup tool.  I expanded the archive and copied the data and logos directories over from my JIRA_HOME folder.  I recreated the archive so it matched the following structure:

jira-7.3.1-backup/
jira-7.3.1-backup/activeobjects.xml
jira-7.3.1-backup/entities.xml
jira-7.3.1-backup/data
jira-7.3.1-backup/data/attachments
jira-7.3.1-backup/data/avatars
jira-7.3.1-backup/logos

I ran through the JIRA Import tool and followed all of the prompts.  The main user is in the jira-administrators and administrators group.  I performed a re-index and the list of projects does not include any of the projects from my old JIRA Server.

Read through the hints and tips and haven't found anything to work.  Thoughts?

Ack 

 

 

Admin Project View.png

Import Wizard 2.png

Alexander Ackerman

Ok, finally figured this out...

I setup a new JIRA Server on my laptop as a temp instance and went through an import procedure of my prior backup.  Discovered it couldn't find the entities.xml file because of the directory structure listed above.  When I created the backup using the following structure it restored the backup with no problems:

archive.zip
- entities.xml
- activeobjects.xml
- data/
- data/attachments
- data/avatars
- logos

So I re-ran the atlassian-xml-cleaner-0.1.jar file on my entities.xml file (I'm coming from a MySQL server) and recreated the archive as listed above.  Low and behold, it worked!  This was a new error though:

JIRA Import-Crowd Server Error.png

 

I had to edit the entities.xml file to find the following entry:

<Directory id="10000" directoryName="Crowd Server" lowerDirectoryName="crowd server" createdDate="2016-05-14 21:35:07.0" updatedDate="2017-02-21 18:25:36.0" active="1" implementationClass="com.atlassian.crowd.directory.RemoteCrowdDirectory" lowerImplementationClass="com.atlassian.crowd.directory.remotecrowddirectory" type="CROWD" position="0"/>

and change the "active" attribute to 0.  I have been using a local Crowd Server on my network to manage users across JIRA and Bitbucket. 

There were a few more errors I had to edit and clean up with duplicate email addresses for users (who had multiple entries for some reason).  I just defaulted to adding extra characters at the end of the email to get it to pass initial validation.  I was able to remove those users (who no longer required access) from the Cloud instance after the import succeeded.

Long winded explanation on what I did, but I hope it helps someone.  I spent WAY to long fighting this.  The bottom line is that when you create your Zip archive, don't put the entities.xml and activeobjects.xml file in a subdirectory (and zip that). Go into the working directory with those files and create the archive there.  

Onward to testing the Cloud instance now...

Ack

JIRA Import-Crowd Server Error.png