Come for the products,
stay for the community

The Atlassian Community can help you and your team get more value out of Atlassian products and practices.

Atlassian Community about banner
Community Members
Community Events
Community Groups

FishEye/Crucible DB lost, login broken after Jira upgrade

After a minor Jira update (7.1=>7.7), all our Atlassian products are Failing, and in different ways. I'm trying to recover them one by one; FishEye/Crucible is just the first. They all fail in different ways, though.

We were running Jira 7.1, using its user management, and had 10 users configured. Half of them didn't need FishEye/Crucible access . This included former employees (account disabled), product managers and an external partner. Non of those had admin access.

As we got an EOL warning on Jira 7.1, we upgraded Jira to 7.7 (disaster in its own right, not the subject of this question). However this broke FishEye - it seems to have lost its entire configuration. When trying to connect it to Jira again, it decided to import 5 out of the 10 accounts there. And unfortunately, it picked not a single admin account, but it *did* pick every single Jira user that should not be a FishEye user.


As a result, we now run into the 5-user limit of Fisheye/Crucible. And we can't fix it, as we can't even log in anymore to delete the offending users!


1 answer

1 accepted

0 votes
Answer accepted
Andy Heinzer Atlassian Team Feb 13, 2018

Hi Michiel,

You should still be able to login with the anonymous admin account that does not count against your license limit by going to


In addition to that URL, you can also follow the steps in How to reset the Administration Page password in FishEye or Crucible to reset the Admin account.

Once you can login to Fisheye as an admin, you can then reconfigure the External User directory in use here.  I suspect there is a particular group membership in Jira that Fisheye might be looking at in order to determine which users should have access to that application.

Worked. Had to reconfigure repositories and perform a DB restore, but it's working again.

Suggest an answer

Log in or Sign up to answer

Atlassian Community Events