Question related to Restore Confluence

[ujkimbitbucket]

Hi.

I'm processing on restore confluence backup procedure.

 - original confluence version: 5.9.4

 - update target confluence version : 6.9.0

I was followed restore confluence backup procedure.

First of all, I was installed with recent confluence version at my test server.

I confirmed with starting up confluence normally, and then replaced with blow files.

  - confluence backup database

  - confluence.cfg.xml

  - attachments files

  - index files

But issue was occured with starting confluence at first.

1) license expired issue(add-on) => refreshed recently evaluation license (this issue was resolved)

2) build number is not matched => following with below contents & this issue was resolved

https://confluence.atlassian.com/confkb/confluence-will-not-start-up-because-the-build-number-in-the-home-directory-doesn-t-match-the-build-number-in-the-database-after-upgrade-376834096.html

next, another issue was occured.

I can show confluence main login page. but I can't login this confluence page.

I assume that my origin confluence account is not match with new confluence version's account. (Where can I get to check now original account info?)

I thought that confluence database import was resolved this issue. but no...

Can I resolve this problem? I can want to know that.

Thank you 

Best Regards,

Uijoong-Kim

[Igor M. · Accepted solution]

Hello,

If you were using Jira or Crowd to manage and login to Confluence, and new instance does not have a proper link back to those authentication services, you can only log in with a local account. The one you set up during initial Confluence installation. If you can't log-in with that account or unable to find it's alias, you can try resetting local account by following instructions in Restore Passwords To Recover Admin User Rights KB. Once logged in, re-create user directories.

Let us know if you encounter troubles.

Regards,

Igor

[ujkimbitbucket]

Hello! @Igor M_ Thank you for help.

I understand that your opinion.

I tried to log in confluence server as recovery-admin. but admin password was not changed for Crowd Synchronize. (originally we are synchronize with jira server as Atlassian crowd, but now we're on backup restore testing and not construct jira server not yet.)

In this situation, Except Atlassian Crowd Can I get some method for activate confluence restore contents?

[Igor M.]

In this situation, you need to

Log-in with recovery admin account,

Go to Cog > General Configuration > User Directories

Using small arrows in Order section move Confluence Internal Directory to the top, so it's a primary authentication directory.

Now go to Users > Add User

Add a new user that will become admin, and make sure this user is a member of confluence-users and confluence-administrators group

Log-out and log-in with new user

You should be able to access all content with your new user

Any other users that were pulled from Jira directory will either have to be manually re-created with the same username, or you have to re-add Jira as directory

Hope this is what you were looking for.

Regards,

Igor

[ujkimbitbucket]

Thank you for your reply!

I understand your opinion, and I'm done that. 

I have a question related to this issue.

The purpose of this issue is successful backup restore of confluence.

I was followed with confluence restore procedure, and I reached this step.

Now, We are impossible check the original confluence page contents. (only we can check the user Information)

Why can we not show confluence page contents?

( We was followed confluence restore step (insert to the new confluence version with files (confluence.cfg.xml, attachements, postgresql db dump import, index)) 

I expected that's why was our confluence not access to jira Atlassian Crowd directory.

(In issue, at Confluence atlassian crowd configuration failed(application failed to authenticate), In server configuration, jira administrator account(application name field) is failed to authenticate)

That is ture? or anything else?  I doubted why this issue occured.

Can I get some information about it?

Thank you for helping me with your sincerity.

[Igor M.]

Hello,

That is most likely the cause. If you use Jira Crowd for user management then all content will be created by Jira users and restricted for jira users, if you remove Jira auth provider, or if you restore data but never allow Jira to resynchronise, your Jira accounts either will not exist or be disabled. So content could even appear to be under "unknown" user.

Same goes for groups, if you use groups to manage access to content and those groups are pulled from Jira Crowd, if you remove or prevent Jira Auth Provider from synchronising or re-synchronising after restore, then you will not be able to access pages, because restrictions can not be read, a workaround for that is to create a local group in Confluence or add confluence-users group to space and page restrictions and make sure local users are members of that group.

Ownership information is stored in DB, so even if you delete Jira Crowd and re-add it later, permissions will be recognised.

Hope this clarified a bit

Regards,

Igor

[ujkimbitbucket]

Thank you for request!

I've been got a good information thanks to you!

have a good day:)

[Yogesh Mude · Accepted solution]

HI @ujkimbitbucket

Are you using the crowd for User management in Confluence v 5.9.4 server? if yes, then can you please try login with Internal directory User.

[ujkimbitbucket]

Hello!

No, I'm not used it. what does mean login with Internal directory user?

[Yogesh Mude]

HI @ujkimbitbucket

The internal directory is enabled by default at installation. When you create the first administrator during the setup procedure, that administrator's username and other details are stored in the internal directory.

Once you restore the Confluence then the user directories(LDAP) are not synchronize, hence the user is not able to log in.

To overcome this issue you need to log in to internal directory user and then you need to synchronize the directory then only other users can able to access the Confluence.

[ujkimbitbucket]

Hello @Yogesh Mude Thank you for your reply.

We are using the atlassian crowd for user management now.

I'm sorry about I was misunderstood at yesterday.

In new confluence version, we don't log in confluence because confluence is synchronize with jira.(but now jira server is not access)

Can I get documentation to login without Atlassian Crowd? Or We never login without alive the atlassian crowd server?

[Yogesh Mude]

HI @ujkimbitbucket

Its ok, you can login without crowd user using internal/local directory user.

If you know the user and password for this internal/local user then you can log in with this and then sync with jira, then it will allow all the users to log in with their own id and password.

If you don't have a password with the internal/local user then we can reset the password with DB and we can log in. 

Here is the documentation for resetting the password.

[ujkimbitbucket]

Thank you @Yogesh Mude!

I will be refer to your documentation. But, I tried to log in original account's ID and password. Nevertheless that is not changed, log in is failed. I don't know why.

(At first, I was installed recently confluence version. is it affected to this issue?)

In environment where the original db(postgres) is inserted, I expect confluence log in is work correctly. what is related to this problem anything else?

Anyway, Thanks for your recommend!

[Yogesh Mude]

Original account means are you talking about the Internal/local user or the user coming from crowd directory. 

Still, if you are not able to access then the best way update password of the internal/local user from DB and then try to log in.

[ujkimbitbucket]

Thank you for request.

I've been got a good information thanks to you.

have a good day!