nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Confluence Import fails

Malachi Burke_0qaupf

Environment: Confluence 6.5.2-x64 on 64 bit debian 9, postgres

I am unable to perform an import of a zipped backup file. The import error is "Import failed. Check your server logs for more information. null". Logs are huge, so I won't post the whole thing here, but pertinent portions include:

a little over 5000 of these:

2017-12-06 15:33:59,162 WARN [Long running task: Importing data] [confluence.importexport.xmlimport.DeferredOperations] reportOutstandingOperations Uncompleted deferred operations waiting for:[class = com.atlassian.confluence.content.ContentProperty, id = 3968162]
 -- url: /longrunningtaskxml.action | referer: http://it.moducom.net/admin/restore.action?synchronous=false | traceId: 5552b07c6cb26fd1 | userName: malachib | action: longrunningtaskxml

two of these (truncated):

2017-12-06 15:34:00,753 WARN [Long running task: Importing data] [confluence.impl.hibernate.ConfluenceHibernateTransactionManager] doRollback Performing rollback. Transactions:
 ->[com.atlassian.confluence.importexport.DefaultImportExportManager.performImport]: PROPAGATION_REQUIRED,ISOLATION_DEFAULT (Session #14613578)
 [com.atlassian.confluence.impl.schedule.caesium.SchedulerClusteredJobDao.delete]: PROPAGATION_REQUIRES_NEW,ISOLATION_DEFAULT (Session #2090290650)

and a half dozen exceptions, the first one (truncated):

java.lang.NullPointerException
 at java.io.File.<init>(File.java:277)
 at com.atlassian.confluence.pages.persistence.dao.BootstrapAttachmentLocationResolver.getFileLocation(BootstrapAttachmentLocationResolver.java:17)

Any clues? Thank you

Accepted answers

R. Fuerst

The issue here was caused by exporting/importing a backup from an older version of Confluence.

In Confluence, it is not recommended to export the instance from a previous version, and import it to the new instance with the newer version. This will often fail due to changes between major versions of Confluence.

The solution was to export the site from the production Confluence 6.3.2 instance, spin up a new 6.3.2 instance and restore the backup there. Once restored, we upgraded to 6.5.2 without issue.

All comments

Shannon S

Hi Malachi,

Could you confirm the Postgres version you're using? Is the import a full site import or a space import? Can you tell us what version of Confluence it came from? I understand you're importing into a 6.5.2 version.

The messages you've sent are just warnings, are you able to review your error log and see if you see any lines that say ERROR rather than WARN and then check to see if there's a Caused by: message?

Thank you!

Regards,

Shannon

Malachi Burke_0qaupf

Hi Shannon and thank you for the quick response!

It's as near as I can tell a full site import. Certainly that's the intention.

It is coming from Confluence 6.3.2, and we're using postgres 9.6.6.

Quite a few errors, including the NullPointerException. Here's a few of them, with the stack trace truncated:

2017-12-06 15:34:00,756 ERROR [Long running task: Importing data] [confluence.importexport.actions.ImportLongRunningTask] runInternal Failure during import
 -- url: /longrunningtaskxml.action | traceId: 5552b07c6cb26fd1 | userName: malachib | referer: http://it.moducom.net/admin/restore.action?synchronous=false | action: longrunningtaskxml
java.lang.NullPointerException
 at java.io.File.<init>(File.java:277)
 at com.atlassian.confluence.pages.persistence.dao.BootstrapAttachmentLocationResolver.getFileLocation(BootstrapAttachmentLocationResolver.java:17)
 at com.atlassian.confluence.pages.persistence.dao.VersionedFileLocationResolver.getConfluenceAttachmentsDirectory(VersionedFileLocationResolver.java:23)
 at com.atlassian.confluence.pages.persistence.dao.VersionedFileLocationResolver.getFileLocation(VersionedFileLocationResolver.java:19)

2017-12-06 15:34:54,401 ERROR [localhost-startStop-3] [hipchat.spacetoroom.service.SpaceToRoomNotificationService] hipChatNotifications Failed to obtain hipchat notifications for event com.atlassian.plugin.event.events.PluginFrameworkShutdownEvent@3766a37a: service proxy has been destroyed

2017-12-06 15:34:54,504 ERROR [localhost-startStop-3] [atlassian.confluence.setup.ConfluenceListableBeanFactory] destroyBean Destroy method on bean with name 'contentPropertyService' threw an exception
com.atlassian.util.concurrent.LazyReference$InitializationException: java.lang.IllegalStateException: Unable to create a tracker when osgi is not running
 at com.atlassian.util.concurrent.LazyReference.getInterruptibly(LazyReference.java:149)
 at com.atlassian.util.concurrent.LazyReference.get(LazyReference.java:112)
 at com.atlassian.confluence.osgi.OsgiServiceRegistryImpl$ServiceTrackingProxy.getServiceTracker(OsgiServiceRegistryImpl.java:75)
 at com.atlassian.confluence.osgi.OsgiServiceRegistryImpl$ServiceTrackingProxy.invoke(OsgiServiceRegistryImpl.java:80)

2017-12-06 15:34:54,532 ERROR [localhost-startStop-3] [atlassian.confluence.setup.ConfluenceListableBeanFactory] destroyBean Destroy method on bean with name 'boostingQueryFactory' threw an exception
com.atlassian.util.concurrent.LazyReference$InitializationException: java.lang.IllegalStateException: Unable to create a tracker when osgi is not running
 at com.atlassian.util.concurrent.LazyReference.getInterruptibly(LazyReference.java:149)
 at com.atlassian.util.concurrent.LazyReference.get(LazyReference.java:112)
 at com.atlassian.confluence.osgi.OsgiServiceRegistryImpl$ServiceTrackingProxy.getServiceTracker(OsgiServiceRegistryImpl.java:75)

There's a few more, if you like I can post them as well. I know they are "just warnings" but when I see 5,000 warnings directly preceding error messages, it piques my interest!

Malachi Burke_0qaupf

It's been one week. It's fortunate that this is a separate-machine upgrade and not an emergency restore. Do you have any recommendations?

Shannon S

Hi Malachi,

We will need more information from you, so I have created a ticket in our support system.

Please reply to that case with a copy of your support zip after re-attempting the import so we can have the full logs.

Thank you!

Regards,

Shannon

R. Fuerst

The issue here was caused by exporting/importing a backup from an older version of Confluence.

The solution was to export the site from the production Confluence 6.3.2 instance, spin up a new 6.3.2 instance and restore the backup there. Once restored, we upgraded to 6.5.2 without issue.