Uploaded image for project: 'Confluence Data Center'
  1. Confluence Data Center
  2. CONFSERVER-14507

Large amounts of deleted content may cause Purge All Trash to fail

      A customer has recently experienced this bug. The initial problem was that their space export kept failing and after further testing and troubleshooting, it was found out that the deleted content in the trash was at fault. Upon trying to purge the trash through the UI, it failed. The workaround was to create an sql script that cleaned up and deleted everything associated to all of the content that was marked as deleted.

      This is apparently a known bug but we were not able to find a bug report for it. Apparently, the TEST space on confluence.atlassian.com also fails to purge.

            [CONFSERVER-14507] Large amounts of deleted content may cause Purge All Trash to fail

            Hi all, 

            How can I use it in confluence 4.2.3 to clean trash during assigned period, thanks a lot.

            Regards, Anderson Hsu

            Deleted Account (Inactive) added a comment - Hi all,  How can I use it in confluence 4.2.3 to clean trash during assigned period, thanks a lot. Regards, Anderson Hsu

            Partha added a comment -

            I believe this is a duplicate of CONF-15233, so I am going to close the issue.

            Please watch the issue CONF-15233.

            Kind Regards,
            Partha Kamal

            Partha added a comment - I believe this is a duplicate of CONF-15233 , so I am going to close the issue. Please watch the issue CONF-15233 . Kind Regards, Partha Kamal

            Igor Minar added a comment -

            Igor Minar added a comment - two related issues: http://jira.atlassian.com/browse/CONF-15232 http://jira.atlassian.com/browse/CONF-15233

            Hi Maleko,

            Am I right in asuming

            java.lang.NullPointerException
            	at com.atlassian.core.spool.SmartSpool.spool(SmartSpool.java:29)
            	at org.springframework.orm.hibernate.support.SpoolingBlobInputStreamType.nullSafeGetInternal(SpoolingBlobInputStreamType.java:52)
            	at org.springframework.orm.hibernate.support.AbstractLobType.nullSafeGet(AbstractLobType.java:137)
            	at net.sf.hibernate.type.CustomType.nullSafeGet(CustomType.java:92)
            

            appears to be the terminating error when trying to delete too much content?

            We need a more specific description of the issue to begin investigation.

            Regards,
            Andrew Lynch

            Andrew Lynch (Inactive) added a comment - Hi Maleko, Am I right in asuming java.lang.NullPointerException at com.atlassian.core.spool.SmartSpool.spool(SmartSpool.java:29) at org.springframework.orm.hibernate.support.SpoolingBlobInputStreamType.nullSafeGetInternal(SpoolingBlobInputStreamType.java:52) at org.springframework.orm.hibernate.support.AbstractLobType.nullSafeGet(AbstractLobType.java:137) at net.sf.hibernate.type.CustomType.nullSafeGet(CustomType.java:92) appears to be the terminating error when trying to delete too much content? We need a more specific description of the issue to begin investigation. Regards, Andrew Lynch

            Additional comments from the customer:

            We have several spaces intended for use as reference material for Call Center staff where we've built 1,000+ pages of content, often through the use of WebDAV. While the content authors were getting familiar with using Confluence and settling on the approach they wanted to take re: structuring the content, a large amount of Trash was collected... 3,000 or more pages in several spaces.

            The inability for us to empty the Trash in any of these spaces means (a) we cannot export the space because the trash is carried forward to the new space on an export and (b) we're using space in our database for stuff we really don't need or want.

            There should be both a long-term fix to the Space Export function to improve its performance and increase its ability to handle large amounts of Trash (see CONF-7905 (Enable removal of trashed pages at certain point) and CONF-4296 (Purging the trash can be a long running task, so provide appropriate feedback)) and, in the absence of those issues being addressed in a future release, Atlassian should provide detailed instructions for manually deleting the Trash including, for instance, whether the Ancestor table needs to be rebuilt, the appropriate sequence in which to do the deletes, etc.

            Perhaps because of the large amount of Trash in these spaces or perhaps CONF-13814 (Error removing space with custom space logo and attachments stored in the database), we are also unable to remove these spaces even when all that is left in them is the Trash.

            Thanks,
            Betsy

            Maleko Taylor (Inactive) added a comment - Additional comments from the customer: We have several spaces intended for use as reference material for Call Center staff where we've built 1,000+ pages of content, often through the use of WebDAV. While the content authors were getting familiar with using Confluence and settling on the approach they wanted to take re: structuring the content, a large amount of Trash was collected... 3,000 or more pages in several spaces. The inability for us to empty the Trash in any of these spaces means (a) we cannot export the space because the trash is carried forward to the new space on an export and (b) we're using space in our database for stuff we really don't need or want. There should be both a long-term fix to the Space Export function to improve its performance and increase its ability to handle large amounts of Trash (see CONF-7905 (Enable removal of trashed pages at certain point) and CONF-4296 (Purging the trash can be a long running task, so provide appropriate feedback)) and, in the absence of those issues being addressed in a future release, Atlassian should provide detailed instructions for manually deleting the Trash including, for instance, whether the Ancestor table needs to be rebuilt, the appropriate sequence in which to do the deletes, etc. Perhaps because of the large amount of Trash in these spaces or perhaps CONF-13814 (Error removing space with custom space logo and attachments stored in the database), we are also unable to remove these spaces even when all that is left in them is the Trash. Thanks, Betsy

              Unassigned Unassigned
              mtaylor@atlassian.com Maleko Taylor (Inactive)
              Affected customers:
              2 This affects my team
              Watchers:
              3 Start watching this issue

                Created:
                Updated:
                Resolved: