bulk delete issues fails with java lang null pointer exception

Question

Technical details

Log's referral number: 151d809c-0c09-4054-872e-553baedd0947

Cause

Referer URL: http://nyc-cbmweb-d29:8080/secure/views/bulkedit/BulkDeleteDetailsValidation.jspa

java.lang.NullPointerException

java.lang.NullPointerException
	at com.atlassian.jira.web.action.issue.bulkedit.BulkOperationProgress.doProgress(BulkOperationProgress.java:76) [classes/:?]

Andy Heinzer · Answer

I suppose it is possible that the rank information from Cloud is having this problem too, but if that is the case, then you should have this same problem in the Cloud instance.   I am not sure if you still have access to the Cloud site in order to test this theory.  But if you do, I would recommend trying to test this to understand if the source of the problem is in the Cloud site or not.

If it is, then I strongly recommend creating a support case with our Cloud support team to have them resolve this.  I suspect it will require some downtime for your site in order to correct this.  You can create a support request for our cloud team at https://support.atlassian.com/contact/

But it is also possible that when you migrate from Cloud to Server, there could be an upgrade (or downgrade) task that is not getting executed correctly.   Or worse, that the import process did not complete the migration of all 100% of that data in the backup.

So I would recommend looking very closely at the $JIRAHOME/log/atlassian-jira.log file at the time the import first took place.  Jira should be logging a percentage when importing all this data.  1-100%.   If you only see the logs indicate that say 90% was imported, and there exists some kind of error, such as a SQL exception, then I think we have a bigger problem here that should be addressed first.

It is possible to follow the KB on fixing the rank field to get past this, but it might be necessary for us to create a support request in order to get your logs first so we can better understand the best next steps to follow.

I can do this on your behalf if you like, or you can also create this on https://support.atlassian.com/contact/ as well.  Please let me know.

Eric Sebian · Answer

Hi Andrew, I verified that there was only 1 rank custom field (other than an obsolete one). One other fun fact that I forgot to mention is that it is only sub tasks that I cant delete. tasks that are NOT sub tasks and do NOT have sub tasks associated with them, can be deleted.

This instance is to help he prepare for a migration from Atlassian cloud to a data center version. so the data is currently from the cloud version. I'm working to verify how certain add-ons are going to react and reconfigure them ahead of time. Is there a chance that the issue is originating on the cloud side? Garbage in. Garbage out? How can I verify that?

I tried restoring the data again but the problem persists. Would it be worth it to re-build the database entirely?

Please advise.

Andy Heinzer · Answer

Ah, that error is much more clear as to the cause.   Please see this KB: Cannot reindex jira due to Expected exactly 2 rows the maximum marker row and the lowest ranked row for rank field.

I suspect that the cause is #3 in that KB, and the KB does provide steps on how to correct this.  You might need to identify if your environment has a duplicate or obsolete rank field that is causing problems here.  Of course I would recommend that you create a backup of your database before you make any direct changes.

If you do not have duplicate or unused ranks fields, then the problem is likely in regards to your existing rank field is either missing a marker, or has more than 2 markers.

Try to follow that KB and let me know if you run into any problems here.

Andy