JSD does not send customer notifications, throws java.lang.IllegalStateException

[Daniel Fietta]

Hello community,

We are running on Jira Core and Software on v7.4.2 and Jira Service Desk on v3.6.2.

In the Service Desk project, everything is working, except the customer notifications. The notifications are defined in the project properties and a request type is set.

All notifications to the agents that are defined in the notification scheme are working, so there is no problem with outgoing mails. Also incoming mails are working, when customers are answering tickets via email.

When an agent comments on a ticket, there is no log entry in the outgoing mails, but in the atlassian-jira.log:

2017-11-14 17:27:38,976 PsmmqAsyncExecutors-job:thread-375 WARN user 1047x808839x1 c7tfz 45.xxx.xxx.xxx,192.168.1.10 /rest/api/2/issue/23649/comment [c.atlassian.ozymandias.SafePluginPointAccess] Unable to run plugin code because of 'java.lang.IllegalStateException - Too many rows found for query on ClusteredJob

Following this entry, there are always these two entries, too:

OfBizClusteredJob[id=15833,delegate=ImmutableClusteredJob[jobId=sd.custom.notification.batch.send,jobRunnerKey=sd.custom.notification.batch.job.runner,schedule=Schedule[type=INTERVAL,intervalScheduleInfo=IntervalScheduleInfo[firstRunTime=Mon Nov 13 10:45:28 CET 2017,intervalInMillis=60000]],nextRunTime=Mon Nov 13 10:45:28 CET 2017,version=1,rawParameters=(null)]]'.
row1:

OfBizClusteredJob[id=15832,delegate=ImmutableClusteredJob[jobId=sd.custom.notification.batch.send,jobRunnerKey=sd.custom.notification.batch.job.runner,schedule=Schedule[type=INTERVAL,intervalScheduleInfo=IntervalScheduleInfo[firstRunTime=Mon Nov 13 10:45:28 CET 2017,intervalInMillis=60000]],nextRunTime=Mon Nov 13 10:45:28 CET 2017,version=1,rawParameters=(null)]]row2: 

I hope I added all needed information and you can help me to find a solution.

Thank you in advance!

Daniel

[Andy Heinzer · Accepted solution]

Hi Daniel,

Sorry for the great delay in response, but the specific error message you are seeing here of

Too many rows found for query on ClusteredJob

is a known bug.   We have a KB that explains how you can work-around this issue in JIRA Fails to Start due to Too many rows found for query on ClusteredJob.  From that KB you should be able to run the following SQL command to find the duplicate entries and their respective ids:

SELECT * FROM clusteredjob WHERE job_id in (SELECT job_id FROM clusteredjob GROUP BY job_id HAVING COUNT(*)>1)

Once you know the id of this duplicate, you can then follow these steps to remove the duplicate from SQL:

Resolution

1. Shut down JIRA and back up the database properly

2. Delete either of the entries:

   delete from clusteredjob where id = 11106;

3. For Oracle DB, ensure run commit; after the above.

4. Restart JIRA

Where in that example one of the duplicate entries had an id of 11106. 

Regards,

Andy

[Daniel Fietta]

Hi Andy, 

that solution is exactly what solved the problem.

Thanks a lot!

Best,
Daniel