Replicate task fails with "the task stopped abnormally"

[billmcevoy]

I have a task that replicates several tables from an Oracle source to SQL Server target.

When I run the task, I get "The task stopped abnormally".  

I turned on verbose logging, but no errors are apparent.  Here is the tail end of the log:

00014760: 2022-07-27T01:08:23 [TARGET_APPLY ]I: Target endpoint 'SQL Server' is using provider syntax 'SQLServer' (provider_syntax_manager.c:879)

00014760: 2022-07-27T01:08:23 [TARGET_APPLY ]I: Restore bulk state. Last bulk last record id - '0', last applied record id - '0', target confirmed record id - '0', sorter confirmed record id - '0' (endpointshell.c:2013)

00014760: 2022-07-27T01:08:23 [TARGET_APPLY ]I: Set Bulk Timeout = 30000 milliseconds (bulk_apply.c:562)

00014760: 2022-07-27T01:08:23 [TARGET_APPLY ]I: Set Bulk Timeout Min = 1000 milliseconds (bulk_apply.c:563)

00014760: 2022-07-27T01:08:23 [TARGET_APPLY ]I: Working in bulk apply mode (endpointshell.c:2021)

00014760: 2022-07-27T01:08:23 [SOURCE_CAPTURE ]I: Source endpoint 'Oracle' is using provider syntax 'Oracle' (provider_syntax_manager.c:873)

00013912: 2022-07-27T01:08:23 [TASK_MANAGER ]I: All stream components were initialized (replicationtask.c:3609)

00008144: 2022-07-27T01:08:23 [SOURCE_CAPTURE ]I: Oracle capture start time: now (oracle_endpoint_capture.c:605)

00012488: 2022-07-27T01:08:23 [SORTER ]I: Sorter last run state: confirmed_record_id = 0, confirmed_stream_position = '' (sorter_transaction.c:3251)

00008144: 2022-07-27T01:08:23 [SOURCE_CAPTURE ]I: Used difference between the Replicate machine UTC time and Oracle Local time is '14365' seconds (oracle_endpoint_capture.c:349)

00008144: 2022-07-27T01:08:23 [SOURCE_CAPTURE ]I: RAC instance 'eipt11' (thread '1') is captured (oracdc_merger.c:576)

00008144: 2022-07-27T01:08:23 [SOURCE_CAPTURE ]I: RAC instance 'eipt12' (thread '2') is captured (oracdc_merger.c:576)

00008144: 2022-07-27T01:08:23 [SOURCE_CAPTURE ]I: RAC instance 'eipt11' (thread '1') is opened (oracdc_merger.c:924)

00008144: 2022-07-27T01:08:23 [SOURCE_CAPTURE ]I: RAC instance 'eipt12' (thread '2') is opened (oracdc_merger.c:924)

[john_wang · Accepted solution]

Hello @billmcevoy ,

Looks like the task crashed, we need additional information to understand the issue: How about If you turn off "Change Processing" in task setting but keep Full Load option ON then what's the result.

If Full Load works fine then the problem is in CDC part. I'd like to suggest you open a support ticket and  with below information: set SOURCE_CAPTURE to Verbose, reproduce the behavior then attach the Diagnostics Packages to the ticket , Support Team will help you on that.

Hope this helps.

Regards,

John.

[SachinB · Accepted solution]

Hello William,

This failure occurs for one of the following reasons:

1. The executed process does not have correct security access to the window station and desktop that are associated with the process.

2. The system ran out of desktop heap.

3.  Also check do you have already running same process ID for particular task in task manager, If yes please kill and resume the task.

Could you please install the debugdiag tool which will give more details in order to check the task crash dump files and create support case in order to investigate it further.

DebugDiag is an official tool from Microsoft for catching an abnormal process termination and creating a dump file for debugging, it can be download from below link :

https://www.microsoft.com/en-us/download/details.aspx?id=58210

Regards,

Sachin B

[Bill_Steinagle · Accepted solution]

Bill,

Hello thank you for the post to the QDI forums. Since the Task is crashing you will need to also check the Replicate Server for Memory and or space related issues with the Replicate\data directory as these would cause the Task to crash as shown. If further investigation is needed as John posted please send us a Support ticket to help address the issue.

Thanks!

Bill

[Dana_Baldwin]

Hi @billmcevoy 

Please follow the steps at this link to collect the required information to troubleshoot the issue, and open a support case, as the solution may involve providing a patch:

https://community.qlik.com/t5/Knowledge/Collecting-Replicate-Process-Dumps/ta-p/1746840

Thanks,

Dana

[danielshryock · Accepted solution]

I was having error messages very similar to the ones listed here.  The problem turned out to be a timeout running a SQL query while trying to do CDC processing.  I was able to get better error details by upgrading Replicate to the latest version.

[john_wang]

Hello @danielshryock ,

Thank you so much for the feedback. The information is valuable for us.

Best Regards,

John.