Changing Task Recovery options for Replicate

[Sonja_Bauernfeind]

The default recovery options can cause the log to spam with a task stopped with a recoverable error. Seen underneath is an example of the logs that can be encountered. The reason the logs are spamming is that the default settings has -1 which is infinite and Replicate will attempt to keep up until the max seconds is reached. With the increase retry interval for long outages is checked, the time when Replicate will attempt the task is doubled each time. 

Example of the settings: 


Maximum Retry Count:

• Make sure the retry account is above -1 as -1 is infinite. 

Interval between retry attempts:

• Determines in seconds how long before Replicate attempts to redo the task.
• If increase retry interval for long outages is checked the seconds will double each attempt that is made to redo the task.

  Example of Logs:

2020-05-01T00:39:25 [SERVER          ]I:  “Task ‘1-  Example task” stopped with recoverable error  (server.a:1214)
2020-05-01T01:09:50 [SERVER          ]I:  Task '1 Example task” stopped with recoverable error  (server.a:1214)
2020-05-01T01:40:15 [SERVER          ]I:  Task '1”  stopped with recoverable error  (server.a:1214)
2020-05-01T02:10:39 [SERVER          ]I:  Task '1 - Example Task stopped with recoverable error  (server.a:1214)
2020-05-01T02:41:04 [SERVER          ]I:  Task '1 - Example Task stopped with recoverable error  (server.a:1214)
2020-05-01T03:11:28 [SERVER          ]I:  Task '1 - Example Task stopped with recoverable error  (server.a:1214)
2020-05-01T03:41:51 [SERVER          ]I:  Task '1 - Example Task stopped with recoverable error  (server.a:1214
2020-05-01T04:12:15 [SERVER          ]I:  Task '1 - Example Task stopped with recoverable error  (server.a:1214)
2020-05-01T04:42:40 [SERVER          ]I:  Task '1 - Example Task stopped with recoverable error  (server.a:1214)
2020-05-01T05:13:02 [SERVER          ]I:  Task '1 - Example Task stopped with recoverable error  (server.a:1214)
Example Task..log:00021896: 2020-05-01T05:12:41 [TASK_MANAGER    ]I:  Task '1 - Example Task running full load and CDC in resume mode after recoverable error, retry #39  (replicationtask.c:1232)
Example Task..log:00012144: 2020-05-01T05:12:51 [TASK_MANAGER    ]W:  Task '1 - Example Task encountered a recoverable error  (repository.c:5189)

Resolution:

Instructions: 

To change the settings go to the Error Handling in the Environmental Errors under SQL to SQL - Settings. Make sure that the Maximum Retry Count is above -1 so that Replicate is not attempting to recover the task infinitely. Also, make sure the Interval between retry attempts is deselected so that the interval between the attempts is not doubled.