We have an AWS DMS CDC replication task replicating data from an EC2 running MongoDB to Postgres. It has been running fine, DMS replication instance resources (RAM, CPU etc) all seem normal, however the task has started to fail, with:
Last Error The task stopped abnormally Stop Reason RECOVERABLE_ERROR Error Level RECOVERABLE
And error from Cloudwatch logs:
at_common_is_supported: load_common_wrapper( common_load_locker ) failed (at_common.c:69)
It is also possible to improve logging output further in the task settings JSON editor (additionally to the 5 fields that you can set in the console UI) by opening the task JSON editor and amending all the additional fields to: "Severity": "LOGGER_SEVERITY_DETAILED_DEBUG"
With the detailed_debug error logging enabled:
sqlite_execute_statement:361: Failed (at_sqlite.c:367)
sqlStatement=ALTER TABLE cdc_status ADD COLUMN read_volume_rollback INTEGER; (at_sqlite.c:325)
object already exists (at_sqlite.c:331)
Has anyone seen/ resolved this previously? Thanks in advance