To Nha Notes | July 14, 2023, 5:18 p.m.
The following are limitations when using Amazon DocumentDB as a source for AWS DMS:
When the _id option is set as a separate column, the ID string can't exceed 200 characters.
Object ID and array type keys are converted to columns that are prefixed with oid and array in table mode.
Internally, these columns are referenced with the prefixed names. If you use transformation rules in AWS DMS that reference these columns, make sure to specify the prefixed column. For example, specify ${oid__id} and not ${_id}, or ${array__addresses} and not ${_addresses}.
Collection names and key names can't include the dollar symbol ($).
Table mode and document mode have the limitations discussed preceding.
Migrating in parallel using autosegmentation has the limitations described preceding.
An Amazon DocumentDB (MongoDB compatible) source doesn’t support using a specific timestamp as a start position for change data capture (CDC). An ongoing replication task starts capturing changes regardless of the timestamp.
When using DocumentDB (MongoDB compatible) as a source, DMS can handle a maximum of 250 records per second.
AWS DMS doesn't support documents where the nesting level is greater than 97.
Source filters aren't supported for DocumentDB.
|
Amazon DocumentDB data types |
AWS DMS data types |
|---|---|
|
Boolean |
Bool |
|
Binary |
BLOB |
|
Date |
Date |
|
Timestamp |
Date |
|
Int |
INT4 |
|
Long |
INT8 |
|
Double |
REAL8 |
|
String (UTF-8) |
CLOB |
|
Array |
CLOB |
|
OID |
String |
https://docs.aws.amazon.com/dms/latest/userguide/CHAP_Reference.DataTypes.html
https://docs.aws.amazon.com/dms/latest/userguide/CHAP_Reference.DataTypes.html