mongosync
. View the
current documentation
for up-to-date guidance on mongosync
and instructions on how to upgrade
to the latest version.To reverse the direction of a sync operation, use the following
procedure to commit your sync and call the
reverse
endpoint.
If you want to keep your source cluster updated with new writes
from your destination cluster post-cutover, you can use reverse
.
For more details, see Mid-sync Considerations.
Before You Begin
Before you can reverse your sync direction, you must configure
mongosync
and use the start
endpoint to set the
following parameters:
reversible
totrue
enableUserWriteBlocking
totrue
.
Warning
Unique indexes on the source cluster must not use the legacy format.
To validate that collection indexes on the source cluster use the proper formatting, see Validate Unique Indexes.
For more information on limitations and requirements of reversing sync,
see reverse
.
Important
If you are migrating from a pre-6.0 source cluster, you cannot reverse sync.
Steps
Use these steps to reverse the direction of your sync:
Finalize your sync.
Follow the Finalize Cutover Process tutorial to enable write blocking and finalize your data migration.
IMPORTANT: mongosync
does not support
Filtered Sync during the reverse sync process.
Call the reverse
endpoint.
Call the reverse
endpoint to reverse the direction of your sync
operation. Your original source cluster becomes your new destination cluster
and your original destination cluster becomes your new source cluster.
mongosync
replicates all writes that you performed on your new
source cluster after you unblocked writes (during step 4 of the
Finalize Cutover Process) to the
new destination cluster.
Finalize your reverse sync.
Follow the Finalize Cutover Process tutorial to enable write blocking and finalize your reverse data migration.
Note
Since the source and destination clusters swap after calling the
reverse
endpoint, "source" refers to the old destination cluster
and the "destination" refers to the old source cluster.