@@ -25,28 +25,6 @@ Starting in 1.4.0, you can set the workload level that
25
25
destination clusters. For more information, see the
26
26
:option:`--loadLevel` option.
27
27
28
- Improved Resiliency for ``oplog`` Rollover
29
- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
30
-
31
- ``mongosync`` 1.4.0 improves :term:`oplog` handling. The ``oplog``
32
- handling improvement adds resiliency to long running operations.
33
-
34
- ``mongosync`` does an initial data copy and then applies updates from
35
- the ``oplog`` to sync the source and destination clusters. In earlier
36
- versions, ``mongosync`` waits to complete the initial sync before it
37
- begins to replicate any changes from the ``oplog``. If there is a large
38
- initial data set, the ``oplog`` can roll over before ``mongosync`` can
39
- complete the initial data transfer. When this happens, the sync has to
40
- be restarted from the beginning.
41
-
42
- Starting in 1.4.0, ``mongosync`` begins to apply changes while the
43
- initial sync is still in progress. Since ``mongosync`` starts to apply
44
- changes earlier, ``mongosync`` maintains a recent position in the
45
- ``oplog``. This mitigates the risk of ``oplog`` roll over and
46
- significantly lowers the risk of restarting the sync.
47
-
48
- For more information, see :ref:`oplog sizing <c2c-oplog-sizing>`.
49
-
50
28
Faster Processing
51
29
~~~~~~~~~~~~~~~~~
52
30
0 commit comments