Page History

...

There is no guarantee that this update updates with already assigned counter will be completed.

For , for example, if during the dht DHT prepare transaction phase some mapped nodes will exit , the transaction may will be rolled back.
When a primary node fails under load situations may arise where may be permanent gaps in the updates on backup nodes in the event of transaction reordering.

...

In this case, the gaps will be closed by special king of message PartitionUpdateCountersRequest on tx rollback, and similarly closed in WAL using RollbackRecord if persistence is enabled.

TODO fig

Switching Importance of atomic primaries switching

If the primary node fell out, fails then one of the backups becomes the new node, and due to the PME protocol atomic switching to the new primary takes place, where hwm becomes equal to the last known lwm for partition (see Figure 2).
It is important to ensure the atomicity of switching the node’s primaries for partitioning in node entry or exit scenarios, otherwise the HWM >= LWM invariant will break, which ensures that if at least one stable copy is available, everyone else can restore its integrity from it. This is achieved due to the fact that transactions on the previous topology are completed before switching (as a phase of the PME protocol) to a new one and the situation when two primaries of a node simultaneously exist in a grid for one partition is impossible.

...