MV_REPLICATION_MANAGEMENT_STATE

This view shows the aggregated state of replication management.

This view is also available as an LMV (local view per node) view.

Column Name

Description

NODE_ID

The node ID of the node that originated this replication management row.

ID

For internal use, it is always 1.

CURRENT_CLUSTER_LSN

The log sequence number of the cluster database for which replication management is currently executing.

SUCCESSFUL_CLUSTER_LSN

The log sequence number of the cluster database for which replication management has most recently succeeded.

SELF_NODE_ID

This node's ID. Should always correspond to NODE_ID.

SUCCESS_STATE

Possible values include: Success, Soft Failure, Hard Failure, Cluster database is missing, Cluster database is out of date, and Node has no self-id.

Success: the most recent iteration of the replication management thread ended successfully, with local state matching metadata state.

Soft Failure: the local state doesn’t match metadata state, but there is no mismatch that impacts the correct functioning of the cluster (for example, if all that’s wrong is a disconnected async replica, then the cluster is technically healthy, although it is replication management’s responsibility to reconnect it, so we’d be in this state), and thus it is not blocking the advancement of SUCCESSFUL_CLUSTER_LSN.

Hard Failure: the local state does not match metadata state in a way that impacts the correct functioning of the cluster (for example, if a sync replica is disconnected), and will block the advancement of SUCCESSFUL_CLUSTER_LSN.

Cluster database is missing: has the same consequences as Failure, but is useful for tracing.Cluster database is out of date: similar to Cluster database is missing, but just has the detail that the cluster database just exists, but is out of date (due to a divergence reprovisioning event / the database went unrecoverable and recovered).

Cluster database is out of date: similar to Cluster database is missing, but just has the detail that the cluster database just exists, but is out of date (due to a divergence reprovisioning event / the database went unrecoverable and recovered).

Node has no self-id: the same consequences as Failure, means the node has no ID, and is useful for tracing.

LAST_REPLAY_LSN

Stores the last replay log sequence number (LSN) that was processed from the cluster database. This is used to determine when the cluster database is out of date, due to a reprovisioning. The LSN is potentially different from CURRENT_CLUSTER_LSN since that LSN is meant to be a lower bound for the LSN of the cluster database that holds the state replication management acted on - so _SYNC_PARTITIONS <LSN> can work correctly, while the LAST_REPLAY_LSN is an upper bound to guarantee that we never act on out of date metadata. The term is used for dealing with divergence scenarios.

LAST_REPLAY_TERM

Stores the last term that was processed from the cluster database. This is used to determine when the cluster database is out of date, due to a reprovisioning.

CLUSTER_GLOBAL_TERM

The most recently read global term attribute from the cluster.

OOM

Whether the most recent iteration of replication management failed with an out of memory error.

ITERATION_COUNTER

Counts iterations of the replication management thread.

Last modified: September 14, 2022

Was this article helpful?