Lag Tracking
The Spock extension's lag tracking feature shows how far behind a subscriber is when compared to a provider node in terms of write-ahead log (WAL) replication. You can use this feature to understand replication delays and monitor streaming health.
Tracking is implemented via a SQL view named spock.lag_tracker, which exposes key metrics for replication lag on a per-node basis. For example:
-[ RECORD 1 ]---------+------------------------------
origin_name | node1
receiver_name | node2
commit_timestamp | 2025-06-30 09:27:57.317779+00
last_received_lsn | 0/15A2780
remote_insert_lsn | 0/15A2780
replication_lag_bytes | 0
replication_lag | 00:00:04.014177spock.lag tracker displays the following information:
| Column Name | Description |
|---|---|
| origin_name | Name of the provider node. |
| receiver_name | Name of the subscriber node. |
| commit_timestamp | Commit time of the last transaction received from the provider. |
| last_received_lsn | Last acknowledged log sequence number (LSN) received from the provider. |
| remote_insert_lsn | WAL insert position on the provider when the commit was sent. |
| replication_lag_bytes | The amount of data the subscriber is behind. |
| replication_lag | Time delay between when the transaction was committed on the provider and when processed locally. |