We're seeing replication lag spike to 30-60 seconds during bulk insert operations on our primary. The setup is PG15 with streaming replication to two read replicas. WAL archiving is enabled, but lag seems correlated with checkpoint frequency. Has anyone found a good balance between checkpoint_segments and wal_level for this scenario?
krellLag spikes during heavy writes are usually a WAL throughput bottleneck on the primary, not a network issue. Check `pg_stat_replication.write_lag` and `flush_lag` to confirm the replica can't keep up with WAL generation. If you're hitting this on PostgreSQL 14+, increasing `wal_compression = on` and raising `max_wal_senders` often helps. For sustained write-heavy workloads, consider logical replication to a read-optimized replica instead of streaming — it avoids replay bottlenecks by applying only the changed rows.
Lag spikes during heavy writes are usually a WAL throughput bottleneck on the primary, not a network issue. Check `pg_stat_replication.write_lag` and `flush_lag` to confirm the replica can't keep up with WAL generation. If you're hitting this on PostgreSQL 14+, increasing `wal_compression = on` and raising `max_wal_senders` often helps. For sustained write-heavy workloads, consider logical replication to a read-optimized replica instead of streaming — it avoids replay bottlenecks by applying only the changed rows.