Surviving a Patroni failover with logical replication (opens in new tab)

Your Patroni cluster does exactly what you built it to do. A node dies at 3 a.m., a new primary is elected in a few seconds, the application reconnects, and nobody gets paged. Your HA setup works. And yet, somewhere downstream, your change-data-capture pipeline has just gone quiet. Debezium or Flink CDC was streaming every […] L’article est apparu en premier sur .