toluaina/pgsync

Replication Slot "_tmp_" is active

virtualaidev opened this issue · 1 comments

PGSync version: 2.5.0

Postgres version: debezium/postgres:15

Elasticsearch version: docker.elastic.co/elasticsearch/elasticsearch:8.7.0

Redis version: 6.2.6

Python version: 3.7

Problem Description:
I'm getting replication slot "tmp" error, I'm running pgsync in AWS Fargate with autoscaling, so basically whenever an instance memory and cpu resources hit the max threshold around 70%, it will scale up by running another instance of pgsync.

Error Message (if any):

psycopg2.errors.ObjectInUse: replication slot "_tmp_" is active for PID 14744

We get same error messages, running on GCP.

Do you already find a way out?

When I kill the replication slot a new process spins up to use it again and I get same error when spinning up pgsync again.