Questions[postgresql-replication](server)

For High Availability on DB was build Active-Standby(replica) cluster on Postgresql-13. Inserting data correctly transfer to 'replica' server, it is ok. Now, I want check case, how i can make change roles on two servers. for this I the /var/lib/postgresql/13/main/standby.signal was delete, also for apply changes restarted postgrsql.service Something tells me what I'm doing wrong. So,

1. how to make a master out of a replica,
2. how then from the old master (which was temporarily not available) to make the master back, without losing data?

Sorry for my English.

I'm testing repmgr tool and I'm a little bit confused. Everything's fine until standby register step.

What should I have configured on standby server? From docs I assumed that standby should't have the cluster, but it didn't configure a new one. When the postgres cluster exists repmgr have a problem with existing data. When I'll stop the cluster service and delete the data manually and run the migration from repmgr I have an error after starting the cluster:

postgres@ip-10-0-34-147:~$ /usr/lib/postgresql/13/bin/pg_ctl -D /var/lib/postgresql/13/main start
waiting for server to start....postgres: could not access the server configuration file "/var/lib/postgresql/13/main/postgresql.conf": No such file or directory
 stopped waiting
pg_ctl: could not start server
Examine the log output.

It's running on Ubuntu, but maybe more applicable will be Debian. But I still have no idea how to make it right.

I have spent quite a bit of time to install postgresql 11 on the high side production, but I just can't quite get it installation and configured correctly after I yum install

rpm -i postgresl11-server-11.1-1PGDG.rhel7.x86_64
rpm -i postgresl11-libs-11.1-1PGDG.rhel7.x86_64 rpm -i rpm -i postgresl11-contrib-11.1-1PGDG.rhel7.x86_64

cd /usr/pgsql-11/bin/postgresql-11-setup initdb

the initializing database ...OK according to the log file that all are ok and success and "You can now start the database server using /usr/pgsql-11/bin/pg_ctl -D /var/lib/pgsql/11/data -l logfile start"

I can't run psql from postgres and from what I could see, the postgres directories seemed to be correctly excepted those missing features below under /usr/psql11/bin when I compared with the development server: clusterdb, created, createuser, dropdb, dropuser, ecpg, pg_basebackup, pg_archivecleanup, pgbench, pg_config, pg_dump, pg_dumpall, pg_isready, pg_receivewal, pg_restore, pg_rewin, pg_test_fsync, pg_test_timing, pg_upgrade, pg_waldump, pgsql, reindexdb and vacuumdb.

Did I miss a step? Also, when I run ps -ef|grep postgres, the server running locally.

How do I get it install and run correctly? I haven't install on the slave server yet.

Thanks for all your help.

Bach-Nga

I've created a postgresql replica database to mirror the content of my production database and want to be sure that queries against the replica can never impact the production server. How safe am I?

Thanks!

I wonder if you can help. I need to replicate a Postgres 8.4 database (master-slave style) from an on-premise server to an RDS instance on AWS running Postgres 9.x. I've read that Postgres > 9.0 has a nice replication mechanism, but before that tools like Slony ruled. The problem I'm seeing with Slony is that it seems to need a process to be running on the target database which is a problem for me because RDS is fully managed.

Can anybody suggest a path to make this happen? I'd like to avoid downtime of the source 8.4 database whilst taking a dump and then importing it on the other side.

Thanks in advance for you help!