0

I want to set up streaming replication on the Slave.

On the Primary Server (Server A. Ubuntu 18.04; Postgres 12) :

  1. sudo nano /etc/postgresql/12/main/postgresql.conf
  2. Set listen_addresses = '*'
  3. The postgresql.conf settings are :

wal_level = "logical"

archive_mode = "ON"

archive_timeout = 3600

checkpoint_timeout = 600

hot_standby = "ON"

max_connections = 700

max_wal_sender = 5

max_wal_size = 16GB

min_wal_size = 2GB

max_worker_processes = 30

synchronous_commit = "OFF"

wal_log_hints = "OFF"

wal_recycle = "ON"

wal_keep_segments = 4000

wal_sync_method = "fdatasync"

This Primary Server has already hosted logical replication to another server (Server B. Ubuntu 18.04; Postgres 12)

On the Slave Server (Server C. Ubuntu 22.04; Postgres 12) :

  • change the data directory
  • set this up as Slave for streaming replication from Server A.

Here is the detail : After installing postgres (fresh install) I make sure it's running, then I change the data directory :

  1. sudo -u postgres psql
  2. SHOW data_directory;. --Output /var/lib/postgresql/12/main
  3. Quit psql : \q
  4. sudo systemctl stop postgresql
  5. sudo systemctl status postgresql
  6. sudo rsync -av /var/lib/postgresql /data/postgresql2
  7. rename the folder : sudo mv /var/lib/postgresql/12/main /var/lib/postgresql/12/main.bak
  8. Edit the postgresql.conf : sudo nano /etc/postgresql/12/main/postgresql.conf
  9. Set to new location : data_directory = '/data/postgresql2/postgresql/12/main'
  10. Start the db again : sudo systemctl start postgresql

The result is good. Postgres is running well on Server C. When I do SHOW data_directory; it gives out : /data/postgresql2/postgresql/12/main

Then I setup streaming replication on Server C :

  1. stop the service on the slave : systemctl stop [email protected]

  2. su - postgres then cp -R /data/postgresql2/postgresql/12/main /data/postgresql2/postgresql/12/data_original

  3. Remove the old directory :

    rm -rf /data/postgresql2/postgresql/12/main/*

  4. Then do base backup :

    pg_basebackup -h 10.100.9.40 -p 5432 -U rep_user -D /data/postgresql2/postgresql/12/main -Fp -Xs -R -P -v

  5. Restart : systemctl start [email protected]

--gives out error :

*`Job for [email protected] failed because the service did not take the steps required by its unit configuration.
See "systemctl status [email protected]" and "journalctl -xeu [email protected]" for details`*
  1. Then I start using another way : sudo -i -u postgres /usr/lib/postgresql/12/bin/pg_ctl start -D /data/postgresql2/postgresql/12/main

It gives out error :

waiting for server to start....postgres: could not access the server configuration file "/data/postgresql2/postgresql/12/main/postgresql.conf": No such file or directory

Please help. What might cause this?

0

1 Answer 1

3

You forgot to set up replication on the standby. At the very minimum, you have to set restore_command in postgresql.conf and create standby.signal.

The whole dance with renaming directories and running rsync is unnecessary, all that is needed is pg_basebackup.

1
  • ok.thanks @LaurenzAlbe . Yes, agreed, the renaming directories and rsync are not part of the replication process. I just have to do it because I have to move the directory due to the tiny size of the original drive.
    – padjee
    Commented Nov 2, 2022 at 11:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.