pgEdge Platform
Installing the pgEdge Platform
Manually Configuring a Cluster

Manually Configuring a Cluster

This tutorial will walk you through the process of manually configuring each node in a replication scenario. In our example, we'll walk you through creating a two-node multi-master pgEdge cluster, and then use pgbench to create some tables and perform some read/write activity on the cluster.

Installing the pgEdge Platform

Before starting this tutorial, you should prepare two (or more) Linux servers running EL9 or Ubuntu 22.04. AWS users can find the Rocky Linux ami here (opens in a new tab). On each machine, you should:

Then, install the pgEdge Platform on each node with the command:

python3 -c "$(curl -fsSL https://pgedge-download.s3.amazonaws.com/REPO/install.py)"

Paste the command into your command line client and press Return.

Configuring a Cluster

After installing the pgEdge Platform, you can use the pgedge CLI to configure a replication cluster. The first command we use will be the setup command. In addition to the clauses we'll use for this tutorial, the setup command allows you to include optional clauses to specify details such as a port number for the cluster (--port=port_number), PostgreSQL version (--pg_ver=version), Spock version (--spock_ver=version), and autostart preferences (--autostart=True|False). The setup command installs:

  • the pgedge Command Line Interface (CLI).
  • PostgreSQL (Postgres).
  • spock (opens in a new tab), an extension that provides logical, asynchronous, multi-master replication.
  • snowflake, an extension that provides robust sequences for distributed clusters.
  • other extensions that support database management on a distributed cluster.

On each server, move into the pgedge directory, and invoke the setup command:

./pgedge setup -U db_superuser_name -P db_superuser_password -d db_name

For example, the following command (invoked by an OS user named rocky) installs the pgEdge CLI and creates a PostgreSQL database named acctg, with a PostgreSQL database superuser named admin, whose password is 1safe_password:

./pgedge setup -U admin -P 1safe_password -d acctg

The setup command creates a PostgreSQL database superuser with the name and password you provide with the command. The database user cannot be the name of an OS superuser, pgedge, or any of the PostgreSQL reserved words (opens in a new tab). In the examples that follow, that user is named admin.

The setup command also creates a PostgreSQL replication role (opens in a new tab) with the same name as the OS user that invokes the setup command, and adds the username and the scrambled password to the ./pgpass file. In our example, that user is named rocky (the name of the OS user invoking the command). In the steps that follow, this replication user is provided in the connection strings used between nodes for replication.

Creating Nodes

After installing pgEdge Platform and running setup, we'll use spock node-create to create a replication node on each host. A replication node is a named collection of databases, tables, and other artifacts that are replicated via a pgEdge subscription. The syntax is:

./pgedge spock node-create node_name 'host=IP_address_of_n1 user=replication_user_name dbname=db_name' db_name

When invoking pgedge spock node-create, provide three arguments:

  • a name for the node (in our examples, n1 and n2).
  • a single-quoted connection string that specifies the node's IP address (for example, 10.0.0.5|10.0.0.6), the name of the PostgreSQL replication user (rocky), and the database name (acctg).
  • the last argument repeats the database name (acctg).

For example, to create node n1:

./pgedge spock node-create n1 'host=10.0.0.5 user=rocky dbname=acctg' acctg

To create node n2:

./pgedge spock node-create n2 'host=10.0.0.6 user=rocky dbname=acctg' acctg

Note: If you use the node naming convention: n1, n2, etc., the postgresql.conf file will be automatically updated to enable Snowflake sequences on your cluster.

Creating Subscriptions

Next, create the subscriptions that connect the nodes to each other. Since this is a multi-master replication system, each node acts as both a subscriber and a publisher node. Use the command:

./pgedge spock sub-create subscription_name 'host=IP_address_of_publisher port=port_number user=replication_user_name dbname=db_name' db_name

When invoking pgedge spock sub-create, provide three arguments:

  • a unique subscription name for each node (in the examples that follow, sub_n1n2 and sub_n2n1).
  • a single-quoted connection string that specifies the IP address of the node you're subscribing to (in our examples, 10.0.0.5 and 10.0.0.6), the PostgreSQL port on the publisher node (5432), the name of the PostgreSQL replication user (rocky), and the database name (acctg).
  • the last argument repeats the database name (acctg).

For example, if you have a two node cluster, you'll create two subscriptions; one that subscribes node 1 to node 2 and one that subscribes node 2 to node 1:

On node n1:

./pgedge spock sub-create sub_n1n2 'host=10.0.0.6 port=5432 user=rocky dbname=acctg' acctg

On node n2:

./pgedge spock sub-create sub_n2n1 'host=10.0.0.5 port=5432 user=rocky dbname=acctg' acctg

Adding Tables to the Replication Set

For replication to begin, you will need to add tables to the replication set; for this example, we'll use pgbench to add some tables. When you open pgbench or psql, specify your database name after the utility name.

On each node, source the PostgreSQL environment variables to add pgbench and psql to your OS PATH; this will make it easier to move between the nodes:

source pgedge/pg16/pg16.env

Then, use pgbench to set up a very simple four-table database. At the OS command line, (on each node of your replication set), create the pgbench tables in your database (db_name) with the pgbench command. You must create the tables on each node in your replication cluster:

pgbench -i db_name

Then, connect to each node with the psql client:

psql db_name

Once connected, alter the numeric columns, setting LOG_OLD_VALUE equal to true. This will make these numeric fields conflict-free delta-apply columns, ensuring that the value replicated is the delta of the committed changes (the old value plus or minus any new value) to a given record:

ALTER TABLE pgbench_accounts ALTER COLUMN abalance SET (LOG_OLD_VALUE=true);
ALTER TABLE pgbench_branches ALTER COLUMN bbalance SET (LOG_OLD_VALUE=true);
ALTER TABLE pgbench_tellers ALTER COLUMN tbalance SET (LOG_OLD_VALUE=true);

Then, exit psql:

db_name=# exit

On the OS command line for each node, use the pgedge spock repset-add-table command to add the tables to the system-created replication set (named default); the command is followed by your database name (db_name):

./pgedge spock repset-add-table default 'pgbench_*' db_name

The fourth table, pgbench_history, is excluded from the replication set because it does not have a primary key.

Checking the Configuration

On the psql command line, check the configuration on each node with the following SQL statements:

db_name=# SELECT * FROM spock.node;
node_id | node_name
---------+----------
673694252 | n1
560818415 | n2
(2 rows)

and:

db_name=# SELECT sub_id, sub_name, sub_slot_name, sub_replication_sets  FROM spock.subscription;
   sub_id   | sub_name |	sub_slot_name 	|                	sub_replication_sets             
------------+----------+----------------------+--------------------------------------------------------
 3293941396 | sub_n1n2 | spk_db_name_n2_sub_n1n2 | {default,default_insert_only,ddl_sql}
(1 row)

The sub_replication_sets column shown above displays the system-created replication sets. You can add custom replication sets with the spock repset-create and spock sub-add-repset commands.

Testing Replication

Now, if you update a row on n1, you should see the update to the same row on n2.

On n1:

db_name=# SELECT * FROM pgbench_tellers WHERE tid = 1;
 tid | bid | tbalance | filler
-----+-----+----------+--------
   1 |   1 |    	0 |
 (1 row)
db_name=# UPDATE pgbench_tellers SET filler = 'test' WHERE tid = 1;
UPDATE 1

Check n2:

db_name=# SELECT * FROM pgbench_tellers WHERE tid = 1;
 tid | bid | tbalance | filler  	 
-----+-----+----------+--------------------------------------------------
   1 |   1 |    	0 | test                               
(1 row)

You can also use pgbench to exercise replication; exit psql:

db_name=# exit

Then, run the following command on both nodes at the same time to run pgbench for one minute.

pgbench -R 100 -T 60 -n db_name

When you connect with psql and check the results on both nodes, you'll see that the sum of the tbalance columns match on both pgbench_tellers tables. Without the conflict-free delta-apply columns, each conflict would have resulted in accepting the first in, potentially leading to sums that do not match between nodes.

n1:

db_name=# SELECT SUM(tbalance) FROM pgbench_tellers;
  sum  |
 ------+
 -84803
(1 row)

n2:

db_name=# SELECT SUM(tbalance) FROM pgbench_tellers;
  sum  |
 ------+
 -84803
(1 row)
Installing the pgEdge PlatformUsing a json File to Deploy a Remote Cluster