How To Install PostgreSQL
Although CentOS contains postgres packages within its default repositories, we will use the packages available from the postgres project itself.
This will ensure that we have access to the latest version of the software without having to install from source.
Exclude CentOS Postgres Packages
Before we install postgres, we must exclude the CentOS version of postgres in order to get the most recent version from the project’s website.
Open the CentOS repository configuration file in a text editor:
At the bottom of the
[base] section, add a line that excludes the postgres packages:
[base] name=CentOS-$releasever - Base mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=os #baseurl=http://mirror.centos.org/centos/$releasever/os/$basearch/ gpgcheck=1 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-6 exclude=postgresql*
Add the same line to the bottom of the
[updates] section to prevent yum from updating postgres from the default repositories:
[updates] name=CentOS-$releasever - Updates mirrorlist=http://mirrorlist.centos.org/?release=$releasever&arch=$basearch&repo=updates #baseurl=http://mirror.centos.org/centos/$releasever/updates/$basearch/ gpgcheck=1 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-6 exclude=postgresql*
Add Postgres Repositories
Now, we’ve told yum to not bother with the default repositories for postgres package interactions.
We need to provide it with an alternative now. We will use the packages provided on the postgres project’s website.
Back in your droplet session, change to your home directory. Type
curl -O and then paste the download link:
cd ~ curl -O http://yum.postgresql.org/9.3/redhat/rhel-6-x86_64/pgdg-centos93-9.3-1.noarch.rpm
Install the package you just downloaded by typing the following command:
rpm -ivh pgdg*
This will allow your system to pull the postgres package from the project’s website.
We can do that now to get the latest version. These packages contain version numbering, so you’ll need to search to see what version is being use:
yum list postgres*
Loaded plugins: fastestmirror Loading mirror speeds from cached hostfile * base: mirror.linux.duke.edu * extras: mirror.linux.duke.edu * updates: mirrors.kernel.org Installed Packages postgresql93.x86_64 9.3.1-1PGDG.rhel6 @pgdg93 postgresql93-libs.x86_64 9.3.1-1PGDG.rhel6 @pgdg93 postgresql93-server.x86_64 9.3.1-1PGDG.rhel6 @pgdg93 Available Packages postgresql93-contrib.x86_64 9.3.1-1PGDG.rhel6 pgdg93 postgresql93-debuginfo.x86_64 9.3.1-1PGDG.rhel6 pgdg93
We can see that in this case, all of the packages have “93” for version 9.3. We want to download the “-server” packages:
yum install postgresql93-server
Change the “93” to whatever the current version of postgres is for your system throughout the rest of this guide.
Initialize the Database Environment
You can initialize the database environment by typing:
service postgresql-9.3 initdb
We will then configure it to start at boot up and we will start the software:
chkconfig postgresql-9.3 on service postgresql-9.3 start
We are now ready to start using postgres.
How to Log Into the Postgres Database
By default, postgres creates a user and a database called
The default security scheme assumes that the postgres user will only be accessible to those who need access.
We interact with the postgres database software through an interface called
psql. It provides a prompt where we can manipulate and query for data.
For now, log into the postgres user like this:
su - postgres
Your prompt will change. You can now connect to the database that matches your username by typing:
Your prompt will change again, this time to indicate that you are interacting with the postgres database.
If you need to exit this interface, you can type this at any time:
Afterwards, to get back to the root user shell, you can type:
Some Helpful psql Commands
Here are a few commands that can help you get an idea of your current environment:
- \?: Get a full list of psql commands, including those not listed here.
- \h: Get help on SQL commands. You can follow this with a specific command to get help with the syntax.
- \q: Quit the psql program and exit to the Linux prompt.
- \d: List available tables, views, and sequences in current database.
- \du: List available roles
- \dp: List access privileges
- \dt: List tables
- \l: List databases
- \c: Connect to a different database. Follow this by the database name.
- \password: Change the password for the username that follows.
- \conninfo: Get information about the current database and connection.
Create a Postgres Table
We can create tables in postgres by using the following syntax:
CREATE TABLE new_table_name ( table_column_title TYPE_OF_DATA column_constraints, next_column_title TYPE_OF_DATA column_constraints, table_constraint table_constraint ) INHERITS existing_table_to_inherit_from;
We will make a table called “popsicles” to store our popsicle varieties and some information about them.
CREATE TABLE popsicles ( pop_id serial PRIMARY KEY, flavor varchar (50) NOT NULL, amount int NOT NULL, size varchar (10) check (size in ('small', 'normal', 'large')) );
Now we can use the
\dt command to see the table:
List of relations Schema | Name | Type | Owner --------+-----------+-------+---------- public | popsicles | table | postgres (1 row)
To see the columns and data structure we just defined, we can type this command:
Table "public.popsicles" Column | Type | Modifiers --------+-----------------------+------------------------------------ ------------------------ pop_id | integer | not null default nextval('popsicles _pop_id_seq'::regclass) flavor | character varying(50) | not null amount | integer | not null size | character varying(10) | . . .
Insert Data Into a Postgres Table
We now have a table in our database, but there’s no data in it yet. We can populate it by using the
INSERT INTO followed by the table name. Then, we type a comma-separated list of column names inside a set of parentheses. We then type
VALUES, followed by a second set of parentheses containing a comma-separated list of values that correspond to the column names.
Let’s try it now. We will insert some grape popsicles into our table:
INSERT INTO popsicles (flavor, amount, size) VALUES ('grape', 10, 'normal');
INSERT 0 1
INSERT 0 1 indicates that our command was accepted successfully. Some things to notice about this command is that the words in our value list are inside of single quotes (‘), and that the command ends with a semicolon (;). Both of these points are important to remember.
Another thing to notice is that we didn’t set the
pop_id column. That is because we defined it as a primary key. This column must have unique values for each entry, so postgres will automatically assign the next available value to each record we create.
Let’s fill in some more data so that we have a more useful table:
INSERT INTO popsicles (flavor, amount, size) VALUES ('orange', 8, 'small'); INSERT INTO popsicles (flavor, amount, size) VALUES ('fudge', 20, 'large'); INSERT INTO popsicles (flavor, amount, size) VALUES ('eclair', 14, 'normal'); INSERT INTO popsicles (flavor, amount, size) VALUES ('rainbow', 4, 'small');
If you recall, when we created the table, we defined the acceptable values for the “size” column. Postgres checks that the value is either “small”, “normal”, or “large”. What happens if we try to provide a different value?:
INSERT INTO popsicles (flavor, amount, size) VALUES ('lime', 6, 'huge');
ERROR: new row for relation "popsicles" violates check constraint "popsicles_size_check" DETAIL: Failing row contains (6, lime, 6, huge).
As you can see, postgres has validated that we have entered the correct data. It has rejected our newest popsicle because it didn’t have one of the values that we defined for size.
Select Data From a Postgres Table
We can query our new table for data by using the
select command. The select command returns data from a table that matches the criteria specified.
To return all of the info from our table, we can use the asterisk (*), which is a special symbol that means “match everything”:
SELECT * FROM popsicles;
We can select by column by replacing asterisk with the columns we want:
SELECT flavor, amount FROM popsicles;
If we want to filter to only show certain results, we can add a
where filter on the end:
SELECT * FROM popsicles WHERE amount <= 10;