Everything You Wanted to Know About Databases (Keith).pdf

Everything You Wanted to Know about Databases as a Developer
but Were Too Afraid to Ask Your DBA
Keith Fiske | Senior Database Engineer @ Crunchy Data
All Things Open | October 2024

whoami?
Keith Fiske | @keithf4
● Senior Database Engineer
● Principle Engineer for
○ pg_partman
○ pgMonitor
○ Other third-party PG Projects at
Crunchy
● Ansible Automation w/ PostgreSQL and
Patroni
● Bird Enthusiast

Introduction
● Databases are essential to most applications
● Most developers aren’t trained in database administration
● DBAs are grumpy busy people

• Database Architecture
• Users and Roles
• Database Objects
• Database Connections
• Database Operations and Transactions
• WAL
• Documentation
Agenda

https://www.postgresql.org/docs/current/tutorial-arch.html
Database Architecture

How do I Install Postgres?
• Install from source code
https://www.postgresql.org/docs/current/install-procedure.html
• Install package for given platform
https://www.postgresql.org/download/
• Choose a managed service
https://crunchybridge.com/register
• Try out the Postgres Playground (Postgres in your browser using WASM):
https://www.crunchydata.com/developers/tutorials

# 1. Configure the PostgreSQL repository
user@my_vm$ sudo sh -c 'echo "deb http://apt.postgresql.org/pub/repos/apt $(lsb_release -cs)-pgdg main"
> /etc/apt/sources.list.d/pgdg.list’
# 2. Import the repository signing key
user@my_vm$ wget --quiet -O - https://www.postgresql.org/media/keys/ACCC4CF8.asc | sudo apt-key add –
# 3. Update the package lists
user@my_vm$ sudo apt-get update
# 3. Install PostgreSQL
user@my_vm$ sudo apt-get -y install postgresql
Ubuntu 20.04 (Focal)

# 1. Enable the pgdg repository
user@my_vm$ sudo dnf install -y
https://download.postgresql.org/pub/repos/yum/reporpms/EL-9-x86_64/pgdg-redhat-repo-latest.noarch.rpm
# 2. Disable the built-in PostgreSQL module
user@my_vm$ sudo dnf -qy module disable postgresql
# 3. Install PostgreSQL
user@my_vm$ sudo dnf install -y postgresql14-server
Alma9 (RHEL)

What is a PostgreSQL Cluster?
https://www.postgresql.org/docs/current/creating-cluster.html data directory
Collection of DBs
managed by single instance

How do I Create a PostgreSQL Cluster?

How do I create a PostgreSQL Cluster?
postgres@my_vm$

postgres@my_vm$ export PATH=$PATH:/usr/pgsql-14/bin
postgres@my_vm$

postgres@my_vm$ export PGDATA=/var/lib/pgsql/14/data
postgres@my_vm$

https://www.postgresql.org/docs/current/app-initdb.html
postgres@my_vm$ initdb

https://www.postgresql.org/docs/current/app-initdb.html
postgres@my_vm$ initdb
...
Success. You can now start the database server using:
/usr/pgsql-14/bin/pg_ctl -D /var/lib/pgsql/14/data -l logfile start

How do I Start (or Stop) PostgreSQL?

How do I Start Postgres?
postgres@my_vm$

https://www.postgresql.org/docs/current/app-pg-ctl.html
postgres@my_vm$ pg_ctl -l logfile start

postgres@my_vm$ pg_ctl -l logfile start
waiting for server to start.... done
server started

postgres@my_vm$ ps -ef|grep postgres
postgres 4236 1 0 11:59 ? 00:00:00 /usr/pgsql-14/bin/postgres
postgres 4237 4236 0 11:59 ? 00:00:00 postgres: logger
postgres 4239 4236 0 11:59 ? 00:00:00 postgres: checkpointer
postgres 4240 4236 0 11:59 ? 00:00:00 postgres: background writer
postgres 4241 4236 0 11:59 ? 00:00:00 postgres: walwriter
postgres 4242 4236 0 11:59 ? 00:00:00 postgres: autovacuum launcher
postgres 4243 4236 0 11:59 ? 00:00:00 postgres: stats collector
postgres 4244 4236 0 11:59 ? 00:00:00 postgres: logical replication launcher
List the Postgres processes

How do I Stop Postgres?
postgres@my_vm$ pg_ctl stop

How do I Stop Postgres?
postgres@my_vm$ pg_ctl stop
waiting for server to shut down.... done
server stopped

How do I Control Postgres using systemd?
root@my_vm$ systemctl start|stop|restart postgresql-14

root@my_vm$ systemctl status postgresql-14

root@my_vm$ systemctl status postgresql-14
● postgresql-14.service - PostgreSQL 14 database server
Loaded: loaded (/usr/lib/systemd/system/postgresql-14.service; disabled; vendor preset: disabled)
Active: active (running) since ...

What is a Database?
https://www.postgresql.org/docs/current/managing-databases.html
● Topmost hierarchical level for organizing database objects
● Databases are isolated from users’ point of view
● Databases are closely bound from an administrative perspective
● Pre-defined databases: postgres, template0, template1
● User-defined databases (optional)

What is a Database?

What is a Database?
postgres template0 template1
Pre-defined databases

Aside: What is psql?
https://www.postgresql.org/docs/current/app-psql.html
● Command line tool
● Execute commands and/or scripts against the database
• SQL
• psql commands: e.g. c to connect to a database d to view table details
● Try it out in your browser at the Crunchy Data Postgres Playground:
https://www.crunchydata.com/developers/tutorials

How do I Create a Database?
https://www.postgresql.org/docs/current/sql-createdatabase.html
postgres@my_vm$

postgres@my_vm$ psql

psql (14.0)
Type "help" for help.
postgres=#

psql (14.0)
postgres=# CREATE DATABASE my_database;

psql (14.0)
postgres=# CREATE DATABASE my_database;
CREATE DATABASE

How do I List my Databases?
postgres=# l

postgres=# SELECT datname FROM pg_database;
https://www.postgresql.org/docs/current/catalogs.html

postgres=# SELECT datname FROM pg_database;
datname
-----------
my_database
postgres
template0
template1
(4 rows)

What is a Tablespace?
And why should I create one?
● Physical location of the database objects
● Default tablespace pg_default
● Allows control of the disk layout of a PostgreSQL installation
● Accessible to the entire cluster

How do I Create a Tablespace?
https://www.postgresql.org/docs/current/sql-createtablespace.html
postgres@my_vm$

postgres@my_vm$ mkdir -p /my_tablespaces/tbsp_1
postgres@my_vm$

postgres@my_vm$ psql -c "CREATE TABLESPACE tablespace_1 location '/my_tablespaces/tbsp_1'"

postgres@my_vm$ psql -c "CREATE TABLESPACE tablespace_1 location '/my_tablespaces/tbsp_1'"
CREATE TABLESPACE

How do I List my Tablespaces?
postgres=# db

postgres=# db+

postgres@my_vm$ psql -c "select spcname from pg_tablespace"

postgres@my_vm$ psql -c "select spcname from pg_tablespace"
spcname
-----------------
pg_default
pg_global
tablespace_1
(3 rows)

What is a Schema?
Collection of objects within the database
● Logical grouping - no impact on physical location of objects
● Schema and owner of objects need not be the same
● Specific to the database in which it is created
● Namespace

How do I Create a Schema?
https://www.postgresql.org/docs/current/sql-createschema.html
postgres@my_vm$

postgres@my_vm$ psql my_database -c "CREATE SCHEMA my_schema"

postgres@my_vm$ psql my_database -c "CREATE SCHEMA my_schema"
CREATE SCHEMA

What’s the Difference
Between a User and a Role?

What’s the Difference Between a User and a Role?
● CREATE USER and CREATE ROLE are synonyms except:
• CREATE USER: LOGIN by default
• CREATE ROLE: NOLOGIN by default
● A role can be considered a user or a group (or both)
● A role is available to the entire cluster
● A role can be granted to another role
● A role can own database objects

How do I Create a Role?
https://www.postgresql.org/docs/current/sql-createrole.html
postgres@my_vm$

postgres@my_vm$ psql -c "CREATE ROLE my_role"

CREATE ROLE
postgres@my_vm$

CREATE ROLE
postgres@my_vm$ psql -d my_database -U my_role

CREATE ROLE
postgres@my_vm$ psql -d my_database -U my_role
psql: error: connection to server on socket "/var/run/postgresql/.s.PGSQL.5432" failed:
FATAL: role "my_role" is not permitted to log in

How do I Create a User?
https://www.postgresql.org/docs/current/sql-createuser.html
postgres@my_vm$

How do I Create a User?
postgres@my_vm$ psql -c "CREATE USER my_user"

How do I Create a User Role?
CREATE ROLE

CREATE ROLE
postgres@my_vm$ psql -d my_database -U my_user

CREATE ROLE
psql (14.0)

What is a Privilege?
Permission to perform certain action(s) on given object(s)

What is a Privilege?
Permission to perform certain action(s) on given object(s)
● Granted by the owner of the object or by a superuser
● Objects : DATABASE, FUNCTION, SCHEMA, TABLE etc
● Permissions: SELECT, INSERT, UPDATE, DELETE, TRUNCATE, CREATE etc
● ALTER DEFAULT PRIVILEGES

Granting Privileges
https://www.postgresql.org/docs/current/ddl-priv.html
postgres@my_vm$ psql -d my_database
my_database=# GRANT CREATE ON DATABASE my_database TO my_user;
GRANT
my_database=# GRANT CREATE ON SCHEMA my_schema TO my_user;
GRANT
my_database=# GRANT CREATE ON TABLESPACE tablespace_1 TO my_user;
GRANT

Database Objects
• View
• Materialized View
• Sequence
• Table
• Index
• Constraint

What is a Table?
• A “relation”
• Data arranged in columns and rows
• Rows are not ordered

How do I Create a Table?
https://www.postgresql.org/docs/current/sql-createtable.html
my_database=>

my_database=> CREATE TABLE my_schema.dept (
my_database(>

my_database(> dept_id integer,
my_database(> dept_name varchar);
my_database->

my_database-> TABLESPACE tablespace_1;

CREATE TABLE

What is a Sequence?
Object that generates a sequence of integers
● Specific to a schema
● Used to generate unique numeric identifiers
● Implemented as a single row table

How do I Create a Sequence?
https://www.postgresql.org/docs/current/sql-createsequence.html
my_database=>

my_database=> CREATE SEQUENCE my_schema.s_dept
my_database->

my_database-> INCREMENT BY 1 MINVALUE 1 NO MAXVALUE;

CREATE SEQUENCE

CREATE SEQUENCE
my_database=> ALTER TABLE my_schema.dept
my_database->

CREATE SEQUENCE
my_database-> ALTER COLUMN dept_id SET DEFAULT nextval('my_schema.s_dept');

CREATE SEQUENCE
my_database-> ALTER COLUMN dept_id SET DEFAULT nextval('my_schema.s_dept');
ALTER TABLE

How do I Auto Generate an ID Column?

my_database=> CREATE TABLE my_schema.emp (
my_database(> emp_id integer,
my_database(> emp_name varchar,
my_database(> dept_id integer)

my_database(> emp_id integer generated always as identity,

my_database(> emp_id integer generated always as identity,
CREATE TABLE

How do I View the Table Definitions?

my_database=> d my_schema.dept

What is an Index?
Ordered list of entries containing a value and a pointer to the table row
● Can create on one or more columns or expressions
● Can speed up searches based on values of column(s) in the index
● Default type is btree
● Various types available in Postgres
https://www.postgresql.org/docs/current/indexes-types.html

How do I Create an Index?
https://www.postgresql.org/docs/current/sql-createindex.html
my_database=>

my_database=> CREATE INDEX emp_dept_id
my_database->

my_database-> ON my_schema.emp (dept_id)
my_database->

CREATE INDEX

What is a Constraint?
https://www.postgresql.org/docs/current/ddl-constraints.html
● NOT NULL constraint
● CHECK constraint
● PRIMARY KEY (PK) constraint
● FOREIGN KEY (FK) constraint

What is a NOT NULL Constraint?
● Column must contain a value

What is a NOT NULL Constraint?
● Column must contain a value
What is a CHECK Constraint?
● Value must conform to certain rules
● for example: dept_name must contain only letters and spaces

What is a Primary Key?
One or more columns that allow a row in a table to be identified uniquely

What is a Primary Key?
One or more columns that allow a row in a table to be identified uniquely
● Enforced through PK constraint + unique index
● A table may have only one PK constraint
● The columns in the PK must be NOT NULL

How do I Create a Primary Key Constraint?

How do I Create a PK Constraint?
https://www.postgresql.org/docs/current/sql-altertable.html
my_database=>

my_database=> ALTER TABLE my_schema.emp
my_database-> ADD CONSTRAINT emp_pk PRIMARY KEY (emp_id)
my_database-> USING INDEX TABLESPACE tablespace_1;

ALTER TABLE

ALTER TABLE
Table "my_schema.emp"
----------+-------------------+-----------+----------+------------------------------
emp_id | integer | | not null | generated always as identity
emp_name | character varying | | |
dept_id | integer | | |
Indexes:
"emp_pk" PRIMARY KEY, btree (emp_id), tablespace "tablespace_1"
"emp_dept_id" btree (dept_id), tablespace "tablespace_1"

my_database-> ADD CONSTRAINT dept_pk PRIMARY KEY (dept_id)
ALTER TABLE
Table "my_schema.dept"
-----------+-------------------+-----------+----------+------------------------------
dept_id | integer | | not null | generated always as identity
dept_name | character varying | | |
Indexes:
"dept_pk" PRIMARY KEY, btree (dept_id), tablespace "tablespace_1"

What is a Foreign Key?
A relationship between two tables

What is a Foreign Key?
A relationship between a “parent” and a “child” table
● A way to enforce “referential integrity”
● References the primary key or another unique key in the parent table

How do I Create a Foreign Key Constraint?

How do I Create a FK Constraint?
my_database=>

my_database-> ADD CONSTRAINT emp_dept_fk FOREIGN KEY (dept_id)
my_database-> REFERENCES my_schema.dept(dept_id);

ALTER TABLE

ALTER TABLE
...
Foreign-key constraints:
"emp_dept_fk" FOREIGN KEY (dept_id) REFERENCES my_schema.dept(dept_id)
...

ALTER TABLE
...
Foreign-key constraints:
"emp_dept_fk" FOREIGN KEY (dept_id) REFERENCES my_schema.dept(dept_id)
...
...
Referenced by:
TABLE "emp" CONSTRAINT "emp_dept_fk" FOREIGN KEY (dept_id) REFERENCES dept(dept_id)
...

Aside: What is a Search Path?
Schema(s) that will be searched if I don’t refer to an object using a fully-qualified object name
my_database=>

my_database=> SHOW search_path;

search_path
-----------------
"$user", public
my_database=>

search_path
-----------------
"$user", public
my_database=> select * from emp;

search_path
-----------------
"$user", public
ERROR: relation "emp" does not exist
LINE 1: select * from emp
^

my_database=> SET search_path = my_schema;

SET
my_database=>

SET

SET
search_path
-------------
my_schema
(1 row)
my_database=>

SET
search_path
-------------
my_schema
(1 row)

SET
search_path
-------------
my_schema
(1 row)
emp_id | emp_name | dept_id
--------+----------+---------
(0 rows)

I can set the search path (and other options) automatically
postgres@my_vm$ echo 'set search_path to my_schema' >> ~/.psqlrc
postgres@my_vm$

SET
psql (14.0)
my_database=>

SET
psql (14.0)

SET
psql (14.0)
search_path
-------------
my_schema
(1 row)

How do I Populate my Tables?
https://www.postgresql.org/docs/current/sql-insert.html
my_database=> INSERT INTO dept (dept_name)
my_database-> VALUES ('Sales'),('Consulting'),('Product'),('HR');
INSERT 0 4

my_database=> INSERT INTO dept (dept_name)
my_database-> VALUES ('Sales'),('Consulting'),('Product'),('HR');
INSERT 0 4
my_database=> SELECT * FROM dept ORDER BY dept_id;
dept_id | dept_name
---------+------------
1 | Sales
2 | Consulting
3 | Product
4 | HR
(4 rows)

my_database=> INSERT INTO emp (emp_name, dept_id)
my_database->VALUES ('Ay Bee',2),('Cee Dee',2),('E.F. Gee',1),('Aitch Eye',null),('Jay Kay',4);
INSERT 0 4

What if I try to Insert an Invalid dept_id?
my_database-> VALUES ('No Good',7);

What if I try to Insert an Invalid dept_id?
my_database-> VALUES ('No Good',7);
ERROR: insert or update on table "emp" violates foreign key constraint "emp_dept_fk"
DETAIL: Key (dept_id)=(7) is not present in table "dept".

What is a View?
A virtual table, based on a query
● Does not take up any space
● Executed in real-time
● Shorthand for a long query
● Present just certain data to certain users

How do I Create a View?
https://www.postgresql.org/docs/current/sql-createview.html
my_database=> CREATE VIEW emp_name_view AS
my_database-> SELECT emp_name AS "employee_name" FROM emp ORDER BY emp_name;

CREATE VIEW

CREATE VIEW
my_database=> SELECT * FROM emp_name_view;
employee_name
---------------
Aitch Eye
Ay Bee
Cee Dee
E.F. Gee
Jay Kay
(5 rows)

What is a Materialized View?
● Query is not executed in real-time
● Can be “refreshed” (query re-executed to gather latest results)
● Useful for aggregating data
● Avoids re-executing long-running/frequently executed queries
A table that contains the results of a query

How do I Create a Materialized View?

https://www.postgresql.org/docs/current/sql-creatematerializedview.html
my_database=> CREATE MATERIALIZED VIEW emp_mview
my_database-> TABLESPACE tablespace_1 AS
my_database-> SELECT emp_id, emp_name AS "employee_name" FROM emp WHERE dept_id in (1,2,3);

SELECT 3

SELECT 3
my_database=> SELECT * FROM emp_mview;
emp_id | employee_name
--------+---------------
1 | Ay Bee
2 | Cee Dee
3 | E.F. Gee
(3 rows)

How do I Refresh a Materialized View?
First, update some data…
my_database=> UPDATE emp SET emp_name = 'CEE DEE' WHERE emp_id = 2;
UPDATE 1

UPDATE 1
my_database=> SELECT emp_name FROM emp WHERE emp_id = 2;
emp_name
----------
CEE DEE

UPDATE 1
my_database=> SELECT emp_name FROM emp WHERE emp_id = 2;
emp_name
----------
CEE DEE
my_database=> SELECT employee_name FROM emp_name_view WHERE employee_name like 'C%';
employee_name
---------------
CEE DEE
(1 row)

my_database=>SELECT employee_name FROM emp_mview where emp_id = 2;
employee_name
---------------
Cee Dee
(1 row)

https://www.postgresql.org/docs/current/sql-refreshmaterializedview.html
employee_name
---------------
Cee Dee
(1 row)
my_database=> REFRESH MATERIALIZED VIEW emp_mview;
REFRESH MATERIALIZED VIEW

https://www.postgresql.org/docs/current/sql-refreshmaterializedview.html
employee_name
---------------
Cee Dee
(1 row)
my_database=> REFRESH MATERIALIZED VIEW emp_mview;
REFRESH MATERIALIZED VIEW
my_database=> SELECT employee_name FROM emp_mview where emp_id = 2;
employee_name
---------------
CEE DEE

Database Connections
● Information Required
● Connecting Django to Postgres
● Client Tools

What Information do I Need?
● host
● port
● database
● username
● password/certificate

What Information do I Need?
name of database to connect to
database host
database port (5432 by default)
database username
Format depends on client tool/driver
psql example: psql -h 127.0.0.1 -p 5432 -U myuser mydatabase
jdbc example: jdbc:postgresql://myuser@127.0.0.1:5432/mydatabase

DATABASES = {
'default': {
'ENGINE': 'django.db.backends.postgresql',
'NAME': 'mydatabase',
'USER': 'myuser',
'HOST': '127.0.0.1',
'PORT': '5432',
'PASSWORD': 'password',
}
}
How do I Configure a Postgres Connection in Django?
https://docs.djangoproject.com/en/4.1/ref/databases/#postgresql-notes
or define a service in .pg_service.conf
settings.py conf file
or provide a password (.pgpass) file

What Client Tools are Available?
Many client tools allow connection to Postgres, including:
● psql
● DBeaver (multi-platform)
https://dbeaver.io/
● pgAdmin4
https://www.pgadmin.org/

Database Operations and Transactions
• SQL, DML, DDL
• Join Types
• Execution Plans
• Transactions
• Commit
• Rollback

What is a Transaction?
● Ends with a COMMIT or a ROLLBACK:
• COMMIT: operations permanently applied to database
• ROLLBACK: operations cancelled
● Changes only visible to other transactions after COMMIT
● Certain locks retained until end of transaction
● ACID (Atomic, Consistent, Isolated, Durable)
A single unit of work that consists of one or more operations

How do I Begin/End a Transaction?

How do I Begin/End a Transaction?
● No need to explicitly BEGIN or COMMIT any transactions
● Each command is a distinct transaction with implicit COMMIT
● To manage a transaction manually:
• Issue BEGIN to start a transaction
• Execute the operations that comprise the transaction
• Issue COMMIT to make the changes permanent or/
• Issue ROLLBACK to undo the changes
Example for psql: default behaviour is autocommit

What is SQL/DML/DDL?
SQL Structured Query Language
https://www.postgresql.org/docs/current/sql.html

DML Data Manipulation Language SELECT, INSERT, UPDATE, DELETE...

DML Data Manipulation Language SELECT, INSERT, UPDATE, DELETE...
DDL Data Definition Language CREATE, ALTER, DROP, RENAME ...

What is a Join?
A way to select from multiple tables in one statement

What are the Different Types of Join?

INNER JOIN
my_database=> SELECT emp.emp_id, emp.emp_name, dept.dept_name
my_database-> FROM emp INNER JOIN dept
my_database-> ON emp.dept_id = dept.dept_id;

INNER JOIN or JOIN (default join type)
my_database-> FROM emp JOIN dept

INNER JOIN or JOIN default join type

LEFT OUTER JOIN
my_database-> FROM emp LEFT OUTER JOIN dept

LEFT OUTER JOIN or LEFT JOIN
my_database-> FROM emp LEFT JOIN dept

RIGHT OUTER JOIN
my_database-> FROM emp RIGHT OUTER JOIN dept

RIGHT OUTER JOIN or RIGHT JOIN
my_database-> FROM emp RIGHT JOIN dept

FULL OUTER JOIN
my_database-> FROM emp FULL OUTER JOIN dept

What is an Execution Plan?
● Access path: full table scan, index lookup
● Join algorithm: nested loop, hash, merge
● Estimated # rows
● Estimated cost
The steps that Postgres takes to execute a SQL statement

How do I Generate an Execution Plan?

https://www.postgresql.org/docs/current/using-explain.html
my_database=> EXPLAIN SELECT * FROM emp_name_view;

my_database=> EXPLAIN SELECT * FROM emp_name_view;
QUERY PLAN
----------------------------------------------------------
Sort (cost=1.11..1.12 rows=5 width=32)
Sort Key: emp.emp_name
-> Seq Scan on emp (cost=0.00..1.05 rows=5 width=32)
(3 rows)

my_database=> EXPLAIN ANALYZE SELECT * FROM emp_name_view;

my_database=> EXPLAIN ANALYZE SELECT * FROM emp_name_view;
QUERY PLAN
--------------------------------------------------------------------------------------------------------
Sort (cost=83.37..86.37 rows=1200 width=32) (actual time=0.225..0.226 rows=5 loops=1)
Sort Key: emp.emp_name
Sort Method: quicksort Memory: 25kB
-> Seq Scan on emp (cost=0.00..22.00 rows=1200 width=32) (actual time=0.009..0.010 rows=5 loops=1)
Planning Time: 0.310 ms
Execution Time: 0.434 ms
(6 rows)

What is WAL?
● Write-ahead log
● Details of changes executed against the Postgres cluster
● Allows crash-recovery, online backups and restores
● WAL files can be archived to allow point in time recovery (PITR)
● DON’T DELETE THEM!
https://www.postgresql.org/docs/current/wal-intro.html

Where are my WAL Files?
postgres@my_vm$ ls -ltr $PGDATA/pg_wal

Where are my WAL Files?
postgres@my_vm$ ls -ltr $PGDATA/pg_wal
total 32768
drwx------. 2 postgres postgres 6 Nov 9 11:00 archive_status
-rw-------. 1 postgres postgres 16777216 Nov 14 08:41 000000010000000000000001

• Database Architecture
• Users and Roles
• Database Objects
• Database Connections
• Database Operations and Transactions
• Documentation
Agenda

Where is the PostgreSQL Documentation?

PostgreSQL Documentation
Current online version: https://www.postgresql.org/docs/current/index.html
https://www.postgresql.org/docs

Where else can I get information?

Slack Channel
● Over 100 channels
● 20k members
https://postgres-slack.herokuapp.com/

Mailing Lists
● Linked to PostgreSQL Community account
● Many different lists
https://lists.postgresql.org/

Thank You!
Karen Jex | @karenhjex | karen.jex@crunchydata.com
Slides: https://karenjex.blogspot.com
Keith Fiske | @keithf4 | keith.fiske@crunchydata.com

Everything You Wanted to Know About Databases (Keith).pdf

More Related Content

Similar to Everything You Wanted to Know About Databases (Keith).pdf (20)

More from All Things Open (20)

Recently uploaded (20)

Everything You Wanted to Know About Databases (Keith).pdf