Oracle Database Express Edition Container / Docker images.
The images are compatible with podman
and docker
. You can use podman
or docker
interchangeably.
21.3.0
,21
,latest
21.3.0-slim
,21-slim
,slim
21.3.0-full
,21-full
,full
18.4.0
,18
18.4.0-slim
,18-slim
18.4.0-full
,18-full
11.2.0.2
,11
11.2.0.2-slim
,11-slim
11.2.0.2-full
,11-full
Run a new database container (data is removed when the container is removed, but kept throughout container restarts):
docker run -d -p 1521:1521 -e ORACLE_PASSWORD=<your password> gvenzl/oracle-xe
Run a new persistent database container (data is kept throughout container lifecycles):
docker run -d -p 1521:1521 -e ORACLE_PASSWORD=<your password> -v oracle-volume:/opt/oracle/oradata gvenzl/oracle-xe
Run a new persistent 11g R2 database container (volume path differs in 11g R2):
docker run -d -p 1521:1521 -e ORACLE_PASSWORD=<your password> -v oracle-volume:/u01/app/oracle/oradata gvenzl/oracle-xe:11
Reset database SYS
and SYSTEM
passwords:
docker exec <container name|id> resetPassword <your password>
We are proud of the following users of these images:
- Airbyte [
6e53a57
] - Apache Spark [
e03afc9
] - Eclipse Vert.x [
ffedbce
] - Flowable [
18c751f
] - GeoTools [
f922f0b
] - Hibernate Search [
173f0b7
] - jOOQ [
#35
] - Liquibase [
c6a31c0
] - Micronaut Data [
ddf11c1
] - Quarkus [
9a63a58
] - Ruby API for Oracle PL/SQL [
63baad0
] - Ruby on Rails ActiveRecord adapter [
afd7a93
] - Rucio by CERN [
80dffbb
] - SchemaCrawler [
08d9b87
] - Sqitch [
8b38027
] - Testcontainers [
99b91b8
] - Upscheme [
954650a
] - utPLSQL [
327110f
]
If you are using these images and would like to be listed as well, please open an issue on GitHub or reach out on Twitter.
The 11gR2 (11.2.0.2) Oracle Database version stores the database data files under /u01/app/oracle/oradata/XE
.
A volume for 11gR2 has to be pointed at /u01/app/oracle/oradata
!
Environment variables allow you to customize your container. Note that these variables will only be considered during the database initialization (first container startup).
This variable is mandatory for the first container startup and specifies the password for the Oracle Database SYS
and SYSTEM
users.
This is an optional variable. Set this variable to a non-empty value, like yes
, to generate a random initial password for the SYS
and SYSTEM
users. The generated password will be printed to stdout (ORACLE PASSWORD FOR SYS AND SYSTEM: ...
).
This is an optional variable. Set this variable to a non-empty string to create a new pluggable database with the name specified in this variable.
Note: this variable is only supported for Oracle Database XE 18c and onwards; 11g does not support pluggable databases.
Note: creating a new database will add to the initial container startup time. If you do not want that additional startup time, use the already existing XEPDB1
database instead.
This is an optional variable. Set this variable to a non-empty string to create a new database schema user with the name specified in this variable. For 18c and onwards, the user will be created in the default XEPDB1
pluggable database. If ORACLE_DATABASE
has been specified, the user will also be created in that pluggable database. This variable requires APP_USER_PASSWORD
or APP_USER_PASSWORD_FILE
to be specified as well.
This is an optional variable. Set this variable to a non-empty string to define a password for the database schema user specified by APP_USER
. This variable requires APP_USER
to be specified as well.
The images can be used as a Service Container within a GitHub Actions workflow. Below is an example service definition for your GitHub Actions YAML file:
services:
# Oracle service (label used to access the service container)
oracle:
# Docker Hub image (feel free to change the tag "latest" to any other available one)
image: gvenzl/oracle-xe:latest
# Provide passwords and other environment variables to container
env:
ORACLE_RANDOM_PASSWORD: true
APP_USER: my_user
APP_USER_PASSWORD: my_password_which_I_really_should_change
# Forward Oracle port
ports:
- 1521:1521
# Provide healthcheck script options for startup
options: >-
--health-cmd healthcheck.sh
--health-interval 10s
--health-timeout 5s
--health-retries 10
After your service is created, you can connect to it via the following properties:
- Hostname:
oracle
(from within another container)localhost
or127.0.0.1
(from the host directly)
- Port:
1521
- Service name:
XEPDB1
- Database App User:
my_user
- Database App Password:
my_password_which_I_really_should_change
If you amend the variables above, here is some more useful info:
- Ports: you can access the port dynamically via
${{ job.services.oracle.ports[1521] }}
. This is helpful when you do not want to specify a given port via- 1521/tcp
instead of- 1521:1521
. Note that theoracle
refers to the service name in the yaml file. If you call your service differently, you will also have to changeoracle
here to that other service name. - Database Admin User:
system
- Database Admin User Password:
$ORACLE_PASSWORD
- Database App User:
$APP_USER
- Database App User Password:
$APP_USER_PASSWORD
- Example JDBC connect string with dynamic port allocation:
jdbc:oracle:thin:@localhost:${{ job.services.oracle.ports[1521] }}/XEPDB1
Flavor | Extension | Description | Use cases |
---|---|---|---|
Slim | -slim |
An image focussed on smallest possible image size instead of additional functionality. | Wherever small images sizes are important but advanced functionality of Oracle Database is not needed. |
Regular | [None] | A well-balanced image between image size and functionality. Recommended for most use cases. | Recommended for most use cases. |
Full | -full |
An image containing all functionality as provided by the Oracle Database installation. | Best for extensions and/or customizations. |
For a full list of changes that have been made to the Oracle Database and OS installation in each individual image flavor, please see ImageDetails.md.
As an alternative to passing sensitive information via environment variables, _FILE
may be appended to some of the previously listed environment variables, causing the initialization script to load the values for those variables from files present in the container. In particular, this can be used to load passwords from Container/Docker secrets stored in /run/secrets/<secret_name>
files. For example:
docker run --name some-oracle -e ORACLE_PASSWORD_FILE=/run/secrets/oracle-passwd -d gvenzl/oracle-xe
This mechanism is supported for:
APP_USER_PASSWORD
ORACLE_PASSWORD
ORACLE_DATABASE
Note: there is a significant difference in how containerization technologies handle secrets. For more information on that topic, please consult the official containerization technology documentation:
If you would like to perform additional initialization of the database running in a container, you can add one or more *.sql
, *.sql.gz
, *.sql.zip
or *.sh
files under /container-entrypoint-initdb.d
(creating the directory if necessary). After the database setup is completed, these files will be executed automatically in alphabetical order.
The *.sql
, *.sql.gz
and *.sql.zip
files will be executed in Sql*Plus as the SYS
user connected to the Oracle instance (XE
). Compressed files will be uncompressed on the fly, allowing for e.g. bigger data loading scripts to save space.
Executable *.sh
files will be run in a new shell process while non-executable *.sh
files (files that do not have the Linux ex
ecutable permission set) will be sourced into the current shell process. The main difference between these methods is that sourced shell scripts can influence the environment of the current process and should generally be avoided. However, sourcing scripts allows for execution of these scripts even if the executable flag is not set for the files containing them. This basically avoids the "why did my script not get executed" confusion.
Note: scripts in /container-entrypoint-initdb.d
are only run the first time the database is initialized; any pre-existing database will be left untouched on container startup.
Note: you can also put files under the /docker-entrypoint-initdb.d
directory. This is kept for backwards compatibility with other widely used container images but should generally be avoided. Do not put files under /container-entrypoint-initdb.d
and /docker-entrypoint-initdb.d
as this would cause the same files to be executed twice!
Warning: if a command within the sourced /container-entrypoint-initdb.d
scripts fails, it will cause the main entrypoint script to exit and stop the container. It also may leave the database in an incomplete initialized state. Make sure that shell scripts handle error situations gracefully and ideally do not source them!
Warning: do not exit executable /container-entrypoint-initdb.d
scripts with a non-zero value (using e.g. exit 1;
) unless it is desired for a container to be stopped! A non-zero return value will tell the main entrypoint script that something has gone wrong and that the container should be stopped.
The following example installs the countries, cities and currencies sample data set under a new user TEST
into the database:
[gvenzl@localhost init_scripts]$ pwd
/home/gvenzl/init_scripts
[gvenzl@localhost init_scripts]$ ls -al
total 12
drwxrwxr-x 2 gvenzl gvenzl 61 Mar 7 11:51 .
drwx------. 19 gvenzl gvenzl 4096 Mar 7 11:51 ..
-rw-rw-r-- 1 gvenzl gvenzl 134 Mar 7 11:50 1_create_user.sql
-rwxrwxr-x 1 gvenzl gvenzl 164 Mar 7 11:51 2_create_data_model.sh
[gvenzl@localhost init_scripts]$ cat 1_create_user.sql
ALTER SESSION SET CONTAINER=XEPDB1;
CREATE USER TEST IDENTIFIED BY test QUOTA UNLIMITED ON USERS;
GRANT CONNECT, RESOURCE TO TEST;
[gvenzl@localhost init_scripts]$ cat 2_create_data_model.sh
curl -LJO https://raw.githubusercontent.com/gvenzl/sample-data/master/countries-cities-currencies/install.sql
sqlplus -s test/test@//localhost/XEPDB1 @install.sql
rm install.sql
As the execution happens in alphabetical order, numbering the files will guarantee the execution order. A new container started up with /home/gvenzl/init_scripts
pointing to /container-entrypoint-initdb.d
will then execute the files above:
docker run --name test \
> -p 1521:1521 \
> -e ORACLE_RANDOM_PASSWORD="y" \
> -v /home/gvenzl/init_scripts:/container-entrypoint-initdb.d \
> gvenzl/oracle-xe:18.4.0-full
CONTAINER: starting up...
CONTAINER: first database startup, initializing...
...
CONTAINER: Executing user defined scripts...
CONTAINER: running /container-entrypoint-initdb.d/1_create_user.sql ...
Session altered.
User created.
Grant succeeded.
CONTAINER: DONE: running /container-entrypoint-initdb.d/1_create_user.sql
CONTAINER: running /container-entrypoint-initdb.d/2_create_data_model.sh ...
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 115k 100 115k 0 0 460k 0 --:--:-- --:--:-- --:--:-- 460k
Table created.
...
Table provided actual
-------------------- -------- ------
regions 7 7
countries 196 196
cities 204 204
currencies 146 146
currencies_countries 203 203
Thank you!
--------------------------------------------------------------------------------
The installation is finished, please check the verification output above!
If the 'provided' and 'actual' row counts match, the installation was successful
.
If the row counts do not match, please check the above output for error messages
.
CONTAINER: DONE: running /container-entrypoint-initdb.d/2_create_data_model.sh
CONTAINER: DONE: Executing user defined scripts.
#########################
DATABASE IS READY TO USE!
#########################
...
As a result, one can then connect to the new schema directly:
[gvenzl@localhost init_scripts]$ sql test/test@//localhost/XEPDB1
SQLcl: Release 20.3 Production on Sun Mar 07 12:05:06 2021
Copyright (c) 1982, 2021, Oracle. All rights reserved.
Connected to:
Oracle Database 18c Express Edition Release 18.0.0.0.0 - Production
Version 18.4.0.0.0
SQL> select * from countries where name = 'Austria';
COUNTRY_ID COUNTRY_CODE NAME OFFICIAL_NAME POPULATION AREA_SQ_KM LATITUDE LONGITUDE TIMEZONE REGION_ID
---------- ------------ ------- ------------------- ---------- ---------- -------- --------- ------------- ---------
AUT AT Austria Republic of Austria 8793000 83871 47.33333 13.33333 Europe/Vienna EU
SQL>
If you would like to perform additional action after the database running in a container has been started, you can add one or more *.sql
, *.sql.gz
, *.sql.zip
or *.sh
files under /container-entrypoint-startdb.d
(creating the directory if necessary). After the database is up and ready for requests, these files will be executed automatically in alphabetical order.
The execution order and implications are the same as with the Initialization scripts described above.
Note: you can also put files under the /docker-entrypoint-startdb.d
directory. This is kept for backwards compatibility with other widely used container images but should generally be avoided. Do not put files under /container-entrypoint-startdb.d
and /docker-entrypoint-startdb.d
as this would cause the same files to be executed twice!
Note: if the database inside the container is initialized (started for the first time), startup scripts are executed after the setup scripts.
Warning: files placed in /container-entrypoint-startdb.d
are always executed after the database in a container is started, including pre-created databases. Use this mechanism only if you wish to perform a certain task always after the database has been (re)started by the container.
If you have questions or constructive feedback about these images, please file a ticket over at github.com/gvenzl/oci-oracle-xe.