EnvVar Reference

Environment Variable Reference

Commons to all applications

Set by the deployment process

These are read-only variables that are generated for each scaler before they build and start your application.

Name	Description	Example value
`INSTANCE_NUMBER`	Allows your application to differentiate each running node on the application level.	0, 1…
`INSTANCE_TYPE`	Whether this instance is a “build” instance or a “production” instance.	build, production
`INSTANCE_ID`	The ID of the current instance (scaler) of your application. It’s unique for each instance of your application and changes every time you deploy it.	518c8d8f-e304-4310-81e0-9c4237d55645
`CC_PRETTY_INSTANCE_NAME`	The random generated string as instance pretty name using pokemon names.	Tiny rhyhorn
`APP_ID`	The ID of your Clever Cloud application	app_649a93d1-6677-44bc-aca7-6f46107d6e02
`APP_HOME`	The absolute path to your application folder	/home/bas/app_649a93d1-6677-44bc-aca7-6f46107d6e02
`CC_DEPLOYMENT_ID`	Internal id of current deployment	f7efaf04-1a63-45a1-8503-0de7c750ee48
`CC_COMMIT_ID`	The id of the commit that’s currently running	d88cd2ae1aaa91923ed2bd689d95d713b6f3f45f
`CC_REVERSE_PROXY_IPS`	A comma separated list of trusted IP addresses. You should only accept requests coming from these IP addresses.	x.y.z.z,x.y.z.z
`ELASTIC_APM_SERVICE_NAME`	Sets the name of your service/application in Elastic APM. Automatically defined when you have linked an Elastic APM service to your application. You can override it by defining it yourself	Your application’s name conforming to Elastic APM naming convention
`CC_APP_NAME`	The customer defined application name	cloud-api-production
`PORT`	The mandatory port value is 8080	8080

Variables you can define

So you can alter the build&start process for your application.

Name	Description	Default value
`APP_FOLDER`	Folder in which the application is located (inside the git repository)
`CC_DISABLE_JVM_AGENT`	Disable JVM agent activation
`CC_GIT_FULL_CLONE`	If set to `false`, the git repository is cloned with `--depth 1` to speed up the deployment	true
`CC_DISABLE_GIT_SUBMODULES`	Disable Git submodules initialization & synchronization
`CC_PREVENT_LOGSCOLLECTION`	Disable Vector service used to collect and send logs	`false`
`CC_HEALTH_CHECK_PATH`	Custom path to validate your application deployment
`CC_SSH_PRIVATE_KEY`	A ssh private key to setup for the user running your application
`CC_SSH_PRIVATE_KEY_FILE`	The name to use for the file containing the private ssh key	id_ed25519
`CC_WORKER_RESTART`	One of `always`, `on-failure` or `no`. Control whether workers need to be restarted when they exit. This setting controls all workers.	on-failure
`CC_WORKER_RESTART_DELAY`	Define a delay in seconds to restart the worker when they exit.	1
`CC_WORKER_COMMAND`	Command to run in background as a worker process. You can run multiple workers.

You can set some tools’ version in any runtime (except Docker):

Name	Description
`CC_HUGO_VERSION`	Set the Hugo version, for example `0.147`
`CC_NODE_VERSION`	Set Node.js version, for example `24`, `23.11` or `22.15.1`

Control build and dependencies cache

Name	Description	Default value
`CC_CACHE_DEPENDENCIES`	Enable caching of your build dependencies to speed up following builds.	false
`CC_DISABLE_BUILD_CACHE_UPLOAD`	Disable creation and upload of cache archive. Restarts won’t be speeded up.	`false`
`CC_IGNORE_FROM_BUILDCACHE`	Allows to specify paths to ignore when the build cache archive is created.
`IGNORE_FROM_BUILDCACHE`	(Deprecated) Allows to specify paths to ignore when the build cache archive is created.
`CC_OVERRIDE_BUILDCACHE`	Allows to specify paths that will be in the build cache. Only those files / directories will be cached

Control the deployment’s behavior

Name	Description	Default value
`CC_RUN_COMMAND`	Custom command to run your application.
`CC_TASK`	If set to true, the deployer runs `CC_RUN_COMMAND` and terminates the instance after running the task. Trigger an execution using `git push` or starting your instance	false
`CC_TROUBLESHOOT`	Enable debug log level, will also keep the VM up after failure for 15 minutes so you can SSH and debug. Don’t forget to cancel deployment if you push a new commit.	false

🐳

CC_RUN_COMMAND has no effect on Docker. To run Docker, use CMD in your Dockerfile.

Deployment hooks

Use these to define commands to run between various steps of the deployment.

Name	Description
`CC_PRE_BUILD_HOOK`	Ran before the dependencies are fetched. If it fails, the deployment fails.
`CC_POST_BUILD_HOOK`	Ran after the project is built, and before the cache archive is generated. If it fails, the deployment fails.
`CC_PRE_RUN_HOOK`	Ran before the application is started, but after the cache archive has been generated. If it fails, the deployment fails.
`CC_RUN_FAILED_HOOK`	Ran once the application has failed to start.
`CC_RUN_SUCCEEDED_HOOK`	Ran once the application has started successfuly.

Configure extra software

Name	Description	Default value
`CC_CLAMAV`	Start the clamav and clamav-freshclam services (the database is updated every 2 hours). WARNING: Clamscan consumes a lot of resources (~ 1GB of memory), make sure you have a scaler with enough memory to avoid OOM.	false
`CC_CLAMAV_MAXQUEUE`	Maximum number of queued items.	100
`CC_CLAMAV_MAXTHREADS`	Maximum number of threads running at the same time.	10
`CC_METRICS_PROMETHEUS_PASSWORD`	Define the password for the basic auth of the Prometheus endpoint
`CC_METRICS_PROMETHEUS_PATH`	Define the path on which the Prometheus endpoint is available	/metrics
`CC_METRICS_PROMETHEUS_PORT`	Define the port on which the Prometheus endpoint is available	9100
`CC_METRICS_PROMETHEUS_RESPONSE_TIMEOUT`	Define the timeout in seconds to collect the application metrics. This value must be below 60 seconds as data are collected every minutes	3
`CC_METRICS_PROMETHEUS_USER`	Define the user for the basic auth of the Prometheus endpoint
`CC_VARNISH_FILE`	The path to the Varnish configuration file, relative to your application root	`/clevercloud/varnish.vcl`
`CC_VARNISH_STORAGE_SIZE`	Configure the size of the Varnish cache.	1G
`CC_WORKER_COMMAND`	Command to run in background as a worker process. You can run multiple workers.

Install tools with Mise package manager

Mise package manager is a powerful tool to install and manage dependencies on Clever Cloud. Just add a mise.toml file at the root of your project or set the CC_MISE_FILE_PATH environment variable. All tools will be installed at the defined version before the build phase and available for your scripts. You can also use Mise to define environment variables and alias (tasks) in a declarative way.

To disable Mise, set CC_DISABLE_MISE environment variable to true.

Learn more about Mise package manager

Name	Description
`CC_RUN_MISE_INSTALL`	Set to `true` to run `mise install` before the build phase

Use Redirection.io as a proxy

Redirection.io can help reduce HTTP traffic issues on your website. It gives a complete control on how HTTP requests are handled, which helps make it SEO-friendly. It can perform redirections and comes with lots of features. You can link any application to a Redirection.io project easily, setting up the proxy mode with following environment variables:

Name	Description	Default value
`CC_ENABLE_REDIRECTIONIO`	Enable Redirection.io support	`false`
`CC_REDIRECTIONIO_PROJECT_KEY`	The Redirection.io project key
`CC_REDIRECTIONIO_FORWARD_PORT`	The listening port of your application
`CC_REDIRECTIONIO_INSTANCE_NAME`	The name of your application (optional)

The Redirection.io agent will start as a service, listen to 8080 port and forward the traffic to your application port.

Learn more about Redirection.io

Redirection.io is not available in Docker and PHP applications

Tailscale support

Tailscale is a managed VPN service based on Wireguard that enable private networking between users, devices or machines. Clever Cloud provides a native integration of Tailscale, by mounting a VPN endpoint for each of your application’s instances.

Note that Reusable keys are required to use multiple instances. You can generate one here.

Name	Description	Default value
`TAILSCALE_ACCEPT_DNS`	Tailscale use its default DNS settings	`true`
`TAILSCALE_ACCEPT_ROUTES`	Tailscale uses its subnets settings	`false`
`TAILSCALE_AUTH_KEY`	Contains your Tailscale Auth key
`TAILSCALE_LOGIN_SERVER`	Contains the login server

How it works?

For a given application with TAILSCALE_AUTH_KEY configured, each instance will be configured to join a Tailscale network. Instances will be named after your configured name, suffixed with the INSTANCE_NUMBER : CC-<NAME>-<INSTANCE_NUMBER>. If you have multiple instances and use one of them for being an admin instance (using INSTANCE_NUMBER), you can match the instance from your deployment to reach it over VPN.

If TAILSCALE_LOGIN_SERVER is provided, the agent will be configured to reach an alternative control server. Note that using your own control server is at your own risks, and Tailscale can’t be responsible. An alternative control server can still be useful to use for constraints environements. Headscale is an example of self-hosted implementation of the Tailscale control server that can run on Clever Cloud.

Docker

Docker Documentation

Name	Description	Default value
`CC_DOCKERFILE`	The name of the Dockerfile to build.	Dockerfile
`CC_DOCKER_BUILDX`	Set to `true` to use`buildx` for building your image	false
`CC_DOCKER_EXPOSED_HTTP_PORT`	Set to custom HTTP port if your Docker container runs on custom port.	8080
`CC_DOCKER_EXPOSED_TCP_PORT`	Set to custom TCP port if your Docker container runs on custom port.	4040
`CC_DOCKER_FIXED_CIDR_V6`	Activate the support of IPv6 with an IPv6 subnet int the docker daemon.
`CC_DOCKER_LOGIN_PASSWORD`	The password of your username.
`CC_DOCKER_LOGIN_SERVER`	The server of your private registry (optional).	Docker’s public registry
`CC_DOCKER_LOGIN_USERNAME`	The username to login to a private registry.
`CC_MOUNT_DOCKER_SOCKET`	Set to true to access the host Docker socket from inside your container.	false

.NET

.NET Documentation

Name	Description	Default value
`CC_DOTNET_PROFILE`	Override the build configuration settings in your project.	Release
`CC_DOTNET_PROJ`	The name of your project file to use for the build, without the .csproj / .fsproj / .vbproj extension.
`CC_DOTNET_TFM`	Compiles for a specific framework. The framework must be defined in the project file. Example : `net5.0`
`CC_DOTNET_VERSION`	Choose the .NET Core version between `6.0`,`8.0`.	8.0
`CC_RUN_COMMAND`	Custom command to run your application.

Elixir

Elixir Documentation

Name	Description	Default value
`CC_ELIXIR_VERSION`	Choose the Elixir version among those supported
`CC_MIX_BUILD_GOAL`	The mix goal to build the application (default compile)
`CC_PHOENIX_ASSETS_DIR`	Folder in which your Phoenix assets are located.
`CC_PHOENIX_DIGEST_GOAL`	Phoenix digest goal.	phx.digest
`CC_PHOENIX_RUN_ECTO_MIGRATE`	Whether to run ‘mix ecto.migrate’ or not.	true
`CC_PHOENIX_SERVER_GOAL`	Phoenix server goal.	phx.server
`CC_RUN_COMMAND`	Custom command to run your application. Replaces the default behaviour.

FrankenPHP

FrankenPHP Documentation

Name	Description	Default value
`CC_FRANKENPHP_PORT`	The port on which FrankenPHP listens for HTTP requests	`8080`
`CC_FRANKENPHP_WORKER`	Path to the worker script, relative to the root of your project (e.g. `/worker/scrip.php`)
`CC_PHP_COMPOSER_FLAGS`	Flags to pass to Composer	`-n --no-dev --no-progress --no-scripts`
`CC_PHP_DEV_DEPENDENCIES`	Set to `install` to install PHP development dependencies during build
`CC_WEBROOT`	Path to the web content to serve, relative to the root of your application	`/`

Go

Go Documentation

Name	Description
`CC_GO_BUILD_TOOL`	Available values: `gomod`, `gobuild`. Build and install your application. `goget` still exists but is deprecated.
`CC_GO_RUNDIR`	Run the application from the specified path, relative to `$GOPATH/src/`, now deprecated.
`CC_GO_PKG`	Tell the `CC_GO_BUILD_TOOL` which file contains the `main()` function, default `main.go`.

Haskell

Haskell Documentation

Name	Description
`CC_HASKELL_STACK_TARGET`	Specify Stack package target.
`CC_HASKELL_STACK_SETUP_COMMAND`	Only use this variable to override the default `setup` Stack step command
`CC_HASKELL_STACK_INSTALL_COMMAND`	Only use this variable to override the default `install` Stack step command
`CC_HASKELL_STACK_INSTALL_DEPENDENCIES_COMMAND`	Only use this variable to override the default `install --only-dependencies` Stack step command
`CC_RUN_COMMAND`	Custom command to run your application.

Java

Java Documentation

Name	Description	Default value	Read Only
`CC_DISABLE_MAX_METASPACE`	Allows to disable the Java option -XX:MaxMetaspaceSize
`CC_EXTRA_JAVA_ARGS`	Define extra arguments to pass to ‘java’ for jars.
`CC_JAR_ARGS`	Define arguments to pass to the jar we launch.
`CC_JAR_PATH`	Define the path to your jar.
`CC_JAVA_VERSION`	Choose the JVM version between 7 to 17 for OpenJDK or graalvm-ce for GraalVM 21.0.0.2 (based on OpenJDK 11.0).	21
`CC_MAVEN_PROFILES`	Define which maven profile to use during default build.
`CC_RUN_COMMAND`	Custom command to run your application. Replaces the default behaviour.
`CC_SBT_TARGET_BIN`	Define the bin to pick in the `CC_SBT_TARGET_DIR`.	The first bin found in the `CC_SBT_TARGET_DIR`.
`CC_SBT_TARGET_DIR`	Define where pick the bin to run.	`.`	Then `/target/universal/stage/bin` is concatenated.
`GRADLE_DEPLOY_GOAL`	Define which gradle goals to run during build.
`MAVEN_DEPLOY_GOAL`	Define which maven goals to run during build.
`NUDGE_APPID`
`PLAY1_VERSION`	Define which play1 version to use between 1.2, 1.3, 1.4 and 1.5
`SBT_DEPLOY_GOAL`	Define which sbt goals to run during build.	stage

Linux

Linux Documentation

Name	Description	Default value
`CC_BUILD_COMMAND`	The command to run during the build phase	``
`CC_RUN_COMMAND`	The command to run during the run phase	``

Use Linux runtime with Mise package manager to install your dependencies easily

Node.js

Node.js Documentation

Name	Description	Default value
`CC_NODE_VERSION`	Set Node.js version, for example `24`, `23.11` or `22.15.1`
`CC_NODE_DEV_DEPENDENCIES`	Control if development dependencies are installed or not. Values are either `install` or `ignore`	`ignore`
`CC_RUN_COMMAND`	Define a custom command. Example for Meteor: `node .build/bundle/main.js <options>`
`CC_NODE_BUILD_TOOL`	Choose your build tool between npm, npm-ci, yarn, yarn2 and custom	npm
`CC_CUSTOM_BUILD_TOOL`	A custom command to run (with `CC_NODE_BUILD_TOOL` set to `custom`)
`CC_NPM_REGISTRY`	The host of your private repository, available values: `github` or the registry host.	registry.npmjs.org
`CC_NPM_BASIC_AUTH`	Private repository credentials, in the form `user:password`. You can’t use this if `CC_NPM_TOKEN` is set
`NPM_TOKEN`	Private repository token

PHP

PHP Documentation

Name	Description	Default value
`ALWAYS_POPULATE_RAW_POST_DATA`
`CC_COMPOSER_VERSION`	Choose your composer version between 1 and 2	2
`CC_CGI_IMPLEMENTATION`	Choose the Apache FastCGI module between `fastcgi` and `proxy_fcgi`	proxy_fcgi
`CC_HTTP_BASIC_AUTH`	Restrict HTTP access to your application. Example: `login:password`. You can define multiple credentials using additional `CC_HTTP_BASIC_AUTH_n` (where `n` is a number) environment variables.
`CC_APACHE_HEADERS_SIZE`	Set the maximum size of the headers in Apache, between `8` and `256`. Effective value depends on deployment region. Ask for a dedicated load balancer for a specific value	8
`CC_LDAP_CA_CERT`
`CC_MTA_AUTH_PASSWORD`	Password to authenticate to the SMTP server
`CC_MTA_AUTH_USER`	User to authenticate to the SMTP server
`CC_MTA_SERVER_AUTH_METHOD`	Enable or disable authentication to the SMTP server	on
`CC_MTA_SERVER_HOST`	Host of the SMTP server
`CC_MTA_SERVER_PORT`	Port of the SMTP server	465
`CC_MTA_SERVER_USE_TLS`	Enable or disable TLS when connecting to the SMTP server	true
`CC_OPCACHE_INTERNED_STRINGS_BUFFER`	The amount of memory used to store interned strings, in megabytes.	Default 4 (PHP5), 8 (PHP7)
`CC_OPCACHE_MAX_ACCELERATED_FILES`	Maximum number of files handled by opcache.	Default depends on the scaler size
`CC_OPCACHE_MEMORY`	Set the shared opcache memory size	Default is about 1/8 of the RAM
`CC_OPCACHE_PRELOAD`	The path of the PHP preload file (PHP version 7.4 or higher).
`CC_PHP_ASYNC_APP_BUCKET`	Mount the default app FS bucket asynchronously. If set, should have value `async`
`CC_PHP_DEV_DEPENDENCIES`	Control if development dependencies are installed or not. Values are either `install` or `ignore`
`CC_PHP_DISABLE_APP_BUCKET`	Disable entirely the app FS Bucket. Values are either `true`, `yes` or `disable`
`CC_PHP_VERSION`	Choose your PHP version among those supported	8.3
`CC_REALPATH_CACHE_TTL`	The size of the realpath cache to be used by PHP	120
`CC_WEBROOT`	Define the `DocumentRoot` of your project	`.`
`ENABLE_ELASTIC_APM_AGENT`	Elastic APM Agent for PHP	`true` if `ELASTIC_APM_SERVER_URL` is defined, `false` otherwise
`ENABLE_GRPC`	Enable the use of gRPC module	false
`ENABLE_PDFLIB`	Enable the use of PDFlib module	false
`ENABLE_REDIS`		false
`HTTP_TIMEOUT`	Define a custom HTTP timeout	180
`LDAPTLS_CACERT`
`MAX_INPUT_VARS`
`MEMORY_LIMIT`	Change the default memory limit
`SESSION_TYPE`	Choose `redis` to use it as session store
`SOCKSIFY_EVERYTHING`
`SQREEN_API_APP_NAME`	The name of your sqreen application.
`SQREEN_API_TOKEN`	organisation token.

Python

Python Documentation

Name	Description	Default value
`CC_HTTP_BASIC_AUTH`	Restrict HTTP access to your application. Example: `login:password`. You can define multiple credentials using additional `CC_HTTP_BASIC_AUTH_n` (where `n` is a number) environment variables.
`CC_NGINX_PROXY_BUFFERS`	Sets the number and size of the buffers used for reading a response from the proxied server, for a single connection.
`CC_NGINX_PROXY_BUFFER_SIZE`	Sets the size of the buffer used for reading the first part of the response received from the proxied server.
`CC_PIP_REQUIREMENTS_FILE`	Allows you to define a custom requirements.txt file	requirements.txt
`CC_PYTHON_BACKEND`	Choose the Python backend to use between daphne, gunicorn, uvicorn and uwsgi	uwsgi
`CC_PYTHON_CELERY_LOGFILE`	Relative path to your Celery logfile: /path/to/logdir
`CC_PYTHON_CELERY_MODULE`	Specify the Celery module you want to start
`CC_PYTHON_CELERY_USE_BEAT`	Set to “true” to activate Beat support
`CC_PYTHON_MANAGE_TASKS`	Comma-separated list of Django manage tasks
`CC_PYTHON_MODULE`	Select which module you want to start with the path to the folder containing the app object. For example, a module called server.py in a folder called /app would be used here as app.server:app
`CC_PYTHON_USE_GEVENT`	Set to true to enable Gevent
`CC_PYTHON_VERSION`	Choose the Python version among those supported	3
`ENABLE_GZIP_COMPRESSION`	Set to `true` to gzip-compress through Nginx
`GZIP_TYPES`	Set the mime types to compress.	text/* application/json application/xml application/javascript image/svg+xml
`HARAKIRI`	Timeout (in seconds) after which an unresponding process is killed	180
`NGINX_READ_TIMEOUT`	Read timeout in seconds	300
`PYTHON_SETUP_PY_GOAL`	Custom setup goal to be launch after requirements.txt have been installed
`STATIC_FILES_PATH`	Relative path to where your static files are stored: path/to/static
`STATIC_URL_PREFIX`	The URL path under which you want to serve static file, usually /public
`STATIC_WEBROOT`
`UWSGI_ASYNC`	Number of cores to use for uWSGI asynchronous/non-blocking modes
`UWSGI_ASYNC_ENGINE`	Select the asynchronous engine for uWSGI (optional)
`UWSGI_INTERCEPT_ERRORS`
`WSGI_BUFFER_SIZE`	Buffer size (in bytes) for uploads.	4096
`WSGI_POST_BUFFERING`	Maximal size (in bytes) for the headers of a request.	4096
`WSGI_THREADS`	Number of threads per worker. (Default: automatically setup with the scaler size)
`WSGI_WORKERS`	Number of workers. (Default: automatically setup with the scaler size)

When your Python application doesn’t use one of the supported backends, with CC_RUN_COMMAND for example, it must listen on port 9000, not 8080.

Ruby

Name	Description	Default value
`CC_ENABLE_SIDEKIQ`	Enable Sidekiq background process	false
`CC_HTTP_BASIC_AUTH`	Restrict HTTP access to your application. Example: `login:password`. You can define multiple credentials using additional `CC_HTTP_BASIC_AUTH_n` (where `n` is a number) environment variables.
`CC_NGINX_PROXY_BUFFERS`	Sets the number and size of the buffers used for reading a response from the proxied server, for a single connection. Nginx documentation
`CC_NGINX_PROXY_BUFFER_SIZE`	Sets the size of the buffer used for reading the first part of the response received from the proxied server. Nginx documentation
`CC_RACKUP_SERVER`	The server to use for serving the ruby application	puma
`CC_RAKEGOALS`	A list of comma-separated rake goals to execute e.g. db:migrate, assets:precompile
`CC_RUBY_VERSION`	Choose the Ruby version to use but we strongly advise to set Ruby version in your Gemfile
`CC_SIDEKIQ_FILES`	Specify a list of Sidekiq configuration files e.g. “./config/sidekiq_1.yml,./config/sidekiq_2.yml”
`ENABLE_GZIP_COMPRESSION`	Set to `true` to gzip-compress through Nginx
`GZIP_TYPES`	Set the mime types to compress.	text/* application/json application/xml application/javascript image/svg+xml
`NGINX_READ_TIMEOUT`	Read timeout in seconds	300
`RACK_ENV`
`RAILS_ENV`
`STATIC_FILES_PATH`	Relative path to where your static files are stored: path/to/static
`STATIC_URL_PREFIX`	The URL path under which you want to serve static file, usually /public
`STATIC_WEBROOT`

Rust

Rust Documentation

Name	Description	Default value
`CC_RUSTUP_CHANNEL`	The rust channel to use. Use a specific channel version with stable, beta, nightly or a specific version like 1.13.0	stable
`CC_RUST_BIN`	The name of the binary to launch once built
`CC_RUST_FEATURES`	The list of features to enable
`CC_RUN_COMMAND`	Custom command to run your application.

Static

Static Documentation

Name	Description	Default value
`CC_HUGO_VERSION`	Set the Hugo version, for example `0.147`
`CC_BUILD_COMMAND`	The command to run during the build phase
`CC_OVERRIDE_BUILDCACHE`	Files and path to put in the build cache, separated by a `:`
`CC_STATIC_AUTOBUILD_OUTDIR`	The output directory of the static site generator (SSG) auto-build, relative to the root of your application	`/cc_static_autobuilt`
`CC_STATIC_CADDYFILE`	Path to the Caddyfile, as expected by `caddy run --config` command	`./Caddyfile`
`CC_STATIC_FLAGS`	Command line flags to pass to the static web server
`CC_STATIC_PORT`	The port on which the static web server listens for HTTP requests	`8080`
`CC_STATIC_SERVER`	The server to use to serve your static website, can be `caddy`	`static-web-server`
`CC_WEBROOT`	Path to the web content to serve, relative to the root of your application, used as default build cache	`/`

Static Site Generators (SSG) Auto-build

If no CC_BUILD_COMMAND is provided, Clever Cloud tries to detect and configure Static Site Generator (SSG), any provided CC_WEBROOT is ignored. See Static Site Generators (SSG) Auto-build for more information.

V (Vlang)

Name	Description	Default value
`CC_V_BINARY`	The name of the output binary file.	`${APP_HOME}/v_bin_${APP_ID}`
`ENVIRONMENT`	Set to `development` to compile without the `-prod` flag

Add-ons environment variables

FS Bucket

FS Bucket Documentation

Name	Description	Read Only
`BUCKET_HOST`	Hostname of the bucket	✓
`CC_FS_BUCKET`	Defines which bucket to mount on which path

MongoDB

MongoDB Documentation

Name	Description	Default value	Read Only
`MONGODB_ADDON_HOST`		Generated upon creation	✓
`MONGODB_ADDON_PORT`		Generated upon creation	✓
`MONGODB_ADDON_DIRECT_HOST`	Points directly to the VM, bypassing reverse proxies	Generated on demand via the dashboard	✓
`MONGODB_ADDON_DIRECT_PORT`	Points to ectly to the VM, bypassing reverse proxies	Generated on demand via the dashboard	✓
`MONGODB_ADDON_DB`		Generated upon creation	✓
`MONGODB_ADDON_PASSWORD`		Generated upon creation	✓
`MONGODB_ADDON_USER`		Generated upon creation	✓

MySQL

MySQL Documentation

Name	Description	Default value	Read Only
`MYSQL_ADDON_HOST`		Generated upon creation	✓
`MYSQL_ADDON_PORT`		Generated upon creation	✓
`MYSQL_ADDON_DIRECT_HOST`	Points directly to the VM, bypassing reverse proxies	Generated on demand via the dashboard	✓
`MYSQL_ADDON_DIRECT_PORT`	Points to ectly to the VM, bypassing reverse proxies	Generated on demand via the dashboard	✓
`MYSQL_ADDON_DB`		Generated upon creation	✓
`MYSQL_ADDON_PASSWORD`		Generated upon creation	✓
`MYSQL_ADDON_ROLE`		Generated upon creation	✓
`MYSQL_ADDON_USER`		Generated upon creation	✓

ProxySQL

ProxySQL Documentation

Name	Description	Default value	Read Only
`CC_ENABLE_MYSQL_PROXYSQL`	Enable the ProxySQL feature	false
`CC_MYSQL_PROXYSQL_MAX_CONNECTIONS`	Defines the maximum number of connections the local ProxySQL will open to your MySQL add-on	10
`CC_MYSQL_PROXYSQL_USE_TLS`	Enable or disable secured connection using TLS to your MySQL add-on	true
`CC_MYSQL_PROXYSQL_SOCKET_PATH`	Contains the path to the Unix Datagram Socket to connect to ProxySQL		✓

PostgreSQL

PostgreSQL Documentation

Name	Description	Default value	Read Only
`POSTGRESQL_ADDON_HOST`		Generated upon creation	✓
`POSTGRESQL_ADDON_PORT`		Generated upon creation	✓
`POSTGRESQL_ADDON_DIRECT_HOST`	Points directly to the VM, bypassing reverse proxies	Generated on demand via the dashboard	✓
`POSTGRESQL_ADDON_DIRECT_PORT`	Points to ectly to the VM, bypassing reverse proxies	Generated on demand via the dashboard	✓
`POSTGRESQL_ADDON_DB`		Generated upon creation	✓
`POSTGRESQL_ADDON_PASSWORD`		Generated upon creation	✓
`POSTGRESQL_ADDON_ROLE`		Generated upon creation	✓
`POSTGRESQL_ADDON_USER`		Generated upon creation	✓

Pgpool-II

Pgpool-II Documentation

Name	Description	Default value
`CC_ENABLE_PGPOOL`	Enables the Pgpool-II feature	false
`CC_PGPOOL_SOCKET_PATH`	Contains the path to the Unix Datagram Socket to connect to Pgpool-II	true
`CC_PGPOOL_PCP_SOCKET_PATH`	Contains the path to the Unix Datagram Socket to connect to PCP	true
`CC_PGPOOL_EXTRA_USERS`	Add new user/password pairs to the pool_passwd file, separated by commas (user1:password,user2:password…)	’ '
`CC_PGPOOL_RESERVED_CONNECTIONS`	Number of reserved connections	0
`CC_PGPOOL_LISTEN_BACKLOG_MULTIPLIER`	Specifies the length of connection queue from frontend to Pgpool-II	2
`CC_PGPOOL_LEADER_WEIGHT`	Weight for backend 0 (Leader)	1
`CC_PGPOOL_NUM_INIT_CHILDREN`	Number of concurrent sessions allowed	16
`CC_PGPOOL_MAX_POOL`	Number of connection pool caches per connection	1
`CC_PGPOOL_CHILD_LIFE_TIME`	Pool exits after being idle for this many seconds	300
`CC_PGPOOL_CHILD_MAX_CONNECTIONS`	Pool exits after receiving that many connections	0 (no exit)
`CC_PGPOOL_CONNECTION_LIFE_TIME`	Connection to backend closes after being idle for this many seconds	0 (no close)
`CC_PGPOOL_CLIENT_IDLE_LIMIT`	Client is disconnected after being idle for that many seconds (even inside an explicit transactions!)	0 (no disconnection)
`CC_PGPOOL_LOG_CONNECTIONS`	Log connections	off
`CC_PGPOOL_LOG_DISCONNECTIONS`	Log disconnections	off
`CC_PGPOOL_LOG_HOSTNAME`	Hostname will be shown in ps status and in logs if connections are logged	off
`CC_PGPOOL_LOG_STATEMENT`	Log all statements	off
`CC_PGPOOL_LOG_PER_NODE_STATEMENT`	Log all statements with node and backend informations	off
`CC_PGPOOL_LOG_CLIENT_MESSAGES`	Log any client messages	off
`CC_PGPOOL_LOG_STANDBY_DELAY`	Log standby delay. Valid values are combinations of always, if_over_threshold and none	if_over_threshold
`CC_PGPOOL_CONNECTION_CACHE`	Activate connection pools	on
`CC_PGPOOL_REPLICATE_SELECT`	Replicate SELECT statements in replication mode	off
`CC_PGPOOL_INSERT_LOCK`	Automatically locks a dummy row or a table with INSERT statements to keep SERIAL data consistency	on
`CC_PGPOOL_LOBJ_LOCK_TABLE`	When rewriting lo_creat command in replication mode, specify table name to lock	’ '
`CC_PGPOOL_REPLICATION_STOP_ON_MISMATCH`	On disagreement with the packet kind sent from backend, degenerate the node which is most likely “minority” If off, just force to exit this session	off
`CC_PGPOOL_FAILOVER_IF_AFFECTED_TUPLES_MISMATCH`	On disagreement with the number of affected tuples in UPDATE/DELETE queries, then degenerate the node which is most likely “minority”. If off, just abort the transaction to keep the consistency	off
`CC_PGPOOL_LOAD_BALANCE_MODE`	Activate load balancing mode	on
`CC_PGPOOL_IGNORE_LEADING_WHITE_SPACE`	Ignore leading white spaces of each query	on
`CC_PGPOOL_READ_ONLY_FUNCTION_LIST`	Comma separated list of function names that don’t write to database (regexp are accepted)	’ '
`CC_PGPOOL_WRITE_FUNCTION_LIST`	Comma separated list of function names that write to database (regexp are accepted)	’ '
`CC_PGPOOL_PRIMARY_ROUTING_QUERY_PATTERN_LIST`	Semicolon separated list of query patterns that should be sent to primary node (regexp are accepted)	’ '
`CC_PGPOOL_DATABASE_REDIRECT_PREFERENCE_LIST`	Comma separated list of pairs of database and node id	’ '
`CC_PGPOOL_APP_NAME_REDIRECT_PREFERENCE_LIST`	Comma separated list of pairs of app name and node id	’ '
`CC_PGPOOL_ALLOW_SQL_COMMENTS`	If on, ignore SQL comments when judging if load balance or query cache is possible	off
`CC_PGPOOL_DISABLE_LOAD_BALANCE_ON_WRITE`	Load balance behavior when write query is issued in an explicit transaction. Valid values are “transaction”, “trans_transaction”, “dml_adaptive” or “always”	transaction
`CC_PGPOOL_DML_ADAPTIVE_OBJECT_RELATIONSHIP_LIST`	Comma separated list of object pairs	’ '
`CC_PGPOOL_STATEMENT_LEVEL_LOAD_BALANCE`	Enables statement level load balancing	off
`CC_PGPOOL_SR_CHECK_PERIOD`	Streaming replication check period	10
`CC_PGPOOL_DELAY_THRESHOLD`	Threshold before not dispatching query to standby node Unit is in bytes	10000000
`CC_PGPOOL_HEALTH_CHECK_PERIOD`	Health check period	20
`CC_PGPOOL_HEALTH_CHECK_TIMEOUT`	Health check timeout	20
`CC_PGPOOL_HEALTH_CHECK_MAX_RETRIES`	Maximum number of times to retry a failed health check before giving up	0
`CC_PGPOOL_HEALTH_CHECK_RETRY_DELAY`	Amount of time to wait (in seconds) between retries	1
`CC_PGPOOL_CONNECT_TIMEOUT`	Timeout value in milliseconds before giving up to connect to backend	10000
`CC_PGPOOL_MEMORY_CACHE_ENABLED`	Use the memory cache functionality	off
`CC_PGPOOL_MEMQCACHE_TOTAL_SIZE`	Total memory size in megabytes for storing memory cache	64
`CC_PGPOOL_MEMQCACHE_MAX_NUM_CACHE`	Total number of cache entries	1000000
`CC_PGPOOL_MEMQCACHE_EXPIRE`	Memory cache entry life time specified in seconds	0
`CC_PGPOOL_MEMQCACHE_AUTO_CACHE_INVALIDATION`	Invalidation of query cache is triggered by corresponding DDL/DML/DCL	on
`CC_PGPOOL_MEMQCACHE_MAXCACHE`	Maximum SELECT result size in kilobytes (must be smaller than MEMQCACHE_CACHE_BLOCK_SIZE)	400
`CC_PGPOOL_MEMQCACHE_CACHE_BLOCK_SIZE`	Cache block size in megabytes	1
`CC_PGPOOL_CACHE_SAFE_MEMQCACHE_TABLE_LIST`	Comma separated list of table names to memcache that don’t write to database (regexp are accepted)	’ '
`CC_PGPOOL_CACHE_UNSAFE_MEMQCACHE_TABLE_LIST`	Comma separated list of table names not to memcache that don’t write to database (regexp are accepted)	’ '

Elastic Stack

Elastic Stack Documentation

Name	Description	Default value	Read Only
`ELASTIC_APM_SERVER_URL`	URI to connect APM Server	Generated upon creation	✓
`ES_ADDON_APM_HOST`	APM Server hostname	Generated upon creation	✓
`ES_ADDON_APM_AUTH_TOKEN`	Authentication token to send metrics to APM Server	Generated upon creation	✓
`ELASTIC_APM_SECRET_TOKEN`	Authentication token to send metrics to APM Server	Generated upon creation	✓
`ES_ADDON_APM_USER`	Username credential used by APM Server to send metrics to Elasticsearch	Generated upon creation	✓
`ES_ADDON_APM_PASSWORD`	Password credential used by APM Server to send metrics to Elasticsearch	Generated upon creation	✓
`ES_ADDON_KIBANA_HOST`	Kibana hostname	Generated upon creation	✓
`ES_ADDON_KIBANA_USER`	Username credential used by Kibana to query Elasticsearch	Generated upon creation	✓
`ES_ADDON_KIBANA_PASSWORD`	Password credential used by Kibana to query Elasticsearch	Generated upon creation	✓
`ES_ADDON_URI`	URI to query Elasticsearch	Generated upon creation	✓
`ES_ADDON_HOST`	Elasticsearch hostname	Generated upon creation	✓
`ES_ADDON_USER`	Username credential to authenticate to Elasticsearch	Generated upon creation	✓
`ES_ADDON_PASSWORD`	Password credential to authenticate to Elasticsearch	Generated upon creation	✓
`ES_ADDON_VERSION`	ElasticSearch Version	7	✓

Blackfire

Name	Description	Default value	Read Only
`ENABLE_BLACKFIRE`	Enable the Blackfire extension	false
`CC_BLACKFIRE_SERVER_TOKEN`	The server token used to authenticate with Blackfire	’ '	✓
`CC_BLACKFIRE_SERVER_ID`	The server id used to authenticate with Blackfire	’ '	✓
`CC_BLACKFIRE_LOG_LEVEL`	Sets the verbosity of Agent’s log output	1	✓
`CC_BLACKFIRE_MEMORY_LIMIT`	Sets the maximum allowed RAM usage (megabytes) when ingesting traces	500	✓
`CC_BLACKFIRE_COLLECTOR`	Sets the URL of Blackfire’s data collector	https://blackfire.io	✓
`CC_BLACKFIRE_TIMEOUT`	Sets the Blackfire API connection timeout	15	✓
`CC_BLACKFIRE_STATSD`	Sets the statsd server to send agent’s statistics	’ '	✓
`CC_BLACKFIRE_STATSD_PREFIX`	Sets the statsd prefix to use when sending data	blackfire	✓

New Relic

Name	Description	Default value	Read Only
`NEW_RELIC_APP_NAME`	Contains the application name
`NEW_RELIC_LICENSE_KEY`	Contains your New Relic account license
`CC_NEWRELIC_BROWSER_MONITORING_AUTO_INSTRUMENT`
`CC_NEWRELIC_DISTRIBUTED_TRACING_ENABLED`
`CC_NEWRELIC_ERROR_COLLECTOR_ENABLED`
`CC_NEWRELIC_TRANSACTION_TRACER_ENABLED`
`CC_NEWRELIC_TRANSACTION_TRACER_RECORD_SQL`	obfuscated

Pulsar

Name	Description	Default value	Read Only
`ADDON_PULSAR_BINARY_URL`	The complete URL to use in your application	Generated upon creation	✓
`ADDON_PULSAR_BINARY_PORT`	The port to connect to the Pulsar Cluster	Generated upon creation	✓
`ADDON_PULSAR_HOSTNAME`	The host to connect to the Pulsar Cluster	Generated upon creation	✓
`ADDON_PULSAR_HTTP_URL`	The complete URL to connect with WebSocket	Generated upon creation	✓
`ADDON_PULSAR_HTTP_PORT`	The port to connect with WebSocket	Generated upon creation	✓
`ADDON_PULSAR_NAMESPACE`	Your add-on Pulsar ID	Generated upon creation	✓
`ADDON_PULSAR_TENANT`	Your Clever Cloud tenant ID	Generated upon creation	✓
`ADDON_PULSAR_TOKEN`	Your Biscuit authentication token	Generated upon creation	✓

Redis

Redis Documentation

Name	Default value	Read Only
`REDIS_HOST`	Generated upon creation	✓
`REDIS_PORT`	Generated upon creation	✓
`REDIS_PASSWORD`	Generated upon creation	✓

VPN

The VPN add-on provides a fixed-ip outgoing node. This can be used to work with services protected by ip address filtering. VPN_ADDON_* variables will be provided by Clever Cloud upon setup, the only configuration you have to provide is a list of CIDRs (eg. 1.2.3.0/24) for which you want the traffic to be routed through the exit node.

Name	Description	Default value
`CC_VPN_DNS_OVERRIDES`	Comma-separated list of DNS IP
`VPN_ADDON_CACRT`	Server CA certificate
`VPN_ADDON_CIPHER`	Cipher to use CIPHER, can be either {cipher_suite}:{hmac_alg} or only {cipher_suite}	DES-EDE3-CBC:SHA1
`VPN_ADDON_CRT`	Client certificate
`VPN_ADDON_DEVTYPE`	Kernel virtual interface kind to use (“tap” or “tun”)	tap
`VPN_ADDON_HOST`	Server host or IP address
`VPN_ADDON_KEY`	Client certificate private key
`VPN_ADDON_PORT`	Server port
`VPN_ADDON_TAKEY`	Pre-shared secret
`VPN_TARGETS`	Comma-separated list of CIDRs for which you want the traffic to be routed through the exit node

Last updated on July 15, 2025

Did this documentation help you ?

Configuration on all instances Product support on tools