Environments
Learn how to manage environments in Laravel Vapor.
Introduction
As you may have noticed, projects do not contain much information or resources themselves. All of your deployments and invoked commands are stored within environments. Each project may have as many environments as needed.
Typically, you will have an environment for “production”, and a “staging” environment for testing your application. However, don’t be afraid to create more environments for testing new features without interrupting your main staging environment.
Creating Environments
Environments may be created using the env
Vapor CLI command:
This command will add a new environment entry to your project’s vapor.yml
file that you may deploy when ready:
In addition to our native runtimes, Vapor supports Docker image deployments. If you would like an environment to use a Docker image runtime instead of the default Vapor runtime, use the --docker
option when creating your environment:
This command will create a my-environment.Dockerfile
file in your application’s root directory.
Opening Environments
Environments may be opened in your default browser using the Vapor CLI’s open
command:
Default Environment
When executing a Vapor CLI command, Vapor CLI uses the staging
environment by default:
However, within your application’s vapor.yml
file, you may define a default-environment
option to change the default environment for your project:
Environment Variables
Each environment contains a set of environment variables that provide crucial information to your application during execution, just like the variables present in your application’s local .env
file.
Vapor’s Default Environment Variables
Vapor automatically injects a variety of environment variables based on your environment’s configured cache, database, etc. As an example, by adding a cache
or database
key to your vapor.yml
file, Vapor will inject the necessary CACHE_*
and DB_*
environment variables.
Here is the full list of environment variables injected by Vapor on your environment:
.env Value | env() Value |
---|---|
APP_ENV | Environment name |
APP_DEBUG | False |
APP_LOG_LEVEL | Debug |
APP_URL | Vanity domain, or custom domain if exists |
ASSET_URL | CloudFront |
AWS_BUCKET | Storage resource if exists |
BROADCAST_DRIVER | Pusher |
CACHE_DRIVER | DynamoDB, or cache (Redis) resource if exists |
DB_* | Database (MySQL, Postgresql, etc) resource if exists |
DYNAMODB_CACHE_TABLE | vapor_cache |
FILESYSTEM_DISK | S3 |
FILESYSTEM_DRIVER | S3 |
FILESYSTEM_CLOUD | S3 |
LOG_CHANNEL | Stderr |
MAIL_DRIVER | LOG, but SES for environments with the name production |
MAIL_MAILER | LOG, but SES for environments with the name production |
MAIL_FROM_ADDRESS | [email protected] or [email protected] if exists |
MAIL_FROM_NAME | your_project_name |
MIX_URL | CloudFront |
QUEUE_CONNECTION | SQS |
SCHEDULE_CACHE_DRIVER | DynamoDB |
SESSION_DRIVER | Cookie |
You will not see these environment variables when you manage your environment via Vapor CLI or Vapor UI, and any variables you manually define will override Vapor’s automatically injected variables.
Updating Environment Variables
You may update an environment’s variables via the Vapor UI or using the env:pull
and env:push
CLI commands. The env:pull
command may be used to pull down an environment file for a given environment:
Once this command has been executed, a .env.{environment}
file will be placed in your application’s root directory. To update the environment’s variables, simply open and edit this file. When you are done editing the variables, use the env:push
command to push the variables back to Vapor:
If you are using the DotEnv library’s variable nesting feature to reference default environment variables that Vapor is injecting, you should replace these references with literal values instead. Since Vapor’s injected environment variables do not belong to the environment file, they can not be referenced using the nesting feature.
After updating an environment’s variables, the new variables will not be utilized until the application is deployed again. In addition, when rolling back to a previous deployment, Vapor will use the variables as they existed at the time the deployment you’re rolling back to was originally deployed.
Due to AWS Lambda limitations, your environment variables may only be 4kb in total. To accommodate Vapor’s own injection of environment variables, users are limited to 2,000 characters of environment variables. You should use encrypted environment files in place of or in addition to environment variables if you exceed this limit.
Reserved Environment Variables
The following environment variables are reserved and may not be added to your environment:
_HANDLER
AWS_ACCESS_KEY_ID
AWS_DEFAULT_REGION
AWS_EXECUTION_ENV
AWS_LAMBDA_FUNCTION_MEMORY_SIZE
AWS_LAMBDA_FUNCTION_NAME
AWS_LAMBDA_FUNCTION_VERSION
AWS_LAMBDA_LOG_GROUP_NAME
AWS_LAMBDA_LOG_STREAM_NAME
AWS_LAMBDA_RUNTIME_API
AWS_REGION
AWS_SECRET_ACCESS_KEY
AWS_SESSION_TOKEN
LAMBDA_RUNTIME_DIR
LAMBDA_TASK_ROOT
TZ
In addition, environment variables should not contain AWS_ACCESS_KEY_ID
, AWS_SECRET_ACCESS_KEY
, or AWS_SESSION_TOKEN
in their names. For example: MY_SERVICE_AWS_SECRET_ACCESS_KEY
.
Encrypted Environment Files
Vapor provides built-in support for Laravel’s encrypted environment files. If Vapor discovers an encrypted environment file while booting your application, it will automatically attempt to decrypt it and inject the resulting variables into the runtime.
To leverage this feature, you must first ensure an encrypted environment file is present at the root of your application during deployment. For example, deploying the production
environment requires a file called .env.production.encrypted
to be present at the root of your application.
Additionally, you should ensure the decryption key is available in the runtime by defining it as the LARAVEL_ENV_ENCRYPTION_KEY
environment variable via the Vapor UI or CLI.
Utilizing encrypted environment files requires your application to be running Laravel >= v9.37 and Vapor Core >= v2.26.
Passport Keys
You may easily add your project’s Passport keys to your encrypted environment file using the env:passport
CLI command:
The env:passport
command will append the contents of your local Passport keys to your .env.production
file. When the command completes, you should re-encrypt the environment file using Laravel’s env:encrypt
command and redeploy your project in order for the changes to take effect.
Maintenance Mode
When deploying a Laravel application using a traditional VPS like those managed by Laravel Forge, you may have used the php artisan down
command to place your application in “maintenance mode”. To place a Vapor environment in maintenance mode, you may use the Vapor UI or the down
CLI command:
To remove an environment from maintenance mode, you may use the up
command:
When an environment is in maintenance mode, the environment’s custom domain will display a maintenance mode splash screen; however, you may still access the environment via its “vanity URL”
Customizing The Maintenance Mode Screen
You may customize the maintenance mode splash screen for your application by placing a 503.html
file in your application’s root directory. In addition, you may also place a 503.json
file in your application’s root directory for requests asking for JSON responses.
Bypassing Maintenance Mode
You may find it useful to be able to access your site on its custom domain rather than the vanity domain while in maintenance mode. To accomplish this, you may provide a secret when invoking the down
command:
You may then access your application using the secret key as the URL path:
Commands
Commands allow you to execute an arbitrary Artisan command against an environment. You may issue a command via the Vapor UI or using the command
CLI command. The command
command will prompt you for the Artisan command you would like to run:
By default, your command will timeout after one minute. You can configure the timeout of your CLI commands using the cli-timeout
option within your vapor.yml
file. This option allows you to specify the maximum number of seconds a CLI command should be allowed to run:
Sometimes you may need to run the previous command again. To accomplish this, you may use the Vapor UI or the command:again
CLI command:
Memory
Vapor (via AWS Lambda) allocates CPU power to your Lambda function in proportion to the amount of memory configured for the application. You may increase or decrease the configured memory using the memory
option in your environment’s vapor.yml
configuration:
When configuring the memory for your Lambda function, you may define a value between 128 MB and 10,240 MB in 64-MB increments.
Concurrency
By default, Vapor will allow your application to process web requests at max concurrency, which is typically 1,000 requests executing at the same time at any given moment across all of your AWS Lambda functions in a given region. If you would like to reduce the maximum web concurrency, you may define the concurrency
option in the environment’s vapor.yml
configuration. Additionally, if you need more than 1,000 concurrent requests, you can submit a limit increase request in the AWS Support Center console.
While maximum performance is certainly appealing, in some situations it may make sense to set this value to the maximum concurrency you reasonably expect for your particular application. Otherwise, a DDoS attack against your application could result in larger than expected AWS costs:
Prewarming
By default, when an environment is deployed, the first request it receives may encounter “cold starts”. These requests typically incur a penalty of a few seconds while AWS loads a serverless container to serve the request. Once a request has been served by that container, it is typically kept warm to serve further requests with no delay.
To mitigate “cold starts” after a fresh deployment, Vapor allows you to define a warm
configuration value for an environment in your vapor.yml
file. The warm
value represents how many serverless containers Vapor will “pre-warm” by making concurrent requests to the newly deployed application before it is activated for public accessibility. Vapor will continue to pre-warm this many containers every 5 minutes while the application is deployed so the specified number of containers are always ready to serve requests:
Firewall
You may instruct Vapor to automatically configure a firewall that provides basic protection against denial-of-service attacks targeting your environment, as well as protection against pervasive bot traffic that can consume your environment’s resources.
Before getting started, keep in mind that Vapor’s managed firewall inspects requests using the IP address from the web request origin. Therefore, this feature should only be used if the requests are not already being reversed proxied through a service such as Cloudflare. If you are already using a reverse proxy, you should not use this feature.
You may use Vapor’s managed firewall by defining the firewall
configuration option within your application’s vapor.yml
file:
rate-limit
When using the rate-limit
option, Vapor’s managed firewall tracks the rate of requests for each originating IP address and blocks IPs with request rates over the given rate-limit
value. In the example above, if the request count for an IP address exceeds 1,000 requests in any 5-minute time span then the firewall will temporarily block requests from that IP address with the 403 Forbidden
HTTP status code.
bot-control
When using the bot-control
option, Vapor’s managed firewall blocks requests from pervasive bots, such as scrapers or search engines. You may customize the “category” of requests the bot-control
should block by providing an array
of categories within your application’s vapor.yml
file:
Here is the list of available categories you may use:
Category | Description |
---|---|
CategoryAdvertising | Blocks requests from bots that are used for advertising purposes. |
CategoryArchiver | Blocks requests from bots that are used for archiving purposes. |
CategoryContentFetcher | Blocks requests from bots that are fetching content on behalf of an end-user. |
CategoryHttpLibrary | Blocks requests from HTTP libraries that are often used by bots. |
CategoryLinkChecker | Blocks requests from bots that check for broken links. |
CategoryMiscellaneous | Blocks requests from miscellaneous bots. |
CategoryMonitoring | Blocks requests from bots that are used for monitoring purposes. |
CategoryScrapingFramework | Blocks requests from web scraping frameworks. |
CategorySecurity | Blocks requests from security-related bots. |
CategorySeo | Blocks requests from bots that are used for search engine optimization. |
CategorySocialMedia | Blocks requests from bots that are used by social media platforms to provide content summaries. Verified social media bots are not blocked. |
CategorySearchEngine | Blocks requests from search engine bots. Verified search engines are not blocked. |
SignalAutomatedBrowser | Blocks requests with indications of an automated web browser. |
SignalKnownBotDataCenter | Blocks requests from data centers that are typically used by bots. |
SignalNonBrowserUserAgent | Blocks requests with user-agent strings that don’t seem to be from a web browser. |
Due to AWS limitations, Vapor’s managed firewall does not support API Gateway v2.
Behind the scenes, Vapor’s managed firewall uses Amazon WAF, creating a Web ACL with one rate-based rule per Vapor environment. Feel free to check out the AWS WAF documentation for more information about WAF and its pricing.
Timeout
By default, Vapor will limit web request execution time to 10 seconds. If you would like to change the timeout value, you may add a timeout
value (in seconds) to the environment’s configuration. API Gateway has a maximum timeout of 30 seconds. Should you require a longer request duration, you may utilize a load balancer. Note that AWS does not allow Lambda executions to process for more than 15 minutes:
Remember, you can use the “Logs” panel of the Vapor dashboard to search for timeout occurrences.
Scheduler
Vapor automatically configures Laravel’s task scheduler and instructs it to use the DynamoDB cache driver to avoid overlapping tasks, so no other configuration is required to begin leveraging Laravel’s scheduled task feature.
Due to the serverless nature of Vapor, you should avoid using the runInBackground
method when scheduling jobs. Doing so may prevent other tasks from running in the event the Lambda container shuts down before the current task completes.
If you would like to disable the scheduler, you may set an environment’s scheduler
option to false
:
Due to Vapor limitations, log messages from scheduled tasks will not appear in AWS CloudWatch or Vapor UI. As a workaround, you should dispatch a queued job from your scheduled tasks and write log messages from your queued job.
Sub-Minute Scheduled Tasks
Due to AWS limitations, there is no guarantee the scheduler will be invoked at the very beginning of any given minute. Therefore, you may find that, by default, sub-minute tasks scheduled early in the schedule:run
process do not run as expected and those which run later in the schedule may not start at the expected time. For example, when scheduling a command using everyThirtySeconds
and assuming the scheduler is invoked by AWS at 12:00:10, you should expect your command to run at 12:00:10 and 12:00:40.
To work around these limitations for sub-minute tasks, you may enable Vapor’s own scheduler engine by adding scheduler: sub-minute
to your application’s vapor.yml
file:
When Vapor’s sub-minute scheduler is enabled, Vapor runs its own scheduler command which waits until the beginning of every minute before calling schedule:run
. Therefore, when enabling this option, you should also ensure your cli-timeout
is >= 120 to allow enough time to both invoke Laravel’s scheduler and allow it to complete all sub-minute tasks. You should also ensure cli-concurrency
is >= 2 to allow for the possibility of overlapping invocations of Vapor’s scheduler.
In addition, the runInBackground
option is not supported on Vapor; therefore, you may find some of your tasks are blocked from running if the previous task runs longer than expected.
Laravel provides a clean, simple email API. And, by default, Vapor will automatically configure your environment to use Amazon SES as the default mail driver by injecting the proper Laravel environment variables during deployment. Of course, you may change the default mail driver by defining a different value for the MAIL_MAILER
environment variable.
If you plan to use Amazon SES as your application’s mail service, you should first ensure the MAIL_MAILER
environment variable is set and it contains the value ses
. Next, you should attach a domain to your environment. Once attached and deployed, Vapor will automatically update the domain’s DNS records so Amazon SES can validate the domain and configure DKIM. These DNS records are necessary to protect your reputation as a sender.
If you self-manage your domain’s DNS records, Vapor will not be able to update the domain’s DNS records automatically. Therefore, you should run the vapor record:list domain-name.com
command to view the records that Vapor indicates are required for your domain and update your DNS records accordingly.
By default, Vapor configures the “From Address” and “From Name” Laravel configuration settings with [email protected]
and Your Project Name
, respectively. Of course, you’re free to modify these values by defining new values for the MAIL_FROM_ADDRESS
and MAIL_FROM_NAME
environment variables.
Finally, if you haven’t used Amazon SES before, your SES account will be in “sandbox” mode. Sandbox mode only allows you to send emails to manually verified domains. To move out of SES sandbox mode, follow these instructions: https://docs.aws.amazon.com/ses/latest/DeveloperGuide/request-production-access.html.
Metrics
A variety of environment performance metrics may be found in the Vapor UI or using the metrics
CLI command:
Alarms
You may configure alarms for all environment metrics using the Vapor UI. These alarms will notify you via the notification method of your choice when an alarm’s configured threshold is broken and when an alarm recovers.
Runtime
The runtime
configuration option allows you to specify the runtime a given environment runs on.
Native Runtimes
The currently supported native runtimes are php-8.1:al2
, php-8.2:al2
, php-8.2:al2-arm
, php-8.3:al2
, and php-8.3:al2-arm
. The runtimes that are suffixed with al2
use Amazon Linux 2 while those without the suffix use Amazon Linux 1. Runtimes without the -arm
suffix run on x86 artchitecture whereas those suffixed with -arm
run on Arm architecture. Arm provides performance benefits and cost savings compared to its x86 equivalent:
Using Amazon Linux 2 (php-8.1:al2
, php-8.2:al2
, php-8.2:al2-arm
, php-8.3:al2
, php-8.3:al2-arm
) is highly recommended, as Amazon Linux 1 is no longer maintained as of December 31, 2020.
The following limitations apply to Vapor native runtimes:
- The application size, including the runtime itself, must not exceed 250MB.
- Additional PHP extensions or libraries (such as
imagick
) can not be installed.
Customizing Core php.ini
Directives
To customize core php.ini
directives on our native runtimes, create a php.ini
file in the php/conf.d
directory of your project with the desired directives. For example, to change the upload_max_filesize
, add the following line to your application’s php.ini
file:
After deploying, the new directive will take effect in your Vapor environment.
Docker Runtimes
Docker based runtimes allow you to package and deploy applications up to 10GB in size and allow you to install additional PHP extensions or libraries by updating the environment’s corresponding .Dockerfile
. For every new Docker based environment, Vapor creates a .Dockerfile
file within your application that uses one of Vapor’s base images as a starting point for building your image. All of Vapor’s Docker images are based on Alpine Linux:
Vapor’s Docker runtimes also support Arm architecture, which provides increased performance and cost savings. If you want to take advantage of Arm architecture, you should update your base image as follows:
If you would like to use a Docker image instead of the Vapor native runtimes, you should set your vapor.yml
configuration file’s runtime
option to docker
(for x86 architecture) or docker-arm
(for Arm architecture):
To avoid errors when compiling an image for the docker-arm
runtime, it is essential to ensure that the environment where the image is built is compatible with arm64
based architecture. To avoid such issues, it may be necessary to emulate the arm64
architecture using a tool such as QEMU. The Docker documentation provides further guidance on this topic.
If you would like to rename the Dockerfile or use a shared Dockerfile across multiple environments, you may define a dockerfile
option within your vapor.yml
file:
When migrating an existing environment to a Docker runtime, please keep in mind that you won’t be able to revert that environment to the default Vapor Lambda runtime later. For that reason, you may want to create an environment for testing the Docker runtime first.
Vapor will build, tag, and publish your environment’s image during your deployments; therefore, you should ensure that you have installed Docker on your local machine. Vapor’s base Docker images are:
laravelphp/vapor:php74
laravelphp/vapor:php80
laravelphp/vapor:php81
laravelphp/vapor:php82
laravelphp/vapor:php82-arm
laravelphp/vapor:php83
laravelphp/vapor:php83-arm
Of course, you are free to modify your environment’s Dockerfile
to install additional dependencies or PHP extensions. Here are a few examples:
Customizing Core php.ini
Directives
To customize core php.ini
directives on our Docker runtimes, create a php.ini
file in the root directory of your project with the desired directives. For example, to change the upload_max_filesize
directive, add the following line to your php.ini
file:
Next, in your Dockerfile, include a new COPY
command that copies the local php.ini
file into the Docker image:
After deploying, the new directive will take effect in your Vapor environment.
Docker Build Arguments
When using Docker, is common to use ARG
instructions in .Dockerfile
files to define build-time variables:
You may set Docker build arguments via the docker-build-args
configuration option within your application’s vapor.yml
file:
Alternatively, you may provide one or multiple --build-arg
options to the deploy
Vapor CLI command to specify the values of the build arguments:
Octane
By default, when Lambda receives HTTP requests, Vapor sends those requests synchronously to PHP-FPM using the FastCGI Protocol. This means every incoming request spawns a PHP-FPM worker and boots your application. This approach ensures Vapor behaves exactly like a traditional web server.
If you wish to reduce the overhead involved in using PHP-FPM, you may opt-in to Laravel Octane support on Vapor. Octane can increase your application’s performance by booting your application once, keeping it in memory, and then feeding that same application instance requests as they are received.
To get started, install Laravel Octane in your project. After installing Octane, don’t forget to review important Octane documentation topics such as dependency injection and managing memory leaks.
Finally, you may instruct Vapor to use Octane by setting the octane
configuration option within your application’s vapor.yml
file:
In addition, if your project uses a database, you may use the octane-database-session-persist
and octane-database-session-ttl
options to instruct Octane that database connections should be reused between requests:
- The
octane-database-session-persist
option indicates that database connections should persist between requests. The main purpose of this option is to reduce the overhead involved on creating a database connection on each request. - The
octane-database-session-ttl
option allows specifying the time (in seconds) the Lambda container should stay connected to the database when the Lambda container is not being used. This option is only available for MySQL fixed size databases.
We recommended that you specify an octane-database-session-ttl
value; otherwise, the Lambda container will stay connected to your database until the Lambda container gets destroyed. This may take several minutes and may result in your database becoming overwhelmed with active connections.
Gateway Versions
By default, Vapor routes HTTP traffic to your serverless applications using AWS API Gateway v1 (REST APIs). Your application may run on either API Gateway v1 (REST APIs) or API Gateway v2 (HTTP APIs). By default, applications deploy using API Gateway v1 as it provides a fuller feature set such as Vapor’s managed Firewall, and more.
However, API Gateway v2 offers a cost reduction per million requests to your application (3.50 per million). If you would like to use API Gateway v2, you may specify the gateway-version
configuration option for a given environment in your vapor.yml
file:
HTTP to HTTPS Redirection With API Gateway v2
If you choose to use API Gateway v2 and would like to support HTTP to HTTPS redirection, we currently suggest using Cloudflare as an external DNS provider for your Vapor application. Cloudflare not only provides DNS, but serves as a reverse proxy to your application and features an option for automatic HTTP to HTTPS redirection.
After a Vapor deployment is completed, Vapor will provide you with CNAME records for the domain(s) associated with your environment. These records will point the domain to your Lambda application. You should manually add these values as CNAME records within the Cloudflare DNS dashboard.
In addition, when using Cloudflare, you should set your Cloudflare SSL / TLS mode to “Full”. The “Always Use HTTPS” configuration option may be found under the SSL / TLS menu’s “Edge Certificates” tab.
Custom VPCs
If you want to place your Lambda functions within a VPC that is not managed by Vapor, you may specify the subnets
and security-groups
configuration options for a given environment in your vapor.yml
file:
Deleting Environments
Environments may be deleted via the Vapor UI or using the env:delete
CLI command: