@capriza/far
v2.4.2
Published
CLI for deploying ApproveSimple connectors to AWS fargate
Downloads
71
Maintainers
Keywords
Readme
far
far
is a CLI used for deploying an ApproveSimple connector running inside a docker container to AWS Fargate.
far
automates the process of building the docker image, setting up the AWS cluster, defining the service and creating task definitions.
Installation
npm install -g @capriza/far
Prerequesits
- A user in ApproveSimple Admin Console with deployment privileges
- Docker installed and running. Get Docker from here
Login
Running commands in far
requires Org Admin credentials to ApproveSimple. These credentials are used to obtain and then provide far
with temporary AWS credentials. The temporary credentials expire after 15 minutes.
Interactive Login
Every time you run a far
command, the default browser will open and you will need to login to the Admin Console. Once logged-in, an automatic process of obtainnig AWS credentials will occur. Once obtained, the temporary credentials will be automaticalyy delivered to far
to continue the command
Unattended Login
An ApproveSimple API Key and Secret are read from a local file, specified by the --creds
command line option. The creds file must be a JSON with apiKey
and secret
keys.
{
"apiKey": "YOUR USER API KEY",
"secret": "THE API KEY SECRET"
}
Getting Started
Assume a connector named my-connector
with a Dockerfile you want to deploy to region eu-west-1
of AWS.
Initialize
Change into the my-connector
directory and run far init --env test
to create and initialize a farconfig file for the test
environment.
$ cd my-connector
$ far init --env test
Service name ["my-connector"]:
AWS Cluster name ["my-connector"]:
AWS region ["eu-west-1"]:
Wrote configuration to farconfig.test
Deploy
Run the far deploy --env test
command to build your docker image, deploy it to AWS and run it as a service.
$ far deploy --env test
test • my-connector • deploy
<... LOGIN THROUGH BROWSER ...>
login ok
Building docker image from Dockerfile
<... OMITTED ...>
At this point your docker image is running as a service named my-connector-test
on AWS fargate.
View Logs
Run the far logs
command to view and follow the logs of our service. Logs are everything that's printed to the console by your docker image.
$ far logs --env test
test • my-connector • logs
login ok
log line 1...
log line 2...
log line 3...
Usage
far <command> [options]
far requires that a farconfig.<env>
file exists in the working directory where far is executed (except for far init
), where env
denotes the environment of the deployment, examples of which include "test" and "prod". You must specify the required environment with every far command.
far operates in the cotext of the current working directory, so it's necessary to cd into the directory containing the farconfig file and to execute the far
command from within the directory.
Commands
To see the list of available commands run far help
. To see help and available options for a specific command run far <command> help
.
$ far help
far <command>
Commands:
far init initialize a new far configuration file in the current directory
far deploy deploy the local docker image and run it as a service
far update update an existing service (cpu, memory, secrets, etc.) without deploying a new docker image
far scale <count> scale up or down the number of instances running in the service (0 will stop the service)
far stop stop all container instances running in the service (same as 'far scale 0')
far status view service status information
far logs view all container instances logs in the service
far terminate terminate the environment and all its resources
Options:
--help Show help [boolean]
--version Show version number [boolean]
--env configuration file environment (e.g. specifying "prod" will load "farconfig.prod")
--name service name
--region aws region
--cluster aws cluster name
--creds file containing AS api key and secret. if specified, will not open browser for interactive login
Common Options
env
- the environment of deploymentcreds
- file containing ApproveSimple api key and secret. if specified, will not open the browser for interactive loginname
- the name of the service. default is the current working directory name.region
- the AWS region where the service is deployed.cluster
- the name if the AWS cluster where the service is deployed.type
- the type of cluster (fargate or ec2). default is fargate.
init
$ far init help
far init
initialize a new far configuration file in the current directory
Options:
--help Show help [boolean]
--version Show version number [boolean]
--env configuration file environment (e.g. specifying "prod" will load "farconfig.prod")
--name service name
--region aws region
--cluster aws cluster name
--creds file containing AS api key and secret. if specified, will not open browser for interactive login
deploy
When deploying, the image is tagged with value of the tag
option. If the tag option is not provided,
an attempt is made to use the version
value from the local package.json.
If the tag already exists in the remote AWS respository then the deploy fails.
If draft
is set to true
then the image tag is suffixed with the current timsestamp so that every time a deploy is executed a unqiue image tag is created for the deployment. It is recommended that for production deployments draft
be set to false
.
It is possible to pass custom docker build options via the dockerBuild
options set. Available docker build options are listed
at Docker Engine API Documentation, under the section "Build image from a Dockerfile".
dockerBuild:
- buildargs: '{ "ENV": "MY_VAR_1"}'
- memory: '4m'
$ far deploy help
far deploy
deploy the local docker image and run it as a service
Options:
--help Show help [boolean]
--version Show version number [boolean]
--env configuration file environment (e.g. specifying "prod" will load "farconfig.prod")
--name service name
--region aws region
--cluster aws cluster name
--tag image tag. default tage name is the version in the package.json file
--repository the docker repository storing the docker image
--draft draft mode. in this mode every image tag is suffixed with a timestamp [boolean]
--subnets the subnets to associate with the deployment [array]
--securityGroups the security groups to associate with the deployment [array]
--cpu vCPU reservation (256|512|1024|2048|4096) [number]
--memory memory reservation (aligned to vCPU) [number]
--secrets list of files to upload to AWS Secrets Manager [array]
--variables environment variables to provide to the service (in the form of name=value) [array]
--logRetention number of days for log retention in CloudWatchLogs (default is 90) [number]
--type deployment type (fargate|ec2)
--role IAM role that containers in this task assume (default is "far-tasks", created automatically)
--dockerfile docker file to use for building the image
--count number of container instances to run in the service
Logs
Logs are automatically stored in AWS Cloud Watch Logs under
the log group name awslogs-far-{name}-{env}
and log stream awslogs-{name}-{env}
in the region of the deployment.
You may specify a custom datetime format for distinguishing between log events by setting the logDatetimeFormat
configuration option (format option can be found here).
Log retention is 90 days by default, but you may specify a different retention policy through the logRetention
configuration option. Possible values are 1, 3, 5, 7, 14, 30, 60, 90, 120, 150, 180, 365, 400, 545, 731, 1827, and 3653 days.
Environment Variables
far
automatically defines three environment variables that are available to the running service:
AWS_REGION
- the AWS region that the service is deployed inAWS_CLUSTER
- the cluster name running the serviceMS_NAME
- the service name suffixed with the environment name, for example, if the service name ismy-connector
and the env isprod
then MS_NAME valus ismy-connector-prod
You may specify additional environment variables in the configuration file under the variables
option
Secrets
The secrets
configuration option is an array list of files to securly make available to the service via the AWS Secrets Manager.
During the deploy
command, far
reads the contents of the secret file and uploads it to the secrets manager in the account,
giving it the name ${MS_NAME}/<SECRET_ID>
. You may specify just the file name and the secret id will be the same as the file name, or specify the secret id and the path to the file containig the secret.
secrets:
# this creates a secret with the name "secrets/mySecretFile.json" with content from that file
- secrets/mySecretFile.json
# this creates a secret with the name "secrets/myOtherFile.json" and content from file "../path/to/other/file.json"
- secrets/myOtherFile.json: ../path/to/other/file.json
Files that are specified as secrets
in the far configuration file are automatically excluded from the built docker image.
In addition, files named caprizaConfig.json
and systemConfig.json
are automativally excluded from the built docker image regardless of wether they are specified as secrets.
You can specify additional files/directories to exclude from the docker image by listing them in your local .dockerignore
file.
Subnets
The subnets
configuration option is an array of subnet ids to assign to the service. You may also specify awsDefaultVpc
to have far automatically use the subnets of the default VPC. If no subnets are specified, far will search amd use all the subnets that have a tag named farSubnet
with a value of true
.
A public IP can be assigned to the deployed container by:
- Setting the
assignPublicIp
configuration option totrue
- Specifying
awsDefaultVpc
as thesubnets
configuration option
Security Groups
The securityGroups
configuration option is an array of security group ids to assign to the service. If no security groups are specified, far searches for all the security groups that have a tag named farSecurityGroup
with a value of true
. If at least one such security group is found, all tagged security groups will be used. If no tagged security group is found, far will not assign any security group to the service.
update
The update command updates an existing service with new settings.
It performs the same steps as the deploy
command except for building and pushing a new docker image.
$ far update help
far update
update an existing service (cpu, memory, secrets, etc.) without deploying a new docker image
Options:
--help Show help [boolean]
--version Show version number [boolean]
--env configuration file environment (e.g. specifying "prod" will load "farconfig.prod")
--name service name
--region aws region
--cluster aws cluster name
--tag image tag. default tage name is the version in the package.json file
--repository the docker repository storing the docker image
--draft draft mode. in this mode every image tag is suffixed with a timestamp [boolean]
--subnets the subnets to associate with the deployment [array]
--securityGroups the security groups to associate with the deployment [array]
--cpu vCPU reservation (256|512|1024|2048|4096) [number]
--memory memory reservation (aligned to vCPU) [number]
--secrets list of files with secrets to make available to the service container instances [array]
--variables environment variables to provide to the service (in the form of name=value) [array]
--logRetention number of days for log retention in CloudWatchLogs (default is 90) [number]
--type deployment type (fargate|ec2)
--role IAM role that containers in this task assume (default is "far-tasks", created automatically)
scale
$ far scale help
far scale <count>
scale up or down the number of instances running in the service (0 will stop the service)
Positionals:
count number of container instances to run in the service [number] [required]
Options:
--help Show help [boolean]
--version Show version number [boolean]
--env configuration file environment (e.g. specifying "prod" will load "farconfig.prod")
--name service name
--region aws region
--cluster aws cluster name
--creds file containing AS api key and secret. if specified, will not open browser for interactive login
stop
$ far stop help
far stop
stop all container instances running in the service (same as 'scale 0')
Options:
--help Show help [boolean]
--version Show version number [boolean]
--env configuration file environment (e.g. specifying "prod" will load "farconfig.prod")
--name service name
--region aws region
--cluster aws cluster name
--creds file containing AS api key and secret. if specified, will not open browser for interactive login
status
$ far status help
far status
view service status information
Options:
--help Show help [boolean]
--version Show version number [boolean]
--env configuration file environment (e.g. specifying "prod" will load "farconfig.prod")
--name service name
--region aws region
--cluster aws cluster name
--creds file containing AS api key and secret. if specified, will not open browser for interactive login
--events number of events to display (0-100)
logs
$ far logs help
far logs
view all container instances logs in the service
Options:
--help Show help [boolean]
--version Show version number [boolean]
--env configuration file environment (e.g. specifying "prod" will load "farconfig.prod")
--name service name
--region aws region
--cluster aws cluster name
--creds file containing AS api key and secret. if specified, will not open browser for interactive login
--start start time of logs to view. default is current time
--end end time of logs to view. if not provided, logs will stream continuously
terminate
$ far terminate help
far terminate
terminate the environment and all its resources
Options:
--help Show help [boolean]
--version Show version number [boolean]
--env configuration file environment (e.g. specifying "prod" will load "farconfig.prod")
--name service name
--region aws region
--cluster aws cluster name
--creds file containing AS api key and secret. if specified, will not open browser for interactive login
License
The MIT License (MIT)