cloudtask-launch-workers: Pre-allocate workers

SYNOPSIS

cloudtask-launch-workers [ -opts ] howmany ( path/to/list-of-ips | - )

DESCRIPTION

Launch new cloud workers and write list of addresses to a file.

cloudtask can launch and destroy cloud workers automatically when needed, but it can also accept a list of pre-allocated workers via the --workers option. This command launches workers and creates a list of addresses which you can feed to cloudtask. This allows you to manage the pool of workers for a task or series of tasks by hand, which may be desirable for various reasons (e.g., launching workers takes time, billing is performed at a 1-hour resolution, etc.)

ARGUMENTS

howmany := number of workers to launch

path/to/list-of-ips := file location to save IP addresses of launched workers. If '-' then print addresses to stdout.

OPTIONS

--hub-apikey: Hub APIKEY
   Environment: HUB_APIKEY | CLOUDTASK_HUB_APIKEY

--backup-id=ID: TurnKey Backup ID to restore on launch

--region: Region for instance launch (default: us-east-1)
   Regions:

    us-east-1 (Virginia, USA)
    us-west-1 (California, USA)
    eu-west-1 (Ireland, Europe)
    ap-southeast-1 (Singapore, Asia)

--size: Instance size (default: m1.small)
   Sizes:

    t1.micro (1 CPU core, 613M RAM, no tmp storage)
    m1.small (1 CPU core, 1.7G RAM, 160G tmp storage)
    c1.medium (2 CPU cores, 1.7G RAM, 350G tmp storage)

--type: Instance type <s3|ebs> (default: s3)
--label: Hub description label for all launched servers

USAGE EXAMPLES

    # create workers.txt file with list of new worker addresses
    cloudtask-launch-workers 10 workers.txt

    # append list of worker addresses to a file
    cloudtask-launch-workers 10 - >> workers.txt