Command line interface Guide

There are two kind of CLI tools for Synaps. Those are AWS CLI and cwutil of boto.

AWS CLI

Using AWS CLI is a good way to manage service.

You need to install AWS CLI and setup your credentials in the config file.

to install AWS CLI,

$ pip install awscli

Configuration file path is ~/.aws/config. It looks like below.

[default]
aws_access_key_id = ACCESS-KEY
aws_secret_access_key = SECRET-KEY
region = us-east-1

Usage example,

$ aws --endpoint-url http://synaps.service.url:3776/monitor cloudwatch list-metrics

For more details, see http://docs.aws.amazon.com/cli/latest/userguide

cwutil(boto) Guide

Using cwutil in the boto, you can also manage service. Use spcs branch of boto. which is added more CLI functions based on the original cwutil of boto.

to install cwutil(spcs branch),

$ pip install https://github.com/spcs/boto/archive/2.9.9.spcs.tar.gz

Usage example,

To list metrics.

$ cwutil ls

To list metrics in SPCS/SYNAPS namespace.

$ cwutil ls SPCS/SYNAPS

cwutil’s usage is,

Usage: cwutil [command]
        delete_alarm -
    Delete Alarm

        disable_alarm_actions -
    Enable alarm actions.
        alarmName: Alarm name to activate action.

        enable_alarm_actions -
    Enable alarm actions.
        alarmName: Alarm name to activate action.

        help -
    Print help message, optionally about a specific function

        history -
    List alarm history

    Action

        ls -
    List metrics, optionally filtering by a specific namespace
        namespace: Optional Namespace to filter on

        ls_alarm -
    Describe list of alarms.

        put -
    Publish custom metrics
        namespace: The namespace to use; values starting with "AWS/" are reserved
        metric_name: The name of the metric to update
        dimensions: The dimensions to use, formatted as Name:Value (such as QueueName:myQueue)
        value: The value to store, mutually exclusive with `statistics`
        statistics: The statistics to store, mutually exclusive with `value`
            (must specify all of "Minimum", "Maximum", "Sum", "SampleCount")
        timestamp: The timestamp of this measurement, default is current server time
        unit: Unit to track, default depends on what metric is being tracked

        put_alarm -
    Put MetricAlarm
        namespace:
        metric_name:
        dimensions:
        alarm_name:
        statistic: default "Average"
        comparison: default ">"
        threshold: default 90
        unit: default None
        period: default 60
        evaluation_periods: default 1 minute
        description: default None
        alarm_actions: default None
        insufficient_data_actions: default None
        ok_actions: default None

        stats -
    Lists the statistics for a specific metric
        namespace: The namespace to use, usually "AWS/EC2", "AWS/SQS", etc.
        metric_name: The name of the metric to track, pulled from `ls`
        dimensions: The dimensions to use, formatted as Name:Value (such as QueueName:myQueue)
        statistics: The statistics to measure, defaults to "Average"
             'Minimum', 'Maximum', 'Sum', 'Average', 'SampleCount'
        start_time: Start time, default to now - 1 day
        end_time: End time, default to now
        period: Period/interval for counts, default to 60 minutes
        unit: Unit to track, default depends on what metric is being tracked