Get human readable costs data from AWS account.
By default, bring each day cost since a month ago. You can custom the period of fetching the data.
Go to the project's root directory and then types:
pip install -r requirements.txt
pip install .
In the command line, type:
awscosts --profile <your_aws_profile_name>
By default, the output is setted to get data costs from a month ago until the current date. But if you want to set a different starting period, do:
awscosts --profile <your_aws_profile_name> --start-time 2022-02-16
Maybe you want to plot the day-by-day costs in a graph. You can change the data output to csv format, so you can use in your own application as data input or to plot a graph i Libreoffice.
awscosts --profile <your_aws_profile_name> --format csv
If you want to get costs from a specific service, do:
awscosts --profile <your_aws_profile_name> --type <service>
Allowed services are:
- ec2
- s3
- workmail
- tax
- cloudwatch
- sns
- route53
- rds
- codecommit
- dynamodb
- ce
- compute
- cloudfront
- efs
- kms
You can get prices from several services at once separated by comma:
awscosts --profile <your_aws_profile_name> --type workmail,sns
Good to know the meaning of each service (following the results of command: aws ce get-dimension-values --dimension SERVICE --time-period Start=2022-04-01,End=2022-04-25):
- ec2: EC2 - Other
- ce: AWS Cost Explorer
- compute: Amazon Elastic Compute Cloud - Compute
- s3: Amazon Simple Storage Service
- workmail: AmazonWorkMail
- tax: Tax
- cloudwatch: AmazonCloudWatch
- sns: Amazon Simple Notification Service
- route53: Amazon Route 53
- rds: Amazon Relational Database Service
- dynamodb: Amazon DynamoDB,
- kms: AWS Key Management Service,
- cloudfront: Amazon CloudFront,
- efs: Amazon Elastic File System
You can at once check all costs for each of above service with the following:
awscosts --profile costs_profile --spread-services
Then you can see something like this:
2022-04-23:
s3: 0.3582866845
workmail: 0.4000000032
tax: 0.0
cloudwatch: 0.0
sns: 0.0
route53: 0.0001528
rds: 1.0765918608
codecommit: 0.0
dynamodb: 0.0
Total in 2022-04-23: 1.8350313485
2022-04-24:
s3: 0.3591142326
workmail: 0.4000000032
tax: 0.0
cloudwatch: 0.0
sns: 0.0
route53: 6.52e-05
rds: 1.0757114117
codecommit: 0.0
dynamodb: 0.0
Total in 2022-04-24: 1.8348908475
BUT CAUTION: Several api calls to AWS may be done at once with the parameter --spread-services. At the time of this writting, AWS charges US$ 0.01 for each api call. So use this parameter with moderation, or you will ended up with unexpected charge rise.
ANOTHER WARNING: Still this way not all services that consumes money are considered. Run this command without --spread-services and probably you still see some difference. The consumption sum may still be underestimated compared to the output that considers all costs from a period.
The script search for a environment variable called CONNECTIONSTRING. If no string are found, the fetched data won`t be cached and no further research over the data is possible. Remember that asking for costs in AWS is relativelly expensive. So consecutives calls with same parameters is better with cache to save some resource. Furthermore, the storaged data can be latter be retrieved, analysed and receve further processing.
The connection string can be of two formats:
One for sqlite: in this case, the connection string must exists in the format:
sqlite:<connection_string_sqlite>
Qhere <connection_string_sqlite> is just a file path for the sqlite database (which may even not exists, it is immediately created if not exists).
If the connection string is to connect to a database, the connection string must be in the folowing format:
mysql:<connection_string_mysql>
Where <connection_string_mysql> mus follows the following format:
USER=my_user_name;PASSWORD=the_secret_passowrd;HOST=database_host;DATABASE_NAME=the_database_name;PORT=the_port_which_is_optional
The environment also can have MONGOCONNECTIONSTRING, in case you want to transfer all data stored on sqlite to a mongodb to a better visualization. There's the command line utility that is installed together to the current cli, called awscosts_tomongo. Execute it, and all data is passed to mongodb.