-
-
Save rcoup/421559 to your computer and use it in GitHub Desktop.
#!/usr/bin/env python | |
""" | |
A script to query the Amazon Web Services usage reports programmatically. | |
Ideally this wouldn't exist, and Amazon would provide an API we can use | |
instead, but hey - that's life. | |
Basically takes your AWS account username and password, logs into the | |
website as you, and grabs the data out. Always gets the 'All Usage Types' | |
report for the specified service. | |
Requirements: | |
* Mechanize: http://wwwsearch.sourceforge.net/mechanize/ | |
You can install this via pip/easy_install | |
Run with -h to see the available options. | |
""" | |
import os | |
import sys | |
from datetime import date | |
import time | |
import mechanize | |
FORMATS = ('xml', 'csv') | |
PERIODS = ('hours', 'days', 'months') | |
SERVICES = ('AmazonS3', 'AmazonEC2', 'AmazonCloudFront', 'AmazonSimpleDB', 'AWSQueueService', 'IngestionService', 'AmazonVPC',) | |
FORM_URL = "https://aws-portal.amazon.com/gp/aws/developer/account/index.html?ie=UTF8&action=usage-report" | |
def get_report(service, date_from, date_to, username, password, format='csv', period='days', debug=False): | |
br = mechanize.Browser() | |
br.set_handle_robots(False) | |
if debug: | |
# Log information about HTTP redirects and Refreshes. | |
br.set_debug_redirects(True) | |
# Log HTTP response bodies (ie. the HTML, most of the time). | |
br.set_debug_responses(True) | |
# Print HTTP headers. | |
br.set_debug_http(True) | |
br.addheaders = [ | |
# the login process 404s if you leave Python's UA string | |
('User-Agent', 'Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1) Gecko/20090701 Ubuntu/9.04 (jaunty) Firefox/3.5'), | |
('Accept', 'text/html, application/xml, */*'), | |
] | |
# login | |
print >>sys.stderr, "Logging in..." | |
try: | |
resp = br.open(FORM_URL) | |
br.select_form(name="signIn") | |
br["email"] = username | |
br["password"] = password | |
resp = br.submit() # submit current form | |
except Exception, e: | |
print >>sys.stderr, "Error logging in to AWS" | |
raise | |
# service selector | |
print >>sys.stderr, "Selecting service %s..." % service | |
br.select_form(name="usageReportForm") | |
br["productCode"] = [service] | |
resp = br.submit() | |
# report selector | |
print >>sys.stderr, "Building report..." | |
br.select_form(name="usageReportForm") | |
br["timePeriod"] = ["Custom date range"] | |
br["startYear"] = [str(date_from.year)] | |
br["startMonth"] = [str(date_from.month)] | |
br["startDay"] = [str(date_from.day)] | |
br["endYear"] = [str(date_to.year)] | |
br["endMonth"] = [str(date_to.month)] | |
br["endDay"] = [str(date_to.day)] | |
br["periodType"] = [period] | |
resp = br.submit("download-usage-report-%s" % format) | |
return resp.read() | |
if __name__ == "__main__": | |
from optparse import OptionParser | |
USAGE = ( | |
"Usage: %prog [options] -s SERVICE DATE_FROM DATE_TO\n\n" | |
"DATE_FROM and DATE_TO should be in YYYY-MM-DD format (eg. 2009-01-31)\n" | |
"Username and Password can also be specified via AWS_USERNAME and AWS_PASSWORD environment variables.\n" | |
"\n" | |
"Available Services: " + ', '.join(SERVICES) | |
) | |
parser = OptionParser(usage=USAGE) | |
parser.add_option('-s', '--service', dest="service", type="choice", choices=SERVICES, help="The AWS service to query") | |
parser.add_option('-p', '--period', dest="period", type="choice", choices=PERIODS, default='days', metavar="PERIOD", help="Period of report entries") | |
parser.add_option('-f', '--format', dest="format", type="choice", choices=FORMATS, default='csv', metavar="FORMAT", help="Format of report") | |
parser.add_option('-U', '--username', dest="username", metavar="USERNAME", help="Email address for your AWS account") | |
parser.add_option('-P', '--password', dest="password", metavar="PASSWORD") | |
parser.add_option('-d', '--debug', action="store_true", dest="debug", default=False) | |
opts, args = parser.parse_args() | |
if len(args) < 2: | |
parser.error("Missing date range") | |
date_range = [date(*time.strptime(args[i], '%Y-%m-%d')[0:3]) for i in range(2)] | |
if date_range[1] < date_range[0]: | |
parser.error("End date < start date") | |
if not opts.service: | |
parser.error("Specify a service to query!") | |
if not opts.username and not os.environ.get('AWS_USERNAME'): | |
parser.error("Must specify username option or set AWS_USERNAME") | |
if not opts.password and not os.environ.get('AWS_PASSWORD'): | |
parser.error("Must specify password option or set AWS_PASSWORD") | |
kwopts = { | |
'service': opts.service, | |
'date_from': date_range[0], | |
'date_to': date_range[1], | |
'format': opts.format, | |
'period': opts.period, | |
'username': opts.username or os.environ.get('AWS_USERNAME'), | |
'password': opts.password or os.environ.get('AWS_PASSWORD'), | |
'debug': opts.debug, | |
} | |
print get_report(**kwopts) |
This is broken. https://gist.github.com/1212148/e794734c3d825c131d4d3097700b0c87e6f1fbc1 works.
Hi,
I tried using your script, but I get this error:
Traceback (most recent call last):
File "log-amazon.py", line 27, in
import mechanize
ImportError: No module named mechanize
why ?
Ok,
I installed Mechanize module, insert: python log-amazon.py AmazonCloudFront 2012-05-01 2012-05-06 -U pippo@pippo.com -P pippo
I get this error:
Traceback (most recent call last):
File "log-amazon.py", line 173, in
date_range = [date(*time.strptime(args[i], '%Y-%m-%d')[0:3]) for i in range(2)]
File "/usr/lib/python2.6/_strptime.py", line 454, in _strptime_time
return _strptime(data_string, format)[0]
File "/usr/lib/python2.6/_strptime.py", line 325, in _strptime
(data_string, format))
ValueError: time data 'AmazonCloudFront' does not match format '%Y-%m-%d'
@maxpar, --help
is your friend.
I suspect you want python log-amazon.py -s AmazonCloudFront 2012-05-01 2012-05-06 -U pippo@pippo.com -P pippo
yeaaaa it works! Thank you.
Thanks for the script, works great!
is it posible to log in with access key, secret key and amazon id? so we dont need to use aws credential login? many thanks.
any idea, why i got this error?
Error logging in to AWS
Traceback (most recent call last):
File "aws_usage.py", line 129, in <module>
print get_report(**kwopts)
File "aws_usage.py", line 56, in get_report
br.select_form(name="signIn")
File "build/bdist.macosx-10.8-intel/egg/mechanize/_mechanize.py", line 499, in select_form
File "build/bdist.macosx-10.8-intel/egg/mechanize/_html.py", line 544, in __getattr__
File "build/bdist.macosx-10.8-intel/egg/mechanize/_html.py", line 557, in forms
File "build/bdist.macosx-10.8-intel/egg/mechanize/_html.py", line 237, in forms
File "build/bdist.macosx-10.8-intel/egg/mechanize/_form.py", line 844, in ParseResponseEx
File "build/bdist.macosx-10.8-intel/egg/mechanize/_form.py", line 981, in _ParseFileEx
File "build/bdist.macosx-10.8-intel/egg/mechanize/_form.py", line 760, in feed
mechanize._form.ParseError: unexpected '\\' char in declaration
This was working great for me until 4/17. I'm now running into the same issue as jackbit. Anyone have any ideas?
And as of 4/29 it's working again... I didn't change a thing. :-/
Running into a similar issue as jackbit. I'm able to signin, but it fails when trying to process the usage report form:
File "aws_usage.py", line 115, in get_report
br.select_form(name="usageReportForm")
File "/mechanize/_mechanize.py", line 499, in select_form
global_form = self._factory.global_form
File "/mechanize/_html.py", line 544, in __getattr__
self.forms()
File "/mechanize/_html.py", line 557, in forms
self._forms_factory.forms())
File "/mechanize/_html.py", line 237, in forms
_urlunparse=_rfc3986.urlunsplit,
File "/mechanize/_form.py", line 844, in ParseResponseEx
_urlunparse=_urlunparse,
File "/mechanize/_form.py", line 981, in _ParseFileEx
fp.feed(data)
File "/mechanize/_form.py", line 760, in feed
raise ParseError(exc)
mechanize._form.ParseError: unexpected '\\' char in declaration
When debugging I see the last response (that should contain the usage report form), this is included in the body:
<div id="message_warning" class="message warning">
<span></span>
<h6>Important Message!</h6>
<p>
To better protect your account, please re-enter your password and then enter the characters as they are shown in the image below.<br />
</p>
</div>
So the script probably fails to login, or loses its session along the way.
In my fork I've updated the login procedure to catch login failures. It will also prompt for captcha inputs. The code is not very nice, but it works for me. See: https://gist.github.com/Bouke/5719905
Simplest usage:
Any comments are welcome here, or drop me an email at robert@coup.net.nz