Use & Examples¶

RIPE Atlas Cousteau wraps the majority of the RIPE Atlas v2 API calls, but not all of them. For some of these calls you will need to have a specific API key that you can create with the API key manager.

Creating Measurements¶

Important

An API key is needed for this function.

You can create multiple measurements with one API request that will share the same start/end times and allocated probes. This means that if you create a ping and a traceroute with one call they will start and finish at the same time and will use same probes.

Measurement Types¶

The first step is to create the measurement specification object. Currently you can use the following measurement types objects:

Ping
Traceroute
Dns
Sslcert
Ntp
Http

You can initialise any of these objects by passing any of arguments stated in the API docs. Keep in mind that this library is trying to comply with what is stated in these docs. This means that if you try to create a measurement that is missing a field stated as required in these docs, the library won’t go ahead and do the HTTP query. On the contrary, it will raise an exception with some info in it. The required fields for each of the above type are:

Ping	Traceroute	Dns	Sslcert	Ntp	Http
af	af	af	af	af	af
description	description	description	description	description	description
target	target	query_class	target	target	target
	protocol	query_type
		query_argument

Examples:

from ripe.atlas.cousteau import (
  Ping,
  Traceroute
)

ping = Ping(
    af=4,
    target="www.google.gr",
    description="Ping Test"
)

traceroute = Traceroute(
    af=4,
    target="www.ripe.net",
    description="Traceroute Test",
    protocol="ICMP",
)

Measurement Sources¶

The second step is to create the measurements source(s). In order to do that you have to create an AtlasSource object using the arguments type, value, requested, and optionally tags. Type as described in the API docs should be one of the “area”, “country”, “prefix”, “asn”, “probes”, “msm”. Value is the actual value of the type and requested is the number of probes that will be selected from this source. Optionally you can use tags argument, which has to be a dictionary like {“include”: [], “exclude”: []}. Examples:

from ripe.atlas.cousteau import AtlasSource

source = AtlasSource(
    type="area",
    value="WW",
    requested=5,
    tags={"include":["system-ipv4-works"]}
)
source1 = AtlasSource(
    type="country",
    value="NL",
    requested=50,
    tags={"exclude": ["system-anchor"]}
)

Create Request¶

The last step is to make the actual HTTP request. Before you do this you need at least to specify if you measurements will be oneoff and you API key. Additional you can have start and stop time defined.

Examples:

from datetime import datetime
from ripe.atlas.cousteau import (
  Ping,
  Traceroute,
  AtlasSource,
  AtlasCreateRequest
)

ATLAS_API_KEY = ""

ping = Ping(af=4, target="www.google.gr", description="testing new wrapper")

traceroute = Traceroute(
    af=4,
    target="www.ripe.net",
    description="testing",
    protocol="ICMP",
)

source = AtlasSource(
    type="area",
    value="WW",
    requested=5,
    tags={"include":["system-ipv4-works"]}
)
source1 = AtlasSource(
    type="country",
    value="NL",
    requested=50,
    tags={"exclude": ["system-anchor"]}
)


atlas_request = AtlasCreateRequest(
    key=ATLAS_API_KEY,
    measurements=[ping, traceroute],
    sources=[source, source1],
    is_oneoff=True
)

(is_success, response) = atlas_request.create()

Changing Measurement Sources¶

Important

An API key is needed for this function.

If you want to add or remove probes from an existing measurement you have to use the AtlasChangeRequest. First step is to create an AtlasChangeSource objects which is similar to AtlasSource object for the creation of measurements. The difference is that here you have to specify an additional action argument. This parameter takes only two values “add” or “remove”. In case of “remove” the type of the source can only be “probes”. For more info check the API docs.

Example:

from ripe.atlas.cousteau import AtlasChangeSource, AtlasChangeRequest

ATLAS_MODIFY_API_KEY = ""

# Add probes
source = AtlasChangeSource(
    value="GR",
    requested=3,
    type="country",
    tags={"include":["system-ipv4-works"], "exclude": ["system-anchor"]},
    action="add"
)
source1 = AtlasChangeSource(
    value="4,5,6",
    requested=3,
    type="probes",
    action="add"
)

# Remove probes
source2 = AtlasChangeSource(
    value="1,2,3",
    type="probes",
    requested=3,
    action="remove"
)

atlas_request = AtlasChangeRequest(
    key=ATLAS_MODIFY_API_KEY,
    msm_id=1000001,
    sources=[source, source1, source2]
)

(is_success, response) = atlas_request.create()

Stopping Measurements¶

Important

An API key is needed for this function.

You can stop a measurement by creating a AtlasStopRequest and specifying measurement ID as shown below:

from ripe.atlas.cousteau import AtlasStopRequest

ATLAS_STOP_API_KEY = ""

atlas_request = AtlasStopRequest(msm_id=1000001, key=ATLAS_STOP_API_KEY)

(is_success, response) = atlas_request.create()

Results¶

Fetching Results¶

Note

If measurement is not public you will need an API key with “download results of a measurement” permission.

You can fetch results for any measurements using AtlasResultsRequest. You can filter them by start/end time and probes. Times can be python datetime objects, Unix timestamps or string representations of dates.

Example:

from datetime import datetime
from ripe.atlas.cousteau import AtlasResultsRequest

kwargs = {
    "msm_id": 2016892,
    "start": datetime(2015, 05, 19),
    "stop": datetime(2015, 05, 20),
    "probe_ids": [1,2,3,4]
}

is_success, results = AtlasResultsRequest(**kwargs).create()

if is_success:
    print(results)

Fetching Latest Results¶

Note

If measurement is not public you will need an API key with “download results of a measurement” permission.

In case you want to download latest results of a measurement or your measurement is an oneoff measurements is easier and faster to use the API for the latest results. Fetching latest results is done by using AtlasLatestRequest and there is an option for filtering by probes.

Example:

from ripe.atlas.cousteau import AtlasLatestRequest

kwargs = {
    "msm_id": 2016892,
    "probe_ids": [1,2,3,4]
}

is_success, results = AtlasLatestRequest(**kwargs).create()

if is_success:
    print(results)

Streaming API¶

Atlas supports getting results and other events through a WebSocket stream to get them close to real time. The AtlasStream class provides an interface to this stream, and supports both callback-based and iterator-based access.

Measurement Results¶

You have to create an AtlasStream object and subscribe to the “result” stream type. More details on the available parameters of the stream can be found on the streaming documentation.

Example using the callback-interface:

from ripe.atlas.cousteau import AtlasStream

def on_result_response(*args):
    """
    Function that will be called every time we receive a new result.
    Args is a tuple, so you should use args[0] to access the real message.
    """
    print(args[0])

atlas_stream = AtlasStream()
atlas_stream.connect()

# Bind function we want to run with every result message received
atlas_stream.bind("atlas_result", on_result_response)

# Subscribe to new stream for 1001 measurement results
stream_parameters = {"msm": 1001}
atlas_stream.subscribe(stream_type="result", **stream_parameters)

# Process incoming events for 5 seconds, calling the callback defined above.
# Make sure you have this line after you start *all* your streams
atlas_stream.timeout(seconds=5)

# Shut down everything
atlas_stream.disconnect()

Connection Events¶

Besides results, the streaming API also supports probe connect/disconnect events. Again you have to create an AtlasStream object, but this time you subscribe to the “probestatus” stream type. More info about additional parameters can be found on the streaming documentation.

Example using the iterator-interface:

from ripe.atlas.cousteau import AtlasStream

atlas_stream = AtlasStream()
atlas_stream.connect()

stream_parameters = {"enrichProbes": True}
atlas_stream.subscribe(stream_type="probestatus", **stream_parameters)

# Iterate over the incoming results for 5 seconds
for event_name, payload in atlas_stream.iter(seconds=5):
    print(event_name, payload)

# Shut down everything
atlas_stream.disconnect()

Using the Sagan Library¶

In case you need to do further processing with any of the results you can use our official RIPE Atlas results parsing library called Sagan. An example of how to combine two libraries is the below:

from ripe.atlas.cousteau import AtlasLatestRequest
from ripe.atlas.sagan import Result

kwargs = {
    "probe_ids": [1,2,3,4]
}

is_success, results = AtlasLatestRequest(**kwargs).create()

if is_success:
    for result in results:
        print(Result.get(result))

Metadata¶

RIPE Atlas API allows you to get metadata about probes and measurements in the system. You can get metadata for a single object or filter based on various criteria.

Single Objects¶

Every time you create a new instance of either Measurement/Probe objects it will fetch meta data from API and return an object with selected attributes.

Measurement¶

Using the Measurement object will allow you to have a python object with attributes populated from specific measurement’s meta data.

Example:

from ripe.atlas.cousteau import Measurement

measurement = Measurement(id=1000002)
print(measurement.protocol)
print(measurement.description)
print(measurement.is_oneoff)
print(measurement.is_public)
print(measurement.target_ip)
print(measurement.target_asn)
print(measurement.type)
print(measurement.interval)
print(dir(measurement)) # Full list of properties

Probe¶

Using the Probe object will allow you to have a python object with attributes populated from specific probe’s meta data.

from ripe.atlas.cousteau import Probe

probe = Probe(id=3)
print(probe.country_code)
print(probe.is_anchor)
print(probe.is_public)
print(probe.address_v4)
print(dir(probe)) # Full list of properties

Filtering¶

This feature queries API for probes/measurements based on specified filters. Available filters can be found in the API docs. Underneath it will follow all next urls until there are no more objects. It returns a python generator that you can use in a for loop to access each object.

Probe¶

The following example will fetch all measurements with Status equals to “Specified”. More info on filters for this call can be found in the API docs.

from ripe.atlas.cousteau import ProbeRequest

filters = {"tags": "NAT", "country_code": "NL", "asn_v4": "3333"}
probes = ProbeRequest(**filters)

for probe in probes:
    print(probe["id"])

# Print total count of found probes
print(probes.total_count)

Measurement¶

The following example will fetch all probes from NL with asn_v4 3333 and with tag NAT. More info on filters for this call can be found in the API docs.

from ripe.atlas.cousteau import MeasurementRequest

filters = {"status": 1}
measurements = MeasurementRequest(**filters)

for msm in measurements:
    print(msm["id"])

# Print total count of found measurements
print(measurements.total_count)

General GET API Requests¶

Using the general AtlasRequest object you can do any GET request to the RIPE Atlas API considering you provide the url path.

Example:

url_path = "/api/v2/anchors"
request = AtlasRequest(**{"url_path": url_path})
result = namedtuple('Result', 'success response')
(is_success, response) = request.get()
if not is_success:
    return False

return result.response["participant_count"]