netskopesdk 0.0.35

SDK to download the Netskope Events

Netskope SDK

Neskope SDK is Python library for dealing with API's to download the Netskope events.

Installation

Use the package manager pip to install NetskopeSDK.

pip install netskopesdk

Rest sdk Usage to pull the Alert & Events

from netskope_api.iterator.netskope_iterator import NetskopeIterator
from netskope_api.iterator.const import Const
from requests.exceptions import RequestException
import time

# Construct the params dict to pass the authentication details 
params = {
        Const.NSKP_TOKEN : "<REST-API-TOKEN>",
        Const.NSKP_TENANT_HOSTNAME : "<HOSTNAME>",
        # Optional param to pass the proxy hosts.
        Const.NSKP_PROXIES : {"<PROXY-HOSTS>"},
        Const.NSKP_EVENT_TYPE : "<EVENT-TYPE>",
        Const.NSKP_ITERATOR_NAME : "<ITERATOR-NAME>",
        Const.NSKP_USER_AGENT : "<SPLUNK-TENANT-HOSTNAME>"
    
        # To query specific alert pass the NSKP_EVENT_TYPE as "alert" and the alert type.
        # Const.NSKP_EVENT_TYPE : Const.EVENT_TYPE_ALERT,
        # Const.NSKP_ALERT_TYPE : Const.ALERT_TYPE_DLP
    }

DEFAULT_WAIT_TIME = 30
RESULT = "result"
WAIT_TIME = "wait_time"

# Create an Iterator
iterator = NetskopeIterator(params)

# Use the next() iterator to download the logs. 
# Consume the message indefinitely in a loop and ingest the data to SIEM
while True:
    response = (iterator.next())
    try:
        if response:
            data = response.json()
            if RESULT in data and len(data[RESULT]) != 0:
                # processData() 
                # sleep() the thread to avoid constant polling
                if WAIT_TIME in data:
                    time.sleep(data[WAIT_TIME])
                else:
                    time.sleep(DEFAULT_WAIT_TIME)
            else:
                print("No response received from the iterator")
                time.sleep(DEFAULT_WAIT_TIME)
    except Exception as e:
        time.sleep(DEFAULT_WAIT_TIME)
        raise RequestException(e)

Rest sdk Usage to pull the Alert & Events in CSV format.

import csv
from io import StringIO
from netskope_api.iterator.netskope_iterator import NetskopeIterator
from netskope_api.iterator.const import Const
from requests.exceptions import RequestException

# Construct the params dict to pass the authentication details 
# Note: For CSV iterator - only the next() api is supported.
params = {
        Const.NSKP_TOKEN : "<REST-API-TOKEN>",
        Const.NSKP_TENANT_HOSTNAME : "<HOSTNAME>",
        # Optional param to pass the proxy hosts.
        Const.NSKP_PROXIES : {"<PROXY-HOSTS>"},
        Const.NSKP_EVENT_TYPE : "<EVENT-TYPE>",
        Const.NSKP_ITERATOR_NAME : "<ITERATOR-NAME>",
        Const.NSKP_USER_AGENT : "<SPLUNK-TENANT-HOSTNAME>"
    
        # To query specific alert pass the NSKP_EVENT_TYPE as "alert" and the alert type.
        # Const.NSKP_EVENT_TYPE : Const.EVENT_TYPE_ALERT,
        # Const.NSKP_ALERT_TYPE : Const.ALERT_TYPE_DLP
    }

# Create an Iterator
iterator = NetskopeIterator(params)

    # Create an Iterator. Next is the only API supported.
    iterator = NetskopeIterator(params)
    # Consume the message indefinitely in a loop and ingest the data to SIEM
    while True:
        try:
            # Use the next() iterator to download the logs.
            response = (iterator.next())
            if response.status_code == 200:
                # Parse the CSV data from the response
                headers = response.headers["schema_headers"]
                csv_data = StringIO(response.text)

                # Create a CSV reader object
                csv_reader = csv.reader(csv_data)
                # Print or use the headers as needed
                print("CSV Headers:", headers)

                # Iterate over each row in the CSV data
                for row in csv_reader:
                    # Process each row as needed
                    print(row)
            else:
                print("API failed with response code: {}, response content {}".format(response.status_code,
                                                                                      response.text))
        except Exception as e:
            raise RequestException(e)

Rest sdk Usage to retrieve tokens used for subscribing to transaction events from PSL.

from requests.exceptions import RequestException
from netskope_api.iterator.const import Const
from netskope_api.token_management.netskope_management import NetskopeTokenManagement

if __name__ == '__main__':
    params = {
        Const.NSKP_TOKEN: "",
        Const.NSKP_TENANT_HOSTNAME: "<HOSTNAME>",
        # Optional param to pass the proxy hosts.
        Const.NSKP_PROXIES : {"<PROXY-HOSTS>"}
    }

    sub_path_response = None
    sub_key_response = None
    try:
        # Create token_management client
        token_management = NetskopeTokenManagement(params)
        token_management_response = token_management.get()
        if token_management_response:
            if "subscription" in token_management_response:
                sub_path_response = token_management_response["subscription"]
            if "subscription-key" in token_management_response:
                sub_key_response = token_management_response["subscription-key"]
    except Exception as e:
        raise RequestException(e)

1. 200 response code means Customer is authorized to create/get subscription key and path.

2. 401 response code means Customer is not authorized to create/get subscription key and path.
   This is a licensed feature, please contact Netskope support to purchase.

3. 403 response code means Netskope token used is not valid. Token is either expired or invalid.

4. 449 response code means Existing customer authorized for transaction events, subscription key and path were already downloaded.
   Use regenerate_and_get() to regenerate the subscription key.
   This is a one time step to onboard the existing customer to the API.

When to use regenerate_and_get() API.

regenerate_and_get() API must only be used if google-cloud-pubsublite throws 401 invalid credentials exception
while using subscription key and path retrieved by using Netskope API. Or to handle 449 response code.

Example error:
google.api_core.exceptions.Unauthenticated: 401 Request had invalid authentication credentials
grpc.aio._call.AioRpcError: <AioRpcError of RPC that terminated with:
status = StatusCode.UNAUTHENTICATED
details = "Request had invalid authentication credentials. Expected OAuth 2 access token, login cookie or other valid authentication credential.

Regenerating subscription key will invalidate the existing key but subscription path will not be updated
so that clients can continue consuming events where they left off.

How to use regenerate_and_get() API.

1. regenerate_and_get() API must only be used if google-cloud-pubsublite throws 401 Request had invalid authentication credentials errors.
2. Retry 3 times with get() API with some Exponential backoff logic with an intial time interval of 60 seconds.
3. If the responses returned by get() API continues to be invalid credentials. Use regenerate_and_get() API and use the new credentials.

Note: Regenerating subscription key will invalidate the existing key.

from requests.exceptions import RequestException
from netskope_api.iterator.const import Const
from netskope_api.token_management.netskope_management import NetskopeTokenManagement

if __name__ == '__main__':
    params = {
        Const.NSKP_TOKEN: "",
        Const.NSKP_TENANT_HOSTNAME: "<HOSTNAME>",
        # Optional param to pass the proxy hosts.
        Const.NSKP_PROXIES : {"<PROXY-HOSTS>"}
    }

    sub_path_response = None
    sub_key_response = None
    try:
        # Create token_management client
        token_management = NetskopeTokenManagement(params)
        token_management_response = token_management.regenerate_and_get()
        if token_management_response:
            if "subscription" in token_management_response:
                sub_path_response = token_management_response["subscription"]
            if "subscription-key" in token_management_response:
                sub_key_response = token_management_response["subscription-key"]
    except Exception as e:
        raise RequestException(e)

Example usage of web txn tokens using Pubsublite client library

Please install Pubsublite client library if not present.

pip install google-cloud-pubsublite

import concurrent.futures
import logging
import os
from concurrent.futures._base import TimeoutError

from google.cloud.pubsublite.cloudpubsub import SubscriberClient
from google.cloud.pubsublite.types import FlowControlSettings, MessageMetadata
from google.pubsub_v1 import PubsubMessage
from requests.exceptions import RequestException

from netskope_api.iterator.const import Const
from netskope_api.token_management.netskope_management import NetskopeTokenManagement

_logger = logging.getLogger()


def callback(message: PubsubMessage):
    message_data = message.data.decode("utf-8")
    metadata = MessageMetadata.decode(message.message_id)
    _logger.info(
        f"Received {message_data} of ordering key {message.ordering_key} with id {metadata}."
    )
    message.ack()


class PSLSubscriberClient:
    def __init__(self):

        configs = {
            "messages_outstanding": 1000,
            "bytes_outstanding": 3.5 * 1024 * 1024,
            "timeout": 60,
            "thread_count": 1
        }
        self.configs = configs

    def make_default_thread_pool_executor(self):
        return concurrent.futures.ThreadPoolExecutor(self.configs.get("thread_count"))

    def stream(self, subscription_path):
        global streaming_pull_future

        per_partition_flow_control_settings = FlowControlSettings(
            # Must be >0.
            messages_outstanding=self.configs.get("messages_outstanding"),
            # Must be greater than the allowed size of the largest message.
            bytes_outstanding=self.configs.get("bytes_outstanding"),
        )

        executor = self.make_default_thread_pool_executor()
        with SubscriberClient(executor=executor) as subscriber_client:
            _logger.info(
                "Listening for messages on the pub sub lite subscription {}".format(subscription_path))
            streaming_pull_future = subscriber_client.subscribe(
                subscription_path,
                callback=callback,
                per_partition_flow_control_settings=per_partition_flow_control_settings,
            )
            try:
                timeout = self.configs.get("timeout")
                if timeout:
                    streaming_pull_future.result(timeout=self.configs.get("timeout"))
                else:
                    streaming_pull_future.result()
            except TimeoutError or KeyboardInterrupt:
                streaming_pull_future.cancel()
                assert streaming_pull_future.done()


if __name__ == '__main__':
    params = {
        Const.NSKP_TOKEN: "",
        Const.NSKP_TENANT_HOSTNAME: "<HOSTNAME>",
        # Optional param to pass the proxy hosts.
        Const.NSKP_PROXIES : {"<PROXY-HOSTS>"}
    }

    sub_path_response = None
    sub_key_response = None
    try:
        # Create token_management client
        token_management = NetskopeTokenManagement(params)
        token_management_response = token_management.get()
        if token_management_response:
            if "subscription" in token_management_response:
                sub_path_response = token_management_response["subscription"]
            if "subscription-key" in token_management_response:
                sub_key_response = token_management_response["subscription-key"]
    except Exception as e:
        raise RequestException(e)

    os.environ['GOOGLE_APPLICATION_CREDENTIALS'] = sub_key_response
    psl_client = PSLSubscriberClient()
    psl_client.stream(subscription_path=sub_path_response)