Skip to content

scenarios.md

Latest commit

 

History

History
102 lines (79 loc) · 3.68 KB

scenarios.md

File metadata and controls

102 lines (79 loc) · 3.68 KB

Scenarios

Create Scenarios

Graph Scenarios

Example

from gremlinapi.config import GremlinAPIConfig as config
from gremlinapi.attack_helpers import GremlinBlackholeAttack, GremlinLatencyAttack, GremlinTargetContainers
from gremlinapi.scenario_graph_helpers import (GremlinScenarioGraphHelper, GremlinScenarioILFINode,
                                               GremlinScenarioDelayNode, GremlinScenarioStatusCheckNode)
from gremlinapi.scenarios import GremlinAPIScenarios as scenarios

my_api_key = "123...xyz"
my_team_guid = 'uuid'

config.api_key = my_api_key
config.team_guid = my_team_guid

description = f'This is a test scenario to illustrate the usage of the new scenario graph model'
hypothesis = f'This should create a functional scenario with multiple attack iterations and a status_check'

blast_radius_steps = [50, 100]  # Percents
latency_magnitude_steps = [100, 500, 1000]  # Latency in milliseconds
delay_time = 5 # Time to delay between steps in seconds

# Status check config
status_check_description="Check if Gremlin.com is Still Up"

endpoint_url = 'https://www.gremlin.com'
endpoint_headers = dict()

# Optionally you can set specific header attributes
# status_check_api_key = 'Key foo...'
# endpoint_headers = {
#     'Authorization': status_check_api_key
# }

evaluation_ok_status_codes = ['200-203', '210']
evaluation_ok_latency_max = 500
evaluation_response_body_evaluation = {
    "op": "AND",
    "predicates": []
}

my_scenario = GremlinScenarioGraphHelper(name='test_scenario_1', description=description, hypothesis=hypothesis)

# Add the status check to the start of the scenario
status_check_node = GremlinScenarioStatusCheckNode(
    description=status_check_description,
    endpoint_url=endpoint_url,
    endpoint_headers=endpoint_headers,
    evaluation_ok_status_codes=evaluation_ok_status_codes,
    evaluation_ok_latency_max=evaluation_ok_latency_max,
    evaluation_response_body_evaluation=evaluation_response_body_evaluation
)
my_scenario.add_node(status_check_node)

# Add a delay step between the status check and the attacks
delay_node = GremlinScenarioDelayNode(delay=delay_time)

#adds node to graph
my_scenario.add_node(delay_node)

#adds edge
my_scenario.add_edge(status_check_node, delay_node)

# Add latency attacks to the scenario
for magnitude_idx, magnitude in enumerate(latency_magnitude_steps):
    for blast_radius_idx, blast_radius in enumerate(blast_radius_steps):
        print(f'Blast radius {blast_radius_idx+1} of {len(blast_radius_steps)} :: {blast_radius}%')
        print(f'Magnitude {magnitude_idx+1} of {len(latency_magnitude_steps)} :: {magnitude}ms')
        new_node = GremlinScenarioILFINode(
            command=GremlinLatencyAttack(delay=magnitude),
            target=GremlinTargetContainers(strategy_type='Random', labels={'owner': 'kyle'}, percent=blast_radius)
        )
        # Passing True as the second variable adds an edge from the last node in the graph to the 
        # node being added
        my_scenario.add_node(new_node, True)
        # Add a delay step between attacks, skip this if it's the last step in the loop
        if not ((magnitude_idx+1 == len(latency_magnitude_steps)) and (blast_radius_idx+1 == len(blast_radius_steps))):
            my_scenario.add_node(GremlinScenarioDelayNode(delay=delay_time), True)

# Let's view the json output
print(my_scenario)

# Send the scenario to Gremlin
my_scenario_guid = scenarios.create_scenario(body=my_scenario)

Halt Scenarios

from gremlinapi.scenarios import GremlinAPIScenarios as scenarios
team_id = 'TEAM_ID/UUID'
scenario_id = 'SCENARIO_ID'
scenario_run = 'SCENARIO_RUN_ID'

confirmation = scenarios.halt_scenario(teamId=team_id, guid=scenario_id, runNumber=scenario_run)