Page tree
Skip to end of metadata
Go to start of metadata

Purpose

This Wiki page is intended to capture and define the integration tests which exercise the Auto Use Cases from a testability perspective.

Scope

This project focuses on somewhere in between L3 and L2 in the following figure.

Edge Cloud

Reference to Auto-UC-01

Martin Klozik

Below  testcases are aiming to provide compatible edge cloud infrastructure with required OPNFV and ONAP features . Testcases mainly covers the VIM , ONAP Edge computing infrastructure and VNF life cycle management to enable and test service provider's management on edge cloud .


Test Section

ID

Title

Description

PrerequisiteStatus

Environment check


auto-edge-env-001

AUTO-31 - Getting issue details... STATUS

Basic OpenStack environment check

Create test script to check basic VIM (OpenStack) services are up and running

 1. Hardware environment in which Edge cloud may be deployed

2.  VIM(OpenStack) with tacker , Telemetry services deployed

 

auto-edge-env-002

AUTO-35 - Getting issue details... STATUS

Basic VNF environment checkCreate test script to check VNF Healthcheck

 1. Hardware environment in which Edge cloud may be deployed

2.  VIM(OpenStack) with tacker , Telemetry services deployed


auto-edge-env-003

AUTO-36 - Getting issue details... STATUS

Basic ONAP environment checkCreate test script to check ONAP components Health check

 1. Hardware environment in which Edge cloud may be deployed

2.  VIM(OpenStack) with tacker , Telemetry services deployed


VNF lifecycle management 

auto-edge-vnf-001

AUTO-32 - Getting issue details... STATUS

VNF Instance Management

Validation of VNF Instance Management which includes VNF instantiation, VNF State Management and termination








1. Hardware environment in which Edge cloud may be deployed

2.  VIM(OpenStack) with tacker , Telemetry services deployed

3. Required VNFs (eg. vFW , vROUTER, vCPE) spawned as per edge cloud requirements


auto-edge-pif-002

AUTO-33 - Getting issue details... STATUS

Tacker Monitoring Driver (VNFMonitorPing)

Write Tacker Monitor driver to handle monitor_call and based on return state value create custom events .

If Ping to VNF fails trigger below events

Event 1 : Collect failure logs from vnf
Event 2 : Soft restart/respawn the VNF


auto-edge-pif-003

AUTO-34 - Getting issue details... STATUS

Integrate with Telemetry

Create TOSCA template policies to implement ceilometer  data collection service

Collect cpu utilization data and check with threshold and perform action accordingly (respawn, scale-in/scale-out)


 

Resiliency Improvements

Richard Elias

AUTO-2 - Getting issue details... STATUS AUTO-4 - Getting issue details... STATUS

The tests below are intended to illustrate an organizational approach to developing tests that demonstrate the value of ONAP closed-loop support for improving platform resiliency. The goal for these tests is to focus first on the most commonly occurring stresses, and to expand toward chaos monkey type testing over time.

Each test case can capture the information before and after the stress or fault is induced to help assess the resilience. These tests can be repeated with and without ONAP closed loop control to demonstrate the improvement in resilience. At the end of the test, the stress or fault should be removed, if it is not already cleared.

Test Section

ID

Title

Description

Status

Physical Infra Failure

auto-resiliency-pif-001 AUTO-9 - Getting issue details... STATUS

Migration upon host failure

Compute host power is interrupted by IPMI command, and affected workloads are migrated to other available hosts.

To inject host power interruption, use commands such as "ipmitool" or "ipmipower -h <host> -u <user> -p <passwd>" with options --stat, --cycle, --reset

 
auto-resiliency-pif-002 AUTO-10 - Getting issue details... STATUS
Migration upon disk failure

Disk volumes are unmounted, and affected workloads are migrated to other available hosts.

To unmount, use command such as "umount <disk>"

 
auto-resiliency-pif-003 AUTO-11 - Getting issue details... STATUS
Migration upon link failure

Traffic on links is interrupted/corrupted by switch admin commands, and affected workloads are migrated to other available hosts.

Use tc (Traffic Control) netem (Network Emulation) commands to manipulate traffic control settings, e.g.
tc qdisc change dev eth0 root netem corrupt 0.1%
tc qdisc change dev eth0 root netem loss 0.1%

 
auto-resiliency-pif-004 AUTO-12 - Getting issue details... STATUS
Migration upon NIC failure

NIC ports are disabled by host commands, and affected workloads are migrated to other available hosts.

To disable NIC port, use command such as "openstack port show/set --disable"

 
Virtual Infra Failure
auto-resiliency-vif-001 AUTO-13 - Getting issue details... STATUS
OpenStack compute host service fail

Core OpenStack service processes on compute hosts are terminated, and auto-restored, or affected workloads are migrated to other available hosts.

To restart service processes on compute host, use commands such as,
openstack compute service set --disable --disable-reason trial log nova nova-compute
openstack compute service set --enable nova nova-compute
service nova-compute restart
openstack volume service list/set --disable <host> <service>
Alternatively restart processes like RabbitMQ/AMQP

 
auto-resiliency-vif-002 AUTO-14 - Getting issue details... STATUS
SDNC service failCore SDNC service processes are terminated, and auto-restored. 
auto-resiliency-vif-003 AUTO-15 - Getting issue details... STATUS
OVS fail

OVS bridges are disabled, and affected workloads are migrated to other available hosts.

To disable OVS bridge, use command such as "ovs-ofctl mod-port <bridge> <iface> down"

 
Security
auto-resiliency-sec-001 AUTO-16 - Getting issue details... STATUS
Host tampering

Host tampering is detected, the host is fenced, and affected workloads are migrated to other available hosts.

Simulate Host tampering, e.g. create tampered images with similar looking name, leading to running instances from attacker images, tamper system logs, binaries, load unauthorized malicious code, etc.

 
auto-resiliency-sec-002 AUTO-17 - Getting issue details... STATUS
Host intrusion

Host intrusion attempts are detected, an offending workload, device, or flow is identified and fenced, and as needed affected workloads are migrated to other available hosts.

Simulate Host intrusion, e.g. using remote buffer overflow (refer to Syntribos tool below).

 
auto-resiliency-sec-003 AUTO-18 - Getting issue details... STATUS
Network intrusion

Network intrusion attempts are detected, and an offending flow is identified and fenced.

Simulate Network intrusion, e.g.

    • Bombard the services with new requests or connections, leading to large number forked processes and threads leading to resource issues on other workloads in the system.

    • Use Syntribos (https://docs.openstack.org/syntribos/latest/index.html), an automated API security testing tool to automatically detect common security defects such as SQL injection, LDAP injection, buffer overflow, etc., and also to identify new security defects by automated fuzzing.
 

Enterprise vCPE

Martin Klozik Richard Elias

Test Section

ID

Title

Description

OPNFV Components

ONAP ComponentsStatus

VNF Management

auto-vcpe-spinup-001

AUTO-19 - Getting issue details... STATUS

Spin up a vCPE instanceSpin up a vCPE instance, by calling NBI of the orchestrator.VNFMSO and VNFMenvironment done, writing test scripts

auto-vcpe-spinup-002


Spin up a vFW instanceSpin up a vFW instance, by calling NBI of the orchestrator.VNFMSO and VNFMTBD
VPN as a Serviceauto-vcpe-vpn-001Subscribe vpn serviceSubsribe vpn service, by calling NBI of the orchestrator.VNFMSO and SDNCenvironment done, writing test scripts
auto-vcpe-vpn-002Unsubscribe vpn serviceUnsubsribe vpn service, by calling NBI of the orchestrator.VNFMSO and SDNCenvironment done, writing test scripts
Internet as a Serviceauto-vcpe-internet-001Subscribe internet serviceSubsribe internet service, by calling NBI of the orchestrator.VNFMSO and SDNC and APPCTBD
auto-vcpe-internet-002Unsubscribe internet serviceUnsubsribe internet service, by calling NBI of the orchestrator.VNFMSO and SDNC and APPCTBD


Test Case database

Spreadsheet to manage test case entries for OPNFV Test Case database (for MS2 in a release):

auto-test-cases-DB-180523.xlsx

(web site with OPNFV Swagger API: http://testresults.opnfv.org/test/swagger/spec.html#!/APIs)

instructions: How do I declare my project in the Test DB?


  • No labels

3 Comments

    1. Is VNF a component of ONAP or OPNFV?