diff options
Diffstat (limited to 'docs/testing/user/userguide')
| -rw-r--r-- | docs/testing/user/userguide/14-nsb-operation.rst | 150 | ||||
| -rw-r--r-- | docs/testing/user/userguide/15-list-of-tcs.rst | 2 | ||||
| -rw-r--r-- | docs/testing/user/userguide/opnfv_yardstick_tc092.rst | 196 | ||||
| -rw-r--r-- | docs/testing/user/userguide/opnfv_yardstick_tc093.rst | 184 |
4 files changed, 530 insertions, 2 deletions
diff --git a/docs/testing/user/userguide/14-nsb-operation.rst b/docs/testing/user/userguide/14-nsb-operation.rst index 2e741822e..a5f3a0cf6 100644 --- a/docs/testing/user/userguide/14-nsb-operation.rst +++ b/docs/testing/user/userguide/14-nsb-operation.rst @@ -84,6 +84,116 @@ In this example we have ``TRex xe0 <-> xe0 VNF xe1 <-> xe0 UDP_Replay`` downlink_0: - xe0 + +Availability zone +^^^^^^^^^^^^^^^^^ + +The configuration of the availability zone is requred in cases where location +of exact compute host/group of compute hosts needs to be specified for SampleVNF +or traffic generator in the heat test case. If this is the case, please follow +the instructions below. + +.. _`Create a host aggregate`: + +1. Create a host aggregate in the OpenStack and add the available compute hosts + into the aggregate group. + + .. note:: Change the ``<AZ_NAME>`` (availability zone name), ``<AGG_NAME>`` + (host aggregate name) and ``<HOST>`` (host name of one of the compute) in the + commands below. + + .. code-block:: bash + + # create host aggregate + openstack aggregate create --zone <AZ_NAME> --property availability_zone=<AZ_NAME> <AGG_NAME> + # show available hosts + openstack compute service list --service nova-compute + # add selected host into the host aggregate + openstack aggregate add host <AGG_NAME> <HOST> + +2. To specify the OpenStack location (the exact compute host or group of the hosts) + of SampleVNF or traffic generator in the heat test case, the ``availability_zone`` server + configuration option should be used. For example: + + .. note:: The ``<AZ_NAME>`` (availability zone name) should be changed according + to the name used during the host aggregate creation steps above. + + .. code-block:: yaml + + context: + name: yardstick + image: yardstick-samplevnfs + ... + servers: + vnf__0: + ... + availability_zone: <AZ_NAME> + ... + tg__0: + ... + availability_zone: <AZ_NAME> + ... + networks: + ... + +There are two example of SampleVNF scale out test case which use the availability zone +feature to specify the exact location of scaled VNFs and traffic generators. + +Those are: + +.. code-block:: console + + <repo>/samples/vnf_samples/nsut/prox/tc_prox_heat_context_l2fwd_multiflow-2-scale-out.yaml + <repo>/samples/vnf_samples/nsut/vfw/tc_heat_rfc2544_ipv4_1rule_1flow_64B_trex_scale_out.yaml + +.. note:: This section describes the PROX scale-out testcase, but the same + procedure is used for the vFW test case. + +1. Before running the scale-out test case, make sure the host aggregates are + configured in the OpenStack environment. To check this, run the following + command: + + .. code-block:: console + + # show configured host aggregates (example) + openstack aggregate list + +----+------+-------------------+ + | ID | Name | Availability Zone | + +----+------+-------------------+ + | 4 | agg0 | AZ_NAME_0 | + | 5 | agg1 | AZ_NAME_1 | + +----+------+-------------------+ + +2. If no host aggregates are configured, please use `steps above`__ to + configure them. + +__ `Create a host aggregate`_ + + +3. Run the SampleVNF PROX scale-out test case, specifying the availability + zone of each VNF and traffic generator as a task arguments. + + .. note:: The ``az_0`` and ``az_1`` should be changed according to the host + aggregates created in the OpenStack. + + .. code-block:: console + + yardstick -d task start\ + <repo>/samples/vnf_samples/nsut/prox/tc_prox_heat_context_l2fwd_multiflow-2-scale-out.yaml\ + --task-args='{ + "num_vnfs": 4, "availability_zone": { + "vnf_0": "az_0", "tg_0": "az_1", + "vnf_1": "az_0", "tg_1": "az_1", + "vnf_2": "az_0", "tg_2": "az_1", + "vnf_3": "az_0", "tg_3": "az_1" + } + }' + + ``num_vnfs`` specifies how many VNFs are going to be deployed in the + ``heat`` contexts. ``vnf_X`` and ``tg_X`` arguments configure the + availability zone where the VNF and traffic generator is going to be deployed. + + Collectd KPIs ------------- @@ -154,7 +264,7 @@ We set the VCPUs and memory using the ``--task-args`` options .. code-block:: console - yardstick task start --task-args='{"mem": 10480, "vcpus": 4, "ports": 2}' \ + yardstick task start --task-args='{"mem": 10480, "vcpus": 4, "vports": 2}' \ samples/vnf_samples/nsut/vfw/tc_heat_rfc2544_ipv4_1rule_1flow_64B_trex_scale-up.yaml In order to support ports scale-up, traffic and topology templates need to be used in testcase. @@ -242,7 +352,7 @@ Baremetal file: /etc/yardstick/nodes/pod.yaml Scale-Out --------------------- +--------- VNFs performance data with scale-out helps @@ -313,3 +423,39 @@ options section. options: tg_0: queues_per_port: 2 + + +Standalone configuration +------------------------ + +NSB supports certain Standalone deployment configurations. +Standalone supports provisioning a VM in a standalone visualised environment using kvm/qemu. +There two types of Standalone contexts available: OVS-DPDK and SRIOV. +OVS-DPDK uses OVS network with DPDK drivers. +SRIOV enables network traffic to bypass the software switch layer of the Hyper-V stack. + +Standalone with OVS-DPDK +^^^^^^^^^^^^^^^^^^^^^^^^ + +SampleVNF image is spawned in a VM on a baremetal server. +OVS with DPDK is installed on the baremetal server. + +.. note:: Ubuntu 17.10 requires DPDK v.17.05 and higher, DPDK v.17.05 requires OVS v.2.8.0. + +Default values for OVS-DPDK: + + * queues: 4 + * lcore_mask: "" + * pmd_cpu_mask: "0x6" + +Sample test case file +^^^^^^^^^^^^^^^^^^^^^ + + 1. Prepare SampleVNF image and copy it to ``flavor/images``. + 2. Prepare context files for TREX and SampleVNF under ``contexts/file``. + 3. Add bridge named ``br-int`` to the baremetal where SampleVNF image is deployed. + 4. Modify ``networks/phy_port`` accordingly to the baremetal setup. + 5. Run test from: + +.. literalinclude:: /submodules/yardstick/samples/vnf_samples/nsut/acl/tc_ovs_rfc2544_ipv4_1rule_1flow_64B_trex.yaml + :language: yaml diff --git a/docs/testing/user/userguide/15-list-of-tcs.rst b/docs/testing/user/userguide/15-list-of-tcs.rst index 678f0f9a9..37ce819f1 100644 --- a/docs/testing/user/userguide/15-list-of-tcs.rst +++ b/docs/testing/user/userguide/15-list-of-tcs.rst @@ -84,6 +84,8 @@ H A opnfv_yardstick_tc057.rst opnfv_yardstick_tc058.rst opnfv_yardstick_tc087.rst + opnfv_yardstick_tc092.rst + opnfv_yardstick_tc093.rst IPv6 ---- diff --git a/docs/testing/user/userguide/opnfv_yardstick_tc092.rst b/docs/testing/user/userguide/opnfv_yardstick_tc092.rst new file mode 100644 index 000000000..895074a85 --- /dev/null +++ b/docs/testing/user/userguide/opnfv_yardstick_tc092.rst @@ -0,0 +1,196 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International +.. License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) OPNFV, Ericsson and others. + +************************************* +Yardstick Test Case Description TC092 +************************************* + ++-----------------------------------------------------------------------------+ +|SDN Controller resilience in HA configuration | +| | ++--------------+--------------------------------------------------------------+ +|test case id | OPNFV_YARDSTICK_TC092: SDN controller resilience and high | +| | availability HA configuration | +| | | ++--------------+--------------------------------------------------------------+ +|test purpose | This test validates SDN controller node high availability by | +| | verifying there is no impact on the data plane connectivity | +| | when one SDN controller fails in a HA configuration, | +| | i.e. all existing configured network services DHCP, ARP, L2, | +| | L3VPN, Security Groups should continue to operate | +| | between the existing VMs while one SDN controller instance | +| | is offline and rebooting. | +| | | +| | The test also validates that network service operations such | +| | as creating a new VM in an existing or new L2 network | +| | network remain operational while one instance of the | +| | SDN controller is offline and recovers from the failure. | +| | | ++--------------+--------------------------------------------------------------+ +|test method | This test case: | +| | 1. fails one instance of a SDN controller cluster running | +| | in a HA configuration on the OpenStack controller node | +| | | +| | 2. checks if already configured L2 connectivity between | +| | existing VMs is not impacted | +| | | +| | 3. verifies that the system never loses the ability to | +| | execute virtual network operations, even when the | +| | failed SDN Controller is still recovering | +| | | ++--------------+--------------------------------------------------------------+ +|attackers | In this test case, an attacker called “kill-process” is | +| | needed. This attacker includes three parameters: | +| | 1. ``fault_type``: which is used for finding the attacker's | +| | scripts. It should be set to 'kill-process' in this test | +| | | +| | 2. ``process_name``: should be set to sdn controller | +| | process | +| | | +| | 3. ``host``: which is the name of a control node where | +| | opendaylight process is running | +| | | +| | example: | +| | - ``fault_type``: “kill-process” | +| | - ``process_name``: “opendaylight-karaf” (TBD) | +| | - ``host``: node1 | +| | | ++--------------+--------------------------------------------------------------+ +|monitors | In this test case, the following monitors are needed | +| | 1. ``ping_same_network_l2``: monitor pinging traffic | +| | between the VMs in same neutron network | +| | | +| | 2. ``ping_external_snat``: monitor ping traffic from VMs to | +| | external destinations (e.g. google.com) | +| | | +| | 3. ``SDN controller process monitor``: a monitor checking | +| | the state of a specified SDN controller process. It | +| | measures the recovery time of the given process. | +| | | ++--------------+--------------------------------------------------------------+ +|operations | In this test case, the following operations are needed: | +| | 1. "nova-create-instance-in_network": create a VM instance | +| | in one of the existing neutron network. | +| | | ++--------------+--------------------------------------------------------------+ +|metrics | In this test case, there are two metrics: | +| | 1. process_recover_time: which indicates the maximun | +| | time (seconds) from the process being killed to | +| | recovered | +| | | +| | 2. packet_drop: measure the packets that have been dropped | +| | by the monitors using pktgen. | +| | | ++--------------+--------------------------------------------------------------+ +|test tool | Developed by the project. Please see folder: | +| | "yardstick/benchmark/scenarios/availability/ha_tools" | +| | | ++--------------+--------------------------------------------------------------+ +|references | TBD | +| | | ++--------------+--------------------------------------------------------------+ +|configuration | This test case needs two configuration files: | +| | 1. test case file: opnfv_yardstick_tc092.yaml | +| | - Attackers: see above “attackers” discription | +| | - Monitors: see above “monitors” discription | +| | - waiting_time: which is the time (seconds) from the | +| | process being killed to stoping monitors the | +| | monitors | +| | - SLA: see above “metrics” discription | +| | | +| | 2. POD file: pod.yaml The POD configuration should record | +| | on pod.yaml first. the “host” item in this test case | +| | will use the node name in the pod.yaml. | +| | | ++--------------+--------------------------------------------------------------+ +|test sequence | Description and expected result | +| | | ++--------------+--------------------------------------------------------------+ +|pre-action | 1. The OpenStack cluster is set up with an SDN controller | +| | running in a three node cluster configuration. | +| | | +| | 2. One or more neutron networks are created with two or | +| | more VMs attached to each of the neutron networks. | +| | | +| | 3. The neutron networks are attached to a neutron router | +| | which is attached to an external network the towards | +| | DCGW. | +| | | +| | 4. The master node of SDN controller cluster is known. | +| | | ++--------------+--------------------------------------------------------------+ +|step 1 | Start ip connectivity monitors: | +| | 1. Check the L2 connectivity between the VMs in the same | +| | neutron network. | +| | | +| | 2. Check the external connectivity of the VMs. | +| | | +| | Each monitor runs in an independent process. | +| | | +| | Result: The monitor info will be collected. | +| | | ++--------------+--------------------------------------------------------------+ +|step 2 | Start attacker: | +| | SSH to the VIM node and kill the SDN controller process | +| | determined in step 2. | +| | | +| | Result: One SDN controller service will be shut down | +| | | ++--------------+--------------------------------------------------------------+ +|step 3 | Restart the SDN controller. | +| | | ++--------------+--------------------------------------------------------------+ +|step 4 | Create a new VM in the existing Neutron network while the | +| | SDN controller is offline or still recovering. | +| | | ++--------------+--------------------------------------------------------------+ +|step 5 | Stop IP connectivity monitors after a period of time | +| | specified by “waiting_time” | +| | | +| | Result: The monitor info will be aggregated | +| | | ++--------------+--------------------------------------------------------------+ +|step 6 | Verify the IP connectivity monitor result | +| | | +| | Result: IP connectivity monitor should not have any packet | +| | drop failures reported | +| | | ++--------------+--------------------------------------------------------------+ +|step 7 | Verify process_recover_time, which indicates the maximun | +| | time (seconds) from the process being killed to recovered, | +| | is within the SLA. This step blocks until either the | +| | process has recovered or a timeout occurred. | +| | | +| | Result: process_recover_time is within SLA limits, if not, | +| | test case failed and stopped. | +| | | ++--------------+--------------------------------------------------------------+ +|step 8 | Start IP connectivity monitors for the new VM: | +| | 1. Check the L2 connectivity from the existing VMs to the | +| | new VM in the Neutron network. | +| | | +| | 2. Check connectivity from one VM to an external host on | +| | the Internet to verify SNAT functionality. | +| | | +| | Result: The monitor info will be collected. | +| | | ++--------------+--------------------------------------------------------------+ +|step 9 | Stop IP connectivity monitors after a period of time | +| | specified by “waiting_time” | +| | | +| | Result: The monitor info will be aggregated | +| | | ++--------------+--------------------------------------------------------------+ +|step 10 | Verify the IP connectivity monitor result | +| | | +| | Result: IP connectivity monitor should not have any packet | +| | drop failures reported | +| | | ++--------------+--------------------------------------------------------------+ +|test verdict | Fails only if SLA is not passed, or if there is a test case | +| | execution problem. | +| | | ++--------------+--------------------------------------------------------------+ + diff --git a/docs/testing/user/userguide/opnfv_yardstick_tc093.rst b/docs/testing/user/userguide/opnfv_yardstick_tc093.rst new file mode 100644 index 000000000..31fa5d3d3 --- /dev/null +++ b/docs/testing/user/userguide/opnfv_yardstick_tc093.rst @@ -0,0 +1,184 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International +.. License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) OPNFV, Intracom Telecom and others. +.. mardim@intracom-telecom.com + +************************************* +Yardstick Test Case Description TC093 +************************************* + ++-----------------------------------------------------------------------------+ +|SDN Vswitch resilience in non-HA or HA configuration | +| | ++--------------+--------------------------------------------------------------+ +|test case id | OPNFV_YARDSTICK_TC093: SDN Vswitch resilience in | +| | non-HA or HA configuration | ++--------------+--------------------------------------------------------------+ +|test purpose | This test validates that network data plane services are | +| | resilient in the event of Virtual Switch failure | +| | in compute nodes. Specifically, the test verifies that | +| | existing data plane connectivity is not permanently impacted | +| | i.e. all configured network services such as DHCP, ARP, L2, | +| | L3 Security Groups continue to operate between the existing | +| | VMs eventually after the Virtual Switches have finished | +| | rebooting. | +| | | +| | The test also validates that new network service operations | +| | (creating a new VM in the existing L2/L3 network or in a new | +| | network, etc.) are operational after the Virtual Switches | +| | have recovered from a failure. | +| | | ++--------------+--------------------------------------------------------------+ +|test method | This testcase first checks if the already configured | +| | DHCP/ARP/L2/L3/SNAT connectivity is proper. After | +| | it fails and restarts again the VSwitch services which are | +| | running on both OpenStack compute nodes, and then checks if | +| | already configured DHCP/ARP/L2/L3/SNAT connectivity is not | +| | permanently impacted (even if there are some packet | +| | loss events) between VMs and the system is able to execute | +| | new virtual network operations once the Vswitch services | +| | are restarted and have been fully recovered | +| | | ++--------------+--------------------------------------------------------------+ +|attackers | In this test case, two attackers called “kill-process” are | +| | needed. These attackers include three parameters: | +| | 1. fault_type: which is used for finding the attacker's | +| | scripts. It should be set to 'kill-process' in this test | +| | | +| | 2. process_name: should be set to the name of the Vswitch | +| | process | +| | | +| | 3. host: which is the name of the compute node where the | +| | Vswitch process is running | +| | | +| | e.g. -fault_type: "kill-process" | +| | -process_name: "openvswitch" | +| | -host: node1 | +| | | ++--------------+--------------------------------------------------------------+ +|monitors | This test case utilizes two monitors of type "ip-status" | +| | and one monitor of type "process" to track the following | +| | conditions: | +| | 1. "ping_same_network_l2": monitor ICMP traffic between | +| | VMs in the same Neutron network | +| | | +| | 2. "ping_external_snat": monitor ICMP traffic from VMs to | +| | an external host on the Internet to verify SNAT | +| | functionality. | +| | | +| | 3. "Vswitch process monitor": a monitor checking the | +| | state of the specified Vswitch process. It measures | +| | the recovery time of the given process. | +| | | +| | Monitors of type "ip-status" use the "ping" utility to | +| | verify reachability of a given target IP. | +| | | ++--------------+--------------------------------------------------------------+ +|operations | In this test case, the following operations are needed: | +| | 1. "nova-create-instance-in_network": create a VM instance | +| | in one of the existing Neutron network. | +| | | ++--------------+--------------------------------------------------------------+ +|metrics | In this test case, there are two metrics: | +| | 1. process_recover_time: which indicates the maximun | +| | time (seconds) from the process being killed to | +| | recovered | +| | | +| | 2. outage_time: measures the total time in which | +| | monitors were failing in their tasks (e.g. total time of | +| | Ping failure) | +| | | ++--------------+--------------------------------------------------------------+ +|test tool | Developed by the project. Please see folder: | +| | "yardstick/benchmark/scenarios/availability/ha_tools" | +| | | ++--------------+--------------------------------------------------------------+ +|references | none | +| | | ++--------------+--------------------------------------------------------------+ +|configuration | This test case needs two configuration files: | +| | 1. test case file: opnfv_yardstick_tc093.yaml | +| | - Attackers: see above “attackers” description | +| | - monitor_time: which is the time (seconds) from | +| | starting to stoping the monitors | +| | - Monitors: see above “monitors” discription | +| | - SLA: see above “metrics” description | +| | | +| | 2. POD file: pod.yaml The POD configuration should record | +| | on pod.yaml first. the “host” item in this test case | +| | will use the node name in the pod.yaml. | +| | | ++--------------+--------------------------------------------------------------+ +|test sequence | Description and expected result | +| | | ++--------------+--------------------------------------------------------------+ +|pre-action | 1. The Vswitches are set up in both compute nodes. | +| | | +| | 2. One or more Neutron networks are created with two or | +| | more VMs attached to each of the Neutron networks. | +| | | +| | 3. The Neutron networks are attached to a Neutron router | +| | which is attached to an external network towards the | +| | DCGW. | +| | | ++--------------+--------------------------------------------------------------+ +|step 1 | Start IP connectivity monitors: | +| | 1. Check the L2 connectivity between the VMs in the same | +| | Neutron network. | +| | | +| | 2. Check connectivity from one VM to an external host on | +| | the Internet to verify SNAT functionality. | +| | | +| | Result: The monitor info will be collected. | +| | | ++--------------+--------------------------------------------------------------+ +|step 2 | Start attackers: | +| | SSH connect to the VIM compute nodes and kill the Vswitch | +| | processes | +| | | +| | Result: the SDN Vswitch services will be shutdown | +| | | ++--------------+--------------------------------------------------------------+ +|step 3 | Verify the results of the IP connectivity monitors. | +| | | +| | Result: The outage_time metric reported by the monitors | +| | is not greater than the max_outage_time. | +| | | ++--------------+--------------------------------------------------------------+ +|step 4 | Restart the SDN Vswitch services. | +| | | ++--------------+--------------------------------------------------------------+ +|step 5 | Create a new VM in the existing Neutron network | +| | | ++--------------+--------------------------------------------------------------+ +|step 6 | Verify connectivity between VMs as follows: | +| | 1. Check the L2 connectivity between the previously | +| | existing VM and the newly created VM on the same | +| | Neutron network by sending ICMP messages | +| | | ++--------------+--------------------------------------------------------------+ +|step 7 | Stop IP connectivity monitors after a period of time | +| | specified by “monitor_time” | +| | | +| | Result: The monitor info will be aggregated | +| | | ++--------------+--------------------------------------------------------------+ +|step 8 | Verify the IP connectivity monitor results | +| | | +| | Result: IP connectivity monitor should not have any packet | +| | drop failures reported | +| | | ++--------------+--------------------------------------------------------------+ +|test verdict | This test fails if the SLAs are not met or if there is a | +| | test case execution problem. The SLAs are define as follows | +| | for this test: | +| | * SDN Vswitch recovery | +| | * process_recover_time <= 30 sec | +| | | +| | * no impact on data plane connectivity during SDN | +| | Vswitch failure and recovery. | +| | * packet_drop == 0 | +| | | ++--------------+--------------------------------------------------------------+ + |