docs/testing/user/userguide/opnfv_yardstick_tc054.rst

   1 .. This work is licensed under a Creative Commons Attribution 4.0 International
   2 .. License.
   3 .. http://creativecommons.org/licenses/by/4.0
   4 .. (c) OPNFV, Yin Kanglin and others.
   5 .. 14_ykl@tongji.edu.cn
   6
   7 *************************************
   8 Yardstick Test Case Description TC054
   9 *************************************
  10
  11 +-----------------------------------------------------------------------------+
  12 |OpenStack Virtual IP High Availability                                       |
  13 |                                                                             |
  14 +--------------+--------------------------------------------------------------+
  15 |test case id  | OPNFV_YARDSTICK_TC054: OpenStack Virtual IP High             |
  16 |              | Availability                                                 |
  17 +--------------+--------------------------------------------------------------+
  18 |test purpose  | This test case will verify the high availability for virtual |
  19 |              | ip in the environment. When master node of virtual ip is     |
  20 |              | abnormally shutdown, connection to virtual ip and            |
  21 |              | the services binded to the virtual IP it should be OK.       |
  22 +--------------+--------------------------------------------------------------+
  23 |test method   | This test case shutdowns the virtual IP master node with     |
  24 |              | some fault injection tools, then checks whether virtual ips  |
  25 |              | can be pinged and services binded to virtual ip are OK with  |
  26 |              | some monitor tools.                                          |
  27 +--------------+--------------------------------------------------------------+
  28 |attackers     | In this test case, an attacker called "control-shutdown" is  |
  29 |              | needed. This attacker includes two parameters:               |
  30 |              | 1) fault_type: which is used for finding the attacker's      |
  31 |              | scripts. It should be always set to "control-shutdown" in    |
  32 |              | this test case.                                              |
  33 |              | 2) host: which is the name of a control node being attacked. |
  34 |              |                                                              |
  35 |              | In this case the host should be the virtual ip master node,  |
  36 |              | that means the host ip is the virtual ip, for exapmle:       |
  37 |              | -fault_type: "control-shutdown"                              |
  38 |              | -host: node1(the VIP Master node)                            |
  39 +--------------+--------------------------------------------------------------+
  40 |monitors      | In this test case, two kinds of monitor are needed:          |
  41 |              | 1. the "ip_status" monitor that pings a specific ip to check |
  42 |              | the connectivity of this ip, which needs two parameters:     |
  43 |              | 1) monitor_type: which is used for finding the monitor class |
  44 |              | and related scripts. It should be always set to "ip_status"  |
  45 |              | for this monitor.                                            |
  46 |              | 2) ip_address: The ip to be pinged. In this case, ip_address |
  47 |              | should be the virtual IP.                                    |
  48 |              |                                                              |
  49 |              | 2. the "openstack-cmd" monitor constantly request a specific |
  50 |              | Openstack command, which needs two parameters:               |
  51 |              | 1) monitor_type: which is used for finding the monitor class |
  52 |              | and related scripts. It should be always set to              |
  53 |              | "openstack-cmd" for this monitor.                            |
  54 |              | 2) command_name: which is the command name used for request. |
  55 |              |                                                              |
  56 |              | e.g.                                                         |
  57 |              | monitor1:                                                    |
  58 |              | -monitor_type: "ip_status"                                   |
  59 |              | -host: 192.168.0.2                                           |
  60 |              | monitor2:                                                    |
  61 |              | -monitor_type: "openstack-cmd"                               |
  62 |              | -command_name: "nova image-list"                             |
  63 |              |                                                              |
  64 +--------------+--------------------------------------------------------------+
  65 |metrics       | In this test case, there are two metrics:                    |
  66 |              | 1) ping_outage_time: which-indicates the maximum outage time |
  67 |              | to ping the specified host.                                  |
  68 |              | 2)service_outage_time: which indicates the maximum outage    |
  69 |              | time (seconds) of the specified Openstack command request.   |
  70 +--------------+--------------------------------------------------------------+
  71 |test tool     | Developed by the project. Please see folder:                 |
  72 |              | "yardstick/benchmark/scenarios/availability/ha_tools"        |
  73 |              |                                                              |
  74 +--------------+--------------------------------------------------------------+
  75 |references    | ETSI NFV REL001                                              |
  76 |              |                                                              |
  77 +--------------+--------------------------------------------------------------+
  78 |configuration | This test case needs two configuration files:                |
  79 |              | 1) test case file: opnfv_yardstick_tc054.yaml                |
  80 |              | -Attackers: see above "attackers" discription                |
  81 |              | -waiting_time: which is the time (seconds) from the process  |
  82 |              | being killed to stoping monitors the monitors                |
  83 |              | -Monitors: see above "monitors" discription                  |
  84 |              | -SLA: see above "metrics" discription                        |
  85 |              |                                                              |
  86 |              | 2)POD file: pod.yaml                                         |
  87 |              | The POD configuration should record on pod.yaml first.       |
  88 |              | the "host" item in this test case will use the node name in  |
  89 |              | the pod.yaml.                                                |
  90 |              |                                                              |
  91 +--------------+--------------------------------------------------------------+
  92 |test sequence | description and expected result                              |
  93 |              |                                                              |
  94 +--------------+--------------------------------------------------------------+
  95 |step 1        | start monitors:                                              |
  96 |              | each monitor will run with independently process             |
  97 |              |                                                              |
  98 |              | Result: The monitor info will be collected.                  |
  99 |              |                                                              |
 100 +--------------+--------------------------------------------------------------+
 101 |step 2        | do attacker: connect the host through SSH, and then execute  |
 102 |              | the shutdown script on the VIP master node.                  |
 103 |              |                                                              |
 104 |              | Result: VIP master node will be shutdown                     |
 105 |              |                                                              |
 106 +--------------+--------------------------------------------------------------+
 107 |step 3        | stop monitors after a period of time specified by            |
 108 |              | "waiting_time"                                               |
 109 |              |                                                              |
 110 |              | Result: The monitor info will be aggregated.                 |
 111 |              |                                                              |
 112 +--------------+--------------------------------------------------------------+
 113 |step 4        | verify the SLA                                               |
 114 |              |                                                              |
 115 |              | Result: The test case is passed or not.                      |
 116 |              |                                                              |
 117 +--------------+--------------------------------------------------------------+
 118 |post-action   | It is the action when the test cases exist.  It restarts the |
 119 |              | original VIP master node if it is not restarted.             |
 120 |              |                                                              |
 121 +--------------+--------------------------------------------------------------+
 122 |test verdict  | Fails only if SLA is not passed, or if there is a test case  |
 123 |              | execution problem.                                           |
 124 |              |                                                              |
 125 +--------------+--------------------------------------------------------------+