1 .. image:: ../../etc/opnfv-logo.png
7 *************************************
8 Yardstick Test Case Description TC025
9 *************************************
11 +-----------------------------------------------------------------------------+
12 |OpenStack Controll Node abnormally shutdown High Availability |
13 +==============+==============================================================+
14 |test case id | OPNFV_YARDSTICK_TC025_HA: OpenStack Controll Node abnormally |
16 +--------------+--------------------------------------------------------------+
17 |test purpose | This test case will verify the high availability of control |
18 | | nodethe. When one of the control node abnormally shutdown, |
19 | | the service provided by it should be OK. |
20 +--------------+--------------------------------------------------------------+
21 |test method | This test case shutdowns a specified control node with some |
22 | | fault injection tools, then checks whether all services |
23 | | provided by the control node are OK with some monitor tools. |
24 +--------------+--------------------------------------------------------------+
25 |attackers | In this test case, an attacker called "host-shutdown" is |
26 | | needed. This attacker includes two parameters: |
27 | | 1) fault_type: which is used for finding the attacker's |
28 | | scripts. It should be always set to "host-shutdown" in |
30 | | 2) host: which is the name of a control node being attacked. |
33 | | -fault_type: "host-shutdown" |
35 +--------------+--------------------------------------------------------------+
36 |monitors | In this test case, one kind of monitor are needed: |
37 | | 1. the "openstack-cmd" monitor constantly request a specific |
38 | | Openstack command, which needs two parameters |
39 | | 1) monitor_type: which is used for finding the monitor class |
40 | | and related scritps. It should be always set to |
41 | | "openstack-cmd" for this monitor. |
42 | | 2) command_name: which is the command name used for request |
44 | | There are four instance of the "openstack-cmd" monitor: |
46 | | -monitor_type: "openstack-cmd" |
47 | | -api_name: "nova image-list" |
49 | | -monitor_type: "openstack-cmd" |
50 | | -api_name: "neutron router-list" |
52 | | -monitor_type: "openstack-cmd" |
53 | | -api_name: "heat stack-list" |
55 | | -monitor_type: "openstack-cmd" |
56 | | -api_name: "cinder list" |
57 +--------------+--------------------------------------------------------------+
58 |metrics | In this test case, there is one metric: |
59 | | 1)service_outage_time: which indicates the maximum outage |
60 | | time (seconds) of the specified Openstack command request. |
61 +--------------+--------------------------------------------------------------+
62 |test tool | None. Self-developed. |
63 +--------------+--------------------------------------------------------------+
64 |references | ETSI NFV REL001 |
65 +--------------+--------------------------------------------------------------+
66 |configuration | This test case needs two configuration files: |
67 | | 1) test case file: opnfv_yardstick_tc019.yaml |
68 | | -Attackers: see above "attackers" discription |
69 | | -waiting_time: which is the time (seconds) from the process |
70 | | being killed to stoping monitors the monitors |
71 | | -Monitors: see above "monitors" discription |
72 | | -SLA: see above "metrics" discription |
74 | | 2)POD file: pod.yaml |
75 | | The POD configuration should record on pod.yaml first. |
76 | | the "host" item in this test case will use the node name in |
78 +--------------+------+----------------------------------+--------------------+
79 |test sequence | step | description | result |
80 | +------+----------------------------------+--------------------+
81 | | 1 | start monitors: each monitor will| The monitor info |
82 | | | run with independently process | will be collected. |
83 | +------+----------------------------------+--------------------+
84 | | 2 | do attacker: connect the host | the host will be |
85 | | | through SSH, and then execute | shutdown. |
86 | | | shutdown script on the host | |
87 | +------+----------------------------------+--------------------+
88 | | 3 | stop monitors after a period of | All monitor result |
89 | | | time specified by "waiting_time" | will be aggregated.|
90 | +------+----------------------------------+--------------------+
91 | | 4 | verify the SLA | The test case is |
92 | | | | passed or not. |
93 +--------------+------+----------------------------------+--------------------+
94 |post-action | It is the action when the test cases exist. It restarts the |
95 | | specified control node if it is not restarted. |
96 +--------------+------+----------------------------------+--------------------+
97 |test verdict | Fails only if SLA is not passed, or if there is a test case |
98 | | execution problem. |
99 +--------------+--------------------------------------------------------------+