1 .. This work is licensed under a Creative Commons Attribution 4.0 International License.
2 .. http://creativecommons.org/licenses/by/4.0
9 Doctor platform provides these features in `Colorado Release <https://wiki.opnfv.org/display/SWREL/Colorado>`_:
11 * Immediate Notification
12 * Consistent resource state awareness for compute host down
13 * Valid compute host status given to VM owner
15 These features enable high availability of Network Services on top of
16 the virtualized infrastructure. Immediate notification allows VNF managers
17 (VNFM) to process recovery actions promptly once a failure has occurred.
19 Consistency of resource state is necessary to execute recovery actions
22 Ability to query host status gives VM owner the possibility to get
23 consistent state information through an API in case of a compute host
26 The Doctor platform consists of the following components:
28 * OpenStack Compute (Nova)
29 * OpenStack Telemetry (Ceilometer)
30 * OpenStack Alarming (Aodh)
35 Doctor Inspector and Monitor are sample implementations for reference.
37 You can see an overview of the Doctor platform and how components interact in
40 .. figure:: ./images/figure-p1.png
44 Doctor platform and typical sequence (Colorado)
46 Detailed information on the Doctor architecture can be found in the Doctor
47 requirements documentation:
48 http://artifacts.opnfv.org/doctor/colorado/requirements/05-implementation.html
53 * A consumer of the NFVI wants to receive immediate notifications about faults
54 in the NFVI affecting the proper functioning of the virtual resources.
55 Therefore, such faults have to be detected as quickly as possible, and, when
56 a critical error is observed, the affected consumer is immediately informed
57 about the fault and can switch over to the STBY configuration.
59 The faults to be monitored (and at which detection rate) will be configured by
60 the consumer. Once a fault is detected, the Inspector in the Doctor
61 architecture will check the resource map maintained by the Controller, to find
62 out which virtual resources are affected and then update the resources state.
63 The Notifier will receive the failure event requests sent from the Controller,
64 and notify the consumer(s) of the affected resources according to the alarm
67 Detailed workflow information is as follows:
69 * Consumer(VNFM): (step 0) creates resources (network, server/instance) and an
70 event alarm on state down notification of that server/instance
72 * Monitor: (step 1) periodically checks nodes, such as ping from/to each
73 dplane nic to/from gw of node, (step 2) once it fails to send out event
74 with "raw" fault event information to Inspector
76 * Inspector: when it receives an event, it will (step 3) mark the host down
77 ("mark-host-down"), (step 4) map the PM to VM, and change the VM status to
80 * Controller: (step 5) sends out instance update event to Ceilometer
82 * Notifier: (step 6) Ceilometer transforms and passes the event to Aodh,
83 (step 7) Aodh will evaluate event with the registered alarm definitions,
84 then (step 8) it will fire the alarm to the "consumer" who owns the
87 * Consumer(VNFM): (step 9) receives the event and (step 10) recreates a new
93 Functest will call the "run.sh" script in Doctor to run the test job.
95 Currently, only 'Apex' and 'local' installer are supported.
97 The "run.sh" script will execute the following steps.
99 Firstly, prepare image for booting VM, then create a test project and test
100 user (both default to doctor) for the Doctor tests.
102 Secondly, boot a VM under the doctor project and check the VM status to verify
103 that the VM is launched completely. Then get the compute host info where the VM
104 is launched to verify connectivity to the target compute host. Get the consumer
105 ip according to the route to compute ip and create an alarm event in Ceilometer
106 using the consumer ip.
108 Thirdly, the Doctor components are started, and, based on the above preparation,
109 a failure is injected to the system, i.e. the network of compute host is
110 disabled for 3 minutes. To ensure the host is down, the status of the host
113 Finally, the notification time, i.e. the time between the execution of step 2
114 (Monitor detects failure) and step 9 (Consumer receives failure notification)
117 According to the Doctor requirements, the Doctor test is successful if the
118 notification time is below 1 second.