Hi All,
As discussed in the summit, "vdu_autoheal" monitoring policy action implementation should be as per ETSI standard HealVnfRequest interface [1] (refer comment given on patch
https://review.openstack.org/#/c/612595/9 )
Started working on the same but have some quires so needs some feedback and inputs.
As per the ETSI doc [1] the data model for HealVnfRequest contains two request parameters (cause anddditionalParams)
(Check section: 5.5.2.9 Type: HealVnfReques from [1])
The Request Payload for HealVnfRequest (ETSI data structure: HealVnfRequest) should be the dictionary as mentioned in [2]
(Check section: Healing VNFs Using ETSI API from [2])
But the “addtionalParams” parameter should be part of "HealVnfOpConfig” and that expects two parameters (cause and parameter) as mentioned in [3]
(Check section: 7.1.5.6 HealVnfOpConfig information element from [3])
So accordingly I have designed the request data for vdu_autoheal monitoring policy action as below:
Heal Request Data for monitoring:
{
"cause": "",
"additionalParams": [
{“parameter”: <Resource name given in the vnfd template for eg. VDU1>, “cause”: [<Reason for failure of the resource>]},