1. Pages
2. …
3. OIMT
4. Core Model
5. OIMT minutes
6. 2021 OIMT Minutes

2021-03-25 OIMT Meeting notes

Daylight Saving Time

• DST USA will start next Sunday 14th March
• See https://www.timeanddate.com/time/dst/2021a.html for other location

Need to arrange virtual face to face meetings.

Sensor model (ONF_T58_Sensor_Model.pptx)

• Analogue, digital, the unit measuring, measured thing (remote)
• IETF have only one of the four things

Equipment model (ONF_T08_EquipmentUpdate.pptx) presentation/discussion

• Chris noted that he had reviewed the standard IETF attributes as a source for consideration, not suggesting that we should change to that model
• Move to augmentation on the basic model as opposed to a big model
• Use of spec and occurrence
• Applying Aggregates pattern
• Expected equipment
• Type invariant detail could be in the class if there is no spec. {{Editor's note: There are many type invariant and clearly most are "conditional optional"...}}
• The presence of the sensor is type invariant (knowledge layer and expectation), but the settings and measures are clearly only present on the actual instance..
• The pattern is being applied extensively in the pack, but it does not mean that it needs to be present for all cases etc.
• Noted that the software revision may be in the software model if such a model is present {{Editor's note: Some details may be applied by alternative conditional decoration/augmentation, e.g., where there is a rich software model then the software properties would be in that model whereas if there is no software model the attributes may appear on the hardware. There does appear to be a general modeling principle here.}}
• Aggregate leaf cannot be decorated. Aggregate leaf is not the same as Yang leaf where decoration via augmentation is normal {{Editor's note: There is potentially a conflation of two concepts in the consideration of aggregates that we do need to discuss. }}

FRU and Non-FRU

• Focus on inventory. What is in the inventory and what not. Chris suggested that we may have focused on the wrong thing. 
• FRU
  • has no holders and cannot be in holders
  • does not have connectors
  • Is a leaf
  • May need a spec
• Physical connector is a root in the model presented
• The equipment is designed to be field replaceable of not
  • This could be in the spec layer 
  • Note that the owner may decide not to ever field replace a non-field replaceable
• Some non-FRUs (e.g., a sensor) will not have a type presented. A non-FRU of this sort cannot have a spec of its own (as there is no reference)
• FRU and non-FRU confused the issue... 
• Slot cover is an FRU but may have no serial number
• Constructed a view of non-FRU/FRU
• Non-FRU is a manufacture choice... "if you remove that part your warrantee is violated"
• The challenge is whether a non-FRU can be in an inventory independent of its containing FRU.
• There could be a distinction between the model of a deployed equipment... "can I pull it out or not" v the inventory.. "is it inventoried or not".
• FRU v non-FRU can be overridden by the owner.... I can choose to work with the assembly and not dismantle it, I can choose to violate the warrantee.
• The Core model has focused on Heavy FRUs
• TAPI is focused on the detectable deployed equipment where that equipment supports transport functions.