Skip to end of metadata
Go to start of metadata

Date

18 September 2018

Attendees

Goals

  • Admin
  • IEEE 802.1Qcx CFM YANG & UML (Bernd & Kam)
  • TAPI OAM (Andrea & Karthik)

Discussion Items

TimeItemWhoNotes
5 minAdminKam
15 minIEEE & ITU-T Ethernet OAMBernd, Kam
  • Q14/15 September 17 IM/DM Modeling Coordination discussion
    • IEEE 802.1Qcx CFM YANG (https://github.com/YangModels/yang/tree/master/standard/ieee/802.1/draft) and UML (cd06r4_BZ__PapyrusIeeeCfmOxygenWorkspace_180731.zip)
    • Inconsistency in having “-Grouping” at the end of the grouping label
      • Marc: There is no best-practice convention. He will make them be consistent
    • Inconsistency in having “-type” at the end of the typedef label
      • Marc: There is no best-practice convention. He will leave the typedef as they are and not to make them consistent
    • Extra level of containers in the CMF YANG (and thus in the literally translated UML) 
      • Marc: In YANG, there is no rule nor best practice convention to have the extra level of containers. Whether have or not have the extra level of container is purely for readability, and readability is subjective.
      • The meeting prefer consistency across and so that can have rules for the translation tool.
      • The meeting incline to removing the middle level of container so that have simpler YANG and UML
  • New version of IEEE 802.1Qcx CFM UML from Bernd has been posted oimt2018.BZ.001.07_PapyrusIeeeCfmOxygenWorkspace_180918.zip.
    • Updates made:
      • Made the association ends from CfmOperation to MaintenanceAssociationGroup and Mep mandatory and marked as PartOfObjectKey


      • Adapted the modelling of the bit set data types to the latest guidelines:

 Model profileBernd
  • Bernd reported problem in migrating the UML model to the latest Profile versions (the official one from the Github repository)
    • The migration of the UML model to the latest Profile versions appeared to be a challenge

      • After applying the new Profile to the model, he was no longer able to delete any artifact from the model.

      • So he was forced to update the Profile – that was already applied to the model – with the bit set properties in parallel à i.e., the IEEE CFM UML model has now a different instance (same content) of the Profile than the official Profile contained in the EAGLE repository

    • Need to investigate what went wrong.

40 minTAPI OAMAndrea, Karthik
  • Andrea present the TAPI OAM for Ethernet

    • He noted that while there was comment that PM history data should be persistence even when the measurement job is deleted, the TAPI model follow the G.8052 approach that current data and History data are lifecycle dependent on the measurement job.
      • The administrativeState of the OamJob is used for pause/resume the OAM job. History data remains persistent while the job is paused (suspended).

    • In response to question raised, he explained the Bin and Bin profile per the requiremnet of MEF 35.1, Basically the MEF bin application is for counting number of occurrence of measurements (e.g., Delay measurement) fall within the range of the defined bins.

    • The TAPI OAM Job model covers both the G.8052 measurement jobs (e.g., delay measurement, loss measurement; proactive, on demand) and maintenance jobs (e.g., loopback, link trace).
    • The type of the OamJob is indicated by EthOamJobType

 

Action Items

  •