openBackhaul / ethernetContainer

Technology specific interface definition for an Ethernet Container
Apache License 2.0
1 stars 0 forks source link

Default value for header-compression-name #31

Closed openBackhaul closed 3 years ago

openBackhaul commented 4 years ago

header-compression-name is a key attribute (default value cannot be assigned according to YANG specifications).

Proposal: Because it is requested that the list shall contain at least a single instance, it is proposed that the header-compression-name for devices, which do not support header compression, shall be defined to be "None" in the comment of the header-compression-name attribute.

openBackhaul commented 4 years ago

During today's 5G-xhaul sub project call a different proposal has been made: Instead of defining a default value for the headerCompressionName attribute in comments, the headerCompressionModeType enumeration shall be amended by an additional value "NO_COMPRESSION".

This value shall close two gaps: 1) In case header compression is not supported at the device, this value of headerCompressionMode shall be used for the only instance of HeaderCompressionKind, which is required in this case. 2) In case header compression is supported by the hardware, but shall be switched off, the configuration attribute for choosing the header compression mode shall point at this instance .

Vendors to please check, whether this modeling would cover their implementations.

openBackhaul commented 4 years ago

Solution: There will be no prescribed default value for header-compression-name. The instance of HeaderCompressionKind that has been created according to solution in issue#29 has to be referenced as a default.

openBackhaul commented 3 years ago

Decision on the DMIP call on 27th of May 2021: In issue#29, it is proposed to always instantiate at least a single instance of the HeaderCompressionKind datatype, which is filled with headerCompressionMode == "NO_COMPRESSION" and a headerCompressionName that is free to be chosen by the vendor. This headerCompressionName is substituting the need for a default value of the headerCompressionName (which is anyway a key attribute, that cannot define a default value). Because no Default value for header-compression-name is required after decision made in regards to issue#29, the issue has been closed without further action.