Hi everyone, I'm currently working on a new open-source IS-IS implementation [1] based on the [email protected] module. Overall, I think the module is very well done, but I'd like to share a few points of feedback that may be worth addressing in future revisions.
1 - The two "discontinuity-time" leaves are implicitly marked as "config write", which is certainly unintended. 2 - The "corrupted-lsp-detected" and other notifications are missing a leaf to specify the LSP level. 3 - The "sequence" leaf within the "lsp-generation" notification has the description: "Sequence number of the received LSP". I believe this should be updated to "Sequence number of the regenerated LSP". 4 - The "circuit-id" typedef is defined but not used anywhere. Additionally, I believe the "interface" list would benefit from a "circuit-id" leaf (I addressed this in Holo using an augmentation). 5 - The "length" leaf within "unknown-tlv" is defined as "uint16" though "uint8" would be more appropriate, unless this is intended for future-proofing. 6 - In "packet-counters", it's unclear how to classify whether an unknown incoming PDU belongs to level 1 or level 2. It seems the "unknown" leaf should be placed outside the "level" list. 7 - In the "ipv6-reachability" container, a leaf indicating whether the external original bit is set is missing. 8 - The "decoded-completed" leaf is described as: "The IS-IS LSP body has been fully decoded". Does "fully decoded" mean that no unknown TLVs or Sub-TLVs are present? Thanks for your time, and I hope you find this feedback useful. [1] https://github.com/holo-routing/holo Best regards, -- Renato Westphal _______________________________________________ Lsr mailing list -- [email protected] To unsubscribe send an email to [email protected]
