Dear GROW Working Group, Thank you very much for your valuable questions and suggestions on our draft "BMP Extension for Configuration Monitoring TLV" during the IETF 126 meeting. Our responses are summarized as follows: Question 1: Can the framework of the REL draft support the use cases of this draft? We believe that REL and this draft have different positioning. They are orthogonal dimensions in the problem space rather than alternatives: REL (draft-ietf-grow-bmp-rel) defines a new BMP message type ― Route Event Logging (REL) for event-driven routing event logging. When a routing event occurs, the router issues an additional, independent log message describing that event. Consumers then correlate this log message with other BMP messages. The event subject of REL messages is NLRI (prefix), meaning the routing change itself is the core object of logging. The Config Monitor TLV (this draft) is used to carry configuration information that has a causal relationship with BGP routing changes. Configuration information itself is the subject, presented in structured fields including configuration session identifier, source, and timestamp. This draft defines an additional annotation: a TLV attached to messages already issued by the router (Route Monitoring, Peer Up/Down) that carries the configuration change context leading to that message. The BMP monitor does not need to perform cross-message correlation, and this annotation cannot be separated from the message it explains due to loss or reordering. Its purpose is to provide traceable evidence for locating routing anomalies and network failures triggered by configuration changes. Meanwhile, when the Log Action TLV (Type 6, Code 1 = Config) in a REL message indicates that the event is triggered by a configuration statement, we also recommend carrying the Config Monitor TLV simultaneously ― from this perspective, the two drafts can work together. In terms of format, the TLV in this draft strictly follows the extended TLV format defined in "BMP v4: Extended TLV Support for BGP Monitoring Protocol (BMP)" (draft-ietf-grow-bmp-tlv). We hope to apply for a new type code for this TLV. Question 2: Regarding the issue of redefining timestamps The timestamp format in this draft strictly follows the timestamp format defined in "BMP v4: Extended TLV Support for BGP Monitoring Protocol (BMP)" (draft-ietf-grow-bmp-tlv). We are not redefining the timestamp encoding method; rather, we are applying for new type and code points for the configuration effective timestamp under the existing framework. We hope the above responses can address your questions. Please feel free to raise any further questions or suggestions at any time. Draft link: https://datatracker.ietf.org/doc/draft-gao-grow-bmp-config-monitor-tlv/ Kind regards, Gao Xing
高星
_______________________________________________ GROW mailing list -- [email protected] To unsubscribe send an email to [email protected]
