Hi Cristian Thanks for your inputs.
@Chongfeng Xie<mailto:[email protected]> Appreciate your inputs & comments. We will handle (1) and (2). On (4) you may wish to refer below mail chain which was discussed earlier in the group. https://mailarchive.ietf.org/arch/msg/opsawg/jeiZUq7BdO0OGjKsG9QYd85OWT0/ Let me know if any questions. Regards —Sriram From: Staicu, Cristian <[email protected]> Date: Friday, 31 October 2025 at 10:22 PM To: Chongfeng Xie <[email protected]> Cc: opsawg-chairs <[email protected]>, opsawg <[email protected]> Subject: [OPSAWG]Re: Comments on draft-ietf-opsawg-ipfix-gtpu Hi Chongfeng, Related to item 3) I am trying to describe the normal 3GPP use case in 5G SA where for a single PDU Session multiple flows with different gtpuQFI will have the same gtpuTEID. Related to item 5) - The use of the IPFIX GTP-U IEs is to identify either a particular PDU Session on a particular slice (using gtpTEid in combination with the gNB and the UPD IP addresses) or all the IP traffic flows for a 3GPP QoS flow on a slice (using gtpuQFI in combination with the gNB and the UPF IP addresses ) or a 3GPP QoS flow for a particular PDU Session on a slice by using both gtpTEid and gtpuQFI in combination with the gNB and UPF IP addresses). Regards, Cristian From: Chongfeng Xie <[email protected]> Date: Friday, October 31, 2025 at 03:52 To: opsawg <[email protected]> Cc: opsawg-chairs <[email protected]> Subject: [EXT][OPSAWG]Comments on draft-ietf-opsawg-ipfix-gtpu Hi authors, I have reviewed draft-ietf-opsawg-ipfix-gtpu, and have the following editorial suggestions and questions, Editorial suggestions, 1) In the abstract section , there is, "Generic Packet Radio Service Tunneling Protocol" It can be changed to, "Generic Packet Radio Service Tunneling Protocol(GTP)" 2) In section 1, there is, "A dedicated header, called GPRS Tunneling Protocol Header (GTP)," It can be changed to "A dedicated header, called GPRS Tunneling Protocol (GTP) Header, " Some questions related to section 4 (Sample Use Cases), 3) In section 4, there is, “It is possible that within the same PDU session there might be multiple IP flows having the same attributes.” What is the key point does this sentence intend to express? What's the relationship of it with the IPFIX GTP-U IEs being defined in this draft? 4) In Section 4, there is, “exporting GTP-U IEs together with transportChecksum IE would help detect misbehaving nodes particularly when using GTP-U over IPv6.” What are the specific particularities in the context of IPv6? Could this be elaborated a little more? 5) In section 4, The second paragrph illustrate how to identify a traffic flow in the case of when a set of UPFs are deployed per 5G slice, the information used is either existing GTP-U fields or the IP addresses of the devices, can you please elaborate on the role of IPFIX GTP-U IEs in this case? Thank you! Best regards Chongfeng
_______________________________________________ OPSAWG mailing list -- [email protected] To unsubscribe send an email to [email protected]
