neels has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-smlc/+/20470 )
Change subject: initial working osmo-smlc implementation ...................................................................... Patch Set 7: (1 comment) https://gerrit.osmocom.org/c/osmo-smlc/+/20470/6/src/osmo-smlc/lb_peer.c File src/osmo-smlc/lb_peer.c: https://gerrit.osmocom.org/c/osmo-smlc/+/20470/6/src/osmo-smlc/lb_peer.c@463 PS6, Line 463: " Dropping message.\n"); > I don't recall all the details of the protocol spec. […] by spec, a connection request can be answered with either a conn confirm or a conn refused. however this here is a weird special case: a conn is already established for a given conn id, but we are receiving another Connection Request for the same conn id. what would that mean? - if it is the same peer sending it, that's a bug / protocol error. - assume the peer has restarted and tries to re-use a previous conn id, it would make sense to discard the previous conn. But then again, there should have been a RESET from that peer that has done just that. - If it is a different peer using the same conn id ... can that happen? in that case a Conn Refused might be in order. so, if it is the same peer, it's a weird corner case that shouldn't happen in practice, not really sure how to deal with it. i think dropping the message like in this patch is the easiest / least invasive way until we see a real situation that suggests a useful way to respond ... ? but let's clarify: is a different peer colliding with another peer's conn id possible? if yes, would it even propagate to this layer? -- To view, visit https://gerrit.osmocom.org/c/osmo-smlc/+/20470 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-smlc Gerrit-Branch: master Gerrit-Change-Id: I917ba8fc51a1f1150be77ae01e12a7b16a853052 Gerrit-Change-Number: 20470 Gerrit-PatchSet: 7 Gerrit-Owner: neels <[email protected]> Gerrit-Reviewer: Jenkins Builder Gerrit-CC: fixeria <[email protected]> Gerrit-CC: laforge <[email protected]> Gerrit-Comment-Date: Fri, 16 Oct 2020 13:36:03 +0000 Gerrit-HasComments: Yes Gerrit-Has-Labels: No Comment-In-Reply-To: neels <[email protected]> Comment-In-Reply-To: laforge <[email protected]> Gerrit-MessageType: comment
