You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The type of segment is known to the endpoint but is not visible
is the path header of data packets. Therefore, a SCION router needs
to explicitly verify that these rules were followed correctly.
this sentence actually comes from the book section "5.5 Path Construction (Segment Combinations)".
We then describe checks at routers in 4.2.2. Processing at Routers and there is no mention that routers verify any of the rules in 1.4.
@jiceatscion do you know if routers do any checks at all? If not, we should just remove the second sentence I guess.
The text was updated successfully, but these errors were encountered:
If my reading of the code is correct, these checks are implemented. Albeit
not exactly mapping 1:1 with how the rules are written:
There is no explicit check that the first segment is up, the second core,
and the third is down. Instead, there is a check on each segment change
that the ingress links are either:
child to core or core to child or child to child. Without a segment change,
the code also rejects valley-use of peering-links or hair-pin segments. I think that ends-up
enforcing the rules since any rule-breaking path would end-up containing a
segment change that fails those checks. Code here:
https://github.com/scionproto/scion/blob/4c929f6b440c568663b4194c125df15cad544bb4/router/dataplane.go#L1552
On Sun, Aug 25, 2024 at 1:43 PM Nicola Rustignoli ***@***.***> wrote:
In section 1.4 we write:
The type of segment is known to the endpoint but is not visible
is the path header of data packets.
*Therefore, a SCION router needs to explicitly verify that these rules
were followed correctly.*
this sentence actually comes from the book section "5.5 Path Construction
(Segment Combinations)".
We then describe checks at routers in 4.2.2. Processing at Routers and
there is no mention that routers verify any of the rules in 1.4.
@jiceatscion <https://github.com/jiceatscion> do you know if routers do
any checks at all?
—
Reply to this email directly, view it on GitHub
<#52>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/BBLEYOBP5XUSMQ3SRWJ2AEDZTG7NNAVCNFSM6AAAAABNCO5KKCVHI2DSMVQWIX3LMV43ASLTON2WKOZSGQ4DKMJZHE3DKMQ>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
In section 1.4 we write:
this sentence actually comes from the book section "5.5 Path Construction (Segment Combinations)".
We then describe checks at routers in 4.2.2. Processing at Routers and there is no mention that routers verify any of the rules in 1.4.
@jiceatscion do you know if routers do any checks at all? If not, we should just remove the second sentence I guess.
The text was updated successfully, but these errors were encountered: