Tim and Dale reported on IETF talk on packet marking, suprising feedback, very positive and constructive comments. We're encouraged to explore HBH in more details as replacement of flow label. Dale mentioned that HBH would be more favourable for routing/forwarding decisions. It can also accommodate more space and thus transfer more information (which in turn would help to propose packet marking as a more generic approach).
Marian commented that Linux API looks easier than flow label, but likely requires root and we would need to run some tests to see if it works as expected (as there are very few examples online). On the other hand, eBPF implementation might be more complicated as we need to add an extra header as opposed to change a single field, Tristan mentioned he can take a look at it.
Following up on comments from GNA-G, we have agreed that we will look for synergies, explore SRv6/SR-MPLS, PolKA and look for approaches that would benefit both projects.
Richard mentioned IETF APN WG - Application-aware Networking (https://datatracker.ietf.org/wg/apn/about/) that might be relevant.
Marcos mentioned https://datatracker.ietf.org/doc/draft-filsfils-spring-srv6-stateless-slice-id/ as an example of bridging flow label and SRv6.
CHEP2023 paper deadline is approaching, draft to be circulated in the next two weeks.
Marian will send a doodle on technical discussion about collectors and dashboards for SC23.
dCache testing has been restarted following re-deployment of the next version (we have also received source code, so it should be easier to follow up).
