IETF MOQ WG: User case or question to Joining Fetch
Zafer Gurel posted to the MOQ working group mailing list on May 20, 2026. Hi Will, You are right. However, the collision problem I mentioned is more about the cache implementation of MOQtail relay. The cache key
Message Snapshot
- List:
moq@ietf.org - Thread subject: User case or question to Joining Fetch
- Message subject: [Moq] Re: User case or question to Joining Fetch
- Author: Zafer Gurel
- Published: 2026-05-20T21:22:23.000Z
- Source tag:
ietf-mailing-list - Message URL: Open in IETF Mail Archive
Summary
Zafer Gurel posted to the MOQ working group mailing list on May 20, 2026. Hi Will, You are right. However, the collision problem I mentioned is more about the cache implementation of MOQtail relay. The cache key
Why It Matters
This message came from the public MOQ working group archive and captures active discussion, meeting coordination, or implementation feedback around Media over QUIC.
Message Excerpt
Hi Will, You are right. However, the collision problem I mentioned is more about the cache implementation of MOQtail relay. The cache key
Plain-Text Body
Hi Will, You are right. However, the collision problem I mentioned is more about the cache implementation of MOQtail relay. The cache key https://github.com/moqtail/moqtail/blob/main/apps/relay/src/server/track_cache.rs#L32 is (relay track id (maps to NS + N), group id). And each cache entity is a vector of objects. So if two publishers publish objects to the same track with the same group Id, a race condition occurs in MOQtail relay. I may need to revisit that logic to implement your idea.
Thanks for the feedback.
Best,
Zafer