Milestone: Improve Provider Record Intervals

[yiannisbot]

eta: 2022-12-05

description: This Milestone focuses on the Replication Factor (k), the Re-Publish and Expiration Intervals in IPFS. More details can be found here: @Provider Record Liveness and here. The (pending) spec PR is here and the new values identified for the Re-Publish and Expiration Intervals are here and here, respectively.

[yiannisbot]

This Milestone has been completed. Linking again here the main outcomes and PRs:

• RFM 17 Report on Provider Record Liveness (includes all the details): https://github.com/protocol/network-measurements/blob/master/results/rfm17-provider-record-liveness.md
• Spec update with new proposed values: New values for DHT Provider Record Republish and Expiration (22h/48h, RFM17) libp2p/specs#451
• PR: increase default Reprovider.Interval: feat: increase default Reprovider.Interval ipfs/kubo#9326
• PR: increase expiration time for Provider Records to 48h: feat: increase expiration time for Provider Records to 48h (RFM17) libp2p/go-libp2p-kad-dht#793
• PR: increase the max record age to 48h (PUT_VALUE): feat: increase the max record age to 48h (PUT_VALUE, RFM17) libp2p/go-libp2p-kad-dht#794

Bonus item RFM 17.1 on Sharing Provider Records with Multiaddress: https://github.com/protocol/network-measurements/blob/master/results/rfm17.1-sharing-prs-with-multiaddresses.md

Excellent work! @cortze please add more pointers if I missed anything.

[cortze]

Thanks, @yiannisbot ! I will link the PR and the Issue that also came along with this Milestone:

• PR: improvements into the DHT codebase to avoid extra lookups: Adjust PeerSet logic in the DHT lookup process libp2p/go-libp2p-kad-dht#802
• Issue: Spotted a stream bottleneck in the DHT's message sender: MessageSender limits concurrent RPCs to remote peer by only keeping one single stream per peer  libp2p/go-libp2p-kad-dht#805