--- layout: post title: Overview and Logs for the tini2p Dev Meeting Held on 2019-10-17 summary: Current project status, Roadmap, Meta issues, and miscellaneous tags: [dev diaries, i2p, crypto] author: el00ruobuob / oneiric --- # Logs **\** 0: Greetings **\** h **\** 1: What's been done **\** last week was the alpha release of tini2p **\** the router is still in its earliest stages, and things will still be changing fairly drastically as refactors and bug fixes are made **\** tini2p routers are capable of building ECIES tunnels, and communicating end-to-end via ECIES-X25519 **\** NTCP2 (a tcp-like transport) is fully implemented **\** cross-implementation testing is still needed to discover and remedy any bugs (if they exist) **\** the router is also currently compiled with a testnet netID to mitigate any communication with mainnet routers **\** using a testnet netID means custom compilation is required to communicate with mainnet I2P routers over NTCP2 **\** I haven't added the necessary build for mainnet because tini2p isn't ready yet, and I don't want to encourage usage which may be disruptive/harmful to the network **\** I also began work on [ElGamal tunnel building](https://gitlab.com/tini2p/tini2p/commits/elg-tunnel) (basic encryption/decryption of BuildRequestRecords is more-or-less done) **\** 2: What's next **\** will continue work on ElGamal tunnel building **\** the next steps are: **\** - serializing/deserializing ElGamal BuildRequestRecords (slightly different from ECIES records) **\** - deriving keys for ChaCha tunnels using the ElGamal shared secret **\** - refactoring the router context to handle ElGamal tunnel building **\** maybe it will go quickly, but it could take a couple weeks **\** will also work on creating test vectors for i2pd and Java I2P for cross-impl ElGamal testing **\** will bring up KDF discussion about deriving record keys using the ElGamal shared secret **\** for AES tunnels, deriving record keys is likely unnecessary, since any saved space would be consumed with random padding **\** for ChaCha tunnels, the saved space is necessary for transmitting the AEAD `receive` key to each hop **\** the additional 32 bytes for the AEAD key will not fit in the current space allocated for "options/random padding" **\** + the bytes for the options header + flags **\** will also be adding fuzz test drivers for data structures with `deserialize` methods as I work on other features **\** if the KDF changes for ElGamal records are unwanted/unworkable, ElGamal tunnels will only be able to use AES for layer encryption **\** which would effectively mean any tunnel with one ElGamal hop must use AES layer encryption for the entire tunnel **\** if the KDF changes are accepted, ElGamal tunnels will be able to use AES and/or ChaCha layer encryption **\** for a review of AES layer encryption: https://geti2p.net/en/docs/tunnels/implementation **\** and ChaCha layer encryption proposal: https://geti2p.net/spec/proposals/153-chacha20-layer-encryption **\** 3: Questions / comments **\** much love for the lurkers **\** 4: Next meeting time **\** 2019-10-31 18:00 UTC **\** @tini2p\_gitlab gavels the gavels like it's never been gavelled before