veilid/CHANGELOG.md
2024-10-11 00:16:39 +00:00

12 KiB

Changed in Veilid 0.3.5 WIP

  • Dialinfo detection issues:

    • Add a publish() as well as a commit() for routing domain editor
    • Should only publish our peer info after we're sure we done editing it (end of public address detection task)
    • Publish should happen after relay selection as well
    • Publish should happen if the relay's peerinfo has changed
    • Publish should not do anything if the peerinfo hasn't changed
    • PeerInfo -> Arc everywhere to minimize deep clones and ensure read-only PeerInfo
    • Routing domain editing is now more atomic
    • When a node selects a relay it now immediately protects its connections.
    • Made dial info port (for port restricted nat) more resilient to changes, in the case there are multiple mappings
    • Relays that drop protected connections should be deprioritized for relay selection (table saturation detection)
    • clear_network_callback in do_public_dial_info_check is a kludge, removed
    • Raised the bar for dialinfo changes when its just the port
    • Pinging node on the same network works again
    • resolve_node() never returns a dead node even when we want to try to communicate with it again
    • Removed 'bad public address' detection as it wasn't working anyway
    • Added separate parallelism lanes for relay keepalive pings from peer liveness check pings, as they are higher priority
    • Change send_data to always check cache for contact method first instead of going with filtered active flows first, avoids choosing UDP when a preferable TCP connection could be made
    • Nodes that are not relay capable should drop relayed packets
  • DHT issues:

    • Make setvalue more likely to succeed by accepting a getvalue consensus if a full setvalue consensus is not reached.
    • Offline subkey writes are cleared too fast and should be thought as 'subkeys not yet synchronized'
    • If set_value is partial / in-flight, it should still be in offline_subkey_writes
    • Make inflight_subkey_writes list and probably some bit for 'written_while_inflight' so we dont clear the offline_subkey_writes until they're really written
  • API Additions:

    • VeilidConfigInner::new parameteriztion for easier config from rust apps
    • Remove veilid-server specific paths from veilid-core defaults
    • Lots more stats about node performance in PeerStats

Changed in Veilid 0.3.4

  • Crates updates
    • Update crates to newer versions
    • Remove veilid-async-tungstenite and veilid-async-tls crates as they are no longer needed
  • Fix startup/shutdown/attach/detach
    • Improved resource accounting
    • Locked startup/shutdown mechanism
    • Perfetto profiler output for ui.perfetto.dev
    • SO_LINGER(0) re-enabled to eliminate TIME_WAIT on restart/detach/attach (this may cause noise for WASM in browsers when websockets are RST dropped rather than handshake closed, we will deal with this later)
  • Community Contributions
    • prototype script to install / run a veilid-server node within a unifyos device (tested on a unify dream machine pro SE) @Vyrus-001

Changed in Veilid 0.3.3

  • Fix set_dht_value and watch_value
    • Watching values incorrectly categorized 'owner' keys as anonymous watchers
    • Setting a dht value with the same sequence number as what is on the network, but with a conflicting value, did not result in the current value being returned to the api caller as it should have been
  • DHT cleanup
    • Proper application of DHT capabilities
    • Fanout debugging log target
    • Performance measurement / timing of veilid_api log target
  • Fix DHT Rust integration test
  • ValueChanged Optional
    • Allow value changed data to be optional in rpc schema
    • Make valuechanged update no longer happen when value hasn't changed or is older
  • Implement closest peers refresh
    • Implement closest peers refresh. Closes issue #372.
    • Find_self/find_target can use capability check
    • Fix offline subkey write reporting to eliminate spurious notifications
    • Add more detail to public address check
  • Improved punishment and state
    • Create 'reasons' for dead and unreliable states
    • Make 'punished' its own state
    • Closes issue #281
    • Fixes an issue with reliable nodes being marked as 'dead' unjustly
  • Community Contributions
    • Fixed memory leak in Windows DNS resolver @kyanha

Changed in Veilid 0.3.2

  • DHT optimization and bugfixes
    • Support for offline write status in DHTRecordReport
    • Fix deprecated functions
    • Improve fanout seeding to ensure records are reached as quickly as possible
  • Native IPV4-IPV6 bridging support
    • fix bug where messages sent to a private route without a safety route would not receive replies
    • fix verbose-tracing feature flag
    • improve route allocation to avoid co-located nodes
    • fix contact method for nodes on the same ipblock
    • add support for maintaining AddressType-translation relays
  • Removed NDK related hotfix, as this has been integrated into cargo-ndk already
  • Open sourced the CICD build scripts
  • Fixes for watchvalue
  • Refactor low level network

Changed in Veilid 0.3.1

  • DHT cleanup
    • Proper application of DHT capabilities
    • Fanout debugging log target
    • Performance measurement / timing of veilid_api log target
  • ValueChanged Optional
    • Allow value changed data to be optional in rpc schema
    • Make valuechanged update no longer happen when value hasn't changed or is older
  • Clippy fixes and cleanup
  • Community Contributions
    • Changed VeilidAPI::parse_as_target to a sync function -- @sashanoraa
    • fix dht rust integration test -- @ssurovsev

Changed in Veilid 0.3.0

  • API BREAKING CHANGES:
    • WatchValue RPC support
    • InspectRecord RPC support
    • RoutingContext now defaults to Reliable and EnsureOrdered modes
    • generate_shared_secret added that abstracts DH and ensures domain separation
  • Closed #357 - AppCall and AppMessage now have private route information
  • Logging: Log facilities now can be enabled and disabled at runtime
  • Logging: Log facility added for DHT, network results, and API calls
  • CLI: Closed #358 - veilid-cli now has 'interactive' (-i), 'log viewer' (-l) and 'execute command' (-e) command line options
  • Testing: veilid-flutter now has integration tests of its own that work like the veilid-python unit tests
  • Network: Failures to hole-punch UDP or reverse-connect TCP or UDP now falls back to inbound relaying
  • Bugfix: Signal handling for unix-like platforms was not handling SIGTERM correctly
  • Bugfix: Restarting veilid-server quickly might result in failures to bind()
  • Bugfix: Closed #359 - Block node identity from DHT record schema owner/writer
  • Bugfix: Closed #355 - Fixed memory error reading macos/ios interfaces list
  • Community Contributions
    • Made private route allocation bidirectional by default @kyanha
    • Use $CI_REGISTRY_IMAGE for the registry path @SalvatoreT
    • Add VeilidConfigInner-based VeilidAPI startup @SalvatoreT
    • rebrand trust-dns-resolver to hickory-resolver @kyanha

Changed in Veilid 0.2.5

  • API BREAKING CHANGES:
    • on RoutingContext: with_privacy() renamed to with_default_safety()
    • on RoutingContext: with_custom_privacy() renamed to with_safety()
    • on RoutingContext: safety() method added that returns the current SafetySelection
    • Routing contexts are now safety-route-enabled by default. To disable, use with_safety() with SafetySelection::Unsafe.
  • WASM now works better with updated connection manager code
  • Async-std flavor of veilid-core now builds correctly again
  • Safety route allocation is bidirectional
  • Connection table LRU cache now has protection for relays and in-use RPC question/answers
  • Dead route notifications are now sent only for manually allocated routes
  • Allocated routes that fail tests now have their nodes marked as 'failure to send' so they go 'unreliable' and get re-tested. Also the same route will not immediately be reallocated as a result.
  • DHT tests ported from Python to Rust
  • Rustls updated to latest release
  • Protected connections (such as relays) that drop result in marking the node as 'failure to send' so a different relay gets chosen

Changed in Veilid 0.2.4

  • Fixed issue with client API failing when ipv6 was disabled
  • Android fixed so it can move out of invalid network state
  • Numerous WASM binding fixes
  • IGD/UPNP fixes for Windows
  • Reduce network downtime when local ip addresses change (ipv6 temporary addresses)
  • Fix support for Android emulator
  • Bootstrap is more robust in environments where some dialinfo won't work, like inbound UDP being firewalled off
  • CLI has timestamps in the log output
  • Base64 fixes for encoding
  • IPv6 capability detection for native platforms

Changed in Veilid 0.2.3

  • Security fix for WS denial of service
  • Support for latest Rust 1.72

Changed in Veilid 0.2.2

  • Capnproto 1.0.1 + Protobuf 24.3
  • DHT set/get correctness fixes
  • Connection table fixes
  • Node resolution fixes
  • More debugging commands (appmessage, appcall, resolve, better nodeinfo, etc)
  • Reverse connect for WASM nodes
  • Better Typescript types for WASM
  • Various script and environment cleanups
  • Earthly build for aarch64 RPM
  • Much improved and faster public address detection

Changes in Veilid 0.2.1

  • Crates are separated and publishable
  • First publication of veilid-core with docs to crates.io and docs.rs
  • Avoid large logs of 127.0.0.1:5959 attack payloads
  • Use getrandom in WASM for RNG
  • Increase privacy for WASM builds by rewriting internal paths
  • Translations
  • Fix python update schema script
  • Earthfile cleanup

Changes in Veilid 0.2.0

  • Rustdoc builds now
  • API visibility changes
  • Android JNI update
  • Fix DHT record data housekeeping
  • Public address detection improvement
  • Manual port forwarding detection
  • lock_api dependency fix
  • DialInfo failover when some dial info does not work

Note: Windows builds may be broken in this release. Please test and let us know by opening an issue.

Changes in Veilid 0.1.10

  • BREAKING CHANGE: ALL MUST UPDATE
    • VLD0 now adds a BLAKE3 hash round on the DH output to further separate it from the raw key exchange
    • Bootstraps are fixed now due to DH issue
  • Windows crate update caused build and nul termination issues for DNS resolver
  • Fix for network key on the veilid-server command line
  • Strict verification for Ed25519 enabled
  • Domain separation for VLD0 signing and crypt

Changes in Veilid 0.1.9

  • SECURITY FIX
    • DESCRIPTION: Decompression was occurring in an unbounded way upon envelope receipt.
    • IMPACT: Node crashes resulting in downtime. There was no risk of RCE or compromise due to Rust's memory protections and no use of unsafe code near the site of the error.
    • INDICATIONS: This resulted in an out-of-memory abort on nodes. Issue first identified on the bootstrap servers.
    • REMEDIATION: Length check added to decompression on envelopes.
  • Earthfile support for generating a debug executable

Changes in Veilid 0.1.8

  • Fix Python Install Instructions
  • Fix to get server version from crate
  • Move dev setup into its own folder
  • Setup support for Fedora
  • Make submodule paths absolute
  • veilid-flutter improvements for crypto and timestamp, and endianness bugfix
  • Offline subkey writes for DHT
  • Fix WASM compilation
  • Improve server port allocation
  • Add more punishments
  • Clap derive refactor for command line args
  • gitignore emacs backup files
  • Various typos
  • Fanout debugging for DHT

Changes in Veilid 0.1.7

  • Fix for connection table crash
  • Fix for incorrect set_dht_value return value
  • Python test updates
  • Various VeilidChat-prompted veilid-flutter updates

Changes in Veilid 0.1.6

  • Fix for 'find_node' too many nodes returned issue

Changes in Veilid 0.1.5

  • Added Changelog
  • Fix detachment issue with suspending network interfaces during operation
  • Fix incorrect punishment on relayed undecryptable messages
  • Minor API feature adds
  • Relay bugfixes