wait on engine down to not only wait for the interface to be down but completely removed. If the waiting loop reaches the timeout we will trigger an interface destroy. On the up command, it now waits until the engine is fully running before sending the response to the CLI. Includes a small refactor of probes to comply with sonar rules about parameter count in the function call
* compile client under freebsd (#1620)
Compile netbird client under freebsd and now support netstack and userspace modes.
Refactoring linux specific code to share same code with FreeBSD, move to *_unix.go files.
Not implemented yet:
Kernel mode not supported
DNS probably does not work yet
Routing also probably does not work yet
SSH support did not tested yet
Lack of test environment for freebsd (dedicated VM for github runners under FreeBSD required)
Lack of tests for freebsd specific code
info reporting need to review and also implement, for example OS reported as GENERIC instead of FreeBSD (lack of FreeBSD icon in management interface)
Lack of proper client setup under FreeBSD
Lack of FreeBSD port/package
* Add DNS routes (#1943)
Given domains are resolved periodically and resolved IPs are replaced with the new ones. Unless the flag keep_route is set to true, then only new ones are added.
This option is helpful if there are long-running connections that might still point to old IP addresses from changed DNS records.
* Add process posture check (#1693)
Introduces a process posture check to validate the existence and active status of specific binaries on peer systems. The check ensures that files are present at specified paths, and that corresponding processes are running. This check supports Linux, Windows, and macOS systems.
Co-authored-by: Evgenii <mail@skillcoder.com>
Co-authored-by: Pascal Fischer <pascal@netbird.io>
Co-authored-by: Zoltan Papp <zoltan.pmail@gmail.com>
Co-authored-by: Viktor Liu <17948409+lixmal@users.noreply.github.com>
Co-authored-by: Bethuel Mmbaga <bethuelmbaga12@gmail.com>
* Adds management, signal, and relay (STUN/TURN) health probes to the status command.
* Adds a reason when the management or signal connections are disconnected.
* Adds last wireguard handshake and received/sent bytes per peer
Add netstack support for the agent to run it without privileges.
- use interface for tun device
- use common IPC for userspace WireGuard integration
- move udpmux creation and sharedsock to tun layer
In case the 53 UDP port is not an option to bind then we hijack the DNS traffic with eBPF, and we forward the traffic to the listener on a custom port. With this implementation, we should be able to listen to DNS queries on any address and still set the local host system to send queries to the custom address on port 53.
Because we tried to attach multiple XDP programs to the same interface, I did a refactor in the WG traffic forward code also.
EBPF proxy between TURN (relay) and WireGuard to reduce number of used ports used by the NetBird agent.
- Separate the wg configuration from the proxy logic
- In case if eBPF type proxy has only one single proxy instance
- In case if the eBPF is not supported fallback to the original proxy Implementation
Between the signature of eBPF type proxy and original proxy has
differences so this is why the factory structure exists
Works only with userspace implementation:
1. Configure host to solve DNS requests via a fake DSN server address in the Netbird network.
2. Add to firewall catch rule for these DNS requests.
3. Resolve these DNS requests and respond by writing directly to wireguard device.
This PR adds supports for the WireGuard userspace implementation
using Bind interface from wireguard-go.
The newly introduced ICEBind struct implements Bind with UDPMux-based
structs from pion/ice to handle hole punching using ICE.
The core implementation was taken from StdBind of wireguard-go.
The result is a single WireGuard port that is used for host and server reflexive candidates.
Relay candidates are still handled separately and will be integrated in the following PRs.
ICEBind checks the incoming packets for being STUN or WireGuard ones
and routes them to UDPMux (to handle hole punching) or to WireGuard respectively.
Small code cleaning in the iface package. These changes necessary to
get a clean code in case if we involve more platforms. The OS related
functions has been distributed into separate files and it has been
mixed with not OS related logic. The goal is to get a clear picture
of the layer between WireGuard and business logic.
Added host configurators for Linux, Windows, and macOS.
The host configurator will update the peer system configuration
directing DNS queries according to its capabilities.
Some Linux distributions don't support split (match) DNS or custom ports,
and that will be reported to our management system in another PR
Before this change, NetBird Agent wasn't handling
peer interface configuration changes dynamically.
Also, remote peer configuration changes have
not been applied (e.g. AllowedIPs changed).
Not a very common cause, but still it should be handled.
Now, Agent reacts to PeerConfig changes sent from the
management service and restarts remote connections
if AllowedIps have been changed.
When one of the peers has a static public host IP
or both peers are in the same local network
we establish a direct Wireguard connection
bypassing proxy usage.
This helps reduce FD usage and improves
performance.
* script to generate syso files
* test wireguard-windows driver package
* set int log
* add windows test
* add windows test
* verbose bash
* use cd
* move checkout
* exit 0
* removed tty flag
* artifact path
* fix tags and add cache
* fix cache
* fix cache
* test dir
* restore artifacts in the root
* try dll file
* try dll file
* copy dll
* typo in copy dll
* compile test
* checkout first
* updated cicd
* fix add address issue and gen GUID
* psexec typo
* accept eula
* mod tidy before tests
* regular test exec and verbose test with psexec
* test all
* return WGInterface Interface
* use WgIfaceName and timeout after 30 seconds
* different ports and validate connect 2 peers
* Use time.After for timeout and close interface
* Use time.After for testing connect peers
* WG Interface struct
* Update engine and parse address
* refactor Linux create and assignAddress
* NewWGIface and configuration methods
* Update proxy with interface methods
* update up command test
* resolve lint warnings
* remove psexec test
* close copied files
* add goos before build
* run tests on mac,windows and linux
* cache by testing os
* run on push
* fix indentation
* adjust test timeouts
* remove parallel flag
* mod tidy before test
* ignore syso files
* removed functions and renamed vars
* different IPs for connect peers test
* Generate syso with DLL
* Single Close method
* use port from test constant
* test: remove wireguard interfaces after finishing engine test
* use load_wgnt_from_rsrc
Co-authored-by: braginini <bangvalo@gmail.com>
* feature: replace RegisterPeer with Login method that does both - registration and login
* test: add management login test
* feature: add WiretrusteeConfig to the Login response to configure peer global config
* feature: add client peer login support
* fix: missing parts
* chore: update go deps
* feature: support Management Service gRPC endpoints [CLIENT]
* feature: finalize client sync with management
* fix: management store peer key lower case restore
* fix: management returns peer ip without a mask
* refactor: remove cmd pkg
* fix: invalid tun interface name on mac
* fix: timeout when calling management client
* fix: tests and lint errors
* fix: golang-test workflow
* fix: client service tests
* fix: iface build
* feature: detect management scheme on startup
* chore: better logs for management
* fix: goreleaser
* fix: lint errors
* fix: signal TLS
* fix: direct Wireguard connection
* chore: verbose logging on direct connection
